THÈSE DE DOCTORAT DE L’UNIVERSITÉ D’AIX-MARSEILLE

Soutenue par

Phuong Chi NGUYEN

Pour obtenir le grade de Docteur de l’Université d’Aix-Marseille Discipline: Biologie Spécialité: Microbiologie

Deciphering antimycobacterial activity of Cyclophostin/Cyclipostins analogs and Oxadiazolones derivatives, two new promising family compounds in the treatment of tuberculosis and mycobacterial-related diseases

Soutenue le 25 Mai 2018 devant le jury :

Dr. Alain BAULARD (INSERM, Lille) Rapporteur Dr. Jean Denis PEDELACQ (IPBS, Toulouse) Rapporteur Dr. Véronique FONTAINE (ULB, Belgium) Examinatrice Pr. Sophie BLEVES (LISM, Marseille) Examinatrice-Présidente Dr. Céline CRAUSTE (IBMM, Montpellier) Examinatrice Dr. Stéphane CANAAN (LISM, Marseille) Directeur de thèse

Acknowledgements I would like to take this opportunity to sincerely thank all people who have supported me during the time of my PhD.

First and foremost, I offer the sincerest gratitude to my supervisor, Dr. Stéphane CANAAN. Thank you for your continuous support and guidance during the time I have spent in the group. I really appreciate your encouragement and excellent ideas whenever I was stuck. It has been 3 years since you offered me to come here. We have overcome many difficulties and now I am going to defense a good thesis. Thank you!

Besides my supervisor, I would like to thank the rest of my thesis committee: Dr. Alain BAULARD and Dr. Jean Denis PEDELACQ, who have accepted to be reporters of this manuscript, Pr. Sophie BLEVÉS who has accepted as the president of the jury, Dr. Véronique FONTAINE and Dr. Céline CRAUSTE, for their acceptance as examiners.

My sincere thanks goes to Dr. Jean-François CAVALIER, for advising me on the aspects of my work especially on biochemistry and chemistry. Thank you for your unwavering support during my time in the laboratory as well as be patient to correct all my manuscripts.

I would take this opportunity to express my sincere gratitude to my wonderful colleagues. Thank you, Isabelle and Vanessa, for being kind and ready for help any time I need, despite of difference in language. Thanks Pierre for all your support through the time of my PhD with a lot of mini details have been discussed and solved to bring good results. Thanks Djalil for working together and produced two papers. I would also thank to two postdocs, Nabil and Ticiana, who shared their own knowledge, experiences and ideas.

I also cannot thank enough to all Vietnamese students in Luminy and Marseille who make me feel at home even we live thousand miles away from our country. Thanks to brother Quan and sister Thuy for being so kind. I would also thank to French and foreigner students, I have had a great fortune of spending time with all of you.

This acknowledgement would not be completed without mentioning endless love, support and encouragement from my husband and my family. Thank you for being always beside me and sharing all the joy and sorrow together. I am eternally thankful to my parents, my husband’s parents and my older brother. Thank you for your continuous encouragement. I am happy to have all beloved like you.

Table of contents

INTRODUCTION ...... 1

CHAPTER 1: General introduction ...... 2

1.1. Tuberculosis, an overview ...... 2

1.2. History timeline of Tuberculosis ...... 3

1.3. Diagnosis and vaccination ...... 4

Diagnosis ...... 4

Vaccines...... 5

1.4. Biological characteristics of M. tb ...... 6

General biology ...... 6

Genome of M. tb ...... 8

Cell envelope ...... 9

1.5. Pathogenic life cycle ...... 12

Transmission ...... 12

Initial infection ...... 14

M. tb intracellular life ...... 14

Immune response to M. tb infection ...... 14

Granulomas ...... 15

1.6. Treatment of TB ...... 17

Standard treatment for TB ...... 17

Front-line drugs ...... 17 1.7. Drug resistance and mechanisms ...... 21

1.8. Treatment in drug-resistance TB and new drugs development ...... 23

Current treatment in drug-resistance TB ...... 24 New drugs in development ...... 25

1.9. Lipid metabolism of M. tb ...... 26

CHAPTER II: Lipolytic enzymes in M. tb – the fundamental of a new drug family development ...... 28

2.1. Serine hydrolase enzymes from M. tb: promising therapeutic targets ...... 28

2.2. Mycobacterial lipolytic enzymes ...... 31

True lipases ...... 32

Carboxylesterases ...... 35

Cutinases ...... 37

Phospholipases ...... 38

CHAPTER III: Inhibitors of lipolytic enzymes ...... 40

3.1. Orlistat ...... 40

3.2. Orlistat-core compounds ...... 41

3.3. β-lactone EZ120 ...... 42

3.4. Lalistat ...... 43

3.5. Oxadiazolone-core compounds ...... 45

3.6. Cyclophostin and Cyclipostins molecules ...... 47

3.7. Activity-based protein profiling (ABPP), powerful chemical proteomic platform applied to investigate targets of CyC analogs and Oxadiazolones derivatives ...... 52

3.8. Objectives of my thesis ...... 54

RESULTS ...... 55

Article 1: Cyclipostins and Cyclophostin analogs as promising compounds in the fight against tuberculosis ...... 56

Article 2: Cyclophostin and Cyclipostins analogs, new promising molecules to treat mycobacterial-related diseases ...... 84 Article 3: Cyclipostins and Cyclophostin analogs inhibit the antigen 85C from Mycobacterium tuberculosis both in vitro and in vivo ...... 93

Article 4: A biochemical and structural characterization of TesA a major thioesterase requires for PDIM and PGL syntheses in M. tuberculosis ...... 110

Article 5: Oxadiazolone derivatives against M. tuberculosis ...... 142

CONCLUSIONS AND PERSPECTIVES ...... 183

Conclusions ...... 184

Perspectives ...... 186

REFERENCES ...... 188

List of Figures Figure 1. Estimated TB incidence rates, 2016...... 3 Figure 2. Global clinical pipeline of TB vaccine candidates...... 6 Figure 3. (A) M. tb H37Rv colonies. (B) M. tb mc26230 visualization under confocal microscopy following Nile-Red staining...... 7 Figure 4. Evolutionary relationship between selected mycobacteria and members of the Mycobacterium tuberculosis complex...... 8 Figure 5. Circular map of the chromosome of M. tb H37Rv...... 9 Figure 6. Cell wall structure of (A) Gram-negative bacteria, (B) Gram-positive bacteria and (C) mycobacteria...... 10 Figure 7. Different forms of mycolic acids in M. tb...... 12 Figure 8. The transmission cycle of M. tb...... 13 Figure 9. Structure and cellular constituents of the tuberculous granuloma...... 16 Figure 10. (A) Chemical structures of front-line anti-TB drugs ...... 18 Figure 10. (B) Mechanism of action of Isoniazid...... 18 Figure 11. Estimated incidence of MDR/RR-TB in 2016...... 22 Figure 12. Current global pipeline of new TB drugs ...... 25 Figure 13. Chemical structure of the two new approved anti-TB drugs...... 26 Figure 14. Canonical secondary structure diagram of the α/β-hydrolase fold...... 29 Figure 15. Superimposition of the crystal structures of bacterial serine hydrolases showing a highly structural fold conservation among them...... 30 Figure 16. The hydrolysis of substrate by a typical serine hydrolase enzyme...... 31 Figure 17. (A) LipY hydrolyzed TAG and (B) LipY is located on mycobacterial surface. .... 33 Figure 18. Comparison of the antibody IgG response to recombinant lipolytic enzymes LipY and Rv0183 in TB patients and healthy individuals...... 34 Figure 19. Colony and aggregation modification attribute to the disruption of MSMEG_0220 - the homolog of Rv0183 in M. tb...... 34 Figure 20. Deletion mutant of CaeA leading to attenuation virulence of M. tb in mouse lung ...... 36 Figure 21. Phospholipase cleavage sites...... 38 Figure 22. Chemical structure of Orlistat (A), its mode of inhibition on lipases and the possible primary degradation product (B)...... 40 Figure 23. Chemical structure of Orlistat-core compounds...... 42 Figure 24. -lactone EZ120 and its proposed mechanism of binding to a serine hydrolase enzyme...... 43 Figure 25. Chemical structures of (A) Lalistat and the new La-1 probe; and (B) mechanism of inhibition of lysomal acid lipase (LAL) by such thiadiazole carbamates...... 44 Figure 26. (A) Activity based protein profiling (ABPP) workflow with lalistat La-1 probe for quantitative proteomics, and (B) enrichment and competition target identification (ABPP) volcano plot representations and corresponding list of common hits...... 45 Figure 27. (A, C, D) Chemical structure of Oxadiazolone core compounds, and (B) their mechanism of action...... 46 Figure 28. Chemical structure of natural (A) Cyclophostin and (B) Cyclipostins ...... 48 Figure 29. Rationale for synthesis of new Cyclophostin and Cyclipostins (CyC) analogs, and related structures investigated...... 49 Figure 30. Mode of action Cyclipostins/Cyclophostin analogs (CyC)...... 50 Figure 31. Chemical structure of fluorophosphonate probes and their mechanism of action on serine hydrolases...... 52 Figure 32. Workflow on identification the “off-targets” of inhibitor of interest among a proteome complex using fluorophosphonate probes...... 53

Scheme 1. (A) Activity of CyC7 and CyC17 against GFP-labelled M. tb replicating in culture medium expressed as normalized relative fluorescence units (RFU%). (B) Activity of CyC7 against M. tb replicating inside murine macrophages Raw264.7...... 57 Scheme 2. Function of antigen 85 complex (FbpABC) in M. tb...... 93 Scheme 3. (A) The DIM gene cluster in M. tb. The tesA gene is in red arrow. (B) Scheme of phthiocerol dimycocerosate (DIM) and PGL biosynthesis in M. tb...... 110

Table Table 1. Categories of TB drugs and mechanisms of resistance ...... 20

Abbreviations

ABP Activity-based probe ABPP Activity-based protein profiling AChE Acetyl cholinesterase AG Arabinogalatan AM Alveolar macrophage AMK Amikacin BCG Bacillus Calmette–Guérin BSSL Bile Salt-Stimulated Lipase CC50 Compound concentration leading to 50% host cell toxicity CEMOVIS Cryo-electron microscopy of vitreous sections CF Cystic fibrosis CFU Colony forming unit CMC Critical micelle concentration CyC Cyclipostins/Cyclophostin DAG Diacylglycerol DAT Diacyltrehaloses DDM Dodecyl β-D-maltoside DEP Diethyl p-nitrophenyl phosphate DGAT Diacylglycerol acyltransferase DMSO Dimethyl sulfoxide dNTPs Deoxynucleotide triphosphates DTNB 5,5′-Dithiobis(2-nitrobenzoic acid) ELISA Enzyme-linked immunosorbent assay FAAH Fatty acid amide hydrolase FM Foamy macrophage FOX Cefoxitin FP Fluorophosphonate FU Fluorescent unit GFP Green fluorescent protein HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HPLC High-performance liquid chromatography hrMGL Human recombinant monoacylglycerol lipase HGL Human gastric lipase HSL Hormone-Sensitive Lipase IC50 50% inhibitory concentration IGRA Interferon Gamma Release Assay ILI Intracytoplasmic lipid inclusions INH Isoniazid IPM Imipenem IPTG Isopropyl β-D-1-thiogalactopiranoside KAN Kanamycin LAL Lysosomal acid lipase LB Lipid bodies LB Luria-Bertani LC-MS/MS Liquid chromatography-tandem mass spectrometry LTBI Latent tuberculosis infection MA Mycolic acid MAME Mycolic acid methyl esters MALDI-TOF Matrix Assisted Laser Desorption/Ionization – Time Of Flight. MBC Minimum bactericidal concentration MDR Multi-drug resistant MIC Minimum inhibitory concentration MM Mycomembrane MMCW Mycomembrane-containing cell wall MmPPOX 5-methoxy-N-3-(meta-phenoxyphenyl)-1,3,4-oxadiazol-2(3H)-one MOI Multiplicity of infection MTBC Mycobacterium tuberculosis complex M. tb Mycobacterium tuberculosis MWCO Molecular weight cut-off NAATs Nucleic acid amplification tests NAG-NAM N-acetyl glucosamine-N-acetyl muramic acid Ni-NTA Nickel-nitrilotriacetic OADC Oletic acid-albumin-dextrose-catalase OX Oxadiazolones PAT Polyacyltrehaloses PBS Phosphate-buffered saline PCR Polymerase chain reaction PDB Protein data bank PDIM Phthiocerol dimycocerosates PE-PGRS Pro-Glu-polymorphic GC-rich sequences PG Peptidoglycan PGL Phenolic glycolipids PKS Polyketide synthase PLA1/A2 Phospholipase A1/A2 PLB/C/D Phospholipase B/C/D pNP para-nitrophenyl POA Pyrazinoic acid PZA Pyrazinamide RD1 Region of difference REMA Resazurin Microliter Assay RFU Relative fluorescence units RGM Rapid growing mycobacteria RPMI Roswell Park Memorial Institute RRDR Rifampicin resistance-determining region RR Rifampicin-resistant SAR Structure-activity relationship SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis SGM Slow growing mycobacteria SL Sulfolipids STR Streptomycin TAG Triacylglycerols TB Terific broth TDM Trihalose dimycolate TDR Total drug resistant TEV Tobacco Etch Virus TFA Trifluoroacetic acid THL Tetrahydrolipstatin THPM THP-1 derived macrophages TLC Thin layer chromatography TMM Trehalose monomycolate TST Tuberculin skin test XDR Extensive-drug resistant

List of publications

1. Nguyen, P. C., V. Delorme, A. Bénarouche, B. P. Martin, R. Paudel, G. R. Gnawali, A. Madani, R. Puppo, V. Landry, L. Kremer, P. Brodin, C. D. Spilling, J.-F. Cavalier & S. Canaan (2017a) Cyclipostins and Cyclophostin analogs as promising compounds in the fight against tuberculosis. Scientific Reports, 7, 11751.

2. Nguyen, P. C., A. Madani, P. Santucci, B. P. Martin, R. R. Paudel, S. Delattre, J.-L. Herrmann, C. D. Spilling, L. Kremer, S. Canaan & J.-F. Cavalier (2017b) Cyclophostin and cyclipostins analogs, new promising molecules to treat mycobacterial-related diseases. Int J Antimicrob Agents, https://doi.org/10.1016/j.ijantimicag.2017.12.001.

3. Viljoen, A., M. Richard, P. C. Nguyen, P. Fourquet, L. Camoin, R. R. Paudal, G. R. Gnawali, C. D. Spilling, J. F. Cavalier, S. Canaan, M. Blaise & L. Kremer (2018) Cyclipostins and cyclophostin analogs inhibit the antigen 85C from Mycobacterium tuberculosis both in vitro and in vivo. J Biol Chem, 293, 2755-2769.

4. Nguyen, P. C., V. S. Nguyen, C. Spilling, L. C. Patrick Fourquet¤, J.-F. Cavalier, C. Cambillau & S. Canaan (2018b) Biochemical and structural characterization of TesA, a major thioesterase required for outer-envelope lipid biosynthesis in M. tuberculosis. J Mol Biol. Submitted.

5. Nguyen, P. C., V. Delorme, A. Bénarouche, A. Guy, V. Landry, J.-M. Galano, P. B. Thierry Durand3, S. Canaan & J.-F. Cavalier (2018a) Oxadiazolone derivatives, new promising multi-target inhibitors against M. tuberculosis. Manuscript preparation.

SUMMARY

Tuberculosis (TB) remains the deadliest infectious disease worldwide. The stagnation in drug development accompanied with the emergence of drug resistance mainly results in treatment failure and death. There is thus an urgent need for novel therapeutic targets and approaches to combat Mycobacterium tuberculosis (M. tb), its etiologic agent.

It is now acknowledged that mycobacterial lipolytic enzymes are involved in many physiological processes of the bacteria life-cycle: virulence as well as growth, dormancy, envelop synthesis, storage and consumption of lipids. Accordingly, finding ways to inhibit such enzyme activity could pave the way for the discovery of new modalities in the TB treatment.

The first part of my work was devoted to the study of a series of new antimycobacterial compounds, namely analogs of Cyclophostin/Cyclipostins (CyCs). The CyCs were first tested for their antitubercular activities against extracellular growing pathogenic M. tb H37Rv as well as intracellularly M. tb infected macrophages. We have demonstrated that these CyCs exhibited anti-tubercular activities with moderate to excellent (500 nM for CyC17) MIC values. These data indicated that these compounds are able to penetrate both the host macrophage (without any cytotoxic effect) and the bacteria, properties required for an anti-TB drug. Using the best

CyC17 extracellular inhibitor, activity-based protein profiling (ABPP) approach allowed identifying 23 potential candidate target proteins, being serine/cysteine enzymes involved in M. tb intracellular lipid metabolism and/or in cell wall lipid biosynthesis.

Moreover, we clearly established that the CyCs are powerful and selective growth inhibitors of mycobacteria, with no apparent effect on Gram-negative or Gram-positive bacteria. Interestingly, these compounds were also active against numerous M. abscessus clinical strains, an opportunistic pathogen also called “antibiotic nightmare”, thus opening the way to find alternative solutions to fight against resistant mycobacterial strains.

Among many potential targets, two important proteins, i.e. Ag85C diacylglycerol acyltransferase and TesA thioesterase, playing key roles in cell wall lipid biosynthesis, have been validated biochemically and structurally as main targets of these CyCs.

In the final part, the same approach as described above for target identification was also applied to a family of Oxadiazolone (OX) derivatives. Among the 18 compounds tested, 6 OXs exhibited moderate MIC values but with similar behavior already observed with the CyCs, being either active against intracellular bacteria and/or against extracellular ones. To sum up, our study brings a very attractive data regarding the future use of CyCs and OXs in TB treatment and mycobacterial-related disease regardless they are drug resistant strains or not. Our compounds promise the development of multi-targets antimycobacterial drugs which obstruct the lipid metabolism with the discovery of novel potential targets compared to those already described.

RESUME

La tuberculose (TB) reste la maladie infectieuse la plus meurtrière au monde. L’absence de développement de médicaments accompagnée de l'émergence de la résistance aux molécules existantes se traduit principalement par l'échec du traitement et la mort. Il y a donc un besoin urgent de découvrir de nouvelles cibles ainsi que de nouvelles d'approches thérapeutiques pour lutter contre Mycobacterium tuberculosis (M. tb), l’agent étiologique de TB.

Il est maintenant reconnu que les enzymes lipolytiques mycobactériennes sont impliquées dans de nombreux processus physiologiques du cycle de vie des bactéries: la virulence, la croissance, la dormance, la synthèse de l’enveloppe ainsi que le stockage et la consommation des lipides. Par conséquence, trouver des moyens d'inhiber ces enzymes pourrait ouvrir la voie à la découverte de nouvelles possibilités dans le traitement de la tuberculose.

La première partie de mon travail a été consacrée à l'étude d'une série de nouveaux composés antimycobactériens, à savoir des analogues de Cyclophostin / Cyclipostins (CyCs). Les CyC ont d'abord été testés pour leurs activités antituberculeuses sur la croissance et la prolifération de la souche pathogène M. tb H37Rv mais aussi sur les macrophages infectés par M. tb. Nous avons démontré que ces CyC présentaient des activités antituberculeuses avec des valeurs de

CMI modérées à excellentes (500 nM pour CyC17). Ces données indiquent que certains de ces composés sont capables de pénétrer à la fois dans le macrophage (sans aucun effet cytotoxique) et dans les bactéries, propriétés requises pour un médicament antituberculeux. En utilisant le meilleur inhibiteur extracellulaire CyC17, l'approche basée sur l'activité du profil protéique (ABPP) a permis d'identifier 23 protéines cibles potentielles, impliquées dans le métabolisme des lipides de M. tb et / ou dans la biosynthèse des lipides de la paroi cellulaire.

Egalement, nous avons clairement établi que les CyCs sont de puissant inhibiteurs de croissance, sélectifs des mycobactéries et sans effet sur les autres bactéries Gram positif ou Gram négatif. De manière intéressante, ces composés étaient également actifs contre de nombreuses souches cliniques de M. abscessus, pathogènes opportunistes également appelés Ç cauchemar pour les antibiotiques È, ouvrant ainsi la voie pour la recherche de solutions alternatives qui permettrait de lutter contre les souches mycobactériennes résistantes.

Parmi de nombreuses cibles identifiées, deux protéines importantes, à savoir Ag85C une diacylglycérol acyltransférase et TesA une thioestérase, jouant des rôles clés dans la biosynthèse des lipides de la paroi cellulaire, ont été validées biochimiquement et structuralement comme cibles principales de ces CyC. Dans la dernière partie, la même approche que celle décrite ci-dessus pour l'identification des cibles a également été appliquée à une famille de dérivés d'oxadiazolone (OX). Parmi les 18 composés testés, 6 OX présentaient des valeurs de CMI modérées mais avec un comportement similaire déjà observé avec les CyCs, étant soit actif contre les bactéries intracellulaires et / ou contre les bactéries extracellulaires.

En résumé, notre étude a permis d’apporter des données très intéressantes concernant l'utilisation future des CyC et des OX dans le traitement de la tuberculose et des maladies liées aux mycobactéries, qu'elles soient ou non des souches pharmacorésistantes. Nos composés pourraient permettre le développement de médicaments antimycobactériens multi-cibles qui inhiberaient le métabolisme des lipides avec la découverte de nouvelles cibles potentielles.

INTRODUCTION Introduction

CHAPTER 1: General introduction 1.1. Tuberculosis, an overview Tuberculosis is a very old disease that has been a scourge for humanity and still remains of great concern. With 10.4 million new cases and 1.7 million deaths widespread across the globe as reported by WHO in 2017, TB is now the deadliest infectious disease around the world and remains a great challenge, especially in sub Saharan Africa, India, Russia and Eastern Europe (WHO, 2017) (Fig 1). TB is mainly caused by Mycobacterium tuberculosis (M. tb) but also by various other species belonging to the Mycobacterium tuberculosis complex (MTBC) (Fig 4) (Daher Ede et al., 2013). The disease mostly affects the lung (pulmonary TB) but in 20% of cases can be found in other parts of the body like pleura, lymph nodes, abdomen, genitourinary tract, skin, joints and bones as well as meninges (extra pulmonary TB) (Lee, 2015). An individual can be infected by inhalation of the bacilli contained in aerosol droplets from another person who has active TB. By this way, about one third of the population globally is believed to carry TB germs (Onyango, 2011; WHO, 2017). Fortunately, most of the infected cases are maintained in latent form of TB, also called latent tuberculosis infection (LTBI). The patients do not show any symptoms and the bacteria persist in the lung in a specific structure called granuloma. Approximately 5-15% of these latter cases take a risk to become active (Narasimhan et al., 2013). Patients with active TB have typical symptoms like chronic cough with blood- containing sputum, fever, night sweats, weight loss, etc… Without proper treatment, up to 70% of active TB cases lead to death. People co-infected with HIV and experienced smoking have significant high rate to progress TB disease than normal. The currently recommended treatment by WHO can save 85% of new cases of drug-susceptible TB, but not in the cases of drug- resistant cases including multi-drug resistant TB (MDR-TB), extensive drug resistant TB (XDR-TB) as well as total drug resistant (TDR-TB) (Edlin et al., 1992; Candido et al., 2014; Gunther, 2014; Gunther et al., 2015). Up to date, India, Indonesia, China, Nigeria, Pakistan and South Africa were the top six countries accounted for 60% of the new cases, made Asia and WHO African Region as the highest incidence areas in the world (WHO, 2017). WHO set a strategy up to 2030, to completely control TB.

2 Introduction

Figure 1. Estimated TB incidence rates, 2016. The annual number of incident TB cases relative to population size (the incidence rate) varied widely from under 10 per 100 000 population in most high-income countries to 150–300 in most of the 30 high TB burden countries (WHO, 2017).

1.2. History timeline of Tuberculosis TB is one of the oldest human infectious diseases ever. There are several evidences demonstrating the long existence of TB, as well as the long accompany along with humankind (Rothschild and Martin, 2003). From the coalition analysis in archaeological, sequencing and comparative genomic sequences, the primary origin of modern strains of M. tb arose at least 70,000 years ago. The geographically expansion of M. tb probably happened when humans first ventured out of Africa about 67,000 years ago (Galagan, 2014). The questions about when the bacilli truly start to infect humans or what is the source of the modern human TB, however, are still controversial. A recent study using genomic reconstruction and metagenomic sequencing on 68 skeleton samples associated with TB infection have revealed that human tuberculosis is actually less than 6,000 years old and the sea mammals were probably the first animals who bring the disease from Africa to South America (Bos et al., 2014). In another research, a 9,000 years old M. tb has been detected by conventional PCR from a human bone sample (Hershkovitz et al., 2008) and the pathogen was also thought to be transmitted from domestic animals to humans (Smith, 2003). In literature, the disease was mentioned under various terms like

3 Introduction schachepheth; cunsumptio; consumption; etc... (Daniel and Daniel, 1999). Hippocrates is one of the first human who mentioned the disease as “phthisis” (Daniel and Iversen, 2015) and described in his text the symptom of the disease like coughing with blood, fever, colorless urine, chest pain, etc… and believed that it was heritable. Until 1834, Johann Lukas Schönlein proposed the name “Tuberculosis” based on the recognition of tubercles in all forms of the disease (Cambau and Drancourt, 2014). For centuries, before scientists figure out that bacteria could be a factor responsible to several infectious diseases including TB, many efforts were spent to discover its causative agent. In 1865, A French military doctor - Jean Antoine Villemin - proved that TB was an infectious disease able to transmit from humans or cattle to rabbits. He also proposed that a specific microorganism caused the disease (Daniel, 2015). The discovery was however denied by most of the contemporary scientists. In 1882, Robert Koch and his team discovered the staining method, which was later finalized by Ziehl and Neelsen in 1885, in an attempt to reveal the presence of “foreign parasitic structures...indicative of the causal agent”. By visualizing the causal agent under microscopy and later isolating it on solid media, Robert Koch announced the identification of the microorganism responsible for TB in a remarkable feat of microbiology. For the first time M. tb was isolated and demonstrated in the guinea pig that this slow-growing mycobacterium was the agent responsible for a human disease (Koch, 1882). Since then, an enormous progress has been made with the launch of the first vaccine Bacillus Calmette–Guérin (BCG) in 1921 and series of effective drugs applied in 1950 – 1960s to prevent and treat TB (Luca and Mihaescu, 2013; Zumla et al., 2014). From 1960s to last decades, the stagnation in drug development accompanied with the emergence of MDR, XDR- TB and TB-HIV coinfection lead TB to alarming situation. New diagnostics, drugs and regimens as well as new vaccines are urgently needed. 1.3. Diagnosis and vaccination Diagnosis It is critical to diagnose TB as soon as possible in order to supply a suitable treatment. Although there are several tests used to diagnosis TB, it is however difficult to confirm rapidly and accurately if a person has been TB positive since the investigations often do not have access to adequate initial diagnosis. Globally in 2016, 37% of the 9.6 million new cases were either undiagnosed or not reported, leading to continuously TB transmission, including of MDR-TB cases (Pai et al., 2016b). Classically, the preliminary diagnosis can be started based on typical symptoms and signs of pulmonary TB such as productive and prolonged cough, blood in the sputum, chest pain, losing weight, etc… Tuberculin Skin Test (TST) can be used to detect the present of M. tb in the body by checking if an individual develop the immune response to a

4 Introduction solution of complex antigen PPD. Another option is the blood test with Interferon Gamma Release Assay (IGRA). This is an Enzyme-Linked ImmunoSorbent Assay (ELISA) which measures the in vitro production of interferon-gamma by sensitized lymphocytes in response to M. tb-specific antigens. Therefore, these tests have no cross reaction with BCG antigens. IGRA is also recommended by WHO since it can be done ex-vivo, and have very few false positive in case the individual have been immunized with BCG vaccine (Detjen et al., 2007; de Lima Corvino and Kosmin, 2017). However, these methods are unable to conclude if a person has latent or active TB infection. In order to identify if the individual has active TB disease, three validated methods including microscopy (sputum test using LED microscopy), nucleic acid amplification tests (NAATs) and cultures were applied. Another test such as chest X-ray is also needed (Im et al., 1993; Eisenhuber, 2002). Each diagnosis has its own disadvantages and limitations (Campbell and Bah-Sow, 2006). Currently, the rapid testing Xpert MTB/RIF¨ assay is recommended by WHO (WHO, 2017). The assay is performed in 2 hours, giving accuracy up to 89% of sensitivity and 99% of specificity for TB detection from studies of patients (adults and children) with presumed pulmonary TB. Another assay, the Urine Lipoarabinomannan Rapid Test, can only be used in HIV-coinfected patients since it gives higher sensitivity in these cases. (Drain et al., 2016; Pai et al., 2016b). Vaccines In order to prevent the spread of an infectious disease like TB, vaccine is an indispensable tool. The first vaccine BCG has been used for nearly a century since 1921 (Brewer and Colditz, 1995). A Mycobacterium bovis strain was attenuated by continuously cultured in rich nutrition media for 230 times over 13 years in The Pasteur Institute in Lille, France (Luca and Mihaescu, 2013) in order to generate the vaccine strain. With three billions of people vaccinated worldwide, BCG is still the one and most common used vaccine worldwide against TB (Andersen and Doherty, 2005). Though the BCG vaccination has made important contributions to the control of TB globally, limitations do exist. It has successfully conquered severe forms of TB in children (Lanckriet et al., 1995; WHO, 2017), and the efficacy can be maintained for 10 years. The vaccine has been provided as part of national childhood immunization programs in 163 countries, among which, 102 countries reported coverage of above 90%. However, BCG offered a disappointed level of preventing adult’s cases from TB either before or after infection. Its efficacy indeed varied from 0-80% in adult pulmonary TB giving a disturbance about the protection capacity. This instability of BCG efficiency could be explained due to several factors like genetic characteristics of the population, geographical variation, genetically different between BCG strains, virulence of M. tb strains, or the pre-exposure to environmental

5 Introduction mycobacteria species (Fine, 1995; Andersen and Doherty, 2005). More noticeably, protection level is the lowest in developing countries, where the TB incidence is the highest. In fact, BCG protects only ~5% of all potentially vaccine-preventable cases in total. New and high effective vaccines are obviously needed and should be one of the priorities of The End TB Strategy program. More particular, a new vaccine should be able to prevent the acquisition of infection as well as the progression from infection to active TB disease in adults. Currently, there are 12 TB vaccines in Phase I, Phase II or Phase III clinical trials (Fig 2). These vaccine candidates result from an attenuated M. tb strain, genetic modification of BCG, or recombinant vaccine antigens (Andersen and Doherty, 2005). However, a new TB vaccine in the near future is still a difficult task.

Figure 2. Global clinical pipeline of TB vaccine candidates. Adapted from (Fletcher and Schrager, 2016).

1.4. Biological characteristics of M. tb General biology M. tb species belong to the genus Mycobacterium, family Mycobacteriaceae, order Actinomycetales, suborder Corynebacterineae, phyllum Actinobacteria (Stackebrandt et al., 1997). The bacteria is slightly curved, rod shape, aerobic, non-spore, non-motile with the size varying from 0.2-0.6 µm width and 1-10 µm long, usually 3-5 µm (Cardona, 2012) (Fig 3A). It divides every 15-20 hrs. This is an extremely slow division compare with other bacteria, e.g., Escherichia coli which can divide every 20 min. (Brennan, 2003). When grown on Lowenstein- Jensen (LJ) solid medium, the colonies appears after 6-8 weeks in creamy white color, dry, rough, raised, irregular with wrinkled surface (Fig 3B). Although M. tb does not form spores, it has the capability of transition from active to dormancy state, which is indicated by no or low

6 Introduction metabolic activity (Gengenbacher and Kaufmann, 2012), to protect itself from host immune system.

A B

Figure 3. (A) M. tb mc26230 visualization under confocal microscopy following Nile-Red staining - Source: this study. (B) M. tb H37Rv colonies - Source: CDC / Dr George P. Kubica, 1979.

M. tb is a slow growing mycobacteria (SGM) strict pathogen belonging to M. tb complex (MTBC), (Tortoli, 2014). The relationship between the various strains in MTBC is depicted in Fig 4. Among them, human-adapted M. tb strains consist of 7 groups, classified based on geographical distribution. Lineage 2 (East-Asian, mainly Beijing family) is one of the most virulent lineages, and lineage 4 (Euro-American) is the most geographically widespread. Although MTBC strains including animal-associated ones have highly conserved genome sequences (up to 99.9% identity), they have clearly different phenotypes (Jia et al., 2017). Analysis on different pathogenic aspects indicated that several genes only present in human strains and the deletion of the four first genes of region of difference (RD1) (from Rv3865 to Rv3868), encoding for two secreted proteins ESAT-6 and CFP-10 which are important for virulence(Stanley et al., 2003; Abdallah et al., 2007), in animal strains might explain why they are attenuated virulence in human.

7 Introduction

Figure 4. Evolutionary relationship between selected mycobacteria and members of the Mycobacterium tuberculosis complex. Adapted from (Galagan, 2014).

Genome of M. tb The complete genome sequence revealed that M. tb genome comprises about 4.4 mega base pairs with a significant high content (≥ 65.6%) of guanine + cytosine (GC) (Fig 5) (Miyoshi- Akiyama et al., 2012). This genome is composed of 3924 open reading frames occupying 91% of the potential coding capacity. The genome examination clearly indicated that M. tb is able to synthesize all the essential amino acids, vitamins and enzyme co-factors itself. Moreover the genome annotation revealed a high number of genes encoding for enzymes involved in lipid metabolism: 250 genes in comparison with only 50 genes in E. coli which has a similar genome size. It is thus not surprising since a diversity of lipophilic molecules, from simple to highly complex ones, has been found in M. tb. The bacteria possess the patterns of every known lipid and polyketide biosynthetic system, even the ones only found in mammals and plants, as well as a significant number of degradative and fatty acid oxidation systems. Besides, the tubercle bacillus can produce all the critical enzymes involved in several pathways including glycolysis, pentose phosphate pathway as well as the tricarboxylic acid and glyoxylate cycle. This information suggested the impressive potency of the bacteria to transit among different metabolisms according the environmental changes (e.g., from aerobic in lung to hypoxia environment in the center of granulomas).

8 Introduction

Figure 5. Circular map of the chromosome of M. tb H37Rv. Adapted from (Cole et al., 1998).

Another remarkable highlight revealed in tubercle genome resides in the fact that a large number of potential resistance factors are encoded. Indeed, several genes encoding for drug- modifying enzymes, for instance β-lactamases, aminoglycoside acetyl transferases and many potential drug-efflux systems were found (Cole et al., 1998; Camus et al., 2002). These features combined with the impermeable cell envelop deliver the bacillus natural resistance to many antibiotics. Cell envelope It is difficult to classify M. tb into Gram-negative or Gram-positive bacteria since its cell wall presents characteristics of both groups but also its own features (Fig 6) (L.M.Fu, 2005). In fact, mycobacterial cell envelope is unique and confers the cell outstanding physiological features. The mycobacterial cell envelope was directly analyzed using cryo-electron microscopy of vitreous sections (CEMOVIS) for the first time by Zuber et al (Zuber et al., 2008). This observation revealed a bilayer architecture which prompted the authors to propose that mycolic acids (MAs) are essential for the formation of outer membrane (latterly called mycomembrane) (Fig 6).

9 Introduction

Figure 6. Cell envelope of M. tb . The plasma membrane has typical bacterial membrane structure. The cell wall is lack of an outer membrane like Gram-positive bacteria, but does not have a thick peptidoglycan layer. Instead of that, it consists of thin layers of peptidoglycan (PG), arabinogalactan (AG), and a thick layer mainly composed of long chain mycolic acids (mycomembrane). Intercalated within this lipid environment are the “free” (noncovalently linked) glycolipids (GL, in brown), and phospholipid (PL, in blue). The capsular-like layer outside the outer membrane of M.tb was shown to mainly consist of polysaccharides and proteins with only minor amounts of lipids (2% – 3% of the material. Adapted from (Marrakchi et al., 2014).

Accordingly, the mycobacterial cell envelop consists of the mycomembrane-containing cell wall (MMCW) (Chiaradia et al., 2017) and the plasma membrane. The mycobacterial plasma membrane has its common basic structure, which does not appear obvious differences with that of other bacterial plasma biological membranes (Mamadou Daffé, 2008). But the most important compartment of the cell envelop is the MMCW, constructed by distinct chemical compositions, giving the cell a couple of advantages including their considerable resistance to many drugs and antibiotics, survival in stressful conditions of hostile environments, transport of solutes and proteins, adhesion to receptors (Jarlier and Nikaido, 1990; Chiaradia et al., 2017). Little is known, however, about the composition and arrangement of mycobacterial cell envelope constituents. The mycomembrane (Fig 6), is the bacterial fence as well as the contact between bacteria and the host, obviously playing a role in directing host-pathogen interplay. First, the outer layer (or capsule) of the cell wall presents various non-covalently attached lipids and glycolipids esterified with multimethyl-branched long chain fatty acids. In SGM like M. tb they can be grouped into two families: the trehalose and phthiocerol lipids. Trehalose ester families include sulfolipids (SL) (Layre et al., 2011), diacyltrehaloses (DAT) and

10 Introduction polyacyltrehaloses (PAT) (Rousseau et al., 2003). The phthiocerol lipid family comprises phthiocerol dimycocerosates (DIM) (Astarie-Dequeker et al., 2009), and phenolphthiocerol dimycocerosates or phenolic glycolipids (PGL) (Sinsimer et al., 2008; Kaur et al., 2009). The biosynthesis pathways of all these pathogenicity-related lipids, involve Polyketide synthases (PKSs) (Guenin-Mace et al., 2009). This lipid complex, present in most of pathogenic mycobacteria at the interface between bacteria and host, has been considered to play roles in virulence (Bailo et al., 2015). Then, this outermost layer combined with inner leaflet is made of very long-chain MAs to create the structure called mycomembrane (MM). The cell wall skeleton determines the size and shape of mycobacteria, and comprises three distinct components including MM, peptidoglycan (PG) and arabinogalactan (AG). PG layer is made of several long polymers of disaccharide N-acetyl glucosamine-N-acetyl muramic acid (NAG- NAM) that are cross-linked by peptides bond. Consecutively, PG is covalently attached to arabinogalactan, which is made of galactan chains. Some galactan chains are modified with arabinan polymers. These arabinan polymers are in turn linked to long-carbon-chain MAs. There are six different forms of MAs including α form and oxygenated forms (i.e., cis-keto, trans-keto, cis-methoxy, trans-methoxy and trans-hydroxy) with a number of carbon atoms varying from 76 to 89 (Fig 7). All of these forms of MAs are required for the virulence of M. tb (Brennan, 2003; Hett and Rubin, 2008; Kieser and Rubin, 2014). The MA-containing compound, trehalose dimycolate (TDM), has been reported to protect the bacilli within macrophages by reducing antibiotics efficacy, and to inhibit the excitation of protective immune responses (Marrakchi et al., 2014). Other virulent activities have been attributed to TDM, including the inhibition of phagosome- lysosome fusion (Indrigo et al., 2003), or induce tissue damage and necrosis when it is released outside and associates with lipid, (Hunter et al., 2009). Additionally, oxygenated MAs were demonstrated to trigger the formation of foamy macrophages (FMs), a shelter for persisting bacilli (Peyron et al., 2008).

11 Introduction

Figure 7. Different forms of mycolic acids in M. tb (Marrakchi et al., 2014).

In summary, the cell wall is essential for M. tb pathogenicity and contributes to the bacterial innate drug resistance capacity. Lipid-rich components create an effective barrier for the penetration of TB drugs and antibiotics, and are responsible for the bacilli capacity to survive and replicate in hostile macrophage. The constituents of the cell wall and the enzymes involved in its synthesis are therefore the primary targets of many antituberculosis drugs at present as well as in the near future. 1.5. Pathogenic life cycle Transmission TB is an airborne disease. It is transmitted from one person to another through the air. The development of M. tb infection can be mainly divided in four stages: transmission, immune response, latency and reactivation (Fig 8). When a person who has active TB coughs, speaks or sneezes, the individual expels the aerosol droplets that contain the bacilli. The droplet nuclei which are small enough in size from 1 to 5 µm, are then capable to enter the lower respiratory tract (Ahmad, 2011). Thanks to its impermeable cell wall, the bacteria can survive and remain suspended in the air for several hours. When a person inhales these aerosol droplets, these survival bacilli reach the lungs and begin to grow (Fig 8, Step 1).

12 Introduction

1 .

Figure 8. The transmission cycle of M. tb. Step 1 – Transmission: once an individual inhales bacteria from other patient, the bacille will be engulfed by the first body’s defense – the alveolar macrophage. Step 2 – Immune response: In case the bacilli is able to avoid the phagocytic destruction and start to replicate, the event will attract T cells and they send the signal to order macrophages to come and kill bacteria, leading to the early formation of granulomas. Step 3 - Latency: In most cases, tubercle bacillus stays inside the granulomas, under the control of host immune system and go to latency phase. Step 4 - Reactivation: If the bacteria prevail over in the competition with the immune system, it will escape the granulomas, spread through the body and make the active disease. Adapted from (Nunes-Alves et al., 2014).

From there, they can be transferred to other parts of the body via the lymphatic system and the blood stream (miliary or extrapulmonary TB) (Smith, 2003). Pulmonary TB can be infectious as the bacilli are able to spread to other people through the airborne. Contrariwise, TB in other parts of the body, such as the kidney or spine, is usually not infectious.

13 Introduction

Initial infection The infection starts when a healthy individual breathes tubercle bacilli from expelled droplets from another infected individual. Once reaching the lungs, M. tb deposits and resides within alveolar macrophages and dendritic cells. Normally, the alveolar macrophages ingest the tubercle bacilli and destroy most of them. Although alveolar macrophages are thought to be an effective barrier to contain pathogens, M. tb has evolved various mechanisms to evade the host immune response and survive in these cells (Sasindran and Torrelles, 2011). Therefore, some can multiply within the macrophage and be released when the macrophage dies. From there, the bacilli can spread to other regions of the body by means of bloodstream or lymphatic system. In general, the TB bacilli which spread through the blood, prefer to gather and multiply in the areas that have high level of oxygen including the apex of the lung, lymph nodes, kidneys, brain, long bones and genital tract. M. tb intracellular life The interplay between M. tb and the human host determines the outcome after infection. In most cases, the initial host response can be completely effective thus killing all bacilli (Fig 8, Step 2). In the other cases, organisms start to multiply and grow immediately after infection, causing clinical disease known as primary progressive TB (Andrews et al., 2012). This form of disease is typically seen in children but sometimes also occurs in adults (Donald et al., 2010). Alternatively, the bacilli may become dormant. Within weeks after infection, the immune system is usually able to halt the multiplication of the tubercle bacilli, preventing further progression. In such case, patient is referred to as latent TB infection (LTBI) (Fig 8, Step 3) (Pai and Rodrigues, 2015). About 5 to 10% chance over a lifetime of individuals with LTBI, the tubercle bacilli overcome the defenses of the immune system and eventually begin to grow, with resultant clinical disease, known as reactivation TB (Fig 8, Step 4). This process may occur shortly after infection or several years later (Flynn and Chan, 2001). In patients with HIV infection, there is perhaps a 10% chance of developing TB each year after the establishment of latent infection (Lin and Flynn, 2010). Immune response to M. tb infection It is not surprising to call M. tb “the world's most successful pathogen” (Hingley-Wilson et al., 2003). In fact, the immune response to M. tb infection is a very complex process which need to be thoroughly characterized. In general, M. tb has evolved several tricks to avoid eradication by the immune system, allowing it to live safely within the host. Particularly, the initial defense of the host against infection with M. tb when it reaches the lower respiratory tract is the alveolar macrophages (AM). These cells are capable of inhibiting growth of the bacillus through

14 Introduction phagocytosis. The phagocytosed M. tb is then transferred to lysosome by the appropriate stimuli (Russell, 2001; Gengenbacher and Kaufmann, 2012). The fate of intracellular bacteria can be influenced by autophagy process, resulting in phagosome maturation and in an increase in its acidification by a fusion with lysosome, therefore leading to the clearance of the bacteria (de Chastellier, 2009). Unfortunately, some bacilli developed the ability to overcome such processes especially by preventing the acidification of the phagosomal compartment and detoxifying oxygen and nitrogen radicals (Fig 8, Step 2) (Vandal et al., 2009), thus adapting themselves to the intracellular environment of the macrophage and creating a niche for survival. M. tb bacilli even will multiply and finally lead to the AM destruction. This event will in turn attract blood monocytes and other inflammatory cells (e.g., neutrophils) to the site of infection and also taking part in the phagocytic process until dendritic cells phagocytize the bacteria. The dendritic cells transport the bacteria to lymph nodes. Then when specific T-cells recognize bacterial antigens on the surface of the infected dendritic cells, they will be activated and migrate to the site of infection. This event starts the early stage of the formation of granuloma, where macrophages become activated and kill intracellular M. tb (Ulrichs and Kaufmann, 2006). Remarkably, another trick that allows the TB bacilli to survive is the ability to delay T- cells response. T-cells play a key role in protecting against TB (Cooper, 2009). Specific T-cells recognize peptides, or antigens derived from M. tb and produce factors called cytokines that empower the infected macrophage to kill M. tb, or at least curb its replication. The bacillus transits to the stage of latency (Fig 8, Step 3). At this stage more than 90% of infected people remain asymptomatic, but M. tb is still within AM. In case of reactivation, M. tb starts replicating and increases in numbers in the lung. Unfortunately, T-cells arrive too late and do not reach the lung in sufficient number to begin slowing bacterial replication leading the infection getting out of control (Fig 8, Step 4). Granulomas The granuloma is one of the highlights of immune response to M. tb infection and a hallmark of TB. The formation of granulomas is mediated by both innate and adaptive immunity. It is a collection of immune cells surrounding the central necrotic area, which contains bacteria and destructive infected macrophages (Fig 9). The immune cells found including multinucleated giant cells, dendritic cells, neutrophils, NK cells, epithelioid cells and most notably, foamy macrophages. All these cells are surrounded by a rim of lymphocytes defining a dense cellular wall that restricts the spread of the M. tb (Sasindran and Torrelles, 2011).

15 Introduction

Figure 9. Structure and cellular constituents of the tuberculous granuloma. Adapted from (Ramakrishnan, 2012).

Granulomas have been supposed to form an immune-active microenvironment, their function are believed as a mechanism for the host to control infection and limit systemic dissemination. Using the optically transparent and genetically tractable zebrafish embryo-Mycobacterium marinum model of TB, Clay et al. clearly showed that innate macrophage can restrict the growth and the spread of the bacilli (Clay et al., 2007). However, the mechanisms regarding why granuloma fails to eliminate mycobacteria even in immune-competent hosts, have remained largely unclear. In fact, many studies indicated that the presence of granulomas has both positive and negative sides. On one hand, granulomas creates an immune microenvironment in which the infection can be controlled. On the other hand, it also provides the mycobacterium with a niche in which it can survive, modulating the immune response to ensure its survival without damage over long periods of time. Furthermore, several studies demonstrate that mycobacteria would rather make the granuloma their home than the extra-granuloma environment (Rubin, 2009; Silva Miranda et al., 2012; Orme and Basaraba, 2014; Marino et al., 2015).

16 Introduction

1.6. Treatment of TB Standard treatment for TB Over a century of TB drug development, only 23 anti-TB drugs are currently recommended by WHO in the treatment of TB. They have been divided into first line TB drugs and second line TB drugs, with the latter is further classified into 4 smaller groups (Table 1). During the middle of the twentieth century, the discovery of numerous antibiotics and antituberculosis drugs including streptomycin (STR), isoniazid (INH), kanamycin (KAN), pyrazinamide (PZA) (Fig 10), etc… indeed brought a revolution in the TB treatment (Zumla et al., 2014). Nowadays, the recommended treatment for new cases of drug-susceptible TB is a 6-month regimen of four first-line drugs: isoniazid, rifampicin (RIF), ethambutol (EMB) and pyrazinamide. The first two months is intensive TB drug treatment with the four drugs taken together, followed by continuation TB drug treatment phase with RIF and INH. With this combination therapy, the complete 6-month course of treatment cost about US$ 40 per patient. Treatment success rates of at least 85% for new cases of drug-susceptible TB are regularly reported to WHO. The four- first line drugs are the most effective available drugs against TB which are used more particularly for treating drug-susceptible cases of TB. Front-line drugs INH or isonicotinic acid hydrazide (Fig 10A and B), is one of the most effective antituberculosis drugs since its first introduction in 1952 (Bernstein et al., 1952). Since then it is always in the list of four-first line TB drugs. Although it was shown that non replicating bacilli are insensitive to INH (Zhang et al., 2012; Fattorini et al., 2013), it is nevertheless also applied as a background treatment in the case of latent TB mainly in combination with RIF (Tang and Johnston, 2017). INH has a complex mechanism of action (Fig 10B). After entering the bacterial cell, INH will be activated by a mycobacterial catalase peroxidase called KatG. The activated INH products are then tightly bound to the enzyme called enoylacyl carrier protein (ACP) reductase (InhA) (Unissa et al., 2016). This process afterwards inhibit the action of intracellular target proteins related to mycolic acid biosynthesis, which interfere with cell wall synthesis, thereby producing a bactericidal effect (Fig 10B)(Debasu et al., 2014). The main target of INH is InhA but several other enzymes including ahpC, kasA, ndh, iniABC and fadE are also interfered (Kremer et al., 2003; Unissa et al., 2016).

17 Introduction

A

INH PZA EMB

RIF STR

B

Figure 10. (A) Chemical structures of front-line anti-TB drugs: INH, Isoniazid; PZA, Pyrazinamide; EMB, Ethambutol; RIF, Rifampicin; STR, Streptomycin. (B) Mechanism of action of Isoniazid. Adapted from (Vilcheze and Jacobs, 2014).

Rifampicin (RIF) (Fig 10A), on the other hand, is an antibiotic with a broad antibacterial spectrum, including the activity on mycobacteria. Typically, RIF inhibits the activity of DNA- dependent RNA polymerase, encoded by the β subunit of rpoB gene. RIF forms a stable complex with the enzyme by binding inside the pocket of the RNA polymerase β subunit. It thus suppresses the initiation of RNA synthesis. It was first identified in 1957 and introduced into TB treatment in 1972 (Sensi, 1983). Although this antibiotic is not specifically active on mycobacterium, the combination of RIF with other first-line drugs help reducing the regiment duration from 18 to 9 months (Mitchison, 1985).

18 Introduction

EMB (Fig 10A), which has specific antituberculosis activity, has been introduced in TB standard treatment since the 1960s (EA/BMRC, 1972). The mode of action of EMB is known to obstruct the transfer of mycolic acids into the cell wall (Takayama et al., 1979). EMB also inhibits the synthesis of both arabinogalactan and lipoarabinomannan (LAM). In particular, the enzyme arabinosyl transferase embC is proved to be inhibited by EMB in M. tb. Both of these effects lead to interruption of the complete form of mycolyl-arabinogalactan-peptidoglycan complex and result in the increase of the mycobacterial cell permeability (Goude et al., 2009). EMB is also an important antimycobacterial drug as it enhances the effect of other companion drugs including aminoglycosides, rifamycins and quinolones (Bassam and Mayank, 2013). Therefore, EMB is used in the treatment of MDR- and XDR-TB cases as well.

PZA (Fig 10A) is a prodrug normally used in treatment of active TB, and not recommended for treatment of latent TB infection. It was synthesized for the first time in the 1950s and formerly used only as salvage therapy. It latterly is in use as first line drugs in chemotherapy regime since it reduces current chemotherapy regimen duration from 9 months of treatment to only 6 months (Steele and Des Prez, 1988). The mechanism of action of PZA was only discovered recently by Shi et al. (Shi et al., 2011). After being hydrolyzed by the enzyme pyrazinamidase (PZase) in to pyrazinoic acid (POA), it inhibits the ribosomal protein S1 (RpsA), a vital protein involved in protein translation and the ribosome-sparing process of trans-translation.

STR (Fig 10A) was the earliest antibiotic used in TB treatment. It was discovered in 1943 and was no longer introduced in TB therapy after that. STR targets to 16s rRNA and ribosomal protein S12, which encoded by rrs and rpsL genes, respectively. The compound binds to the phosphate backbone of the 16s rRNA of the 30S subunit of the bacterial ribosome and then interrupts the protein synthesis (Corper and Cohn, 1949; Almeida Da Silva and Palomino, 2011). STR showed to have a good bactericidal activity comparable to EMB and can be used as the substitute for EMB in standard TB treatment in specific cases. Nevertheless, the use of STR as the mono anti-TB drug at the time of its discovery rapidly led to the occurrence of STR resistance isolates in 1947 (Pyle, 1947). STR efficiently kills growing bacillus, but not non- growing or intracellular bacilli.

19 Introduction

Table 1. Categories of TB drugs and mechanisms of resistance

Genes involved Mutation Name Drug targets Mechanism of action Proteins in resistance frequency (%)

Group 1: First-line oral antituberculosis drugs katG Catalase peroxidase Isoniazid FAS-II ER Inhibit mycolic acid biosynthesis 40 - 97 inhA Fatty acid enoyl acyl carrier protein reductase A Rifampicin RNA polymerase β-subunit RpoB Inhibit RNA synthesis rpoB β-subunit of RNA polymerase 90 - 100 Ethambutol Arabinosyl transferase EmbB, EmbC Obstruct the formation of the cell wall embB Arabinosyl transferase 47 - 89 Pyrazinamide Ribosomal protein S1 RpsA Inhibit trans-translation pncA Pyrazinamidase 44 -97 Ribosomal protein S12 and 16S rrs 16S rRNA 12 - 26 Streptomycin rRNA components of 30S ribosomal Inhibit protein synthesis rpsL S12 ribosomal protein 40 - 68 subunit gidB 7-Methylguanosine Methyltransferase 5 - 13 Group 2: fluoroquinolones Ofloxacin, Levofloxacin, DNA gyrase, type II and IV gyrA DNA gyrase subunit A 70 - 90 Inhibit bacterial cell division Moxifloxacin topoisomerase gyrB DNA gyrase subunit B 0 - 11 Group 3: Aminoglycosides - injectable antituberculosis drugs Kanamycin 30S ribosomal subunit Inhibit protein synthesis eis Aminoglycoside acetyltransferase 28 - 80 Amykacin 30S ribosomal subunit Inhibit protein synthesis rrs 16S rRNA 40 - 90 Capreomycin Interbridge B2a between 30S and 50S Inhibit protein synthesis tlyA 2’-Omethyltransferase 4 - 13 ribosomal subunits Group 4: less effective second-line antituberculosis drugs inhA acyl carrier protein reductase A 33 - 62 Ethionamide/ Enoyl-acyl-carrier-protein reductase Inhibit mycolic acid biosynthesis ethA Monooxygenase 46 - 72 Prothionamide ethR Transcriptional repressor EthR 0 - 4 alr Alr protein Cycloserine/ D‑alanine racemase and ligase Inhibit peptidoglycan synthesis ald Ald protein NA Terizidone cycA Dalaninetransporter thyA Thymidylate synthase A Paminosalicylic acid Dihydropteroate synthase Inhibit folate biosynthesis 37 - 50 folC FolC protein Group 5: newly developed drugs and repurposed agents F1F0 proton atpE ATP synthase c chain Bedaquiline ATP depletion NA ATP synthase rv0678 Transcription repressor for transporter MmpL5 Inhibition of mycolic acid synthesis, ddn Deazaflavin-dependent nitroreductase Pretomanid, Delamanid F420 cofactors NA production of reactive nitrogen species fdg1 F420-dependent glucose-6-phosphate dehydrogenase SQ109 MmpL3 Inhibit mycolic acid biosynthesis mmpL3 Membrane transporter NA Transcription repressor for transporter Production of reactive oxygen species, rv0678 Transcriptional regulator of MmpL5-MmpS5 Clofazimine NA MmpL5 inhibition of energy production pepQ PepQ protein Linezolid 23S rRNA Inhibit protein synthesis rrn ribosomal L4 protein 2 - 11 NA: not available. Adapted from (Caminero et al., 2010; Palomino and Martin, 2014; Zhang and Yew, 2015).

20 Introduction

It is undeniable that current standard therapy for drug-susceptible TB is highly effective. Several drawbacks have however arisen, including long treatment duration, various side effects, high rates of non-adherence and increased mortality if standard regimens are not strictly followed. Accordingly, shorter and more effective treatment regimens are needed to reduce the burden of infectious cases. 1.7. Drug resistance and mechanisms Nolonger after the introduction of the front-line antituberculosis drugs, drug resistance had been dawned. One of the earliest cases of TB-drug resistance described, is the streptomycin-resistant case in 1947 (Pyle, 1947). Over three decades, only occasional cases of drug resistance were reported and mostly did not drive the attention of contemporary medical field. Until early of 1990s, several clusters of multidrug-resistant TB strains to the most effective TB drugs including INH and RIF has emerged leading to a high mortality rate (Edlin et al., 1992; Fischl et al., 1992; Park et al., 1996). A bunch of millions of dollars has been spent in order to control the exposition of the outbreaks in the US (Frieden et al., 1995). Nevertheless, the continuation of selection and widespread of MDR-TB led to the rise of extensively drug resistant strains (XDR-TB) in the 2000s (Dahle, 2006; Lawn and Wilkinson, 2006); the turning point being the recent appearance of untreatable TDR-TB (Velayati et al., 2013). Nowadays, TB drug resistance remains as the main critical challenge. In low and middle-income countries, TB still caused numerous morbidity and mortality, and drug resistant TB is a major concern. Globally, about 20% of active TB cases were diagnosed as active drug-resistant TB (WHO, 2017). This issue is of major, importance since it means that these cases are not fully susceptible to any available antibiotics. The traditional regimen for drug–susceptible TB cases cannot be applied. To sum up, it is supposed that mismanagement of TB treatment and person-to-person transmission are responsible for the high rate of deaths. Beside the intrinsic drug resistance due to the possession of diversity of efflux pumps and its complex impermeable cell wall (Balganesh et al., 2012), tubercle bacillus mostly develops drug resistance ability through genetic mutations. The genes which associated with drug-resistance types and their consequences were briefly noted in Table 1. Typically, TB drug resistance have been grouped mainly into three categories. First, multidrug-resistant TB (MDR-TB) refers to active TB cases that does not respond to at least INH and RIF, the most effective drugs. In 2016, 600,000 incident cases of MDR-TB/rifampicin-resistant TB (RR-TB) and 240,000 drug- resistant TB-related deaths have indeed been reported (WHO, 2017). The highest incidences (47% of the global total) have been found in India, China and Russia federation (Fig 11). The

21 Introduction update treatment success rate in 2015 was 83% among patients who adhere to complete treatment procedure.

Figure 11. Estimated incidence of MDR/RR-TB in 2016. Data shown only countries with >1000 incident cases. Areas that are not applicable are in grey (WHO, 2017).

The mechanism of MDR is mainly due to katG, inhA and rpoB mutations. The genetic alteration on katG and inhA genes lead to either the reduction or deletion of catalase-peroxidase activity and structural modification of the target of isoniazid, and both result in resistance to INH (Tseng et al., 2015). Meanwhile, mutations in rpoB have been found in remarkable diverse types (e.g., single-nucleotide changes, deletions and insertions). Mutations happened mainly in the 81-bp hot spot region called the RIF resistance-determining region (RRDR) (Telenti et al., 1993; Schilke et al., 1999; Heep et al., 2001), and have been found strongly associated with RIF resistance. Resistance to RIF alone is rare. RIF-resistant isolates normally associated with INH resistance and therefore it has been proposed that resistance to RIF could be used as a surrogate marker for MDR-TB (Caws et al., 2006). The most concerned is extensive drug-resistant (XDR-TB) infection, recognized when the treatment with second-line drugs are ineffective due to drugs misuse or mismanagement. XDR- TB strains are resistant to either INH and RIF along with any fluoroquinolones and at least one of the second-line injectable drugs – i.e., capreomycin, kanamycin, and amikacin (Caminero et

22 Introduction al., 2010). By the end of 2014, 105 countries reported the occurrence of XDR-TB cases, with the highest incidence in Belarus, Lithuania, Latvia and Georgia (WHO, 2015). In 2015, 7579 XDR-TB cases were reported by 74 countries (Prasad et al., 2017). Approximate 9.5% of MDR- TB cases are XDR-TB, as previously estimated (9.7% in 2014 and 9.0% in 2013). XDR-TB patients are curable, but with the current drugs available, the rate of success is low. Besides genetic modification related to INH and RIF, mutations on targets of fluoroquinolone and second-line injectable drugs have been found in XDR isolates. Fluoroquinolones were introduced into clinical practice in the 1980s, not only in TB treatment. They target mycobacterial topoisomerases II (DNA gyrase), encoded by gyrA and gyrB genes, and therefore inhibit the DNA supercoiling. Missense mutations within the conserved region of gyrA and gyrB have been identified that are highly associated with fluoroquinolone resistance (Ginsburg et al., 2003). The second-line injectable drugs (i.e., kanamycin, amikacin, capreomycin and viomycin) share the same mechanism of action by inhibiting the protein synthesis process (Alangaden et al., 1998; Sirgel et al., 2012). In particular, kanamycin and amikacin bind to 16S rRNA in the 30S small ribosomal subunit and inhibit protein synthesis. On the other hand, capreomycin and viomycin bind to the site at the interface of small and large subunits of the ribosome, resulting in the inhibition of the protein synthesis (Palomino and Martin, 2014). The kanamycin and amikacin resistance related to the mutation of rrs gene while capreomycin and viomycin resistance associated with the tlyA mutation. Last but not least, the development of the most dangerous TB-resistant strains in recent years known as totally drug resistant (TDR-TB) is really alarming. Although it has not been recorded largely, the deadly TDR-TB does exist. TDR-TB strains showed in vitro resistance to all first and second line drugs tested (Velayati et al., 2013). At least four different countries have reported the occurrence of these strains including Iran, India, Italy and South Africa (Migliori et al., 2007; Velayati et al., 2009; Udwadia et al., 2012; Slomski, 2013), and so far, remains untreatable. It is reasonable to say that we may face a new outbreak of untreatable TB. The management of TB needs to be conformed to international standards including appropriate diagnosis and treatment. No less importantly, the development of new, more effective TB drugs to improve treatment outcomes is urgently needed.

1.8. Treatment in drug-resistance TB and new drugs development Current treatment in drug-resistance TB The development of TB-drug resistance is more alarming than ever. Interventions in the fight against TB is urgently needed. Therapies for active drug-resistant TB have poor evidence of

23 Introduction success. Around 20% of active TB cases were diagnosed as active drug-resistant TB (WHO, 2016), which means that these cases are not fully susceptible by any available antibiotics. An effective regimen is indispensable for the best treatment outcomes as well as to prevent the further development and enlargement of drug resistance. However, the identification of suitable regimen for specific case of drug resistance is complicated and depends on numerous factors including the feature of the patients (patient with HIV coinfection for example), and the drug susceptibility profile of the resistant strains (Caminero et al., 2010; Zumla et al., 2013). Standardized treatment of MDR-TB requires 18-24 months with the use of second-line antituberculosis drugs. As the result, the treatment is longer, more expensive and toxic, less potent associated with several side effects (Mukherjee et al., 2017). Until early 2016, a shorter MDR-TB regimen of 9-12 months is now recommended for all patients (excluding pregnant women) with pulmonary MDR-TB that is not resistant to second-line drugs. The cost of a shortened drug regimen is about US$ 1000 per person compare to US$ 2000–5000 for the standardized treatment. The shortened treatment is divided in two phases with the intensive phase consists of kanamycin, high-dose moxifloxacin, clofazimine, EMB, high-dose INH, PZA and prothionamide given daily for 4 months, followed by the second phase of high-dose moxifloxacin, clofazimine, EMB, PZA given daily for an additional 5 months (WHO, 2016). A study in Bangladesh showed a higher treatment success with this new regimen than routine MDR-TB treatment (Moodley and Godec, 2016). In XDR-TB, the treatment regimen depends on the extent of the drug resistance, the severity of the disease and whether the patient’s immune system is compromised. Mortality in patients co-infected with HIV is higher. It is important to detect early and accurate diagnosis to provide an efficient treatment as soon as possible. Effective treatment requiring a good selection of second-line drugs is available to clinicians who have special expertise in treating such cases (Caminero et al., 2010; Gunther, 2014). Currently, drugs used in treatment of TB are categorized in five groups and could be apply depending on each cases of drug resistant. The addition of high doses of INH, EMB, PZA (group 1) and fluoroquinolones (group 2) to therapeutic regimens is recommended but these are not considered as one of the four basic drugs of the regimen in patients with XDR-TB. The four basic drugs should be chosen among: i) injectable antituberculosis drugs (group 3) in which one could be selected, ii) drugs in group 4 or so-called second-line drugs in which all four basic drugs could be selected, and iii) drugs in group 5 in which one could be counted as half of one of the four drugs that formulate a treatment regimen when needed (Caminero et al., 2010).

24 Introduction

New drugs in development The treatments for MDR and XDR TB are experiencing several difficulties including the complexity and length of drug-sensitive regimens. There is no unified therapeutic option for all cases. Besides, the regimen time is still long and easy leading to further drug resistance if the regimen is not strictly followed during 6-12 months of treatment. Shorter, simpler and effective regimens for all cases of TB are mandatory. In order to achieve this goal, it is necessary to develop new drugs that will shorten and simplify the treatment. Ideally, these new drugs should target novel mechanisms of action that are equally effective against MDR, XDR and drug- sensitive strains of TB with no toxicity to the patients. They also need to have minimal drug- drug interactions in case of combination with other antibiotics if TB patients are co-infected with HIV. Last but not least, they need to be given at a low cost. Taking a look into the current drug classes available, most of first- and second-line drugs were all discovered between the 1940s and the 1970s. With the outbreaks of MDR during 1990s and the emergency of XDR during 2000s, the trend in new TB drug development was only restrated since few last decades. Currently in the TB drug pipeline (Fig 12), there are no more than ten drugs in advanced phases of clinical trials for the treatment of drug-susceptible TB, drug-resistant TB or LTBI. Most of these compounds belong to three chemical classes: oxazolidinones, nitroimidazoles or fluoroquinolones (Pai et al., 2016a).

Figure 12. Current global pipeline of new TB drugs (Pai et al., 2016a).

Bedaquiline (Fig 13A) is the only compound approved by US FDA as the new antituberculous drug after 40 years and is currently in use for treatment of drug resistant TB. Bedaquiline is an inhibitor of mycobacterial ATP synthase. Nevertheless, molecular mechanism of mycobacterial growth inhibition is still poorly understood and needed to be improved (Hards et al., 2015).

25 Introduction

Besides bedaquiline, delamanid – a novel inhibitor of mycolic acid synthesis (Fig 13B), has been approved by European Commission for treatment of resistant TB with very limited evidence (Thakare et al., 2015).

A B Figure 13. Chemical structure of the two new approved anti-TB drugs. (A) Bedaquiline; (B) Delamanid.

The crisis of microbial resistance against antibiotics is a serious global health issue, not only in TB but also in many other infectious diseases. In the case of TB, the lack of new drugs is one of the major factors leading to the drug resistant problems. During nearly a century, only two compounds have been approved as new antituberculosis drugs with poor clinical evidences. In the summary information noted in Table 1, most of anti TB drugs final purpose is to interfere with either bacterial cell wall conformation or bacterial protein synthesis machinery. Only the new drug bedaquiline is likely to target a new biological process, i.e., ATP synthesis. It is noteworthy that most of the available drugs, including the new ones in discovery and pre- clinical development phases as depicted in the new TB drugs pipeline (Fig 12), don’t really focus on lipid metabolism – one of the most important metabolic pathway of tubercle bacilli, especially lipid consumption and accumulation pathways (Dedieu et al., 2013; Lovewell et al., 2016). 1.9. Lipid metabolism of M. tb M. tb genome encodes 250 putative enzymes involved in lipid metabolism compared to only 50 in Escherichia coli (Cole et al., 1998; Camus et al., 2002). This suggests that lipids and lipid metabolism are essential for bacterial viability. The hallmark of tuberculosis, granuloma, is composed of a core of infected alveolar macrophages, surrounded by additional types of macrophages e.g., monocytes, multinucleated giant cells, epithelioid cells, and most notably, foamy macrophages (FM) (Russell et al., 2009a). It is indeed well established that some of the infected macrophages in granulomas accumulate lipids in lipid bodies (LB), which are mainly composed of triacylglycerols (TAG) surrounded by a phospholipid layer (Dedieu et al., 2013), giving the cells a foamy appearance (Peyron et al., 2008). The induction of foamy macrophages

26 Introduction packed with LB has been reported in many pathogens, and not only in TB. But the remarkable point is that within FM, phagocytosed M. tb preferentially metabolizes lipids rather than carbohydrates (Wheeler and Ratledge, 1988) and up-regulates several mycobacterial genes involved in lipid metabolism (McKinney et al., 2000). Indeed, in such foamy macrophages the bacilli accumulate lipids and can persist in a non-replicating state for decades, but can also be reactivated to cause acute disease (Cardona et al., 2000; Saunders and Cooper, 2000). To persist inside FM, M. tb hydrolyzes host lipids TAG from LB and the resulting fatty acids are stored as newly synthesized TAGs within intracytoplasmic lipid inclusions (ILI) (Peyron et al., 2008; Russell et al., 2009a; Russell et al., 2009b; Daniel et al., 2011; Podinovskaia et al., 2013; Caire- Brandli et al., 2014). During the reactivation phase, these ILIs are hydrolyzed by M. tb and used to fuel the regrowth of mycobacteria during their exit from the hypoxic non-replicating state (Low et al., 2009). A better understanding of how the bacilli persist inside lipid-rich FM could therefore help finding new ways to fight the disease. Above all, these results strongly suggest a key role of intra and extra-cellular lipolytic enzymes in the capacity of survival of M. tb within the infected host. Therefore such mycobacterial lipolytic enzymes, strongly involved in the host-pathogen cross-talk, are thought to play critical roles in the physiopathology of the disease by participating in the entry into a non-replicating dormant state within host granulomas and/or in dormancy escape, leading to the reactivation of the disease (Dedieu et al., 2013). Moreover, a number of lipolytic enzymes were found embedded within the cell wall, and besides their potential role in enzymatic degradation of host cell lipids, they should be considered as structural components of the membrane (Cotes et al., 2007; Mishra et al., 2008). Surface- exposed localization also suggests that they may participate in modulation of the immune response (Shanahan et al., 2010; Brust et al., 2011; Shen et al., 2012), in invasion of host cells and in virulence (Bakala N'goma J et al., 2010; Dedieu et al., 2013).

27 Introduction

CHAPTER II: Lipolytic enzymes in M. tb – the fundamental of a new drug family development 2.1. Serine hydrolase enzymes from M. tb: promising therapeutic targets Serine hydrolase is a large and diverse enzyme classes that possesses a nucleophilic serine in their active site and contains the α/β-hydrolase fold as secondary structure (Fig 14) (Ollis et al., 1992; Carr and Ollis, 2009). Such α/β-hydrolase fold consists of 5-11 parallel and antiparallel β-strands (blue arrows), normally 8, flanked on both sides by α-helices (red rectangles) in an alternate manner. Serine hydrolases catalyze one of the most basic biochemical reactions: the cleavage of amide and ester bonds, thus make it becomes one of the largest and most diverse enzyme family with several catalytic functions such as lipases, esterases, phospholipases, thioesterases, proteases, peptidases, amidases, etc…(Ortega et al., 2016b). Among different known catalytic functions, lipase and esterase constitute a large category (Nardini and Dijkstra, 1999; Carr and Ollis, 2009). Following the update ESTHER database, the α/β hydrolase was subdivided into 148 families, some of them being grouped in 47 superfamilies. One of the main reason made this group of enzyme has such large number of families due to the difference of aligned sequences of the family (Lenfant et al., 2013b). These enzymes did not share any significant sequence similarity nor did they catalyze on similar substrates or use the same nucleophile. In contrast, they shared a noticeable structural similarity. Furthermore, the arrangement of the catalytic residues is quite preserved, suggested that they had possibly evolved from a common ancestor (Nardini and Dijkstra, 1999). Within the α/β-hydrolase fold superfamily, 75% contain a Ser-His-Asp (Glu) catalytic triad. The catalytic serine residue usually appears in the conserved pentapeptide Gly-Xaa-Ser-Xaa-Gly (GXSXG). The other two active site residues are an acidic residue (Glu or Asp) and a His as a general acid–base catalyst (Rauwerdink and Kazlauskas, 2015). The alternative nucleophilic residue could be found is a Cys in dienelactone hydrolases and in haloalkane dehalogenases, it is an Asp. The nucleophile elbow, which is highly conserved with the active nucleophilic serine positioned on sharp turn linking an α-helix to a β-sheet and harboring the nucleophile active site residue at its tip (Lenfant et al., 2013a). This unique α/β-hydrolase fold structure therefore brings the serine, glutamic/aspartic acid, and histidine in proximity allowing the formation of the correct hydrogen-bond network to create the catalytic triad for catalysis.

28 Introduction

Figure 14. Canonical secondary structure diagram of the α/β-hydrolase fold. Helices and strands are represented by pink cylinders and blue arrows, respectively. The location of the catalytic triad is indicated by black dots. Dashed lines indicate the location of possible insertions. Adapted from (Nardini and Dijkstra, 1999).

The α/β-hydrolase fold refers to the catalytic domain. This domain contains the core catalytic machinery i.e., the Ser-His-Asp catalytic triad and the oxyanion hole, the main part of substrate binding site. In case α/β-hydrolase fold enzymes possess the lid or cap domain, it forms the rest of the substrate-binding site, contribute to substrate specificity. The cap or lid domain has diverged amino acid sequences, normally is ~100 amino acid residues with mostly helical secondary structure inserted between strands β6 and β7 of the catalytic domain. However, the lid domain can be missing completely as in case of cutinase (Rauwerdink and Kazlauskas, 2015). Whereas, a large number of lipases have a mobile lid domain located over the active site. Lid protects the active site and hence responsible for catalytic activity. The lid is mostly closed and open partially only in presence of a hydrophobic factor. Hence, the lid controls the enzyme activity (Khan et al., 2017). In total, α/β-hydrolase enzymes vary in their amino acid sequences as well as in their broad range of substrates, they however share a core (part without a potential lid) with a same fold (Fig 15).

29 Introduction

Figure 15. Superimposition of the crystal structures of bacterial serine hydrolases showing a highly structural fold conservation in catalytic domains among them. Acetylxylan esterase from Talaromyces purpureogenus represented in pink; Fusarium solani cutinase from Nectria haematococca in cyan; cutinase-like protein from Cryptococcus in green; MSMEG_6394 lipase from M. smegmatis in red; and lysin B from Mycobacteriophage D29 in orange. Adapted from (Sultana et al., 2011).

Most importantly, the catalytic triad is strongly conserved (Nardini and Dijkstra, 1999; Holmquist, 2000), leading to a similar mechanism of action between all members of this superfamily (Fig 16). This point can be considered as a key to design efficient inhibitors of such enzymes. The serine hydrolases constitute diverse central functions on all levels of M. tb physiology (Vandalet al., 2008) and obviously represent a gold mine awaiting to be exploited as new therapeutic targets against TB. The following paragraph will describe in more details the importance of such proteins in the bacilli life cycle.

30 Introduction

Figure 16. The hydrolysis of substrate by a typical serine hydrolase enzyme. Adapted from (Raza et al., 2001).

2.2. Mycobacterial lipolytic enzymes With the significant high number of lipolytic enzymes in proteome, M. tb indeed has sufficient tool to fuel the resumption in order to supply the energy source for replication and reactivation. 105 α/β-hydrolases predicted proteins have been identified in M. tb H37Rv genome (Johnson, 2017), among which approximately 30 enzymes have been annotated as putative esterases or lipases based on the presence of the consensus sequence G-x-S-xG (Singh et al., 2010; Delorme et al., 2012). Based on their primary sequences, 28 of 31 M. tb lipolytic enzymes are classified into different families: 1 belonging to the Candida parapsilosis lipase family; 1 to the human Bile Salt-Stimulated Lipase (BSSL) family; 12 to the human Hormone-Sensitive Lipase (HSL) family; 7 to the Fusarium solani cutinase (Cut) family; 3 to the monoglyceride lipase family; 3 not belonging to the above mentioned lipase family and 4 to the phospholipases C (PLC) family. Except for the PLCs, all other enzymes possess an active site with the classic Ser-Glu/Asp-His catalytic triad and belong to the α/β hydrolases fold superfamily.

Among the 21 lipolytic enzymes that have been characterized biochemically, only 6 have solved structures (Johnson, 2017). It is however clear that such α/β hydrolase enzymes take part in a

31 Introduction diversity of essential physiological mechanisms, including cell envelope biosynthesis and maintenance (Belisle et al., 1997; Sassetti et al., 2003; Meniche et al., 2009; Alibaud et al., 2011); detoxification (Canaan et al., 2004b; Lamichhane et al., 2005; Newman et al., 2005); host nutrition scavenging (Cotes et al., 2007; Mishra et al., 2008; Schue et al., 2010) and modulation of the immune response (Chakhaiyar et al., 2004; Pethe et al., 2004; Leyten et al., 2006; Beaulieu et al., 2010); and thus play a major role in the pathogenic life cycle of the mycobacteria. The following paragraphs will describe the diversity of mycobacterial lipolytic enzymes based on the nature and the specificity of their corresponding substrates.

True lipases

Among the 12 genes in the so-called Lip-HSL gene family (LipC to LipZ) (Delorme et al., 2012) only the Rv3097c gene, named as LipY, exhibited a true lipase activity on long-chain TAGs (Deb et al., 2006). The authors further demonstrated that when mycobacteria enter the dormancy-like state, lipY was up-regulated significantly higher than the other lipase/esterase genes. Moreover, they reported that in a lipY-deficient mutant of M. tb, TAG hydrolysis was violently decreased under nutrient deprived condition (Fig 17A). In fact, up to date, only few enzymes have been characterized as true lipases in M. tb. LipY is indeed the first and sole true lipase belonging to this HSL family which has been biochemically characterized since 2006. The sequence analysis indicates that N-terminus of LipY displays sequence homology with the Pro-Glu-polymorphic GC-rich sequences (PE-PGRS) protein family meanwhile the C-terminal domain (i.e., catalytic domain) possesses amino acid domains homologous with the HSL family (Deb et al., 2006). More recently, Daleke et al. showed by immunoblotting and electron microscopy (Fig 17B) that LipY was secreted to the bacterial surface in an ESX-5-dependent fashion; and after transport, the PE domains was removed by proteolytic cleavage (Daleke et al., 2011). Moreover, using a whole cell lipase assay, they demonstrated that the secreted LipY was able to hydrolyze extracellular lipids.

32 Introduction

A B

Figure 17. (A) LipY hydrolyzed TAG. Thin layer chromatography of lipids extracted from cultures of M. tb wild type (WT) strain or LipY-deficient mutant strain (∆-lipY mutant) showing the level of TAGs (arrow) was comparable when the two strains were grown in 7H9 media (lanes 1 and 3; controls, but in the condition of nutrient- deprived in phosphate-buffered saline (PBS) for 6 h, the WT strain utilized TAGs far more efficiently than the ∆- lipY mutant. (B) LipY is located on mycobacterial surface. Electron microscopy showing immunogold localization of LipY on the cell surface in a M. marinum WT strain (left panel) but not in a M. marinum ESX-5 mutant strain (type 7 secretion system) (right panel). Adapted from (Abdallah et al., 2006; Deb et al., 2006; Daleke et al., 2011; Dedieu et al., 2013).

Taken all these data into account, LipY is thus suggested to be not only involved in the hydrolysis of lipids from host LBs, but also responsible for the utilization of accumulated TAG within ILIs during dormancy, leading to the reactivation. The latter involvement in LB hydrolysis is in agreement with the upregulation of lipY found in M. tb recovered from THP-1 derived macrophages (THPM) (Daniel et al., 2011). Another aspect resides in the fact that a robust humoral response was detected in patients with active TB using anti-LipY antibodies, therefore implying that LipY might interact with the host immune system and trigger a specific humoral response (Fig 18) (Mishra et al., 2008; Brust et al., 2011).

Taken together, these studies highlighted the central role of LipY in TAG hydrolysis during nutrient starvation or acquisition of host lipids by degrading TAGs present in the LBs.

Similarly to LipY, the Rv0183 lipolytic enzyme was shown to induce a strong and specific immune response of the host upon M. tb active infection (Brust et al., 2011). (Fig 18). Sequence analysis and homology modeling shows that Rv0183 has 36% similarity with human monoglyceride lipase and is highly conserved among mycobacterial species (Saravanan et al., 2012).

33 Introduction

Figure 18. Comparison of the antibody IgG response to recombinant lipolytic enzymes LipY and Rv0183 in TB patients and healthy individuals. Horizontal lines indicate the mean value, whereas dotted lines indicate the cut-off values, derived from the ROC curves. BCG+BD, BCG vaccinated blood donors (n = 50), TB, active TB patients (n = 105). Adapted from (Brust et al., 2011; Tuaillon et al., 2011).

Accordingly, Rv0183 displays a strong substrate selectivity for monoacylglycerols compared to diacylglycerols or triacylglycerols. Additional immunolocalization studies revealed that this enzyme is present only either in culture medium or in bacterial cell wall (Cotes et al., 2007). Moreover, an intriguing change in the colony morphology, from rough, irregular to smooth and round shape, was observed (Fig 19) with a disrupted mutant of its ortholog MSMEG_0220 in M. smegmatis, which were partly restored in the complemented mutant (Dhouib et al., 2010).

Figure 19. Colony and aggregation modification attribute to the disruption of MSMEG_0220 - the homolog of Rv0183 in M. tb. The deletion of MSMEG_0220 leading to smooth colony formation in the deletion mutants Ms∆0220 (B) and the recovery of rough colonies alike WT strain (A) in complemented strains ComMs∆0220 (C), and ComMs∆0220S111A (D). Deletion strain Ms∆0220 do not form the aggregation (F) meanwhile the other strains do (E, G, H. corresponding to aggregation seen in WT strains, complemented strains ComMs∆0220 and ComMs∆0220S111A). Adapted from (Dhouib et al., 2010).

34 Introduction

Moreover, studies of the antimicrobial susceptibility of such MSMEG_0220 strain showed that this mutant is more sensitive to rifampin and more resistant to isoniazid than the wild-type strain, pointing to a critical structural role of this enzyme in mycobacterial cell wall architecture, in addition to its function in the hydrolysis of host cell lipids.

Carboxylesterases

Carboxylesterases catalyze the hydrolysis of soluble ester substrates rather than water-insoluble long chain TAGs. The main characterized enzymes belonging to this familly are LipH (Rv1399c) (Canaan et al., 2004a), LipF (Rv3487c) (Zhang et al., 2005; Richter and Saviola, 2009), LipC (Rv0220) (Shen et al., 2012), Rv0045c (Guo et al., 2010; Zheng et al., 2011; Savas et al., 2013) and CaeA (Rv2224c) (Lun and Bishai, 2007; Lun et al., 2014) and lipL (Cao et al., 2015; Singh et al., 2016). Most of them are required for bacterial optimal growth (Sassetti et al., 2003). Among them, Rv0045c, LipH, LipL and LipF efficiently hydrolyze short chain esters whereas LipC and CaeA prefer substrates with intermediate carbon chain length. LipH may participate in the detoxification pathway of the intracellular lipid metabolism, its “real” role in bacterial life cycle still remains to be clearly elucidated.

In contrast, various studies pointed out the importance of LipF in bacterial pathogenesis. Indeed, the lipF gene was not only upregulated by exposure to growth media at pH 4.5 (Saviola et al., 2003) but it was also demonstrated that its promoter has been transcriptionally upregulated specifically by acidic stress (Richter and Saviola, 2009). Moreover, a transposon insertion between the promoter region for lipF and the gene resulted in a reduced ability for the bacteria to grow in mouse’s lung (Camacho et al., 1999). These results imply that LipF might contribute to the survival of the tubercle bacilli under the acidic environment of the phagosome.

Regarding LipC, this enzyme is a cell surface-associated esterase that is present in both the cell wall and the capsule of M. tb. Consistent with this localization, LipC induces pro-inflammatory cytokine and chemokine responses from macrophages and pulmonary epithelial cells, suggesting strong immunogenic properties during active M. tb infection (Shen et al., 2012).

CaeA (also named Hip1 for hydrolase important for pathogenesis 1) is a cell wall-associated carboxylesterase involved in cell wall biosynthesis and/or integrity (Lun and Bishai, 2007). It is possible to consider CaeA as a protective factor of the bacteria by degrading toxic lipids (Trivedi et al., 2005). CaeA was also found to play important roles in virulence, multidrug- resistance (Lun et al., 2014) and innate immunity (Rengarajan et al., 2005). The absence of

35 Introduction

CaeA enhanced host innate immune responses and compromised the intracellular survival of M. tb in macrophages (Rengarajan et al., 2008). Accordingly, the deletion mutant of caeA gene in M. tb resulted in the attenuation of bacterial growth in mouse lung and spleen and to reduce virulence of the mycobacteria in infected mice (Fig 20).

Figure 20. Deletion mutant of CaeA leading to attenuation virulence of M. tb in mouse lung. BALB/c mice infected with M. tb CDC1551 WT, CaeA deletion mutant (∆caeA) or complemented stain (COM). More granulomas were found at day 56 post-infection in WT complemented strain than in ∆caeA deletion mutant. Adapted from (Lun and Bishai, 2007).

Recently, the Rv1497 (LipL) from M. tb H37Rv was annotated as putative esterase and predicted to be involved in lipid metabolism of the bacterium (Cao et al., 2015). In fact, Rv1497 demonstrated both esterase and β-lactamase activities (Singh et al., 2016). The Ser88 located within consensus β-lactamase motif S-x-x-K was identified as catalytic residue by site-directed mutagenesis in both esterase and β-lactamase enzymatic activities. The enzyme also demonstrated preference for the short chain para-nitrophenylbutyrate ester. The results of the subcellular localization experiment suggested that LipL was mainly located in the cell membrane and cell wall, and to a lesser extent in the cytoplasm of the bacteria. The expression of lipL gene was significantly up-regulated during acidic stress as compared to normal conditions in in vitro culture of M. tb. Given its potential surface location in mycobacteria, LipL was further demonstrated to induce a strong humoral immune response and activate a CD8+ T cell-mediated response (Cao et al., 2015). Overall, these results suggest that LipL could be considered as potential target of clinical diagnostic as well as a vaccine candidate.

36 Introduction

Cutinases

Cutinases not only degrade cutin - which is a polyester protecting plant leaves, but they are also able to hydrolyze several other substrates such as TAGs or phospholipids as in case of Fusarium solani pisi cutinase (Carvalho et al., 1998; Longhi and Cambillau, 1999; Egmond and de Vlieg, 2000). Seven proteins homologous to cutinase (Cut1 to Cut 7) have been identified from their position on the M. tb genome (Cole et al., 1998). However these proteins have also been named Cutinase like proteins with the following correspondence: Rv1758 or Cut1/Culp5, Rv1984c or Cfp21/Culp1, Rv2301 or Cut2/Culp2, Rv3451 or Cut3/Culp3, Rv3452 or Cut4/Culp4, Rv3724 or Cut5/Culp7 and Rv3802c or Cut6/Culp6 (West et al., 2009). They all belong to the α/β hydrolase fold family proteins and contain the conventional Ser-Asp-His catalytic triad, with Ser located in the conserved G-x-S-x-G sequence. Except Cut1/Culp5, the remaining 6 other enzymes have putative secretion signals (West et al., 2009). Cfp21 and Cut4 are indeed secreted into the culture filtrate, and Cut6 is located in the cell wall fraction (Weldingh et al., 1998; Parker et al., 2007; Meniche et al., 2009). These features support the view that several Cut might interact with the host membranes and participate in the virulence/infection process by degrading host lipids. Enzymatically, while four enzymes displayed weak or no activity at all towards the various substrates tested (West et al., 2009), the three remaning Cut i.e., Cfp21 (Rv1984), Cut4 (Rv3452) and Cut6 (Rv3802c) exhibited various enzymatic activities. In particular, Cfp21 and Cut4 showed distinct substrate specificities, despite of the 50% identity between their amino acid sequences and their similar overall predicted structures. Cfp21 preferentially hydrolyzes medium-chain carboxylic esters and monoacylglycerols, whereas

Cut4 behaves like a phospholipase A2, and is able to induce macrophage lysis (Schue et al., 2010). In contrast, Cut6 is able to act on various substrates, exhibiting a low level of activity on long chain pNP ester substrates (West et al., 2009), as well as a thioesterase activity (Parker et al., 2007; Parker et al., 2009). This latter activity is consistent with the fact that the cut6 gene is located in a gene cluster encoding essential enzymes (i.e., AccD4, Pks13, FadD32, Ag85D and Ag85A) which have been demonstrated to play a crucial role in mycolic acid biosynthesis. Moreover, Cut6 has also been found to be essential to the growth of M. tb (Sassetti et al., 2003; Meniche et al., 2009). One hypothesis should be that thanks to its ability to hydrolyze thioesters, Cut6 would release the nascent mycolic acids while docked via thioester bonds on Pks13 (Parker et al., 2009). Finally, Cut6 is also regarded as a potential vaccine antigen since it was able to trigger a specific T-lymphocyte response and to protect mice from a challenge with M.

37 Introduction tb. (Shanahan et al., 2010). Despite of the high identity with cutinase from Fusarium solani pisi, none of these 7 M.tb cutinases exhibited cutinase activity.

Phospholipases

Phospholipases comprises 4 major groups classified according to the position of the bond they hydrolyze, including phospholipase A (PLA); phospholipase B (PLB); phospholipase C (PLC) and phospholipase D (PLD) (Fig 21).

Figure 21. Phospholipase cleavage sites. PLA1 hydrolyses the acyl ester bond at the sn-1 and PLA2 at the sn-2 position; PLB has a combined PLA1 and PLA2 activity; PLC hydrolyses the glycerol-oriented and PLD the alcohol- oriented phosphodiester-bond. Crosses indicate cleavage site location. PL, phospholipase; R1/R2, non-polar fatty acid chain; X, denotes the phospholipid head group. Adapted from (Flammersfeld et al., 2017).

The respective proteins expressed PLA, PLC and PLD activities have been found in M. tb (Johansen et al., 1996). PLAs, which cleave fatty acyl chain either at the sn-1 (i.e., PLA1) or sn-2 (i.e., PLA2) position of the phospholipid core, are known to play a significant role in human inflammatory states and disease pathogenesis (Yedgar et al., 2000; Istivan and Coloe, 2006). As mentioned above, Cut4 and Cut6, two members of the Cut family, possess PLA activity, and contribute to bacterial virulence of the pathogen (Parker et al., 2007; Schue et al., 2010).

While PLC activity was found only in virulent mycobacterial species, PLD activity was recovered in both pathogen and non-pathogen species, suggesting that PLD may be important in the life-cycle of this genus, while PLC activity is required for virulence. The M. tb genome carries four types of phospholipase C (PLC) genes, plcA, plcB and plcC and plcD, in which the

38 Introduction first three genes are adjacently located and lack in M. bovis (Gordon et al., 1999; Matsui et al., 2000). Due to their ability to hydrolyze host cell membrane phospholipids, all four recombinant enzymes induced cytotoxic effect on mouse macrophages (Bakala N'goma J et al., 2010; Bakala N'Goma et al., 2015).

Overall, all the data collected from the last ten years on M. tb lipolytic enzymes have clearly emphasized their major role in the life-cycle of the bacilli (Dedieu et al., 2013; Johnson, 2017). It is now acknowledged that such microbial lipolytic enzymes are involved in bacterial growth, cell wall biosynthesis, carbon sources management and virulence during infections by M. tb. Consequently, these enzymes have not only become the focus of intense researches, but above all they have been recognized as powerful biomarkers (Brust et al., 2011; Shen et al., 2012) and many investigations seem to prove these enzymes could become new therapeutic targets (Kremer et al., 2002; West et al., 2011; Delorme et al., 2012; Point et al., 2012a; Lehmann, 2016; Johnson, 2017).

39 Introduction

CHAPTER III: Inhibitors of lipolytic enzymes Finding ways to inhibit the activity of mycobacterial lipolytic enzymes could pave the way for discovering new modalities for TB treatment. In order to achieve this strategy, a reconsideration on available related lipolytic enzymes inhibitors is needed. 3.1. Orlistat Among the potent lipolytic enzyme inhibitors, β-Lactones represent an important class of compounds bearing the strained 2-oxetanone 4-membered ring that presents potent inhibitory activity against serine hydrolases. The most representative member of this family of inhibitors, is the FDA approved anti-obesity drug Orlistat (also known as tetrahydrolipstatin, THL, Fig 22A), that inhibits the human digestive lipases (Borgström, 1988; Hadvary et al., 1991). Orlistat inhibits gastric and pancreatic lipases in the lumen of the gastrointestinal tract therefore leading to a significant decrease of systemic absorption of dietary fats (Heck et al., 2000).

Figure 22. Chemical structure of Orlistat (A), its mode of inhibition on lipases and the possible primary degradation product (B). Panel B shows a schematic representation of the nucleophilic attack of the β-lactone ring of Orlistat by the lipase active site serine residue leading to the formation of the long-lived acyl-enzyme complex 1. Hydrolysis of this covalent adduct releases the active enzyme and provides the β-hydroxy carboxylic acid 2 as the primary degradation product. Adapted from (Benarouche et al., 2014).

Orlistat is an active site-directed inhibitor that forms a stoichiometric long-lived acyl-enzyme complex with serine (or cysteine) enzymes after the nucleophilic attack of the catalytic serine residue on the -lactone group (Hadvary et al., 1991; Luthi-Peng et al., 1992). However, a

40 Introduction factor that is needed to refer is the reversibility of lipase inhibition by Orlistat (Fig 22B). Although it is often written in literature reviews that Orlistat is an irreversible lipase inhibitor, it was shown in early studies that such inhibition was reversible therefore leading to several degradation products (Stalder et al., 1990; Stalder et al., 1992; Benarouche et al., 2014). Since 1997, Orlistat was however known to be also able to inhibit some microbial lipases (Haalck and Spener, 1997). Functioning as a versatile serine/cysteine hydrolase inhibitor, Orlistat was indeed found to inhibit the Cut family including the mycobacterial phospholipase/thioesterase Cut6 (Parker et al., 2009; Crellin et al., 2010; West et al., 2011), as well as the Lip-HSL family enzymes (Delorme et al., 2012). Consequently, when tested as a possible anti-mycobacterial agents, Orlistat was found to block M. tb growth on solid medium with a minimum inhibitory concentration (MIC99) of around 40M (Delorme et al., 2012). Using activity-based protein profiling (ABPP) approach on whole cell lysates of M. bovis BCG culture, Wenk’s group identified 14 enzymes targeted by Orlistat in which 10 were lipolytic enzymes: LipD, G, H, I, M, N, O, V, W, and TesA. Cellular overexpression of LipH and TesA further led to a decreased susceptibility of the bacteria to Orlistat, thus providing targets validation (Ravindran et al., 2014). Another interesting recent work conducted with Orlistat highlighted its synergetic antimycobacterial action in combination with Vancomycin (Rens et al., 2016). Lipids analysis confirmed that Orlistat destabilized the outer membrane of the cell envelope by reducing the amount of phthiocerol dimycocerosate (PDIM) content in the mycobacterial cell wall, facilitating the vancomycin action. To sum up, although Orlistat should be considered as an interesting antimycobaterial drug candidate, it however presents the major disadvantage/drawback to inhibit a large panel of human lipolytic enzymes.(Yang et al., 2010). 3.2. Orlistat-core compounds In an attempt to improve the specificity of Orlistat toward the essential M. tb lipolytic enzyme Cut6 (Rv3802c), West et al. synthesized a first generation of inhibitor library (4-20) based on the Orlistat pharmacophore (Fig 23). Particularly, the β-lactone ring was kept as the reactive site of the compounds. Additionally, a diverse side chains were incorporated in order to investigate first structure-activity relationships regarding the selective inhibition of Cut6. These included lipophilic side chains (4-7), aromatic and hetero-aromatic rings (8-11) and a number of flexible non-aromatic heterocyclic side chains (12-20) (West et al., 2011). As a result, several compound exhibited improved inhibitory activities in comparison with Orlistat. In specific compound 12, bearing an L-thiazolidyl ester side chain, and analogs 17-20 bearing L- and D- prolyl ester side chains displayed IC50 value of 0.2-0.8 µM toward Cut6 compared with 3.8 µM for Orlistat. A good correlation between the activities displayed against Cut6 and M. tb growth

41 Introduction inhibition, was also observed. The latter active inhibitors showed antibacterial effects ranging from 1.3 to 4.0 µM which is nearly 4 to 10 times lower than for Orlistat (15 µM). Although Rv3802 is considered as an essential gene for the survival of M. tb in vitro (Sassetti et al., 2003) and was inhibited by Orlistat-core compounds, the target identification experiments need to be performed in order to clarify the mode of action of these molecules to inhibit bacterial growth.

Figure 23. Chemical structure of Orlistat-core compounds. Adapted from (West et al., 2011).

3.3. β-lactone EZ120 Lehmann et al. in an argument that several serine hydrolases in the mycomembrane biosynthesis machinery showed their ability of binding to mycolates to transfer, synthesize or release these mycolate, suggest the idea that aliphatic molecules with an electrophilic group like -lactones could be attractive serine inhibitors (Lehmann, 2016). They are reminiscent of the signature of mycolic acid β-hydroxy motif, form covalent binding to the active site of these enzymes through ring opening and acylation form. A collection of -lactones was then examined for their antibacterial activity. The compound EZ120 (Fig 24) displayed an effective and exclusive antibacterial activity against M. tb H37Rv with MIC value of 1.6 µM and low cytotoxicity to mouse macrophages (Lehmann, 2016). By applying ABPP approach using the alkyne-modified probe EZ120P, the two main targets pks13 and Ag85A were determined.

42 Introduction

However, the mechanism of inhibition of EZ120 on these targets remains unclear and need to be confirmed.

Figure 24. -lactone EZ120 and its proposed mechanism of binding to a serine hydrolase enzyme. Adapted from (Lehmann, 2016).

3.4. Lalistat Similar to Orlistat, Lalistat (Fig 25A) is a mammalian lipase inhibitor which has also been tested on M. tb growth. Lalistat is a thiadiazole carbamate initially designed to inhibit the human lysosomal acid lipase (LAL) (Rosenbaum et al., 2010). This enzyme located in cellular late endosomes hydrolyzes cholesterol esters and TAGs from incoming lipoproteins. Lalistat, similarly to Orlistat, binds covalently to the active site serine of lipolytic enzymes and thereby inhibits their activity. Given the structural and electronic properties of Lalistat, it was anticipated that such inhibitor will react with diverse enzymes which may deviate from the Orlistat target spectrum. Such selectivity may be attribute to the carbamate and the 3,4- disubstituted thiadiazole moieties. With regards to its mechanism of action, while Orlistat inhibit lipases through the acylation of active Ser via nucleophilic attack on the lactone ring, Lalistat do it through a reversible carbamoylation of the active Ser residue (Rosenbaum et al., 2010) (Fig 25B). Recently, in order to perform in situ target identification via activity based protein profiling (ABPP), Lehmann et al. synthesized a Lalistat probe analog bearing a terminal alkyne tag (La-1) (Fig 25B) (Lehmann, 2016).

43 Introduction

Figure 25. Chemical structures of (A) Lalistat and the new La-1 probe; and (B) mechanism of inhibition of lysomal acid lipase (LAL) by such thiadiazole carbamates. Adapted from (Rosenbaum et al., 2010; Lehmann, 2016).

Next, target identification experiments conducted on M. tb cells coupled with proteomics allow identifying 20 target enzymes, including 8 Lip-HSL enzymes (LipH, -N, -I, -R, -M, -G, -T and -O) (Fig 26). Interestingly, comparison of these results with those previously reported with Orlistat proteome labeling studies in M. bovis BCG revealed Lip M, O, N, I and G as shared targets (Ravindran et al., 2014), while LipR and T were additionally captured by Lalistat (Lehmann, 2016). Six additional enzymes with lipolytic activity were also detected: Rv0183, Cfp21 (Rv1984), Rv0045c as well as two uncharacterized proteins Rv2715 and Rv1192.

Lalistat also inhibited M. tb growth with MIC50 values of 25-50 µM, comparable with that of Orlistat ~ 30 µM (Kremer et al., 2005). Interestingly, Lalistat displayed a strong synergetic effect in combination with Vancomycin resulting in a MIC dropt of around 4- and 16-times for Lalistat and Vancomycin, respectively. Again, when tested on M. tb infected human macrophages, Lalistat was found to substantially reduce bacterial load by 55% compared to the untreated control suggesting that this compound could even address intracellular bacteria. This molecule, however possesses the obvious disadvantage, like Orlistat, of inhibiting also mammalian lipases.

44 Introduction

Figure 26. (A) Activity based protein profiling (ABPP) workflow with lalistat La-1 probe for quantitative proteomics, and (B) enrichment and competition target identification (ABPP) volcano plot representations and corresponding list of common hits. Adapted from (Lehmann, 2016).

3.5. Oxadiazolone-core compounds Among the wide range of lipolytic enzymes inhibitors, 3-aryl-5-methoxy-1,3,4-oxadiazol- 2(3H)-one compounds have been reported to exhibit very interesting inhibitory properties depending on the nature of the N-3 substituent. Compounds 1 (Desmoras et al., 1974) and 2 (Tieman, 1981) (Fig 27A) were first described by Huang et al. (Huang and Bushey, 1987) to be efficient housefly head acetylcholinesterase (AchE) inhibitors as well as potent insecticides. In the early 2000’s, high-throughput screening procedure at Aventis Pharma was used to identify various classes of hormone sensitive lipase (HSL) inhibitors (Petry et al., 2004). This led to the development of 3-phenyl oxadiazolone derivatives (e.g., MmPPOX formerly known as compound 7600, and the commercial CAY10499 – Fig 27A and B), based on their in vivo high inhibitory activity on partially purified HSL from rat epididymal adipose tissue (Schoenafinger et al., 2001). In vitro tests also showed that MmPPOX specifically inhibited the HSL family member proteins, whereas under the same experimental conditions no significant inhibition was observed with other lipolytic and non-lipolytic carboxylester hydrolases that were not members of the HSL family (Ben Ali et al., 2006). It was also shown that the inhibition process of HSL family member proteins involved a nucleophilic attack by the hydroxyl group of the enzyme’s catalytic serine on the carbon atom of the carbonyl moiety (C(carb)) of the

45 Introduction oxadiazolone ring (Fig 27B) (Ben Ali et al., 2012). Such N-3 substituted oxadiazolone HSL inhibitors (including MmPPOX and CAY10499) were also identified as reversible inhibitors (Muccioli et al., 2008; Minkkila et al., 2009; Kiss et al., 2011) of endocannabinoid-hydrolyzing enzymes, namely fatty acid amide hydrolase (FAAH) and human recombinant monoacylglycerol lipase (hrMGL); both involved in several physiological processes, such as pain sensation and inflammation (Lambert and Fowler, 2005).

Figure 27. (A, C, D) Chemical structure of Oxadiazolone core compounds, and (B) their mechanism of action. The covalent bond is formed due to the nucleophilic attack of the catalytic serine or cysteine residue on the oxadiazolone ring. Adapted from (Ben Ali et al., 2006; Point et al., 2012b; Point et al., 2016).

More recently, in the lab, the MmPPOX compound as well as ten 5-Alkoxy-N-3-(3- PhenoxyPhenyl)-1,3,4-Oxadiazol-2(3H)-one derivatives (RmPPOX – Fig 27C) have been evaluated as selective and potent inhibitors of mammalian digestive lipases (Point et al., 2012b; Point et al., 2016). These compounds were found to strongly discriminate classical pancreatic lipases (poorly inhibited) from gastric lipase (fully inhibited). Among them, the BemPPOX (Fig 27C) was identified as the most potent inhibitor of gastric lipase, even more active than Orlistat. Further in vitro test meal digestion and in vivo experiments with a mesenteric lymph duct cannulated rat model show that such compound was able to slowed down the overall lipolysis process and led to a subsequent reduction of around 55% of the intestinal absorption of fatty acids compared to controls. All these data promote BemPPOX as a potent candidate to efficiently regulate the gastrointestinal lipolysis and therefore to develop new strategies to “fight against obesity”.

46 Introduction

The oxadiazolone core is also present in several compounds gifted with anti-mycobacterial activity. Compound S57 (Fig 27D), initially described in 1954 by Stempel et al. (Aeschlimann and Stempel, 1954) and Wilder Smith et al. (Wilder Smith, 1954) as being highly active against TB (Wilder Smith and Brodhage, 1961; Wilder Smith et al., 1962; Wilder Smith, 1966), was recently used as lead compound by Mamolo et al. for the synthesis of new series of 1,3,4- oxadiazole-2-one derivatives (Mamolo et al., 2005; Zampieri et al., 2009). These latter molecules were found to exhibit very interesting anti-mycobacterial activity against M. tb H37Rv, with MIC values in the range 1.25-2.5 µg/mL as compared with reference drugs isoniazid (MIC 0.5 µg/mL) and rifampicin (MIC 1 µg/mL). The minimal inhibitory concentration (MIC) was defined as the minimum concentration required to inhibit 99% of the growth. Moreover by molecular modeling, the authors showed that such class of molecules possessed all necessary features to anchor the active site (via van der Waals, electrostatic, hydrophobic interactions and critical hydrogen bonds) leading to block the enzymatic activity of the mycobacterial cytochrome P450-dependent 14a-sterol demethylase (P45014DMs) (Bellamine et al., 1999), a target for antifungal (Sheehan et al., 1999) drug design. These findings thus suggest the potential use of such oxadiazolone derivatives as alternative tuberculosis therapeutic agents (McLean et al., 2007). In the lab, we also reported that the OX compound MmPPOX was also able to inhibit the growth of M. tb with MIC values determined on solid medium of around 15-25 µg/mL (Delorme et al., 2012). Keeping in mind its strong affinity toward the HSL family member proteins, we further investigated the in vitro inhibition of pure recombinant M. tb Lip-HSL enzymes. As expected, all purified Lip-HSL proteins were strongly inhibited by MmPPOX which reacted with the catalytic Serine residue by forming a covalent but reversible bond. Such an inhibitor could then be considered as a long-life substrate rather than a true inhibitor. 3.6. Cyclophostin and Cyclipostins molecules Natural Cyclophostin (Fig 28A), a bicyclic organophosphate molecule, was isolated from a fermentation solution of Streptomyces lavendulae (strain NK901093) during a search for natural insecticides (Kurokawa et al., 1993). This natural product showed potent inhibition of acetyl cholinesterase (AChE) from housefly and the brown plant hopper with reported IC50 of 0.76 nM. The unusual bicyclic enol-phosphate moiety is also found in a second family of structurally related natural products, named the Cyclipostins (Fig 28B) isolated from the fermentation of Streptomyces sp. DSM 13381 (Vertesy et al., 2002; Wink et al., 2002).

47 Introduction

Figure 28. Chemical structure of natural (A) Cyclophostin (R = Alkyl = CH3) and (B) Cyclipostins (R = fatty alcohols C14 to C16; Alkyl chain C1 to C3). Adapted from (Kurokawa et al., 1993; Wink et al., 2002).

The Cyclipostins possess a core structure similar to that of Cyclophostin but are phosphate esters of long chain lipophilic alcohols of various lengths and structures. All identified Cyclipostins have been described to be potent inhibitors of hormone-sensitive lipase (HSL) (Vertesy et al., 2002) and have also been reported to inhibit the growth of various mycobacteria including M. smegmatis, M. phlei, Nocardia abcessus, and Corynebacterium diphteriae (Seibert et al., 2008). MIC obtained were similar or even lower than those of the well-known antibiotics rifampicin or penicillin G. These recent results strongly suggest that Cyclophostin and Cyclipostins compounds can inhibit serine hydrolases produced by these organisms, including mycobacterial lipases. Given these results, we investigated the antibacterial properties of monocyclic analogs to Cyclophostin and Cyclipostins (renamed CyC – Fig 29).

48 Introduction

Figure 29. Rationale for synthesis of new Cyclophostin and Cyclipostins (CyC) analogs, and related structures investigated. Natural Cyclophostin molecule (1) and phosphonate analogs (2); monocyclic phosphonate analogs to either Cyclophostin (3-4, 5-10, 15) or Cyclipostins (11−14, 16-17); and natural Cyclipostins P (18). Adapted from (Point et al., 2012a; Nguyen et al., 2017).

Indeed, to increase the specificity and activity of these molecules, 26 CyC analogs (Fig 29) were synthesized by our collaborator, Dr. Christopher Spilling’s group (University of Missouri- St Louis, USA), by varying the nature of the alkyl group either at the C-5 carbon atom (i.e., R2 – Fig 29) or at the phosphorous center (i.e., R3 – Fig 29). In particular, phosphonate compounds

(X=CH2 – Fig 30) have garnered a reputation for being excellent structural mimics of natural phosphates. It is also well-known that, for such phosphonate analogs, the mode of inhibition

49 Introduction involves the formation of a non-hydrolysable covalent bond between the phosphorus atom and the catalytic Oγ serine residue of the digestive lipases enzyme active site (Cavalier et al., 2000).

Figure 30. Mode of action Cyclipostins/Cyclophostin analogs (CyC). CyCs are able to form a covalent bond with the catalytic serine or cysteine residue at the active site of α/β-hydrolase fold family proteins. These compounds were also best described by the relationship between the OR3 on phosphorus and the H substituent on the C-5 carbon atom as being either in a trans (α-isomer) or cis (β-isomer) relationship. Adapted from (Point et al., 2012a; Point et al., 2013).

These analogs were further tested on available pure mycobacterial lipolytic enzymes, i.e., LipY and Rv0183 from M. tb, Fusarium solani pisi cutinase, as well as the mammalian digestive lipases available in the lab. From the data obtained, we pointed out that such modulation of the lipophilicity strongly impacted the inhibitory efficiency of the CyCs and could be exploited to significantly either attenuate or increase the affinity of one inhibitor to target a specific lipase over others (Lenfant et al., 2013a; Point et al., 2013). More importantly, the obtained molecules were not only powerful mycobacterial enzyme inhibitors, but above all they had lost their effects on mammalian enzymes initially targeted by natural molecules (Point et al., 2012a; Martin et al., 2015; Vasilieva et al., 2015). This fundamental characteristic has led us to consider the use of these CyCs as specific anti-mycobacterial agents. In brief all these latter mycobacterial enzymes impacted belong to the α/β hydrolases fold family of proteins. The highly structural conservation of this family of enzymes along with their essentiality to M. tb life cycle indeed offer a very promising multi-targets drugs which i) toward new targets, new metabolism, ii) broadly applicable against various strains e.g., drug-susceptible and multi-drug resistant strains. These potential giving us a strong motivation of development of candidates with high specificity to microbial enzymes from the α/β hydrolases fold family, solve the encountered major problem when using mammalian lipases/esterases inhibitors as well as improve the inhibitory efficacy on microbial enzymes. Taking into account these results, the design and synthesis of compounds that inhibit the activity of selected microbial α/β-hydrolases fold enzymes is of a fundamental value for understanding not only the molecular mechanisms involved in their catalytic activities, but above all for deciphering the role of mycobacterial enzymes in M. tb lipid metabolism and virulence.

50 Introduction

In summary, all of these compounds initially designed as mammalian enzyme inhibitors, have been evaluated for their potential anti-mycobacterial activity. Some of them has been synthesized purposely as new antitubercular agents (i.e., Orlistat-core compounds), or showed a significant synergy when used in combination with other antibiotics towards M. tb (i.e., Lalistat or Orlistat). These first studies have not only emphasized the importance of lipolytic enzymes in the life cycle of the mycobacteria but above all, they have demonstrated the need of developing selective and specific inhibitors designed towards microbial lipolytic enzymes to fight the disease. In particular, the first results obtained with the OX and CyC compounds, support the assumption that these molecules might target several lipolytic enzymes involved in various physiological pathways, acting thus as potential multi-target candidates. Such multi- targets drug is nowadays a trend of drug research and development (Lu et al., 2012; Medina- Franco et al., 2013). For many years, it was believed that a specific drug molecule acting only against a single target was the right direction for drug development. Nowadays, with the achievement of knowledge in both clinical studies and biological systems, single-target drugs have revealed also disadvantages. Particularly, infection diseases have complex and varied causative factors. Thus, the inhibition of a single target might be inadequate to achieve desired effect to target biological system. In the earlier period of anti-TB drug development, the introduction of streptomycin (STR), discovered in 1943, as the very first mono anti-TB drug and nolonger after that, lead to the occurrence of STR resistance in 1947 (Pyle, 1947). The combination of several drugs in regimen is then needed to overcome such issue. For over 50 years, a quadruple cocktail of antibiotics was applied to eradicate M. tb. However, even when multiple drugs were applied, the genetic resistance to anti-tuberculosis drugs due to spontaneous chromosomal mutations still happened, leading to the occurrence of MDR, XDR and TDR isolates. In this context the development of multi-target drug candidates is of major interest to fight the disease.

51 Introduction

3.7. Activity-based protein profiling (ABPP), powerful chemical proteomic platform applied to investigate targets of CyC analogs and Oxadiazolones derivatives. One of the major hurdles in drug development resides in the identification of the target(s) of small molecules selected from whole cell screens. In this context, activity-based protein profiling (ABPP) approach for targets identification represent an useful technique. The so- called activity-based probes (ABPs), following labelling and enrichment procedures, allows to isolate selective sets of low-copy-number enzymes in complex proteomic mixtures through the chemical recognition of a specific catalytic mechanism without interference from the more represented proteins (Cravatt et al., 2008). ABPs typically contain (i) a reactive group which forms a covalent and irreversible adduct with the target; (ii) a linker region that allows to control the specificity of the probe; and (iii) a tag for visualization (fluorescent tag) (Patricelli et al., 2001) and/or enrichment and isolation (Liu et al., 1999) of the covalently labelled proteins. As suitable probe, fluorophosphonate (FP) ABPs, bearing either a fluorophore (i.e., rhodamine for TAMRA-FP) or a biotin (i.e., Desthiobiotin-FP) reporter tag, were selected (Kidd et al., 2001; Leung et al., 2003) (Fig 31). Due to their mechanism of action leading to irreversible enzyme inhibition (Liu et al., 1999; Okerberg et al., 2005) (Fig 31), such ABPs have been exploited to screen for reversible and irreversible inhibitors of drug targets (Lehmann, 2016; Ortega et al., 2016a; Tallman et al., 2016).

Figure 31. Chemical structure of fluorophosphonate probes and their mechanism of action on serine hydrolases. Each probe forms an irreversible covalent bond with an active site serine (or cysteine) for irreversible enzyme labelling. Adapted from (Phuong Chi Nguyen, 2017).

In one hand, TAMRA-FP bears a fluorescent tag which allow to in situ visualize labelled proteins using fluorescent gel imaging (Patricelli et al., 2001). In the other hand, Desthiobiotin-

52 Introduction

FP brings a Desthiobiotin tag which consequently can perform enrichement step using biotin- affinity beads (Simon and Cravatt, 2010) prior to mass spectrometry analysis. The efficiency of these probes in M. tb proteomes has been proved by Tallman et al. Using these small molecules, the group revealed several functional esterases in active, dormant, and reactivating cultures of M. tb. (Tallman et al., 2016). In another study, by using serine-reactive fluorophosphonate probe, 78 proteins with serine hydrolase activity out of the 186 putative M. tb serine hydrolases were identified (Ortega et al., 2016a). Those studies have confirmed the high specific property of these probes towards serine/cysteine hydrolase enzymes and suggest that they are compatible for being used as competitive molecules for identifying the potential targets of new inhibitors of α/β-hydrolase fold enzymes as described in Fig 32.

Figure 32. Workflow on identification the “off-targets” of inhibitor of interest among a proteome complex using fluorophosphonate probes. A lysate containing proteome is incubated with the inhibitor of interest before adding the FP probes. The proteins which covalently bind to the inhibitor, will not react with the probe. By comparison the bands intensity on SDS-PAGE gel with that of control sample (without inhibitor), the off-targets of inhibitor will be revealed. The mass spectrometry analysis can be performed subsequently in order to pinpoint exactly these impacted targets. Adapted from “User guide: ActivX Serine hydrolase Probes – Thermo Fisher”.

In brief, the proteome of interest, which is prepared by cell lysis procedure, is incubated with the inhibitor in a large molar excess. The proteins which covalently bind to the inhibitors will thus lost the opportunity to bind to the ABPP probes which is added subsequently. Another control tube which is treated only with ABPP probes is prepared in parallel. By the comparison of the protein patterns of the two samples, the off-targets will be revealed directly by in-gel fluorescence scanning in case of using probes with fluorophore reporter tag. For the purpose of targets identification using mass spectrometry analysis, the probes with biotin reporter tag were used. With the high affinity of biotin to streptavidin (Holmberg et al., 2005), the biotin probe- binding proteins were then isolated from those of non-binding proteins in proteome and sent to mass spectrometry for downstream analysis.

53 Introduction

3.8. Objectives of my thesis The emergence of TB drug resistance is undoubtedly warning a new TB outbreak worldwide. New drugs towards new targets and/or new metabolisms are urgently needed. Lipid metabolism has recently raised the notice to play critical roles in pathogenesis of the bacilli. The enzymes involved thus becoming a gold mine to exploit new therapeutic targets for developing future drugs in the fight against TB. In this context, the main strategy was to examine the ability of our newly synthesized compounds as potential lipolytic enzymes inhibitors, to block mycobacterial growth in different physiological conditions and pinpoint the highest potential compounds as well as their real targets, thus set up primary steps in the process of TB drugs development. In order to reach this strategy, the goal of my PhD thesis has been divided in 4 major points: Point 1 - Susceptibility testing: Both families of inhibitors, i.e., oxadiazolone derivatives and CyC analogs, have been tested for i) their capacity to inhibit in vitro mycobacterial growth (determinations of MIC and MBC (minimum bactericidal concentration)); ii) their anti- tuberculous activity on infected macrophages using fluorescent and electron microscopy techniques; and iii) their cytotoxicity towards non-infected macrophages. Point 2 - Target identification: Inhibitors allowing best mycobacterial clearance both in vitro and ex vivo in infected macrophages have been used to identify target proteins. Activity-based protein profiling (ABPP) approach using available probes have been applied to enable selective labeling and enrichment of captured enzymes prior to mass spectrometry identification. Confirmatory studies of candidate targets were done by performing overexpressing mutants. Point 3 - Biochemical study of essential identified targets: In order to validate the identified targets as well as deciphering their comprehensive roles, biochemical studies including enzyme activity, inhibition study as well as crystal structure resolution have been done on pure recombinant proteins. Point 4 - Check the selective inhibition on mycobacteria: Since the molecules are supposed to act on several unique targets of mycobacteria, these compounds should be considered to treat mycobacterial related diseases. Growth inhibition effect on several bacterial strains including Gram positive, gram negative as well as mycobacterial strains was therefore investigated.

54

RESULTS Results – Article 1

Article 1: Cyclipostins and Cyclophostin analogs as promising compounds in the fight against tuberculosis

As described above, we have developed two new families of molecules; namely the Oxadiazolone derivatives (OX – Fig 27), and the Cyclipostins & Cyclophostin analogues (CyC – Fig 28); which exhibit both potent anti-lipolytic enzyme activity and interesting antituberculosis effects (Delorme et al., 2012; Point et al., 2012a; Point et al., 2012b; Point et al., 2013). Regarding the latter CyC inhibitors, modulation of their lipophilicity significantly attenuate or increase the affinity of one inhibitor towards a specific enzyme (Point et al., 2012a; Point et al., 2013). As a consequence, the obtained derivatives have not only proved to be powerful mycobacterial enzyme inhibitors (Point et al., 2012a; Point et al., 2013), but above all they had lost their activity towards mammalian enzymes initially targeted by natural molecules (Point et al., 2012a; Martin et al., 2015; Vasilieva et al., 2015). These results set up a major premise to explore the potential antitubercular properties of the CyCs. In this first article, the 26 CyC analogs have been evaluated for their anti-mycobacterial activity on a M. tb H37Rv-GFP strain, using a high-content screening assay based on the fluorescence measurement of GFP-expressing bacteria (Astarie-Dequeker et al., 2009; Christophe et al., 2010). In vitro growth of M. tb H37Rv-GFP was monitored by directly measuring GFP in presence of increasing concentrations of candidate inhibitors. Intracellular growth of M. tb H37Rv-GFP was also assessed following a 5-day exposure of infected Raw264.7 murine macrophage cell line to the different compounds. In the latter case, the percentage of infected cells and the number of living host cells allowed to determine the values of both MIC50 (compound concentration leading to 50% growth inhibition) and CC50 (compound concentration leading to 50% host cell toxicity) (Christophe et al., 2009; Flipo et al., 2011; Neres et al., 2015).The purpose is to elucidate which compounds can be active towards intracellular or extracellular bacilli. Among all tested molecules, 8 potential CyC candidates exhibited very promising anti- tubercular activities with moderate (16-40 µM) to good (1.7 µM) or to excellent (500 nM for

CyC17) MIC50 values. Beside their activity against bacterial growth inhibition, significantly, these inhibitors exhibited very low toxicity towards host macrophages (CC50 > 100 µM). Interestingly, our data showed that these compounds can be divided based on two types of antibacterial activity. An important proportion of the compounds has higher activity against intracellular bacteria than against extracellular ones: e.g., CyC7β and CyC8β with extracellular

56 Results – Article 1

MIC50 of 16.6 and >100, respectively, vs. intracellular MIC50 values of 3.1 and 11.7, respectively. This means that the intracellular mode of action of these compounds may be different than the one killing bacteria extracellularly. Alternatively, this could suggest that the physiological role of the bacterial target(s) of these CyCs is more critical for intracellular than for extracellular bacteria, or in other words, the vulnerability of the target(s) of these compounds is higher during the intracellular life of M. tb. The second type of compounds shows some activity against extracellular bacteria and poor or no activity against intracellular TB. This kind of property has already been observed and was correlated to limited bioavailability and to the hydrophilicity of the compounds (Flipo et al., 2011). Other explanation could be the targets of these molecules were less critical for bacteria when they are in macrophages.

Scheme 1. (A) Activity of CyC7 and CyC17 against GFP-labelled M. tb replicating in culture medium expressed as normalized relative fluorescence units (RFU%). (B) Activity of CyC7 against M. tb replicating inside murine macrophages Raw264.7.

CyC17, which exhibited the best antitubercular activity on extracellular M. tb growth, was selected for competitive labelling/enrichment assay against the activity-based probe Desthiobiotin-FP. (Tallman et al., 2016) Using M. tb mc26230 strain, an avirulent M. tb strain (RD1 knockout and pantothenate auxotroph (Sambandamurthy et al., 2006)) and an activity based protein profiling (ABPP) approach, total cell lysate were prepared and labelled with

Desthiobiotin in absence or presence of various amount of CyC17. The comparative proteomic analysis between the control sample (i.e., total lysate treated with Desthiobiotin-FP probe) and the CyC17-pretreated sample (i.e., total lysate pre-incubated with CyC17 prior to Desthiobiotin- FP treatment) resulted in mass spectrometry identification of 23 potential target enzymes. As expected, all identified proteins were serine or cysteine enzymes. Most of them (18 enzymes) are involved in M. tb lipid metabolism and in cell wall biosynthesis; 4 of them, namely Ag85A

57 Results – Article 1

(Belisle et al., 1997; Backus et al., 2014), TesA (Rao and Ranganathan, 2004; Waddell et al., 2005; Alibaud et al., 2011), CaeA (Rengarajan et al., 2005; Lun and Bishai, 2007) and HsaD (Lack et al., 2010), have been reported as essential for in vitro growth of M. tb and/or its survival within macrophages (Sassetti et al., 2003). 2 Drug susceptibility testing of CyC17 against M. tb mc 6230 overexpression strains of six enzymes (i.e., Ag85A, Ag85C, Rv0183, LipH, CaeA and HsaD) chosen as representative candidates for their importance during the bacterial life cycle, was performed and compared to wild type strain. As a result, only overexpression of LipH and HsaD led to an increased resistance level to CyC17 compared to wild type strain. Finally, in silico molecular docking was conducted on HsaD available crystallographic structure with CyC17 in order to predict the position of this molecule inside the active pocket of the enzyme and thus explained its good inhibitory activity.

All these findings strengthen the assumption that CyC17, and certainly all other selected CyC compounds, can be considered as two novel classes of multi-target inhibitors, therefore impairing the activity of various enzymes involved in several important processes of M. tb pathogenic life cycle without any cytotoxicity towards host cells. By blocking extracellular and/or intracellular M. tb growth, we anticipate these compounds could prevent the entry of M. tb in the persistence phase and/or reactivation of dormant bacilli residing within the granuloma and the foamy macrophages. The dual activity of the CyC inhibitors is of major importance as it may affect the different stages of the infection process. Because lipid storage in bacteria is thought to drive the infection process, CyC inhibitors can also be viewed as attractive candidates to further dissect the fate of the bacteria in the context of infected foamy macrophages.

58 www.nature.com/scientificreports

OPEN Cyclipostins and Cyclophostin analogs as promising compounds in the fght against tuberculosis Received: 5 June 2017 Phuong Chi Nguyen1, Vincent Delorme4,7, Anaïs Bénarouche1, Benjamin P. Martin2, Accepted: 30 August 2017 Rishi Paudel2, Giri R. Gnawali2, Abdeldjalil Madani1, Rémy Puppo3, Valérie Landry4, Published: xx xx xxxx Laurent Kremer 5,6, Priscille Brodin 4, Christopher D. Spilling2, Jean-François Cavalier 1 & Stéphane Canaan 1

A new class of Cyclophostin and Cyclipostins (CyC) analogs have been investigated against Mycobacterium tuberculosis H37Rv (M. tb) grown either in broth medium or inside macrophages. Our compounds displayed a diversity of action by acting either on extracellular M. tb bacterial growth only, or both intracellularly on infected macrophages as well as extracellularly on bacterial growth with very

low toxicity towards host macrophages. Among the eight potential CyCs identifed, CyC17 exhibited the best extracellular antitubercular activity (MIC50 = 500 nM). This compound was selected and further used in a competitive labelling/enrichment assay against the activity-based probe Desthiobiotin-FP in order to identify its putative target(s). This approach, combined with mass spectrometry, identifed 23 potential candidates, most of them being serine or cysteine enzymes involved in M. tb lipid metabolism and/or in cell wall biosynthesis. Among them, Ag85A, CaeA and HsaD, have previously been reported as essential for in vitro growth of M. tb and/or survival and persistence in macrophages. Overall, our

fndings support the assumption that CyC17 may thus represent a novel class of multi-target inhibitor leading to the arrest of M. tb growth through a cumulative inhibition of a large number of Ser- and Cys- containing enzymes participating in important physiological processes.

Mycobacterium tuberculosis (M. tb) the causative agent of tuberculosis (TB) has become the number one global public health emergency worldwide. With 10.4 million new cases and 1.8 million deaths caused by M. tb, as reported by WHO in 20161, TB is now the deadliest infectious disease around the world and remains a great challenge, especially in sub Saharan Africa, Russia and Eastern Europe. Te emergence of multiple drug-resistant (MDR), extensively drug-resistant (XDR) and totally drug-resistant (TDR)2,3 strains over the years and the con- troversial results of the Gates-backed TB vaccine (MV85A)4 highlight the pressing need for novel therapeutic approaches5,6. Te key feature in the success of M. tb as a pathogen is its ability to evade host immunity and to establish a chronic and persistent infection7. Several unusual characteristics contribute to this success, the frst one being its unique lipid-rich cell wall8. Indeed, the mycobacterial waxy coat, essential for bacterial viability and patho- genicity, possesses unique features. Te complex architecture and impermeability of the cell wall are responsible for the inherent resistance of M. tb to many antibiotics9. Most current available drugs including frst-line drugs such as isoniazid and ethambutol inhibit cell wall biosynthetic enzymes5. Te same comment remains true for new antituberculosis/antibiotics currently evaluated in clinical phase II or III trials, comprising either repur- posed drug or new analogues of known anti-mycobacterial drugs6,10. A posteriori, such target-specifcity may

1i-arseie ni I I 3479 arseie rance. Department of emistr an iocemistr niersit of issourit. ouis One niersit ouear t. ouis issouri 311 nite tates. 3Aix Marseille ni Institut e icroiooie e a iterrane 3479 ate-forme arseie rotomiue a arseie rance. 4I 1019 - 804 Institut asteur e ie niersit e ie 1 rue u rofesseur amette ie rance. 5Institut e ecerce en Infectiooie e ontpeier II 9004 niersit e ontpeier ontpeier rance. II I 3493 ontpeier rance. 7resent aress: uercuosis esearc aorator Institut asteur orea eonnam-si eoni-o 13488 epuic of orea. orresponence an reuests for materias sou e aresse to .-.. emai: fcaaierimm.cnrs.fr or .. emai: canaan imm.cnrs.fr

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 1 www.nature.com/scientificreports/

not address sufciently nor efciently the global spreading of the disease. Following this point of view, in 2013, Zumla et al. stated that “there is growing awareness of the need for drugs that can kill M. tuberculosis in its diferent physiological states”10. Another important issue resides in the fact that current treatments consist in a quadritherapy for 2 months, which has to be extended with a 4- to 7-months bitherapy to prevent latent TB infections (i.e., persisting bacilli) from turning into active TB disease5. Te inherent difculty to be compliant to such long treatments is in part responsible for the emergence of resistant strains and represents a new challenge to achieve control of the disease. In this context, continuous eforts for developing innovative chemotherapeutic approaches to treat TB are needed. Analogues of natural Cyclophostin (CyC1) and Cyclipostins (e.g., natural Cyclipostins P: CyC18(β)) (CyC compounds - Fig. 1) appear as prime candidates to be tested against M. tb. Tese natural compounds, isolated from fermentation of Streptomyces sp.11,12, have been reported to inhibit growth of various mycobacteria such as Mycobacterium smegmatis, Mycobacterium phlei, Nocardia abcessus as well as Corynebacterium diphteriae with similar minimum inhibitory concentrations (MIC) than those of rifampicin and penicillin G13. From a chemical point of view, Cyclipostins family members possess a bicyclic enol-organophosphorus core structure similar to that of Cyclophostin, but are phosphate esters of long chain lipophilic alcohols (Fig. 1A). These natural compounds were also shown to be potent inhibitors of either acetylcholinesterase (i.e., Cyclophostin)11,14 or human hormone-sensitive lipase (i.e., Cyclipostins)12,15. 15 We have previously reported the total synthesis of natural Cyclophostin (CyC1) and Cyclipostins P (CyC18) and their respective biological activity against purifed lipolytic enzymes. Similar studies were conducted with 14,16–19 their phosphate (CyC16-17) and phosphonate (CyC2-15) analogs (Fig. 1B). Tese studies led to the conclusion that, upon nucleophilic attack by a catalytic serine or cysteine residue, a covalent bond is formed between the enol-phosphorous atom and the catalytic residue as depicted in Fig. 1C 16,17. Moreover, modulation of the lipophilicity by varying the nature and chain length of the alkyl group either at the C-5 carbon atom (i.e., R2 group – Fig. 1) or at the phosphorous center (i.e., R1 group – Fig. 1), strongly impacted the inhibitory efciency of the CyC and could be exploited further to either decrease or increase the afnity of one inhibitor to target a specifc enzyme over others17. Consequently, these CyC analogs have not only proved to be powerful mycobacterial enzyme inhibitors; but above all they had lost their inhibitory activity on acetylcholinesterase and human hormone-sensitive lipase, which correspond to the mammalian enzymes initially targeted by the natural CyC compounds17–19. Tis promises a great potential for these cyclic enolphosph(on)ate analogs of Cyclophostin (and the Cyclipostins) as a new class of selective serine/cysteine enzyme inhibitors in mycobacteria. Te selectivity of the CyC derivatives to inhibit the mycobacterial but not the human enzymes, is therefore highly valuable and prompted us to consider these compounds as potential antitubercular agents. Herein, each CyC molecule has been tested against M. tb for i) its capacity to inhibit in vitro growth; ii) its antitubercular activity on M. tb-infected macrophages, and iii) its eventual cytotoxicity towards macrophages. Unexpectedly, whereas few analogs were found to inhibit M.tb growth in vitro and in macrophages similarly to isoniazid, they all showed absence of toxicity in mammalian cells. Importantly, potential targets of CyC17, the most potent inhibitor, were identifed via an activity-based protein profling (ABPP) approach, and further vali- dated by the constructions of overexpressing mycobacterial strains. Results Synthesis of CyC analogs. To further complete the already available library of 26 CyC compounds (i.e., 14,15,17–19 CyC1-12, 14-18) and to signifcantly improve the lipophilicity, CyC13 was synthesized by introducing simul- 1 2 taneously a C16-side alkyl chain (i.e., R group) and a C10-side alkyl chain at the C-5 carbon atom (i.e., R group), leading to an hybrid compound between CyC7 and CyC11 (Fig. 1).

Antitubercular activity and toxicity of the CyC compounds. Te set of 27 CyC analogs were frst evaluated for their antitubercular activity in a high-content screening assay based on H37Rv-GFP reporter strain20. In vitro growth of M. tb H37Rv-GFP was monitored by directly measuring fuorescence emission afer 5 days at 37 °C in the presence of increasing drug concentrations. Intracellular growth of M. tb H37Rv-GFP was also assessed following a 5-day exposure of infected Raw264.7 murine macrophages to the diferent compounds. In the latter case, the percent of infected cells and the number of living host cells allowed to simultaneously determine the MIC50 (concentration leading to 50% growth inhibition) and the CC50 (concentration leading to 50% host cell toxicity) as reported earlier20,21. Among the 27 analogs, eight potential candidates exhibited very promising antitubercular activities (Table 1 and Fig. 2). Interestingly, CyC7(β) and CyC8(α) exhibited mod- erate (16–40 µM) and good (3–4 µM) activity against extracellular and intramacrophagic M. tb, respectively. In contrast, CyC6(β), CyC7(α) and CyC8(β) appeared to be active only on infected macrophages; whereas CyC17 and Cyclipostins P, i.e. CyC18(α) and CyC18(β), impaired selectively M. tb growth in culture broth medium with MIC50 up to the nanomolar range (MIC50 ≅ 500 nM for CyC17). More particularly, both (α) and (β) isomers of CyC7 as well as CyC8(α) were found to exhibit similar or higher MIC50 values towards intramacrophagic bacilli than the frst line antibiotics used as references (Table 1). Beside antibacterial activity, signifcantly, all the latter inhibitors displayed very low toxicity towards host mac- rophages, with cytotoxic concentration (CC50) >100 µM, similarly to isoniazid (CC50 > 150 µM) and ethionamide (CC50 ≥ 120 µM), two potent antitubercular agents. Regarding the newly synthesized analog, varying at the same time both R1 and R2 alkyl side chains did not yield any signifcant antibacterial activity of the resulting CyC13 compared to the parent CyC7 compound.

Targets identifcation - Activity-based protein profling (ABPP) approach. One of the major hur- dles in drug development resides in the identifcation of the target(s) of small molecules selected from whole cell screens. Te abovementioned results with the CyC analogs, acting either against extracellular and/or intracellular

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 2 www.nature.com/scientificreports/

Figure 1. Chemical structure of CyC compounds. Structure of (A) natural Cyclophostin (CyC1), Cyclipostins P (CyC18(β)) and its trans diastereoisomer (CyC18(α)); as well as (B) the related enolphosphorus analogues: Cyclophostin phosphonate analogs (CyC2); monocyclic enolphosphorus analogs to either Cyclophostin (CyC3- 10;15-16) or Cyclipostins (CyC11-14;17). CyC5-10 and CyC13 were best described by the relationship between the OMe on phosphorus and the H-substituent on the C-5 carbon atom as being either in a trans (α-isomer) or cis (β-isomer) relationship. (C) Mode of action of CyC analogs. All CyC compounds are able to form a covalent adduct with the nucleophilic serine or cysteine catalytic residues present at the active site of α/β-hydrolase enzymes family.

mycobacteria, suggest the possibility of several mechanism of action. Tis would also imply that multiple enzymes may be targeted by these compounds, resulting in the inhibition of bacteria growth. Tis prompted us to apply an Activity-based protein profling (ABPP) approach22 for targets identifcation. Te so-called activity-based probes (ABPs), following labelling and enrichment procedures, allows to isolate

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 3 www.nature.com/scientificreports/

Extracellular Intracellular growth macrophage growthb

Compounds MIC50 (µM) MIC50 (µM) CC50 (µM) Isoniazid (INH)c 1.2 1.2 >150 Ethionamide (ETO)c 6.0 6.0 120 Rifampicin (RIF)c 0.01 2.9 24

CyC6(β) No efect 12.6 >100

CyC7(α) 92.6 4.5 >100

CyC7(β) 16.6 3.1 >100

CyC8(α) 40.4 4.0 >100

CyC8(β) >100 11.7 >20

CyC17 0.50 No efect >100

CyC18(α) 24.4 No efect >100

CyC18(β) 1.7 No efect >100

Table 1. Antibacterial activities of the most active CyC analogsa. aExperiments were performed as described in Materials and Methods. MIC50: compound minimal concentration leading to 50% growth inhibition. CC50: compound concentration leading to 50% host cell toxicity. Te best MIC50 obtained are highlighted in bold. Values are means of three independent assays performed in triplicate (CV% < 5%). bRaw264.7 macrophages were infected by M. tb H37Rv-GFP at a MOI of 2. cData from20.

selective sets of low-copy-number enzymes in complex proteomic mixtures through the chemical recognition of a specifc catalytic mechanism without interference from the more represented proteins22. ABPs typically contain (i) a reactive group which forms a covalent and irreversible adduct with the target; (ii) a linker region that allows to control the specifcity of the probe; and (iii) a tag for visualization (fuorescent tag)23 and/or enrichment and isolation24 of the covalently labelled proteins. Considering the structure and mode of action of the 8 selected CyC analogs on catalytic serine or cysteine active residues (Fig. 1C), chemically relevant fuorophosphonate (FP) ABPs, bearing either a fuorophore (i.e., rhodamine for TAMRA-FP) or a biotin (i.e., Desthiobiotin-FP) reporter tag, were selected (Figure S3)25,26. Due to their mechanism of action leading to irreversible enzyme inhibition (Figure S3), such ABPs have been exploited to screen for reversible and irreversible inhibitors of drug targets27–30. Here, compound CyC17, exhibiting the best antitubercular activity on extracellular M. tb growth, was selected for competitive probe labelling/enrichment assay by Desthiobiotin-FP using crude lysates of M. tb mc26230 (Fig. 3A–C). In parallel, TAMRA-FP labelling (Fig. 3D) was used to reveal most, if not all, serine/cysteine enzymes present in the lysate, presumably reacting with CyC17. Ten distinct bands unraveled by TAMRA-FP labelling were clearly visible in the fuorescence readout (Fig. 3E – lane D) and could also be detected in Coomassie blue staining after capture/enrichment of total lysate by Desthiobiotin-FP (Fig. 3E – lane B). In contrast, pre-treatment with CyC17 (Fig. 3A) resulted in a decrease in intensity of bands 1, 2 and 8; or disapearance of bands 3-7 and 9. (Fig. 3E – lane A). Indeed, the enzymes previ- ously inactivated by CyC17 inhibitor will thus be unable to further react with the FP-ABP. Proteins corresponding to bands 1-9 were then excised from the gel, digested with trypsin and the resulting peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) for subsequent protein identifcation. To overcome the potential overlap of proteins, the proteins that were also present at the same position in the control experiment (i.e., lane D: DMSO alone for unspecifc binding to streptavidin-magnetic beads) have not been taken into account, therefore leading to 23 distinct protein candidates (Table 2). Each protein was assigned on the basis of the numbers of unique peptides, the total number of identifed peptide spectra matched, and the correspond- ing molecular weight (Table S1). As expected from previous ABPP studies on M. tb proteome29,30, the FP probe recognized a wide range of serine and cysteine enzymes. Here, the identifed enzyme candidates ranged in their functional category from

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 4 www.nature.com/scientificreports/

Figure 2. In vitro and ex vivo dose-response activity of the CyC analogs against M. tb H37Rv. (A) Activity of CyC7(α), CyC7(β), CyC17, and CyC18(β) against GFP-expressing M. tb replicating in broth medium, expressed as normalized relative fuorescence units (RFU%). (B) Activity of CyC7(α) and CyC7(β) against M. tb replicating inside Raw264.7 macrophages. Results are expressed as the percentage of infected macrophages afer 5 days post-infection. For each concentration, data are means ± SD of at least two independent assays performed in duplicate. Te MIC50 of CyC17, CyC18(β), CyC7(β) and CyC7(α) replicating in culture broth medium were 0.5 µM, 1.7 µM, 16.6 µM and 92.6 µM, respectively. Te MIC50 of CyC7(α) and CyC7(β) replicating inside macrophages were 4.5 µM and 3.1 µM, respectively. Values are means ± SD of three independent assays performed in triplicate (CV% < 5%).

intermediary metabolism/respiration (13 proteins), lipid metabolism (5 proteins), cell wall/cell processes (4 pro- teins), and virulence/detoxifcation/adaptation (1 protein) (Table 2). Enzymes involved in metabolic processes included the alcohol dehydrogenase AdhB (Rv0761c) thought to catalyze the reversible oxidation of ethanol to acetaldehyde with the concomitant reduction of NAD; the putative L-lactate dehydrogenase LidD2 (Rv1872c)31; the methyltransferase SerA1 (Rv2996c) involved in the L-serine bio- synthetic process32; glyA1 (Rv1093) annotated as a serine hydroxymethyltransferase with possible role in serine to glycine conversion33; and UmaA (Rv0469) a S-adenosyl-L-methionine-dependent methyltransferase capable of catalyzing the conversion of phospholipid-linked oleic acid to essential tuberculostearic acid34, a major constit- uent of mycobacterial membrane phospholipids. Little is known about the catalytic reactions of these enzymes. However, our results, in line with previous fndings using fuorophosphonate ABPs29,30, suggest the presence of at least one nucleophilic (catalytic?) serine or cysteine residue involved in the formation of a covalent adduct with CyC inhibitors. Te remaining 18 enzymes belong to the serine/cysteine hydrolase family proteins. Among them, a few hydro- lases were identifed: two putative β-lactamases Rv1730c (currently annotated as a possible penicillin-binding protein) and Rv1367c, both possibly involved in cell wall biosynthesis; two amidases AmiC (Rv2888c) and AmiB2 (Rv1263); and BpoC a possible peroxidase (Rv0554)35 recently proposed as being a functional serine hydrolase30. Five members of the lipase family Lip (LipE, LipH, LipM, LipN36, and LipV) were detected; a number signif- icantly lower than the 13 active M. tb Lip enzymes reported using Desthiobiotin-FP29 or the 8 lipases using an alkyne-PEG-FP probe30. Among the fve captured Lip proteins, LipH (Rv1399c)37 and LipV (Rv3203)38 had been functionally characterized previously. LipH is known to hydrolyze short-chain ester and may participate in the detoxifcation pathway of the intracellular lipid metabolism while LipV posseses a broad range substrate specifcity and is also active at low pH, suggesting a role in M. tb’s adaptation to acidic conditions into the pha- gosome. Beside members of the Lip family, six additional enzymes with lipolytic activity were isolated: Rv0183, a

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 5 www.nature.com/scientificreports/

Figure 3. Activity based protein profling (ABPP) workfow for the identifcation of the proteins covalently 2 bound to CyC17 inhibitor. Cell lysates of M. tb mc 6230 were either (A) pre-treated with CyC17 prior to incubation with Desthiobiotin-FP probe or (B) incubated with Desthiobiotin-FP alone. Both samples were further treated with streptavidin-magnetic beads for the capture and enrichment of labelled proteins. (C) Uncompetitive binding assay using streptavidin-magnetic beads on cell lysate. (D) Detection of all potential serine/cysteine enzymes in total cell lysate using fuorescent TAMRA-FP probe. (E) Equal amounts of proteins obtained in A to D were separated by SDS-PAGE and visualized by Coomassie staining (right panel – lanes A–C) or in-gel fuorescence (lef panel - lane D: TAMRA detection). Enzymes whose labelling is impeded because of the presence of CyC17 in the active-site are circled in red and shown by arrowheads. Te corresponding bands were excised form the gel and subjected to triptic digestion and tandem mass spectrometry analysis. Te SDS gel presented in panel E is representative of three independent ABPP experiments.

monoacylglycerol lipase that degrades host-cell lipids39,40; Cfp21 (Rv1984c), a cutinase-like protein preferentially active against medium-chain carboxylic esters and monoacylglycerols41,42; as well as the esterase Rv0045c43 pro- posed to participate in lipid hydrolysis.

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 6 www.nature.com/scientificreports/

Rv Functional Band Protein name number kDa Essentiality Function Categoryb ref. Amidase AmiC Rv2888c 50.9 Amidase IM/R — Amidase AmiB2 Rv1263 49.1 Amidase IM/R — D-3-phosphoglycerate 1 Rv2996c 54.5 in vitro Methyltransferase IM/R 32 dehydrogenase (PGDH) SerA1 Macrophage and in Carboxylesterase A CaeA Rv2224c 55.9 Lipase/esterase CW/CP 49,51 vitro growth Penicillin-binding protein Rv1730c 55.8 β-lactamase CW/CP 30 2 Serine hydroxymethyltransferase Rv1093 46.2 in vitro Methyltransferase IM/R 33 1 (SHM1) glyA1 L-lactate dehydrogenase LldD2 Rv1872c 45.3 Dehydrogenase IM/R 31 3 Esterase LipM Rv2284 46.7 Lipase/esterase IM/R — Lipase LipE Rv3775 45.2 Lipase/esterase IM/R — hypothetical protein LH57_07490 Rv1367c 43.7 β-lactamase CW/CP — 4 Lipase/esterase LipN Rv2970c 40.1 Lipase/esterase IM/R 36 Alcohol dehydrogenase AdhB Rv0761c 39.7 Dehydrogenase IM/R — Lipase LipH Rv1399c 34.0 Lipase/esterase IM/R 37 Secreted antigen 85-A FbpA Rv3804c 35.7 in vitro Lipase/esterase LM 45,46 5 Ag85A Secreted antigen 85-C FbpC Rv0129c 36.7 Lipase/esterase LM 45,47 Ag85C Putative hydrolase Rv0045c 32.1 Lipase/esterase LM 43 Mycolic acid synthase UmaA Rv0469 33.1 Methyltransferase LM 34 6 Macrophages Hydrolase hsaD Rv3569c 32.1 and growth on Hydrolase IM/R 53,54 cholesterol Monoglyceride lipase Rv0183 30.2 Lipase/esterase IM/R 39,40 7 Tioesterase tesA Rv2928 29.1 in vitro Lipase/esterase LM 48 Lipase LipV Rv3203 27.9 Lipase/esterase IM/R 38 8 Putative non-heme Rv0554 28.4 Hydrolase V/D/A 35 bromoperoxidase BpoC 9 Cutinase Culp1 Rv1984c 21.8 Lipase/esterase CW/CP 41, 42

2 a a Table 2. CyC17 target proteins identifed in M. tb mc 6230 lysate by LC-ESI-MS/MS . Te 9 excised bands from the typical SDS-PAGE gel depicted in Fig. 3E were digested by trypsin followed by LC-MS/MS analysis. Only proteins not present in control incubations (DMSO alone for unspecifc binding to streptavidin-magnetic beads) were included in this list. Positive hits were selected as described in Materials and Methods. bIM/R: Intermediary metabolism/respiration; CW/CP: cell wall/cell processes; LM: Lipid metabolism; V/D/A: Virulence, detoxifcation, adaptation.

Five additional lipolytic enzymes appeared as highly promising target candidates of CyC17: the antigen 85 complex Ag85A (Rv3804c) and Ag85C (Rv0129c), the thioesterase TesA (Rv2928), the carboxylesterase CaeA (Rv2224c) and the hydrolase HsaD (Rv3569c); the latter two proteins being annotated as essential enzymes44. Ag85A and Ag85C express both a mycolyl transferase activity. Tey catalyze the transfer of mycolic acids from trehalose monomycolate (TMM) to produce trehalose dimycolate (TDM) and are also responsible for the covalent attachment of mycolic acids to arabinogalactan45,46. Moreover, inhibition of Ag85C was found to block TDM synthesis and to disrupt the integrity of the cell envelope47. Similarly, TesA has been found to be required for the synthesis of both phenolic glycolipids and phthiocerol dimycocerosate (PDIM). Inactivation of TesA in M. marinum was correlated with an important decrease in virulence and increase susceptibility to drugs48. CaeA (also named Hip1 for hydrolase important for pathogenesis 1) is a cell wall-associated carboxylesterase involved in cell wall biosynthesis and/or integrity49. CaeA was also found to play important roles in virulence49, multidrug-resistance50 and innate immunity51. Te absence of CaeA enhanced host innate immune responses and compromised the intracellular survival of M. tb in macrophages52. Te hydrolase HsaD was frst described as participating in cholesterol catabolism53 and then found to be essential for intramacrophage survival of M. tb51. HsaD has recently been proposed as a novel therapeutic target and awaits further developments54. Functional Validation: Overexpression of Target Proteins Leads to Reduced Susceptibility to CyC17. Genes encoding Ag85A, Ag85C, Rv0183, LipH, TesA and HsaD were cloned and overexpressed in M. tb (Table S2). Tese six genes were choosen as representative candidates for their involvement in mycobacterial lipid metabolism and/or for their importance during the bacteria life cycle. Overexpression of each individual protein was confrmed by Western blotting as compared to the WT strain (Figure S4). To examine whether these six overexpression strains were afected on their susceptibility to CyC17, MIC50 of CyC17 were determined for each strain. Whereas overexpression of Ag85A, Ag85C, Rv0183 or TesA did not show signifcant changes in MIC50 compared to the vector control and parental strain (WT) (Table 3), overexpression

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 7 www.nature.com/scientificreports/

MIC50 ratio Overexpression strains MIC50 (µM) mutant/WT M. tb mc26230 WT 0.55 ± 0.023‡,† 1.00 M. tb mc26230-empty 0.52 ± 0.010 0.95 vector M. tb mc26230-Ag85A 0.55 ± 0.014 1.00 M. tb mc26230-Ag85C 0.54 ± 0.009 0.98 M. tb mc26230-Rv0183 0.44 ± 0.013 0.80 M. tb mc26230-LipH 0.72 ± 0.020*,‡ 1.31 M. tb mc26230-TesA 0.52 ± 0.012 0.95 M. tb mc26230-HsaD 1.20 ± 0.026*,† 2.18

2 a Table 3. MIC50 of CyC17 against M. tb mc 6230 overexpression strains. Experiments were performed as described in Materials and Methods. MIC50: compound minimal concentration leading to 50% growth inhibition. Values are mean of at least two independent assays performed in triplicate (CV% < 5%). MIC50 values with a commun symbol (*,‡,†) are signifcantly diferent (p-value < 0.001; ANOVA followed by Fisher’s test).

Figure 4. In silico molecular docking experiments. (A) In silico molecular docking of CyC17 into the crystallographic structure of HsaD in a van der Waals surface representation. Hydrophobic residues (alanine, leucine, isoleucine, valine, tryptophan, tyrosine, phenylalanine, proline and methionine) are highlighted in white. (B) Superimposition of the top-scoring docking position of CyC17 (yellow) with the crystal structure of 3,5-dichloro-4-hydroxybenzoic acid (cyan) found to bind in the vicinity of the catalytic Ser114 of HsaD. Each inhibitor is in stick representation with the following atom color-code: oxygen, red; phosphorus, orange; carbon, yellow or cyan; chloride, green. Te catalytic Serine residue is colored in magenta. Structures were drawn with PyMOL Molecular Graphics System (Version 1.4, Schrödinger, LLC) using the PDB fle 5JZS54. (C) Ligplot + analyses results: 2D representation of schematic ligand-protein interactions of CyC17 in HsaD active site showing both hydrogen-bonds and hydrophobic interactions.

of either LipH or HsaD was associated with increased resistance levels to CyC17. Compared to WT strain (MIC50 = 0.55 ± 0.023 µM), overexpression of LipH caused a slight increase in the MIC50 value of around 1.3-fold (0.72 ± 0.020 µM; p-value < 0.001), while overexpression of HsaD led to a signifcant 2.2-fold increase in MIC50 value (1.20 ± 0.026 µM; p-value < 0.001).

Modelling the potential CyC17 binding site in HsaD. The increased MIC50 value of the strain overexpressing HsaD prompted us to explore the potential interactions occurring at the enzyme’s active

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 8 www.nature.com/scientificreports/

site following CyC17 binding. In silico molecular docking experiments were conducted, as described pre- viously17 using the recently reported crystal structures of HsaD bound to three different inhibitors54: 3,5-dichloro-4-hydroxybenzoic acid (PDB id: 5JZS), 3,5-dichloro-4-hydroxybenzenesulphonic acid (PDB id: 5JZ9) and 3,5-dichloro-benzenesulfonamide (PDB id: 5JZB). Te best scoring position obtained (i.e., lowest energy complex) indicated that the reactive seven-membered monocyclic enolphosphorus ring adopted a productive orientation (Fig. 4). Te reactive phosphorous atom of the inhibitor was indeed found in a position facilitating the occurrence of a reaction with the catalytic Ser114 (d[Ser-Oγ/P = O] distance <2.5 Å) and thus the formation of a covalent bond. It is also noteworthy that a high level of concordance was observed between this favorable docked conformation of CyC17 and the structure of the 3,5-dichloro-4-hydroxybenzoic acid found to bind in the vicinity of the HsaD active site (Fig. 4B). 55 Te docked CyC17-HsaD complex was then subjected to interactions analysis using Ligplot + v.1.4 (Fig. 4C). Te Ligplot + diagram schematically depicts the hydrogen bonds and hydrophobic interactions between the 114 241 269 ligand (i.e., CyC17) and the active site residues Ser -Asp -His of the protein during the binding process. Te Ligplot + analysis clearly shows that the reactive phosphorous atom is stabilized by H-bonding with Asn244 and His269 residues (Fig. 4C). Moreover, 17 hydrophobic contacts could be detected and appear critical to stabilize the inhibitor inside the HsaD active site (Fig. 4A and C). Te C16-side alkyl chain perfectly accommodate the hydrophobic pocket opposite to the catalytic Ser114 residue, and interacts with Gly44, Pro47, Asn54, Gly75, Tyr76, Leu115, Leu158, Ser162, Ser201, Tr205, Metn and Valn residues. Te seven-membered enolphosphate ring, located in a distinct pocket, is stabilized by two H-bonding with Glyn and Trp270, and interacts with Glyn, Ala49, Phe173, Met177 and Arg192 residues. From these fndings, CyC17 may thus bind to HsaD in a very similar orientation and with clear overlapping areas of interaction than the previously reported HsaD-bound inhibitors (i.e., 3,5-dichloro-4-hydroxybenzoic acid, 3,5-dichloro-4-hydroxybenzenesulphonic acid and 3,5-dichloro-benzenesulfonamide)54. Specifcally, all res- idues involved in H-bonding and hydrophobic contacts in each of the above mentioned complex structures are also present in stabilizing CyC17, therefore reinforcing the accuracy of our model. Taken together, this network of interactions presumably allows the formation of a stable and productive binding mode, and might provide a clear picture of the inhibition of HsaD by CyC17. Discussion Drug discovery developments to generate new lead compounds along with their corresponding targets and mode of action represent a major need in the “fght” against TB. Herein, we have evaluated the anti-tubercular activi- ties of a set of 27 CyC analogs (Fig. 1), that were initially designed to inhibit mycobacterial lipolytic enzymes17. It is now well established that lipolytic enzymes, involved in the host-pathogen cross-talk, play critical roles in the physiopathology of the disease and participate in the entry into a non-replicating dormant state within host granulomas and/or in dormancy escape, leading to reactivation of the disease and virulence56–58. Indeed, M. tb triggers the formation of lipid bodies (LB) inside infected macrophages, providing the cells a foamy appearance59. In foamy macrophages (FM), bacilli accumulate lipids within intracytoplasmic lipid inclusions (ILI)7, which allow the bacteria to persist in a non-replicating state. To persist inside FM, M. tb hydrolyzes host lipids triacylglycer- ols (TAG) from LB into fatty acids that are reprocessed as lipid reserves within ILI7,59. During the reactivation phase, these ILIs are hydrolyzed by M. tb and used to fuel the replication of mycobacteria during their exit from the hypoxic non-replicating state60. Terefore, fnding ways to inhibit the activity of such mycobacterial lipolytic enzymes may pave the way for discovering new modalities for TB treatment. Some known lipase inhibitors such as the oxadiazolone MmPPOX compound61, Orlistat61,62, or more recently the human lysosomal acid lipase inhibitor Lalistat28, have already been described to block M. tb growth with MICs ranging from 25-50 µM. Despite these moderate inhibition activities, a strong synergistic efect on in vitro M. tb growth was reached for the combined application of both latter inhibitors with vancomycin, resulting in a MIC drop of 16-fold for Orlistat (MIC~6 µM)62 and 4-fold for Lalistat (MIC~6–12 µM)28. In our study, among the 27 tested CyC analogs, eight showed moderate (16–40 µM), potent (3–4 µM) or very good (0.5 µM) activity as judged by their MIC50 values (Table 1). Unexpectedly, this set of 8 analogs can be divided into two classes. CyC6(α), CyC7(α,β) and CyC8(α,β) showed a clear preference against intracellularly-replicating mycobacteria. Tis supposes that the intracellular mode of action of this class of inhibitors difers from that of those acting exclusively on extracellularly-replicating bacilli. It can therefore be hypothesized that vulnerability of the corresponding target(s) of these inhibitors becomes more apparent and critical during the intracellular lifestyle of M. tb. A specifc response of the macrophage induced by the action of these compounds and leading to bacterial death cannot however be excluded. In contrast, CyC17 and CyC18(α,β) showed high activity exclusively on extracellular bacteria, a property already observed previously for 1,2,4-Oxadiazole EthR Inhibitors21 and was correlated to limited bioavailability and to the hydrophilicity of these compounds. From a structure-activity rela- tionships (SAR) perspective, some trends have emerged with respect to the efects of the CyC analogs tested. Regarding the natural Cyclophostin (CyC1) and its phosphonate derivatives CyC2-10;15-16, it is noteworthy that identifed bioactive compounds CyC7(α,β) and CyC8(α,β) bearing medium C10- and C12-side alkyl chains, respec- tively, are corresponding to the most potent and also “less” selective inhibitors of various bacterial enzymes, as compared to the other CyC analogs which were found to exhibit a greater selectivity towards pure recombinant 17 mycobacterial lipase over human counterparts . With Cyclipostin P, the potent antibacterial activity of CyC18(α,β) is in good agreement with the in vitro growth inhibition reported earlier on various mycobacteria13. Moreover, the fact that only the monocyclic enolphosphate CyC17 displays antituberculous activity when compared to the non-active enolphosphonate derivatives CyC11, CyC12 and CyC14, emphasizes the specifc need for a phosphate moiety in such heptacyclic analogs to exhibit bactericidal activity against M. tb growth in vitro. Another interesting finding of this work is related to the fact that among the 8 most active CyC tested (Table 1), only the phosphonates CyC6-8 were found active against M. tb in macrophages. It is indeed well known

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 9 www.nature.com/scientificreports/

that phosphates are susceptible to hydrolysis by alkaline phosphatases, whereas the corresponding phosphonates are stable to unwanted hydrolysis, which increases their lifetimes in vivo63. Consequently, the apparent diference between the extracellular and the intracellular modes of action may simply rely on chemical properties of the phosphate vs. phosphonate chemical groups. Based on the aforementioned results, one can assume that CyC inhibitors would profoundly alter the outcome of the infection by impairing mycobacterial growth within host cells. In addition, they may also afect the entry of bacilli into the persistence phase and/or interfere with reactiva- tion of dormant bacilli in macrophages. In contrast to other lipase inhibitors such as Orlistat or Lalistat, a major improvement of the CyC molecules resides in the fact that they may react exclusively with bacterial enzymes17–19 with no cytotoxic efects towards host macrophages. To gain access to the mechanisms of action, an ABPP approach was successfully applied allowing the iden- tifcation of mycobacterial enzymes impaired by the inhibitors during mycobacterial growth. Selective labelling and enrichment of captured enzymes using appropriate fuorophosphonate probes in combination with CyC17 resulted in the identifcation of 23 potential target enzymes (Fig. 3 and Table 2). As anticipated, all identifed proteins were serine or cysteine enzymes, thus validating the approach. All these 23 enzymes have already been identifed from ABPP experiments on M. tb lysates with non-specifc fuorophosphonate probes29,30. It is also noteworthy that the later three essential enzymes (i.e., Ag85A, CaeA and HsaD) were also captured from M. bovis BCG lysates using an Orlistat-alkyne analog and click chemistry for targets enrichment27; they were not detected, however, when a Lalistat-like probe was directly incubated with M. tb cells prior to lysis and chemical proteomics. Such fnding is in agreement with a complementary target profle exerted by each lipase inhibitor given their respective physico-chemical properties. Since the MIC of these two lipase inhibitors towards M. tb growth (around 25-50 µM) was however 50- to 100-times higher than that of CyC17 (0.5 µM), it is thus tempting to speculate that the shared preference for a specifc set of enzymes is responsible for the high growth inhibitory potency of our CyC monocyclic enolphosphate. To validate the targets of CyC17, genes encoding the identifed targets (Ag85A, Ag85C, Rv0183, LipH, HsaD or TesA) were overexpressed in M. tb mc26230. Whereas overexpression of LipH or HsaD led to slight, but statisti- cally signifcant increased resistance levels, thereby suggesting that these two lipolytic enzymes could be efective drug targets; overexpression of Rv0183, Ag85A, Ag85C or TesA did not change the susceptibility/resistance pro- fle to CyC17 (Table 3). Tis further strengthens the hypothesis that this inhibitor, and presumably the other CyC analogs, represent multi-target agents. Consequently, individual overexpression of single potential target enzyme is unlikely to generate high resistance level. Accordingly, by blocking at the same time the activities of various lipolytic enzymes, such as LipH, Rv0183 and HsaD, on the one hand; and those of TesA, Ag85A and Ag85C on the other hand CyC17 would strongly interfere with the acquisition and consumption of host cell-derived lipids by the mycobacteria, and also destabilize the cell envelope assembly. In such conditions, such a large spectrum of inhibitory efects exerted by our CyC analogs cannot be considered as a weakness if only M. tb is impacted, and on the contrary can open new avenues for the treatment of TB. Above all, this work led to the identifcation of very promising anti-TB candidates that should be able to act against bacteria in various physiological stages, thus allowing a faster sterilization. Conclusion A priority for new drug-development to efciently treat TB must be focused on the discovery of novel therapeutic targets and approaches. In this work, we evaluated the antitubercular activities of a series of Cyclipostins and Cyclophostin (CyC) analogs both in vitro, and ex vivo in infected macrophages. Tis led to the selection of a set of promising CyC candidates that are devoid to cytotoxic properties towards host cells. By targeting multiple enzymes either involved in lipid metabolism and/or in cell wall biosynthesis, these compounds are emerging as a novel class of multi-target anti-TB candidates which should open up new chemotherapeutic opportunities in the fght against TB. By blocking extracellular and/or intracellular M. tb growth, we anticipate these compounds could prevent the entry of M. tb in the persistence phase and/or reactivation of dormant bacilli residing within the granuloma and the foamy macrophages. To our knowledge, there is no other family of compounds able to target and impair replicating bacteria as well as intracellular bacteria. Te dual activity of the CyC inhibitors is of major importance as it may afect the diferent stages of the infection process. Because lipid storage in bacteria is thought to drive the infection process, CyC inhibitors can also be viewed as attractive candidates to further dissect the fate of the bacteria in the context of infected foamy macrophages. Materials and Methods 15 Synthesis of Cyclophostin and Cyclipostins molecules. Te synthesis of natural Cyclophostin CyC1 , 14 16 their phosphonate analogs CyC2(α) and CyC2(β) , the monocyclic enolphosphonates CyC3-4 and the trans-(α) 17 15 and cis-(β) diastereoisomers CyC5-10 ; as well as the trans-(α) and cis-(β) Cyclipostin P CyC18 and the corre- 17 18,19 sponding monocyclic phosphonate CyC11-12 , difuorophosphonate CyC14-15 and phosphate CyC16-17 analogs were obtained at 98% purity as described previously. Stock solutions (10 mM) in which the CyC compounds were found to be completely soluble in dimethyl sulfoxide (DMSO), were prepared prior to extracellular and intracel- lular drug susceptibility testing. Te new lipophilic enolphosphonate CyC13 was prepared via a transesterifcation reaction from racemic CyC7 19 using established techniques already reported for CyC16-17 synthesis , giving desired compound as a mixture of diastereoisomers. Briefy a solution of CyC7 (27 mg, 0.072 mmol) in 1,4-dioxane (360 µL) was added to a fask containing tetrabutylammonium iodide (TBAI; 2.7 mg, 0.0073 mmol, 0.1 equiv.) followed by hexadecyl bromide (220 µL, 0.72 mmol, 10 equiv.). Te fask was placed in an oil bath preheated to 105 °C. Afer 4.5 hours, the solution was cooled and concentrated in vacuo. Te residue was purifed by column chromatography (SiO2, 8% EtOAc in hexane) to give the oily product (38.4 mg, 91% yields) as a mixture of trans-(α) and cis-(β) diastereoisomers. Te two isomers were further separated by preparative reversed phase HPLC (C18 column, 100% MeOH) as follows.

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 10 www.nature.com/scientificreports/

Preparative HPLC Specifcations and Conditions. Manual preparative injector: Rheodyne 1700 (3725i-119) with 20 mL loop; Solvent A – MeOH; Solvent B – H2O; Varian ProStar Model 210 pumps equipped with 25 mL/min Rainin/Gilson type pump heads. Kromasil 100-10C18-2025 column; 10 µm particle diameter; 250 mm × 20 mm i.d. Spectra-Physics Spectra 100 UV detector with prep cell. LKB 2211 Superac fraction collector. 100% MeOH at a fow rate of 10 mL/min. Te HPLC data were supported by careful analysis of the1H, 13C, and particularly the 31P NMR spectra, and high resolution mass spectrometry (Figures S1–S2). −1 1 Fast eluting isomer CyC13(β) (15.8 mg). HPLC RT 38 min; IR (neat, NaCl) 2293, 2853 1718, 1651 cm ; H NMR (300 MHz, CDCl3) δ 4.19 (1 H, m), 4.06 (1 H, m), 3.75 (3 H, s), 2.90 (1 H, m), 2.17 (3 H, d, JHP = 2.1 Hz), 2.15–1.85 13 (4 H, m), 1.75–1.45 (4 H, m), 1.30 (42 H, m), 0.88 (6 H, overlapping t, JHH = 6.3 Hz); C NMR (75.4 MHz, CDCl3) δ 169.3, 155.1 (d, JCP = 9.0 Hz), 123.1 (d, JCP = 4.6 Hz), 66.5 (d, JCP = 6.5 Hz), 51.9, 37.5, 32.1 (d, JCP = 1.5 Hz), 31.3, 30.7 (d, JCP = 5.5 Hz), 29.9–29.7 (multiple overlapping peaks), 29.6 (d, JCP = 2.0 Hz), 29.4, 27.9, 25.7, 25.6 (d, 31 JCP = 7.6 Hz), 23.9, 22.8, 22.1, 21.3 (d, JCP = 2.5 Hz), 14.3; P NMR (121.4 MHz, CDCl3) δ 22.1 ppm; HRMS (FAB, + NBA, MH ) calcd for C34H66O5P: 585.4648, found 585.4664. Slow eluting isomer CyC13(α) (17.8 mg). HPLC RT −1 1 48 min; IR (neat, NaCl) 2923, 2853, 1718, 1652 cm ; H NMR (300 MHz, CDCl3) δ 4.13 (2 H, m), 3.73 (3 H, s), 2.98 (1 H, m), 2.22 (3 H, d, JHP = 1.6 Hz), 2.15–1.85 (4 H, m), 1.75–1.45 (4 H, m), 1.30 (42 H, m), 0.88 (6 H, over- 13 lapping t, JHH = 6.9 Hz); C NMR (75.4 MHz, CDCl3) δ 169.2 (d, JCP = 1.7 Hz), 156.1 (d, JCP = 7.3 Hz), 123.2 (d, JCP = 5.1 Hz), 66.1 (d, JCP = 7.0 Hz),, 52.0, 37.4, 32.1 (d, JCP = 1.1 Hz), 30.9, 30.6 (d, JCP = 6.1 Hz), 29.9–29.7 (multi- 31 ple overlapping peaks), 29.5 (d, JCP = 2.1 Hz), 29.3, 27.8, 25.7, 25.1 (d, JCP = 6.7 Hz), 23.3, 22.9, 21.6, 21.5, 14.3; P + NMR (121.4 MHz, CDCl3) δ 24.9 ppm; HRMS (FAB, NBA, MH ) calcd for C34H66O5P: 585.4648, found 585.4634.

Bacterial strains and growth conditions. For intra and extracellular assays, M. tb H37Rv expressing GFP20 was grown for 14 days in 7H9 medium (Difco) supplemented with 10% oleic acid-albumin-dextrose-catalase (OADC, BD Difco), 0.5% glycerol, 0.05% Tween 80 and 50 µg/mL hygromycin B (Euromedex). For target iden- tifcation, the experiments were conducted using M. tb mc26230 (H37Rv ∆RD1 ∆panCD) a derivative of H37Rv which contains a deletion of the RD1 region and panCD, resulting in a pan(−) phenotype64. M. tb mc26230 was grown in 7H9 medium supplemented with 10% OADC (BD Difco), 0.5% glycerol, 0.05% Tween 80 and 24 µg/mL D-panthothenate (Sigma-Aldrich). Cultures were kept at 37 °C without shaking.

Intracellular assay. Te growth of M. tb H37Rv-GFP strain in macrophages was monitored by automated fuorescence confocal microscope (Opera, Perkin-Elmer) as already described20. Briefy, bacteria were washed twice with PBS and resuspended in RPMI 1640 medium (Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Invitrogen). Murine (Raw264.7) macrophages were infected at a multiplicity of infection (MOI) of 2:1 and incubated 2 hours at 37 °C in RPMI 1640 medium containing 10% FBS. Cells were then washed, treated with 50 µg/mL amikacin (Euromedex) for 1 hour at 37 °C to kill all extra-cellular bacteria, washed again and fnally seeded in 384-well plates (5 × 105 cells/mL), containing 2-fold dilutions of compounds in DMSO. Te fnal volume of DMSO was kept under 0.3%. Plates were incubated for 5 days at 37 °C, 5% CO2. Infected cells were stained for 30 min using Syto60 dye (Invitrogen) at a fnal concentration of 5 µM before reading using fuorescence confocal microscope (20X water objective; GFP: λex 488 nm, λem 520 nm; Syto60: λex 640 nm, λem 690 nm). Sigmoidal dose-response curves were ftted using Prism sofware (sigmoidal dose-response, variable slope model). Te concentration required to inhibit 50% of M. tb intracellular growth (MIC50) was determined using ten-point dose-response curves as an average of the MIC50 of all parameters, the ratio of infected cells and the bacterial area per infected cell.

Extracellular assay. A 14 days old culture of M. tb H37Rv-GFP was washed twice with PBS and resuspended in 7H9 medium containing 10% OADC, 0.5% glycerol, 0.05% Tween 80 and 50 µg/mL hygromycin B. Bacteria were seeded in 384 well plates (7 × 105 bacteria/mL) containing 2-fold dilutions of the compounds in DMSO. Te fnal volume of DMSO was kept under 0.3%. Plates were incubated at 37 °C, 5% CO2 for 5 days. Bacterial fuores- cence levels (RFU) were recorded using a fuorescent microplate reader (Victor × 3, Perkin-Elmer). Te MIC50 of all tested compounds were determined using ten-point dose-response curves. Activity-Based Protein Profiling (ABPP) approach for target enzymes identifica- tion. Preparation of lysates for ABPP experiments. From 1 L of culture at the logarithmically growth stage 2 (OD600~1), M. tb mc 6230 cells were harvested by centrifugation at 4,000 g for 15 min. Pellets were washed 3 times with PBS containing 0.05% Tween 80. Te cell pellets were resuspended in PBS bufer at a 1:1 (w/v) ratio. Te bacterial cells were then mixed with the same volume of 0.1 mm diameter glass beads (BioSpec) and disrupted during 4 min of violent shaking using Mini-Beadbeater-96 (BioSpec). Te lysate was then centrifuged at 4 °C and at 12,500 g for 15 min to remove the cell debris and unbroken cells. Supernatants were adjusted to a concentration of 2 mg/mL of total proteins, snap frozen in liquid nitrogen and stored at −80 °C until further use.

In-gel detection of total M. tb potential target enzymes using TAMRA-FP probe. M. tb mc26230 lysates (50 µL–100 µg total proteins) were incubated with 2 µM ActivX TAMRA-FP probe (TermoFisher Scientifc) or DMSO (unlabelled control) for 90 min at room temperature and in absence of light. Te reaction was stopped by adding 5X Laemmli reducing sample bufer and boiling at 95 °C for 5 min. Te labelled proteins were further analyzed by SDS-PAGE electrophoresis (12% Bis-Tris gel) followed by fuorescent gel scanning (TAMRA: λex 557 nm, λem 583 nm) and detection using the Cy 3 flter of a ChemiDoc MP Imager (Bio-Rad). Alternatively, the gel was stained with Coomassie blue R250 staining® solution and was destained with solution of 10% ethanol and 30% acetic acid.

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 11 www.nature.com/scientificreports/

2 Identifcation of M. tb potential target enzymes of CyC17 using Desthiobiotin-FP probe. M. tb mc 6230 lysates (500 µL–1 mg total proteins) were incubated with 2 µM ActivX Desthiobiotin-FP probe (TermoFisher Scientifc) or DMSO (unlabelled control) for 90 min at 37 °C. For inhibitor studies, lysates were pre-incubated with 580 µM CyC17 at 37 °C for 90 min prior to Desthiobiotin-FP treatment. Te reaction was next stopped by adding 0.3 g of urea (10 M fnal concentration) to denature proteins completely. Unreacted probes were removed using Zeba Spin desalting column (7 K MWCO, TermoFisher Scientifc) and labelled proteins were further captured by 200 µg Nanolink streptavidin magnetic beads 0.8 µm (Solulink), according to the manufacturer’s instructions. First, 20 µL of a 10 mg/mL NanoLink streptavidin magnetic beads was transferred into a 1.5 mL Eppendorf tube. Te Wash Bufer (50 mM Tris-HCl, 150 mM NaCl, 0.05% Tween 20, pH 8.0) was then added to bring the fnal volume to 250 µL and the resulting mixture was mixed gently to resuspend and wash the beads. Te tube was placed on a magnetic stand for 2 min. and the supernatant was discarded. Te tube was removed from the magnetic stand and the beads were washed two more times with the Wash Bufer (250 µL). Each M. tb mc26230 treated-lysate sample was enriched for labelled proteins by transfer to the previously washed beads (around 200 µg). Te lysate/beads suspensions were incubated for 1 hour at room temperature with mild shaking. Te tubes were then placed on the magnetic stand for 2 min to collect the beads, and the supernatant was removed. Te beads containing bound, biotinylated proteins were washed three time carefully with the Wash Bufer, as described above, and resuspended in 25 µL PBS bufer pH 7.4 containing 50 mM free biotin. Te resulting solu- tion was mixed with 5X Laemmli reducing sample bufer, and heated at 95 °C for 5 min. Tis step allowed the recovery of the captured labelled proteins by exchanging the initially captured desthiobiotin/streptavidin complex to the greater afnity biotin/streptavidin complex. Te released proteins were resolved by SDS-PAGE at 160 V for 1 hour. Te gel was stained with Coomassie blue R250 staining solution and was destained with solution of 10% ethanol; 30% acetic acid. To check for unspe- cifc binding, a DMSO-treated lysate sample was incubated only with the streptavidin-magnetic beads in absence of Desthiobiotin-FP probe treatment, and processed as described above.

Target enzymes identifcation via mass spectrometry analyses. Peptide analysis by mass spec- trometry. Te bands of interest were frst excised from gels. Classical steps of washes (100 mM ammonium bicarbonate/acetonitrile, 50:50 v/v) were followed by reduction (10 mM dithiothreitol for 1 h. at 56 °C), alkylation (55 mM iodoacetamide for 30 min at room temperature) and digestion by a trypsin solution (10 ng/µL, Promega) containing ProteaseMAX 0.025% (w/v) (Promega) in 50 mM ammonium bicarbonate overnight at 37 °C. Tryptic peptides were extracted by 0.1% TFA in water/acetonitrile (50:50 v/v) and dried into a speed vacuum. Mass spec- trometry was performed on a Q Exactive Plus mass spectrometer (TermoFisher Scientifc, Bremen, Germany) equipped with a nanospray ion source and coupled to an Ultimate 3000 nano UPLC (Dionex, TermoScientifc, Sunnyvale, CA, USA). Dried tryptic peptides were dissolved in 2% acetonitrile/0.05% TFA in water and desalted on a C18 µ-precolumn (PepMap100, 300 µm × 5 mm, 5 µm, 100 Å, Dionex) before elution onto a C18 column (Acclaim PepMap, RSLC, 75 µm × 150 mm, 2 µm, 100 Å, Dionex). Peptides were eluted with a linear gradient from 6 to 40% of mobile phase B (20% water, 80% acetonitrile/0.1% formic acid) in A (0.1% formic acid in water) for 52 min. Peptides were detected with a workfow combining full MS (350- 1900 m/z range at 70000 resolution)/ data dependent MS/MS Top 10 (high collision dissociation, 150 –2250 m/z range).

Database searching for identifcation of CyC17 target enzymes. Mass spectra were processed using Proteome Discoverer sofware v. 2.1.0.81 (TermoFisher Scientifc) based on SequestHT algorithm. Te following param- eters were used: organism, UniProt M. tuberculosis H37Rv database (GI TaxID = 83332, v2016-08-20, 5535 entries); enzyme, trypsin; missed cleavages, 2; dynamic modifcation, Oxidation Met + 15.995 Da; static modifca- tion, Carbamidomethyl Cys + 57.021 Da; minimum length of peptides, 6 amino acids; precursor mass tolerance, 10 ppm; fragment mass tolerance, 0.02 Da. Proteins were considered as identifed by at least two unique peptides passing the high confdence flter (Relaxed Target FDR:0.05 and Maximum Delta Cn: 0.05). For more details about proteins identifcation, i.e. sequence coverage and number of identifed peptides see Table S1.

Functional validation of selected target enzymes. Construction of M. tb mc26230 strains overexpress- ing Ag85A, Ag85C, Rv0183, LipH, HsaD or TesA. ag85A (Rv3804), ag85C (Rv0129c), rv0183, lipH (Rv1399c), hsaD (Rv3569c) and tesA (Rv2928) were amplifed by PCR from M. tb H37Rv genomic DNA. Specifc primers (listed in Table S2) were used to integrate either the NdeI (for rv0183, lipH, hsaD and tesA) or the MscI (for ag85A and ag85C) restriction site at the 5′ end and BamHI at the 3′ end for all the genes. Amplicons were digested with the corresponding restriction enzymes (TermoFisher Scientifc), gel purifed using Nucleospin Gel and PCR Clean-up kit (Macherey-Nagel) and cloned into proper restriction sites of pMV261 (for ag85A, ag85C) or pVV16 in frame with a C-Terminus 6-His tag (for rv0183, lipH, hsaD and tesA), both harbouring the hsp60 promoter. Te DNA sequences of each insert were confrmed by DNA sequencing (GATC Biotech).

Preparation of competent cells. M. tb mc26230 electrocompetent cells were prepared as described previously by 65 2 Goude et al. . Briefy, 100 mL of M. tb mc 6230 cells were cultivated up to mid-log phase (i.e., OD600~0.6) and glycine was added to a fnal concentration of 0.2 M and incubated during 16 hours. Cells were harvested, washed four times with 10% glycerol solution at room temperature and fnally resuspended in 1/100 of the original vol- ume. 200 µL of competent cells were mixed with 1 µg of DNA and transferred to a 2 mm gap electroporation cuvette. A single pulse of 2.5 kV, 25 µF with resistance set at 600 Ω was provided. Culture media was immedi- ately added to the mycobacterial suspension and then incubated during 24 hours at 37 °C. Bacteria were plated on 7H10 Middlebrook agar supplemented with 10% OADC and 50 µg/mL of both kanamycin and hygromycin. Plates were incubated at 37 °C during 3 weeks. Positive transformants were further grown in liquid medium up to OD 1 and the overexpression of the recombinant proteins was checked by Western blot using either the specifc

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 12 www.nature.com/scientificreports/

monoclonal antibody Mab 32/15 kindly provided by Dr. K. Huygen directed against the M. tb Ag85 complex66, specifc rabbit polyclonal antibodies directed against Rv018339, or HisProbe HRP conjugated (TermoFisher Scientifc) for the other proteins.

Resazurin microtiter assay (REMA) for drug susceptibility. Susceptibility testing was performed in 7H9 medium supplemented with 10% OADC, 0.5% glycerol, 0.05% Tween 80, 24 µg/mL D-panthothenate and kanamycin (50 µg/mL) when needed. Assays were carried out in triplicate. MICs of each M. tb mc26230 mutant strains over- expressing Ag85A, Ag85C, Rv0183, LipH, HsaD or TesA were determined in 96-well fat-bottom Nunclon Delta Surface microplates with lid (TermoFisher Scientifc, ref. 167008) using the resazurin microtiter assay (REMA67,68). 6 Briefy, log-phase bacteria (i.e., OD600 ~ 1–1.5) were diluted to a cell density of 5 × 10 CFU/mL. Ten 100 µL of the above inoculum was added to each well containing 100 µL 7H9 medium, serial two-fold dilutions of CyC17 or controls to a fnal volume of 200 µL (fnal bacterial charge of 2.5 × 106 CFU/mL per well). Growth controls containing no inhibitor (i.e., bacteria only = B), inhibition controls containing 50 µg/mL isoniazid (Euromedex) and sterility controls (i.e., medium only = M) without inoculation were also included. Plates were incubated at 37 °C in a humidity chamber69 to prevent evaporation. Afer 10–14 days of incubation, 20 µL of a 0.020% (w/v) resazurin (Sigma-Aldrich) solution was added to each well, and the plates were incubated at 37 °C for 24 hours for color change from blue to pink or violet and for a reading of fuorescence units (FU). Fluorescence corresponding to the resazurin reduction was quantifed using a Tecan Spark 10 M multimode microplate reader (Tecan Group Ltd, France) with excitation at 530 nm and emission at 590 nm. For fuorometric MIC determinations, a back- ground subtraction was performed on all wells with a mean of M wells. Relative fuorescence unit was defne as: RFU% = (test well FU/mean FU of B wells) × 100. MIC values were determined by ftting the RFU% sigmoidal dose-response curves in Kaleidagraph 4.2 sofware (Synergy Sofware). Te lowest drug concentrations inhibiting 50% of growth were defned as the MIC50. References 1. WHO. Global tuberculosis report. http://www.who.int/tb/publications/global_report/en/ (2016). 2. Acosta, C. D., Dadu, A., Ramsay, A. & Dara, M. Drug-resistant tuberculosis in Eastern Europe: challenges and ways forward. Public Health Action 4, S3–S12, https://doi.org/10.5588/pha.14.0087 (2014). 3. Gunther, G. Multidrug-resistant and extensively drug-resistant tuberculosis: a review of current concepts and future challenges. Clin Med (Lond) 14, 279–285, https://doi.org/10.7861/clinmedicine.14-3-279 (2014). 4. Kaufmann, S. H. E., Weiner, J. & von Reyn, C. F. Novel approaches to tuberculosis vaccine development. Int J Infect Dis. 56, 263–267, https://doi.org/10.1016/j.ijid.2016.10.018 (2017). 5. Pai, M. et al. Tuberculosis. Nat. Rev. Dis. Primers 2, 16076, https://doi.org/10.1038/nrdp.2016.76 (2016). 6. Riccardi, G., Old, I. G. & Ekins, S. Raising awareness of the importance of funding for tuberculosis small-molecule research. Drug Discov. Today 22, 487–491, https://doi.org/10.1016/j.drudis.2016.09.012 (2017). 7. Santucci, P. et al. Experimental Models of Foamy Macrophages and Approaches for Dissecting the Mechanisms of Lipid Accumulation and Consumption during Dormancy and Reactivation of Tuberculosis. Front Cell Infect Microbiol. 6, 122, https://doi. org/10.3389/fcimb.2016.00122 (2016). 8. Brennan, P. J. Structure, function, and biogenesis of the cell wall of Mycobacterium tuberculosis. Tuberculosis (Edinb) 83, 91–97, https://doi.org/10.1016/S1472-9792(02)00089-6 (2003). 9. Hett, E. C. & Rubin, E. J. Bacterial growth and cell division: a mycobacterial perspective. Microbiol Mol Biol Rev 72, 126–156, https:// doi.org/10.1128/MMBR.00028-07 (2008). 10. Zumla, A., Nahid, P. & Cole, S. T. Advances in the development of new tuberculosis drugs and treatment regimens. Nat Rev Drug Discov 12, 388–404, https://doi.org/10.1038/nrd4001 (2013). 11. Kurokawa, T. et al. Cyclophostin, acetylcholinesterase inhibitor from Streptomyces lavendulae. J Antibiot (Tokyo) 46, 1315–1318, https://doi.org/10.7164/antibiotics.46.1315 (1993). 12. Vertesy, L. et al. Cyclipostins, novel hormone-sensitive lipase inhibitors from Streptomyces sp. DSM 13381. II. Isolation, structure elucidation and biological properties. J Antibiot (Tokyo) 55, 480–494, https://doi.org/10.7164/antibiotics.55.480 (2002). 13. Seibert, G., Toti, L. & Wink, J. Treating mycobacterial infections with cyclipostins. Sanof-Aventis Deutschland GmbH patent WO/2008/025449 (2008). 14. Bandyopadhyay, S., Dutta, S., Spilling, C. D., Dupureur, C. M. & Rath, N. P. Synthesis and biological evaluation of a phosphonate analog of the natural acetyl cholinesterase inhibitor cyclophostin. J Org Chem 73, 8386–8391, https://doi.org/10.1021/jo801453v (2008). 15. Malla, R. K., Bandyopadhyay, S., Spilling, C. D., Dutta, S. & Dupureur, C. M. Te frst total synthesis of (+/−)-cyclophostin and (+/−)-cyclipostin P: inhibitors of the serine hydrolases acetyl cholinesterase and hormone sensitive lipase. Org Lett 13, 3094–3097, https://doi.org/10.1021/ol200991x (2011). 16. Dutta, S., Malla, R. K., Bandyopadhyay, S., Spilling, C. D. & Dupureur, C. M. Synthesis and kinetic analysis of some phosphonate analogs of cyclophostin as inhibitors of human acetylcholinesterase. Bioorg. Med. Chem. 18, 2265–2274, https://doi.org/10.1016/j. bmc.2010.01.063 (2010). 17. Point, V. et al. Synthesis and kinetic evaluation of cyclophostin and cyclipostins phosphonate analogs as selective and potent inhibitors of microbial lipases. J Med Chem 55, 10204–10219, https://doi.org/10.1021/jm301216x (2012). 18. Martin, B. P., Vasilieva, E., Dupureur, C. M. & Spilling, C. D. Synthesis and comparison of the biological activity of monocyclic phosphonate, difuorophosphonate and phosphate analogs of the natural AChE inhibitor cyclophostin. Bioorg Med Chem 23, 7529–7534, https://doi.org/10.1016/j.bmc.2015.10.044 (2015). 19. Vasilieva, E. et al. Rat hormone sensitive lipase inhibition by cyclipostins and their analogs. Bioorg Med Chem 23, 944–952, https:// doi.org/10.1016/j.bmc.2015.01.028 (2015). 20. Christophe, T. et al. High content screening identifies decaprenyl-phosphoribose 2′ epimerase as a target for intracellular antimycobacterial inhibitors. PLoS Pathog 5, e1000645, https://doi.org/10.1371/journal.ppat.1000645 (2009). 21. Flipo, M. et al. Ethionamide boosters: synthesis, biological activity, and structure-activity relationships of a series of 1,2,4-oxadiazole EthR inhibitors. J Med Chem. 54, 2994–3010, https://doi.org/10.1021/jm200076a (2011). 22. Cravatt, B. F., Wright, A. T. & Kozarich, J. W. Activity-based protein profling: from enzyme chemistry to proteomic chemistry. Annu Rev Biochem 77, 383–414, https://doi.org/10.1146/annurev.biochem.75.101304.124125 (2008). 23. Patricelli, M. P., Giang, D. K., Stamp, L. M. & Burbaum, J. J. Direct visualization of serine hydrolase activities in complex proteomes using fuorescent active site-directed probes. Proteomics 1, 1067–1071, https://doi.org/10.1002/1615-9861(200109)1:9<1067::AID- PROT1067>3.0.CO;2-4 (2001).

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 13 www.nature.com/scientificreports/

24. Liu, Y., Patricelli, M. P. & Cravatt, B. F. Activity-based protein profling: the serine hydrolases. Proceedings of the National Academy of Sciences 96, 14694–14699, https://doi.org/10.1073/pnas.96.26.14694 (1999). 25. Kidd, D., Liu, Y. & Cravatt, B. F. Profling Serine Hydrolase Activities in Complex Proteomes. Biochemistry 40, 4005–4015, https:// doi.org/10.1021/bi002579j (2001). 26. Leung, D., Hardouin, C., Boger, D. L. & Cravatt, B. F. Discovering potent and selective reversible inhibitors of enzymes in complex proteomes. Nat Biotech 21, 687–691, https://doi.org/10.1038/nbt826 (2003). 27. Ravindran, M. S. et al. Targeting lipid esterases in mycobacteria grown under diferent physiological conditions using activity-based profling with tetrahydrolipstatin (THL). Mol Cell Proteomics 13, 435–448, https://doi.org/10.1074/mcp.M113.029942 (2014). 28. Lehmann, J. et al. Human lysosomal acid lipase inhibitor lalistat impairs Mycobacterium tuberculosis growth by targeting bacterial hydrolases. MedChemComm 7, 1797–1801, https://doi.org/10.1039/C6MD00231E (2016). 29. Tallman, K. R., Levine, S. R. & Beatty, K. E. Small Molecule Probes Reveal Esterases with Persistent Activity in Dormant and Reactivating Mycobacterium tuberculosis. ACS Infect. Dis. 2, 936–944, https://doi.org/10.1021/acsinfecdis.6b00135 (2016). 30. Ortega, C. et al. Systematic Survey of Serine Hydrolase Activity in Mycobacterium tuberculosis Defnes Changes Associated with Persistence. Cell Chem Biol 23, 290–298, https://doi.org/10.1016/j.chembiol.2016.01.003 (2016). 31. Osório, N. S. et al. Evidence for Diversifying Selection in a Set of Mycobacterium tuberculosis Genes in Response to Antibiotic- and Nonantibiotic-Related Pressure. Mol Biol Evol. 30, 1326–1336, https://doi.org/10.1093/molbev/mst038 (2013). 32. Kumar, A., Saigal, K., Malhotra, K., Sinha, K. M. & Taneja, B. Structural and Functional Characterization of Rv2966c Protein Reveals an RsmD-like Methyltransferase from Mycobacterium tuberculosis and the Role of Its N-terminal Domain in Target Recognition. Te Journal of Biological Chemistry 286, 19652–19661, https://doi.org/10.1074/jbc.M110.200428 (2011). 33. Chaturvedi, S. & Bhakuni, V. Unusual Structural, Functional, and Stability Properties of Serine Hydroxymethyltransferase from Mycobacterium tuberculosis. Te Journal of Biological Chemistry 278, 40793–40805, https://doi.org/10.1074/jbc.M306192200 (2003). 34. Meena, L. S., Chopra, P., Vishwakarma, R. A. & Singh, Y. Biochemical characterization of an S-adenosyl-l-methionine-dependent methyltransferase (Rv0469) of Mycobacterium tuberculosis. Biol. Chem. 394, 871–877, https://doi.org/10.1515/hsz-2013-0126 (2013). 35. Johnston, J. M., Jiang, M., Guo, Z. & Baker, E. N. Structural and functional analysis of Rv0554 from Mycobacterium tuberculosis: testing a putative role in menaquinone biosynthesis. Acta Crystallogr D Biol Crystallogr. 66, 909–917, https://doi.org/10.1107/ S0907444910025771 (2010). 36. Jadeja, D., Dogra, N., Arya, S., Singh, G. & Kaur, J. Characterization of LipN (Rv2970c) of Mycobacterium Tuberculosis H37Rv and its Probable Role in Xenobiotic Degradation. J Cell Biochem. 117, 390–401, https://doi.org/10.1002/jcb.25285 (2016). 37. Canaan, S. et al. Expression and characterization of the protein Rv1399c from Mycobacterium tuberculosis. A novel carboxyl esterase structurally related to the HSL family. Eur J Biochem 271, 3953–3961, https://doi.org/10.1111/j.1432-1033.2004.04335.x (2004). 38. Singh, G. et al. Characterization of an acid inducible lipase Rv3203 from Mycobacterium tuberculosis H37Rv. Mol Biol Rep 41, 285–296, https://doi.org/10.1007/s11033-013-2861-3 (2014). 39. Côtes, K. et al. Characterization of an exported monoglyceride lipase from Mycobacterium tuberculosis possibly involved in the metabolism of host cell membrane lipids. Biochem J. 408, 417–427, https://doi.org/10.1042/BJ20070745 (2007). 40. Dhouib, R., Laval, F., Carriere, F. & Dafe, M. & Canaan, S. A monoacylglycerol lipase from Mycobacterium smegmatis Involved in bacterial cell interaction. J Bacteriol 192, 4776–4785, https://doi.org/10.1128/JB.00261-10 (2010). 41. Schue, M. et al. Two cutinase-like proteins secreted by Mycobacterium tuberculosis show very diferent lipolytic activities refecting their physiological function. FASEB J 24, 1893–1903, https://doi.org/10.1096/f.09-144766 (2010). 42. Dedieu, L. & Serveau-Avesque, C. & Canaan, S. Identifcation of residues involved in substrate specifcity and cytotoxicity of two closely related cutinases from Mycobacterium tuberculosis. PLoS One 8, e66913, https://doi.org/10.1371/journal.pone.0066913 (2013). 43. Guo, J. et al. Characterization of a Novel Esterase Rv0045c from Mycobacterium tuberculosis. PLoS One 5, e13143, https://doi. org/10.1371/journal.pone.0013143 (2010). 44. Sassetti, C. M., Boyd, D. H. & Rubin, E. J. Genes required for mycobacterial growth defned by high density mutagenesis. Molecular Microbiology 48, 77–84, https://doi.org/10.1046/j.1365-2958.2003.03425.x (2003). 45. Sacchettini, J. C. & Ronning, D. R. Te mycobacterial antigens 85 complex–from structure to function: response. Trends Microbiol 8, 441, https://doi.org/10.1016/S0966-842X(00)01843-6 (2000). 46. Backus, K. M. et al. Te Tree Mycobacterium tuberculosis Antigen 85 Isoforms Have Unique Substrates and Activities Determined by Non-active Site Regions. Te Journal of Biological Chemistry 289, 25041–25053, https://doi.org/10.1074/jbc.M114.581579 (2014). 47. Warrier, T. et al. Antigen 85C inhibition restricts Mycobacterium tuberculosis growth through disruption of cord factor biosynthesis. Antimicrob Agents Chemother. 56, 1735–1743, https://doi.org/10.1128/AAC.05742-11 (2012). 48. Alibaud, L. et al. A Mycobacterium marinum TesA mutant defective for major cell wall-associated lipids is highly attenuated in Dictyostelium discoideum and zebrafish embryos. Molecular Microbiology 80, 919–934, https://doi. org/10.1111/j.1365-2958.2011.07618.x (2011). 49. Lun, S. & Bishai, W. R. Characterization of a Novel Cell Wall-anchored Protein with Carboxylesterase Activity Required for Virulence in Mycobacterium tuberculosis. Te Journal of Biological Chemistry 282, 18348–18356, https://doi.org/10.1074/jbc. M700035200 (2007). 50. Lun, S. et al. Synthetic Lethality Reveals Mechanisms of Mycobacterium tuberculosis Resistance to β-Lactams. mBio 5, e01767, https://doi.org/10.1128/mBio.01767-14 (2014). 51. Rengarajan, J., Bloom, B. R. & Rubin, E. J. Genome-wide requirements for Mycobacterium tuberculosis adaptation and survival in macrophages. Proceedings of the National Academy of Sciences 102, 8327–8332, https://doi.org/10.1073/pnas.0503272102 (2005). 52. Rengarajan, J. et al. Mycobacterium tuberculosis Rv2224c modulates innate immune responses. Proceedings of the National Academy of Sciences 105, 264–269, https://doi.org/10.1073/pnas.0710601105 (2008). 53. Lack, N. A. et al. Characterization of a carbon-carbon hydrolase from Mycobacterium tuberculosis involved in cholesterol metabolism. Te Journal of Biological Chemistry 285, 434–443, https://doi.org/10.1074/jbc.M109.058081 (2010). 54. Ryan, A. et al. Investigation of the mycobacterial enzyme HsaD as a potential novel target for anti-tubercular agents using a fragment-based drug design approach. British Journal of Pharmacology 174, 2209–2224, https://doi.org/10.1111/bph.13810 (2017). 55. Laskowski, R. A. & Swindells, M. B. LigPlot+: Multiple Ligand–Protein Interaction Diagrams for Drug Discovery. J Chem Inf Model 51, 2778–2786, https://doi.org/10.1021/ci200227u (2011). 56. Deb, C. et al. A Novel Lipase Belonging to the Hormone-sensitive Lipase Family Induced under Starvation to Utilize Stored Triacylglycerol in Mycobacterium tuberculosis. Te Journal of Biological Chemistry 281, 3866–3875, https://doi.org/10.1074/jbc. M505556200 (2006). 57. Dedieu, L., Serveau-Avesque, C. & Kremer, L. & Canaan, S. Mycobacterial lipolytic enzymes: a gold mine for tuberculosis research. Biochimie 95, 66–73, https://doi.org/10.1016/j.biochi.2012.07.008 (2013). 58. Johnson, G. Te α/β Hydrolase Fold Proteins of Mycobacterium tuberculosis, with Reference to their Contribution to Virulence. Curr Protein Pept Sci. 18, 190–210, https://doi.org/10.2174/1389203717666160729093515 (2017). 59. Peyron, P. et al. Foamy macrophages from tuberculous patients’ granulomas constitute a nutrient-rich reservoir for M. tuberculosis persistence. PLoS Pathog 4, e1000204, https://doi.org/10.1371/journal.ppat.1000204 (2008).

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 14 www.nature.com/scientificreports/

60. Low, K. L. et al. Triacylglycerol utilization is required for regrowth of in vitro hypoxic nonreplicating Mycobacterium bovis bacillus Calmette-Guerin. J Bacteriol 191, 5037–5043, https://doi.org/10.1128/JB.00530-09 (2009). 61. Delorme, V. et al. MmPPOX inhibits Mycobacterium tuberculosis lipolytic enzymes belonging to the hormone-sensitive lipase family and alters mycobacterial growth. PLoS One 7, e46493, https://doi.org/10.1371/journal.pone.0046493 (2012). 62. Rens, C. et al. Efects of lipid-lowering drugs on vancomycin susceptibility of mycobacteria. Antimicrob Agents Chemother 60, 6193–6199, https://doi.org/10.1128/AAC.00872-16 (2016). 63. Engel, R. Phosphonates as analogues of natural phosphates. Chemical Reviews 77, 349–367, https://doi.org/10.1021/cr60307a003 (1977). 64. Sambandamurthy, V. K. et al. Mycobacterium tuberculosis DeltaRD1 DeltapanCD: a safe and limited replicating mutant strain that protects immunocompetent and immunocompromised mice against experimental tuberculosis. Vaccine 24, 6309–6320, https://doi. org/10.1016/j.vaccine.2006.05.097 (2006). 65. Goude, R., Roberts, D. M., Parish, T. Electrophoresis of Mycobacteria. Mycobacteria protocols. Humana Press 117–130 (2015). 66. Kremer, L., Maughan, W. N., Wilson, R. A., Dover, L. G. & Besra, G. S. The M. tuberculosis antigen 85 complex and mycolyltransferase activity. Lett Appl Microbiol 34, 233–237, https://doi.org/10.1046/j.1472-765x.2002.01091.x (2002). 67. Palomino, J. C. et al. Resazurin microtiter assay plate: simple and inexpensive method for detection of drug resistance in Mycobacterium tuberculosis. Antimicrob Agents Chemother 46, 2720–2722, https://doi.org/10.1128/AAC.46.8.2720-2722.2002 (2002). 68. Martin, A., Camacho, M., Portaels, F. & Palomino, J. C. Resazurin Microtiter Assay Plate Testing of Mycobacterium tuberculosis Susceptibilities to Second-Line Drugs: Rapid, Simple, and Inexpensive Method. Antimicrob Agents Chemother. 47, 3616–3619, https://doi.org/10.1128/aac.47.11.3616-3619.2003 (2003). 69. Walzl, A. et al. A Simple and Cost Efcient Method to Avoid Unequal Evaporation in Cellular Screening Assays, Which Restores Cellular Metabolic Activity. Int. J. Appl. Sci. Technol. 2, 17–21 (2012). Acknowledgements Financial support for this work was provided by the CNRS, the European Community (ERC-STG INTRACELLTB n° 260901, MM4TB n°260872), the Agence Nationale de la Recherche (ANR-10-EQPX-04-01), the Feder (12001407 (D-AL) Equipex Imaginex BioMed), the Région Nord Pas de Calais (convention n° 12000080). P.C. Nguyen was supported by the PhD Training program from the University of Science and Technology of Hanoi. A. Bénarouche post-doctoral fellowship was supported by an ATER position from Aix-Marseille University. A. Madani was supported by a PhD fellowship from the Association Grégory Lemarchal and Vaincre la Mucoviscidose (projet n°RF20160501651). Dr. William R. Jacobs, Jr., is acknowledged for providing M. tb mc26230 and Kris Huygen for providing the Ag85 monoclonal antibody. Author Contributions Conceptualization, P.B., C.D.S., S.C. and J.-F.C.; Methodology, P.B., S.C. and J.-F.C.; Investigation, P.C.N., V.D., A.B., A.M., V.L. and R.P.; Resources, B.P.M., R.P., G.R.G. and C.D.S.; Writing – Original Draf, P.C.N.; Writing – Review & Editing, V.D., R.P., L.K., P.B., C.D.S., S.C. and J.-F.C.; Funding Acquisition, C.D.S., P.B. and S.C.; Supervision, P.B., C.D.S., S.C. and J.-F.C. Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-017-11843-4. Competing Interests: Te authors declare that they have no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. Te images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

© Te Author(s) 2017

Scientific RepoRts | 7: 11751 | DOI:10.1038/s41598-017-11843-4 15 Supporting Information

Cyclipostins and Cyclophostin analogs as promising compounds in the fight against tuberculosis

Phuong Chi Nguyen1, Vincent Delorme4 #, Anaïs Bénarouche1, Benjamin P. Martin2, Rishi Paudel2,

Giri R. Gnawali2, Abdeldjalil Madani1, Rémy Puppo3, Valérie Landry4, Laurent Kremer5,6, Priscille

Brodin4, Christopher D. Spilling2, Jean-François Cavalier1* and Stéphane Canaan1*

1 Aix-Marseille Univ, CNRS, EIPL, IMM FR3479, Marseille, France 2 Department of Chemistry and Biochemistry, University of Missouri−St. Louis, One University Boulevard, St. Louis, Missouri 63121, United States 3 Aix Marseille Univ, CNRS, Institut de Microbiologie de la Méditerranée FR3479, plate-forme Marseille Protéomique (MaP), Marseille, France 4 INSERM U1019 CNRS-UMR 8204, Institut Pasteur de Lille, Université de Lille, 1 rue du Professeur Calmette, Lille, France 5 Institut de Recherche en Infectiologie de Montpellier (IRIM), CNRS, UMR 9004, Université de Montpellier 6 IRIM, INSERM, 34293, Montpellier, France

* Correspondance: [email protected] (S. Canaan); [email protected] (J.-F. Cavalier)

# Current Address: Tuberculosis Research Laboratory, Institut Pasteur Korea, Seongnam-si, Gyeonggi-do, 13488 Republic of Korea.

Contents

1 13 31 H, C NMR and P NMR spectra of CyC13() (Figure S1) S2

1 13 31 H, C NMR and P NMR spectra of CyC13() (Figure S2) S5

Primers used in this study (Table S2) S8

Chemical structures of TAMRA-FP and Desthiobiotin-FP (Figure S3) S9

Western blot analysis of M. tb mc26230 overexpression strains (Figure S4) S10

S1

1 Figure S1A. H NMR spectrum of CyC13() recorded at 300 MHz, and using CDCl3 (δ = 7.27 ppm) as an internal standard of solvent.

S2

13 Figure S1B. C NMR spectrum of CyC13() recorded at 75 MHz, and using CDCl3 (δ = 77.23 ppm) as an internal standard of solvent.

S3

31 Figure S1C. P NMR spectrum of CyC13() recorded at 121 MHz and referenced to external 85% H3PO4 (0 ppm).

S4

1 Figure S2A. H NMR spectrum of CyC13() recorded at 300 MHz, and using CDCl3 (δ = 7.27 ppm) as an internal standard of solvent.

S5

13 Figure S2B. C NMR spectrum of CyC13() recorded at 75 MHz, and using CDCl3 (δ = 77.23 ppm) as an internal standard of solvent.

S6

31 Figure S2C. P NMR spectrum of CyC13() recorded at 121 MHz and referenced to external 85% H3PO4 (0 ppm).

S7

Table S2. Primers used in this study. Restriction sites if present are underlined.

Restriction Gene Primer Sequence (5′-3′) site Ag85A-F Forward 5’-CCCAGCTTGTTGACAGGGTTCGTG -3’ Ag85A-R Reverse 5’-ACCATGGATCCCTAGGCGCCCTGGGGCGCG-3’ BamHI Ag85C-F Forward 5’-CCACGTTCTTCGAACAGGTGCGAAG-3’ Ag85C-R Reverse 5’-ACCATGGATCCTCAGGCGGCCGGCGCAGCAG-3’ BamHI Rv0183-F Forward 5’-GGAAATCATATGACTACCACCCGGACTG-3’ NdeI Rv0183-R Reverse 5’-CGGCGGGATCCCCGACAACCGCTCGGTGAGCC-3’ BamHI LipH-F Forward 5’-GGAAATCATATGACAGAGCCGACCGTCG-3’ NdeI LipH-R Reverse 5’-CGGCGGGATCCCTGATGCGTGCAACGCCCTCTTC-3’ BamHI TesA-F Forward 5'-CCAGCATATGCTGGCCCGTCACGGACCACG-3' NdeI TesA-R Reverse 5'-CCAGAAGCTTAGCTCGATCATGCCATTGGAGTGTT-3' HindIII HsaD-F Forward 5'-CCAGCATATGACAGCTACCGAGGAATTGACGT-3' NdeI HsaD-R Reverse 5'-CCAGAAGCTTTCTGCCACCTCCCAGAAATTCAATC-3' HindIII

S8

Figure S3. Chemical structures of TAMRA-FP for in-gel fluorescence detection and Desthiobiotin-FP for target enrichment. Each probe forms an irreversible covalent bond with an active site serine (or cysteine) for irreversible enzyme labelling.

S9

Results – Article 2

Article 2: Cyclophostin and Cyclipostins analogs, new promising molecules to treat mycobacterial-related diseases

From the optimistic results obtained in our previous publication and showing that such Cyclophostin/Cyclipostins monocyclic analogs efficiently inhibited M. tuberculosis growing either extracellularly or within macrophages, without signs of toxicity towards the host cells, we further have evaluated the efficacy and selectivity of the CyCs against various bacteria and mycobacteria. These included 6 Gram-negative, 5 Gram-positive bacteria, 1 rapid-growing mycobacteria (M. abscessus) and 3 slow-growing mycobacteria (M. marinum, M. bovis BCG and M. tb mc26230). After a first screening assay using the agar plate method, the 26 CyCs, compounds exhibiting a 50-100% growth inhibition rate were selected to determine their MIC using the REMA assay.

Among the 26 CyCs tested, 10 were active and their inhibitory activity was exclusively restricted to mycobacteria (slow or fast growing strains). As already observed in the case of M. tb H37Rv, the best candidate identified was CyC17, which is able to efficiently impair all mycobacterial growth with good MIC values. Of interest, this inhibitor was also found active on M. abscessus. This mycobacterium represents the most frequent mycobacterial species isolated from cystic fibrosis (CF) patients with pulmonary infections (Roux et al., 2009). M.

84 Results – Article 2 abscessus is intrinsically resistant to a broad range of antibiotics including most antitubercular drugs, and is considered the most pathogenic and chemotherapy-resistant rapidly growing mycobacterium (Candido et al., 2014) being nicknamed “a new antibiotic nightmare” (Nessar et al., 2012). From these findings, CyC17 was further tested on 26 clinical strains belonging to the M. chelonae-abscessus clade. With MIC values <2 up to 40 µg/mL, this compound displayed comparable MIC to those of most classical antibiotics used to treat M. abscessus infections. All these results therefore support that fact that such CyCs represent a new family of potent and selective inhibitors against mycobacteria. This is of particular interest for future chemotherapeutic developments against mycobacterial-associated infections, especially against M. abscessus the most drug-resistant mycobacterial species.

85 ARTICLE IN PRESS

International Journal of Antimicrobial Agents ■■ (2018) ■■–■■

Contents lists available at ScienceDirect

International Journal of Antimicrobial Agents

journal homepage: www.elsevier.com/locate/ijantimicag

Short Communication Cyclophostin and Cyclipostins analogues, new promising molecules to treat mycobacterial-related diseases Phuong Chi Nguyen a,1, Abdeldjalil Madani a,1, Pierre Santucci a, Benjamin P. Martin b, Rishi R. Paudel b, Sandrine Delattre c, Jean-Louis Herrmann c,d, Christopher D. Spilling b, Laurent Kremer e,f, Stéphane Canaan a,*,2, Jean-François Cavalier a,*,2 a Aix-Marseille Univ., CNRS, LISM, IMMFR3479, Marseille, France b Department of Chemistry and Biochemistry, University of Missouri−St Louis, One University Boulevard, St Louis, MO 63121, USA c AP-HP, Hôpitaux Universitaires Ile de France Ouest, Ambroise Paré, Boulogne and Raymond Poincaré, Garches, France d 2I, UVSQ, INSERM UMR1173, Université Paris–Saclay, Versailles, France e Institut de Recherche en Infectiologie de Montpellier (IRIM), CNRS, UMR 9004, Université de Montpellier, 1919 route de Mende, 34293 Montpellier, France f IRIM, INSERM, 34293 Montpellier, France

ARTICLE INFO ABSTRACT

Article history: The progression of mycobacterial diseases requires the development of new therapeutics. This study evalu- Received 22 September 2017 ated the efficacy and selectivity of a panel of Cyclophostin and Cyclipostins analogues (CyCs) against various Accepted 2 December 2017 bacteria and mycobacteria. The activity 26 CyCs was first assayed by the agar plate method. Compounds Editor: Dr Ben Gold exhibiting 50–100% growth inhibition were then selected to determine their minimum inhibitory con- centrations (MICs) by the resazurin microtiter assay (REMA). The best drug candidate was further tested Keywords: against clinical mycobacterial isolates and bacteria responsible for nosocomial infections, including 6 Gram- Cyclipostins negative bacteria, 5 Gram-positive bacteria, 29 rapid-growing mycobacteria belonging to the Mycobacterium Cyclophostin Antimycobacterial agents chelonae–abscessus clade and 3 slow-growing mycobacteria (Mycobacterium marinum, Mycobacterium bovis Drug susceptibility BCG and Mycobacterium tuberculosis). Among the 26 CyCs tested, 10 were active and their inhibitory ac- Mycobacterium abscessus tivity was exclusively restricted to mycobacteria. The best candidate (CyC17) was further tested against 26 clinical strains and showed high selectivity for mycobacteria, with MICs (<2–40 μg/mL) comparable with those of most classical antimicrobials used to treat M. abscessus infections. Together, these results support the fact that such CyCs represent a new family of potent and selective inhibitors against myco- bacteria. This is of particular interest for future chemotherapeutic developments against mycobacterial- associated infections, especially against M. abscessus,themostdrug-resistantmycobacterialspecies. © 2017 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.

1. Introduction the presence of a complex lipid-rich cell wall [2] whose general com- position and architecture is shared both by RGM and SGM and which The Mycobacterium genus comprises more than 200 species clas- contributes to its low permeability to many antibiotics, thus lim- sified mainly based on their pathogenicity and growth rates. iting therapeutic options. In addition, the treatment duration is long, Mycobacteria can be separated into rapidly-growing mycobacte- ranging from 6–9 months for tuberculosis (TB) and up to 2 years ria (RGM), which include saprophytic species and opportunistic for several mycobacterial lung infections due to atypical mycobac- species such as the Mycobacterium abscessus complex, Mycobacte- teria such as Mycobacterium kansasii or M. abscessus.Theemergence rium chelonae and Mycobacterium fortuitum,andslow–growing of multidrug-resistant (MDR) mycobacteria such as M. abscessus or mycobacteria (SGM) comprising strict and opportunistic patho- M. tuberculosis (responsible for MDR-TB) strongly impacts treat- gens such as the Mycobacterium tuberculosis complex and the ment success rates, with an increased incidence of treatment failure Mycobacterium avium complex, respectively [1]. Treatment of my- and death [3]. Another major issue relates to the emergence of in- cobacterial infections remains challenging, essentially because of fections caused by RGM [4]. Among them, the difficult-to-manage M. abscessus complex represents one of the most drug-resistant for *Correspondingauthors.LIMSUMR7255,CNRS,LipolysisandBacterial which standardised chemotherapeutic regimens are still lacking [5]. Pathogenicity team, 31 Chemin Joseph Aiguier, Marseille 13402 Cedex 20, France. Therefore, more efficient antimycobacterial agents are needed. E-mail addresses: [email protected] (S. Canaan); [email protected] Previously, we evaluated natural Cyclophostin and Cyclipostins (J.-F. Cavalier). analogues (CyCs) for their activity against M. tuberculosis and dem- 1 These two authors contributed equally to this work. 2 Present address: LIMS UMR7255, CNRS, Lipolysis and Bacterial Pathogenicity team, onstrated that they efficiently inhibited M. tuberculosis growing either 31 Chemin Joseph Aiguier, 13402 Marseille Cedex 20, France. extracellularly or within macrophages [6].Ofmajorimportance,no https://doi.org/10.1016/j.ijantimicag.2017.12.001 0924-8579/© 2017 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.

Please cite this article in press as: Phuong Chi Nguyen, et al., Cyclophostin and Cyclipostins analogues, new promising molecules to treat mycobacterial-related diseases, Interna- tional Journal of Antimicrobial Agents (2018), doi: 10.1016/j.ijantimicag.2017.12.001 ARTICLE IN PRESS

2 P.C. Nguyen et al./International Journal of Antimicrobial Agents ■■ (2018) ■■–■■ cytotoxicity towards host mammalian cells has been observed for Table 1 these CyC compounds at concentrations up to 100 μM [6].Using Bacterial strains used in the study. competitive labelling/enrichment assays and mass spectrometry, 23 Species Bacterial strains Source putative CyC targets were identified, all belonging to the serine/ Gram-negative bacteria cysteine hydrolase family proteins known to play a role in the Escherichia coli DH10B Reference strain lifecycle and/or pathogenesis of M. tuberculosis [7]. E. coli Clinical isolate [11] In this study, the selectivity and activity of a set of 26 CyC ana- Pseudomonas aeruginosa PA01 Reference strain P. aeruginosa Clinical isolate [11] logues (Supplementary Fig. S1) against a variety of Gram-positive Enterobacter aerogenes Clinical isolate [11] and Gram-negative bacteria as well as a large panel of mycobac- Serratia marcescens Clinical isolate [11] terial clinical isolates was examined. The results strongly support Stenotrophomonas maltophilia Clinical isolate [11] the potent and specific activity of CyCs against mycobacteria. These Burkholderia cepacia complex Clinical isolate [11] molecules constitute an unexploited chemical class of active com- Gram-positive bacteria Staphylococcus aureus Clinical isolate [11] pounds with promising translational development possibilities for Enterococcus faecalis Clinical isolate [11] the treatment of mycobacterial infections. Streptococcus pneumoniae Clinical isolate [11] Staphylococcus epidermidis Clinical isolate [11] 2. Materials and methods Enterococcus faecium Clinical isolate [11] Mycobacterium spp. M. bovis BCG Pasteur Reference strain 2.1. Synthesis of Cyclipostins and Cyclophostin analogues (CyCs) M. tuberculosis mc26230 Reference strain [12] M. marinum M Reference strain The 26 CyCs were previously synthesised and obtained at 98% M. smegmatis mc2155 Reference strain Ta purity as reported previously [8–10]. Stock solutions (4 mg/mL) of M. abscessus CIP 104536 Reference strain M. abscessus (RPC95) CF isolate [11] each CyC were prepared in dimethyl sulfoxide (DMSO) prior to sus- M. abscessus (RPC96) CF isolate [11] ceptibility testing. M. abscessus (RPC98) CF isolate [11] M. abscessus (RPC101) CF isolate [11] 2.2. Bacterial strains and growth conditions M. abscessus (RPC102) CF isolate [11] M. abscessus (RPC104) CF isolate [11] M. abscessus (RPC105) CF isolate [11] A total of 6 Gram-negative bacteria, 5 Gram-positive bacteria, M. abscessus (RPC106) CF isolate [11] 29 RGM (mainly clinical isolates) and 3 SGM were included in this M. abscessus (RPC109) CF isolate [11] study (Table 1). The clinical isolates were collected during an epi- M. abscessus (RPC110) CF isolate [11] demiological study performed in 2004 [11] and include M. massiliense (RPC99) CF isolate [11] M. massiliense (RPC100) CF isolate [11] representatives of the M. chelonae–abscessus clade (10 M. abscessus, M. massiliense (RPC107) CF isolate [11] 4 Mycobacterium massiliense,2Mycobacterium bolletii and 10 M. M. massiliense (RPC103) CF isolate [11] chelonae strains) obtained from either cystic fibrosis patients or non- M. bolletii (RPC97) CF isolate [11] cystic fibrosis patients. Mycobacterium smegmatis mc2155 strain was M. bolletii (RPC108) CF isolate [11] M. chelonae (RPC128) CF isolate [11] routinely grown in Middlebrook 7H9 broth (BD Difco, Le Pont-de- M. chelonae (RPC129) CF isolate [11] Claix, France) supplemented with 0.2% glycerol and 0.05% Tween M. chelonae (RPC130) CF isolate [11] 80 (Sigma-Aldrich, Saint-Quentin Fallavier, France) (7H9-S). Myco- M. chelonae (RPC131) Non-CF isolate [11] bacterium marinum ATCC BAA-535/M, Mycobacterium bovis BCG M. chelonae (RPC132) Non-CF isolate [11] Pasteur, M. abscessus CIP 104536T with either a smooth (S) or rough M. chelonae (RPC057) Non-CF isolate [11] 2 M. chelonae (RPC059) Non-CF isolate [11] (R) morphotype, and M. tuberculosis mc 6230 (H37Rv ΔRD1 ΔpanCD M. chelonae (RPC061) Non-CF isolate [11] [12])strainsweregrownin7H9-Smediumsupplementedwith10% M. chelonae (RPC063) Non-CF isolate [11] oleic acid–albumin–dextrose–catalase (OADC enrichment) (BD Difco) M. chelonae (RPC066) Non-CF isolate [11] OADC 2 (7H9-S ). In the case of M. tuberculosis mc 6230, 24 μg/mL CF, cystic fibrosis. d-pantothenate (Sigma-Aldrich) was also added to the 7H9-SOADC a Smooth (S) or rough (R) morphotypes. medium. All cultures were kept at 37 °C without shaking, except M. marinum which was grown at 32 °C. The five Gram-positive bac- screening plate contained negative (DMSO) and positive (50 μMan- terial strains (Staphylococcus aureus, Enterococcus faecalis, tibiotic) controls as well as one well for sterility control (i.e. medium Streptococcus pneumoniae, Staphylococcus epidermidis and Entero- alone). For the 100% inhibition control, 50 μM kanamycin (Sigma- coccus faecium)andthesixGram-negativebacterialstrains Aldrich) was used for M. marinum, M. abscessus, M. smegmatis, M. (Escherichia coli, Pseudomonas aeruginosa, Enterobacter aerogenes, Ser- bovis BCG, M. tuberculosis and E. coli and 50 μMcarbenicillin(Sigma- ratia marcescens, Stenotrophomonas maltophilia and Burkholderia Aldrich) was used for P. aeruginosa.Eachwellwasspottedwith10μL cepacia complex) were grown at 37 °C in LB Broth Base medium of a bacterial culture at 5 × 105 cells/mL. The incubation time varied (Thermo Fisher Scientific, Illkirch, France). from 1 day to 2 weeks depending on the strain tested. The CyC com- pounds leading to a minimum of 50% growth inhibition were selected 2.3. Drug susceptibility testing on solid medium for subsequent minimum inhibitory concentrations (MIC) deter- mination using the resazurin microtiter assay (REMA). Drug susceptibility testing was performed in 24-well suspen- sion culture plates (Greiner Bio-One, Courtaboeuf, France) as 2.4. Resazurin microtiter assay (REMA) for minimum inhibitory described previously [13]. Escherichia coli and P. aeruginosa were concentration determination grown on LB agar medium at 37 °C. All mycobacteria were grown at either 32 °C (M. marinum) or 37 °C (M. smegmatis, M. bovis BCG, Susceptibility testing was performed using the Middlebrook 7H9 M. abscessus SandRvariants,andM. tuberculosis mc26230) on broth microdilution method. All assays for each strain were carried Middlebrook 7H10 agar (BD Difco) supplemented with 10% OADC out at least in triplicate. MICs of the CyCs, selected from solid medium and 24 μg/mL d-pantothenate (M. tuberculosis mc26230). The wells screening against the various bacterial strains, were determined in were filled with 1 mL of the appropriate medium containing each 96-well flat-bottom NunclonTM Delta surface microplates with lid of the CyC analogues at a single 30 μMfinalconcentration.Each (Thermo Fisher Scientific; ref. 167008) using the REMA assay [14,15].

Please cite this article in press as: Phuong Chi Nguyen, et al., Cyclophostin and Cyclipostins analogues, new promising molecules to treat mycobacterial-related diseases, Interna- tional Journal of Antimicrobial Agents (2018), doi: 10.1016/j.ijantimicag.2017.12.001 ARTICLE IN PRESS

P.C. Nguyen et al./International Journal of Antimicrobial Agents ■■ (2018) ■■–■■ 3

Table 2 Antibacterial activities of the selected active Cyclophostin and Cyclipostins analogues (CyCs) compared with four standard antimicrobial agents against a panel of myco- bacterial strains.

a b Compound MIC90 (μg/mL) CC50 (μg/mL)

M. smegmatis mc2155 M. abscessus CIP 104536T M. marinum M. bovis BCG M. tuberculosis mc26230

Smooth Rough

CyC7(α) 11.8 25.9 >40 CyC7(β) 2.2 17.7 13.9 >40 CyC8(α) 0.63 8.5 18.6 >40 CyC8(β) 4.2 9.3 >10 CyC9(β) 7.8 26.2 >45 CyC10(α) 11.4 >50 CyC11 9.0 >45 CyC17 0.81 6.4 0.18 0.74 0.58 1.2 >45 CyC18(α) 7.8 9.6 19.4 >45 CyC18(β) 6.7 6.0 4.9 11.2 11.5 2.6 >45 Isoniazid 5.6 9.3 0.10 0.15 >20 Amikacin 4.7 7.9 1.7 0.48 0.63 Imipenem 1.9 8.9 Cefoxitin 4.0 12.0

a Minimum inhibitory concentration leading to 90% growth inhibition (MIC90) as determined by the resazurin microtiter assay (REMA) expressed as the mean value of two independent assays performed in triplicate (CV% < 5%). b Cytotoxic concentration of compound leading to 50% cell toxicity (CC50) determined on murine (RAW 264.7) macrophages (data from [6]).

Briefly, log-phase bacteria were diluted to a cell density of 5 × 106 3.2. Minimum inhibitory concentration determination cells/mL in 7H9-SOADC (plus 24 μg/mL d-pantothenate when needed). 5 Then, 100 μLoftheaboveinoculum(5× 10 cells per well) was added The potency of the selected CyCs [CyC7(α,β), CyC8(α,β), CyC9(β), CyC10(α), OADC to each well containing 100 μLof7H9-S medium, serial two-fold CyC11, CyC17 and CyC18(α,β)]wasnextconfirmedbydeterminingtheir dilutions of the selected CyC analogue or controls to a final volume MICs towards each respective mycobacterial strain using the REMA of 200 μL. Growth controls containing no inhibitor [i.e. bacteria only assay [14,15] (Table 2). Nearly all selected CyCs were active against (B)], inhibition controls containing 50 μg/mL kanamycin and ste- M. marinum and M. bovis BCG growth with moderate (4.2–25.9 μg/ rility controls [i.e. medium only (M)] without inoculation were also mL) to good (0.6–2.2 μg/mL) MIC90 values. Moreover, the obtained included. Plates were incubated at 37 °C (32 °C for M. marinum) in MICs against M. tuberculosis mc26230 were consistent with those ahumiditychamber[16] to prevent evaporation for either 3–5 days recently reported against M. tuberculosis H37Rv extracellular growth

(M. smegmatis and M. abscessus)or10–14days(M. marinum, M. bovis [6]. Among the 10 CyCs tested, only CyC18(β) and CyC17 inhibited BCG and M. tuberculosis mc26230). Then, 20 μL of a 0.025% (w/v) the growth of all mycobacteria investigated with MICs ranging from resazurin solution was added to each well and the plates were in- 0.18–11.5 μg/mL (Table 2). Of importance, CyC17 was also highly cubated at 37 °C for colour change from blue to pink or violet and active both against R and S variants of M. abscessus with similar or for reading of fluorescence units (FU). Fluorescence corresponding lower MICs (6.4 μg/mL and 0.18 μg/mL against M. abscessus S and to resazurin reduction to its metabolite resorufin was quantified using R, respectively) than AMK, IPM and FOX used as reference drugs aTecanSpark10Mmultimodemicroplatereader(TecanGroupLtd., (Table 2) as well as most conventional antibiotics used in clinical Lyon, France) with excitation at 530 nm and emission at 590 nm. settings [17]. For fluorometric MIC determinations, a background subtraction was performed on all wells with a mean of M wells. Relative fluorescence 3.3. Potency and selectivity of the best inhibitor (CyC17)against units (RFUs) were defined as: RFU% = (test well FU/mean FU of B clinical isolates wells) × 100. MICs were determined by fitting the RFU% sigmoidal dose–response curves [15] in KaleidaGraph 4.2 software (Synergy The efficiency/selectivity of CyC17 was investigated further by Software, Reading, PA). The lowest drug concentration inhibiting 90% testing its activity towards 26 clinical isolates belonging to the M. of growth was defined as the MIC90.Isoniazid,amikacin(AMK), chelonae–abscessus clade as well as several Gram-negative and Gram- imipenem (IPM) and cefoxitin (FOX) were used as reference drugs. positive bacterial species (Table 3). As anticipated, CyC17 was only active against M. abscessus and M. chelonae isolates (Table 3). Of in-

3. Results and discussion terest, MIC50 values for M. abscessus and M. chelonae (10 μg/mL and 40 μg/mL, respectively) were comparable with those of AMK (12.5 μg/ 3.1. Screening of CyC susceptibility on solid medium mL) [19], FOX (32 μg/mL) or IPM (16 μg/mL) [18]. Mycobacterium abscessus complex isolates were also more sensitive than M. chelonae To explore the efficacy and antibacterial spectrum of the CyC ana- isolates, with M. bolletii being the most susceptible organism logues, a preliminary screen on solid medium was first performed (Table 3). with each of the 26 CyCs at a fixed 30 μMfinalconcentrationusing Overall, these results confirm the potential of CyCs as promis- aselectedpanelconsistingofsixmycobacterialspecies(M. ing antimycobacterial candidates. This selective activity against smegmatis, M. marinum, M. abscessus R and S, M. bovis BCG and M. mycobacteria might be related to the cell envelope composition [2], tuberculosis mc26230) and two Gram-negative bacteria (E. coli and which is unique, and/or to the increased ability of these hydropho- P. aeruginosa). Following a period of incubation (from 1 day to 2 bic compounds to cross this lipid-rich cell wall barrier. This weeks), 10 of the 26 CyCs were found to inhibit growth in the range hypothesis is relevant with the recent identification of potential of 50–100% relative to the positive growth control (i.e. bacteria targets for CyC17,whicharemostlyinvolvedinM. tuberculosis lipid without antibiotics). Remarkably, this effect was restricted to my- metabolism and/or in cell wall biosynthesis, such as the antigen 85 cobacteria only (Supplementary Table S1), whilst under the same complex playing a key role in mycolic acid metabolism, which is conditions the growth of E. coli and P. aeruginosa was not impacted. restricted to mycobacteria [6].

Please cite this article in press as: Phuong Chi Nguyen, et al., Cyclophostin and Cyclipostins analogues, new promising molecules to treat mycobacterial-related diseases, Interna- tional Journal of Antimicrobial Agents (2018), doi: 10.1016/j.ijantimicag.2017.12.001 ARTICLE IN PRESS

4 P.C. Nguyen et al./International Journal of Antimicrobial Agents ■■ (2018) ■■–■■

Table 3 Appendix A. Supplementary data Minimum inhibitory concentrations (MICs) of CyC17 on a panel of M. abscessus and M. chelonae clinical strains isolated from cystic fibrosis (CF) and non-CF patients. Supplementary data associated with this article can be found, Clinical isolates (no. of strains) No. of strains with MIC (μg/mL)a in the online version, at doi:10.1016/j.ijantimicag.2017.12.001. indicated MIC (in μg/mL)

<2 5 10 20 40 80 MIC50 MIC90 References Mycobacterium spp. All (26) 2 2 8 5 4 5 10 80 [1] Tortoli E. Microbiological features and clinical relevance of new species of the M. abscessus (10) 1 2 3 3 1 10 40 genus Mycobacterium. Clin Microbiol Rev 2014;27:727–52. M. massiliense (4) 2 2 10 10 [2] Brennan PJ, Nikaido H. The envelope of mycobacteria. Annu Rev Biochem M. bolletii (2) 1 1 <2 10 1995;64:29–63. M. chelonae (10) 2 2 1 5 40 80 [3] Dheda K, Gumbo T, Maartens G, Dooley KE, McNerney R, Murray M, et al. The Gram-negative bacteria epidemiology, pathogenesis, transmission, diagnosis, and management of multidrug-resistant, extensively drug-resistant, and incurable tuberculosis. Escherichia coli >820 Lancet Respir Med 2017;5:291–360. Pseudomonas aeruginosa >820 [4] Griffith DE, Aksamit TR. Understanding nontuberculous mycobacterial lung Serratia marcescens >820 disease: it’s been a long time coming. F1000Res 2016;5:2797. Stenotrophomonas maltophilia >820 [5] Seaworth BJ, Griffith DE. Therapy of multidrug-resistant and extensively Burkholderia cepacia complex >820 drug-resistant tuberculosis. Microbiol Spectr 2017;5. Gram-positive bacteria [6] Nguyen PC, Delorme V, Bénarouche A, Martin BP, Paudel R, Gnawali GR, et al. Staphylococcus aureus >820 Cyclipostins and cyclophostin analogs as promising compounds in the fight Staphylococcus epidermidis >820 against tuberculosis. Sci Rep 2017;7:11751. Streptococcus pneumoniae >820 [7] Johnson G. The alpha/beta hydrolase fold proteins of Mycobacterium Enterobacter aerogenes >820 tuberculosis, with reference to their contribution to virulence. Curr Protein Pept Enterococcus faecalis >820 Sci 2017;18:190–210. Enterococcus faecium >820 [8] Point V, Malla RK, Diomande S, Martin BP, Delorme V, Carriere F, et al. Synthesis and kinetic evaluation of cyclophostin and cyclipostins phosphonate analogs a MICs are expressed as the mean value of triplicate experiments. as selective and potent inhibitors of microbial lipases. J Med Chem Respective MIC50/MIC90 values of standard antimicrobial agents against M. chelonae– 2012;55:10204–19. M. abscessus clinical isolates: imipenem, 16/32 μg/mL; and cefoxitin, 32/64 μg/mL [9] Martin BP, Vasilieva E, Dupureur CM, Spilling CD. Synthesis and comparison [18]. of the biological activity of monocyclic phosphonate, difluorophosphonate and phosphate analogs of the natural AChE inhibitor cyclophostin. Bioorg Med Chem 2015;23:7529–34. [10] Vasilieva E, Dutta S, Malla RK, Martin BP, Spilling CD Dupureur CM. Rat hormone sensitive lipase inhibition by cyclipostins and their analogs. Bioorg Med Chem 4. Conclusion 2015;23:944–52. [11] Roux AL, Catherinot E, Ripoll F, Soismier N, Macheras E, Ravilly S, et al. Multicenter study of prevalence of nontuberculous mycobacteria in patients Taking into account all of these results, we thus propose the CyCs with cystic fibrosis in France. J Clin Microbiol 2009;47:4124–8. analogues to be considered as selective inhibitors of mycobacteria [12] Sambandamurthy VK, Derrick SC, Hsu T, Chen B, Larsen MH, Jalapathy KV, et al. and attractive candidates to be further exploited in the fight against Mycobacterium tuberculosis DeltaRD1 DeltapanCD: a safe and limited repli- mycobacterial-related diseases. In addition, CyC was found to be cating mutant strain that protects immunocompetent and immunocompromised 17 mice against experimental tuberculosis. Vaccine 2006;24:6309–20. active against multiresistant species of the M. abscessus complex. [13] Blanco-Ruano D, Roberts DM, Gonzalez-Del-Rio R, Álvarez D, Rebollo MJ, Further work to decipher the physiological target(s) of CyC17 and Pérez-Herrán E, et al. Antimicrobial susceptibility testing for Mycobacterium to elucidate its mode of action in M. abscessus is currently in progress. sp. Methods Mol Biol 2015;1285:257–68. [14] Palomino JC, Martin A, Camacho M, Guerra H, Swings J, Portaels F. Resazurin microtiter assay plate: simple and inexpensive method for detection of drug resistance in Mycobacterium tuberculosis. Antimicrob Agents Chemother Acknowledgments 2002;46:2720–2. [15] Rybniker J, Vocat A, Sala C, Busso P, Pojer F, Benjak A, et al. Lansoprazole M. Gimenez, Dr B. Ize and Prof. S. Bleves are acknowledged for is an antituberculous prodrug targeting cytochrome bc1. Nat Commun 2015;6:7659. providing the electronic microscopy picture of Pseudomonas [16] Walzl A, Kramer N, Mazza G, Rosner M, Falkenhagen D, Hengstschläger M, et al. aeruginosa displayed in the graphical abstract. A simple and cost efficient method to avoid unequal evaporation in cellular Funding: This study was supported by Centre national de la re- screening assays, which restores cellular metabolic activity. Int J Appl Sci Technol cherche scientifique (CNRS). PCN was supported by a PhD Training 2012;2:17–25. [17] Singh S, Bouzinbi N, Chaturvedi V, Godreuil S, Kremer L. In vitro program from the University of Science and Technology of Hanoi evaluation (Vietnam). AM was supported by a PhD fellowship from the Asso- of a new drug combination against clinical isolates belonging to the ciation Grégory Lemarchal and Vaincre la Mucoviscidose [project Mycobacterium abscessus complex. Clin Microbiol Infect 2014;20:O1124–7. [18] Lavollay M, Dubée V, Heym B, Herrmann JL, Gaillard JL, Gutmann L, et al. In no. RF20160501651], and PS received financial support for his PhD vitro activity of cefoxitin and imipenem against Mycobacterium abscessus fellowship from the Ministère de l’Enseignement Supérieur et de complex. Clin Microbiol Infect 2014;20:O297–300. la Recherche (France). [19] Halloum I, Viljoen A, Khanna V, Craig D, Bouchier C, Brosch R, et al. Resistance to thiacetazone derivatives active against Mycobacterium abscessus involves Competing interests: None declared. mutations in the MmpL5 transcriptional repressor MAB_4384. Antimicrob Ethical approval: Not required. Agents Chemother 2017;61:e2509–16.

Please cite this article in press as: Phuong Chi Nguyen, et al., Cyclophostin and Cyclipostins analogues, new promising molecules to treat mycobacterial-related diseases, Interna- tional Journal of Antimicrobial Agents (2018), doi: 10.1016/j.ijantimicag.2017.12.001 Supplementary material

Cyclophostin and Cyclipostins analogs, new promising molecules to treat mycobacterial-related diseases Phuong Chi Nguyen1¤, Abdeldjalil Madani1¤, Pierre Santucci1, Benjamin P. Martin2, Rishi R. Paudel2, Sandrine Delattre3, Jean-Louis Herrmann3,4, Christopher D. Spilling2, Laurent Kremer5,6, Stéphane Canaan1* and Jean-François Cavalier1*

1 Aix-Marseille Univ, CNRS, EIPL, IMM FR3479, Marseille, France 2 Department of Chemistry and Biochemistry, University of Missouri−St. Louis, One University Boulevard, St. Louis, Missouri 63121, USA. 3 AP-HP, Hôpitaux Universitaires Ile de France Ouest, Ambroise Paré, Boulogne and Raymond Poincaré, Garches, France. 4 2I, UVSQ, INSERM UMR1173, Université Paris-Saclay, Versailles, France. 5 Institut de Recherche en Infectiologie de Montpellier (IRIM), CNRS, UMR 9004, Université de Montpellier, 1919 route de Mende, 34293 Montpellier, France. 6 IRIM, INSERM, 34293, Montpellier, France.

¤ Authors have contributed equally to this work

Corresponding authors: J.-F. Cavalier ([email protected]) and S. Canaan ([email protected]) EIPL UMR7282, CNRS, 31 Chemin Joseph Aiguier, 13402 Marseille Cedex 20, France. Tel.: +33 491164093.

Figure S1. Structure of the 26 CyCs investigated

Table S1. Results of the antimicrobial susceptibility testing of the 26 CyCs towards selected (myco)bacteria by the agar plate method. a M. abscessus P. aeruginosa E. coli M. smegmatis M. bovis M. tuberculosis Compounds CIP 104536T M. marinum PA01 DH10B mc2155 BCG mc26230 Smooth Rough

CyC1

CyC2(α)

CyC2(β)

CyC3

CyC4

CyC5

CyC6(α)

CyC6(β)

CyC7(α)

CyC7(β)

CyC8(α)

CyC8(β)

CyC9(α)

CyC9(β)

CyC10(α)

CyC10(β)

CyC11

CyC12

CyC13(α)

CyC13(β)

CyC14

CyC15

CyC16

CyC17

CyC18(α)

CyC18(β) a Cells format: white cells = no effect; blue cells = 50-80% growth inhibition; green cells = 90-100% growth inhibition. Results – Article 3

Article 3: Cyclipostins and Cyclophostin analogs inhibit the antigen 85C from Mycobacterium tuberculosis both in vitro and in vivo

Two members of the antigen 85 complex from M. tb, i.e. Ag85A and Ag85C are identified as the main targets of the most active extra-cellular molecule CyC17 in our first article. The antigen 85 complex consists of three abundantly secreted proteins (Ag85A, Ag85B, Ag85C) which plays an essential role in the mycobacterial cell wall biosynthesis.

Scheme 2. Function of antigen 85 complex (FbpABC) in M. tb. Trehalose monomycolate (TMM) has been synthesized intracellularly, then transferred through MmpL3 transporter to the FbpABC. The FbpABC, in turn, is responsible for the transmission of mycolic acid (MA) from TMM to either another one located on the outermembrane or to arabinogalactan (AG) layer. Adapted from (Grzegorzewicz et al., 2012). Particularly, the trehalose monomycolate (TMM) which is synthesized inside the cytoplasm is transported to the outermembrane by MmpL3 transporter. At this stage, the antigen 85 complex is responsible for the biosynthesis of TMM and TDM as well as the covalent attachment of mycolic acids to arabinogalactan (AG) layer (Scheme 2). Deletion of fbpC2, encoding Ag85C, resulted in a 40% decrease in the AG-bound mycolic acids but failed to affect the production of non-covalently linked mycolates (Jackson et al., 1999), while deletion of fbpA or fbpB genes, encoding Ag85A and Ag85B, respectively, lead to reduced TDM levels (Nguyen et al., 2005;

93 Results – Article 3

Hunter et al., 2009). These studies implied that although a level of functional redundancy exists in vivo between the three members, the contribution of each one is significant. In this third article, we used biochemical and structural approaches to validate the Ag85 complex as a pharmacological target of three CyC analogs: CyC7 which exhibits a strong activity against extracellular and intracellular mycobacteria (MIC50 of 16.6 and 3.1 µM, respectively); CyC8 mostly active against intracellular bacteria only (MIC50 ≈11.7 µM); and

CyC17 the most potent inhibitor of extracellular in vitro growth (MIC50 ≈ 0.5 µM) with no activity against intracellular bacilli. We found that each CyC investigated bind covalently to the catalytic Ser124 residue in Ag85C and inhibit in vitro mycolyltransferase activity, and in vivo TDM biosynthesis as well as the mycolylation of arabinogalactan in M. tb. Additionally, this study demonstrated that Ag85C also exhibited strong diacylglycerol acyltransferase (DGAT) activity and confirmed that the three CyC analogs were also able to block this DGAT activity in vitro. Importantly, we showed that exposure to CyC17 inhibited TAG biosynthesis in a dose-dependent manner, thus implying that the de novo biosynthesis of TAG is also targeted by CyC17. Another main result obtained in this article is the co-crystal structure of Ag85C in complex with

CyC8β solved at the resolution of 1.8 Å. This 3D structure revealed clearly that CyC8β occupies Ag85C substrate-binding pocket close to the catalytic Ser124 residue. To summary, this study brings a comprehensive exploration not only on the physiological role of Ag85C in the bacterial cell wall biosynthesis or in its contribution to the formation of TAG, but also provides compelling evidence that CyC analogs can inhibit the activity of the Ag85 complex in vitro and in mycobacteria, thus opening the door to a new strategy against M. tb. In this article, I contributed to the work of creating the Ag85C-overexpression strain and Ag85CSer124A– overexpression strains, and performed the fluorescent microscopy experiments on these strains in compare with the wild type strain in order to reveal that the biosynthesis of TAG was inhibited by CyC17. The strains were also used in the experiment to indicate the dependence of the TAG biosynthesis on the level of expression of Ag85C. Besides, I get involved in the mass spectrometry experiment to detect that CyC compounds indeed covalently linked to Ser124 of Ag85C catalytic triad by showing the mass of modification in total mass analysis when the recombinant Ag85C was treated with molecules.

94 ARTICLE cro

Cyclipostins and cyclophostin analogs inhibit the antigen 85C from Mycobacterium tuberculosis both in vitro and in vivo Received for publication, November 2, 2017, and in revised form, December 5, 2017 Published, Papers in Press, January 4, 2018, DOI 10.1074/jbc.RA117.000760 Albertus Viljoen‡1, Matthias Richard‡1, Phuong Chi Nguyen§¶2, Patrick Fourquetʈ, Luc Camoinʈ, Rishi R. Paudal**, Giri R. Gnawali**, Christopher D. Spilling**, Jean-François Cavalier§¶, Stéphane Canaan§¶, Mickael Blaise‡3, and Laurent Kremer‡ ‡‡4 From the ‡Institut de Recherche en Infectiologie de Montpellier (IRIM), Université de Montpellier, CNRS UMR9004, 34293 Montpellier, France, ‡‡INSERM, IRIM, 34293 Montpellier, France, §Aix-Marseille Universite´, CNRS, EIPL, IMM FR3479, 13009 Marseille, France, ¶Aix-Marseille Universite´, CNRS, LISM, IMM FR3479, 13009 Marseille, France, ʈAix Marseille Universite´, CNRS, INSERM, Institut Paoli-Calmettes, CRCM, Marseille Protéomique, 13009 Marseille, France, and the **Department of Chemistry and Biochemistry, University of Missouri, St. Louis, Missouri 63121 Edited by Chris Whitfield

An increasing prevalence of cases of drug-resistant tubercu- With 10.4 million new cases and 1.8 million deaths in 2016, 5

losis requires the development of more efficacious chemother- tuberculosis (TB) continues to be a major global health prob- Downloaded from apies. We previously reported the discovery of a new class of lem. TB is caused by Mycobacterium tuberculosis, a resilient cyclipostins and cyclophostin (CyC) analogs exhibiting potent microorganism that persists through long courses of antibiotics activity against Mycobacterium tuberculosis both in vitro and in and years of dormancy within the host. The emergence of mul- infected macrophages. Competitive labeling/enrichment assays tidrug-resistant and extensively drug-resistant TB has contrib- combined with MS have identified several serine or cysteine uted to the difficulties in treating this bacterial infection (1). http://www.jbc.org/ enzymes in lipid and cell wall metabolism as putative targets of Chemotherapeutic treatments against TB remain very chal- these CyC compounds. These targets included members of the lenging and complicated, essentially because of the slow rate of antigen 85 (Ag85) complex (i.e. Ag85A, Ag85B, and Ag85C), growth of the bacilli and the presence of a thick, greasy, and responsible for biosynthesis of trehalose dimycolate and myco- relatively drug-impermeable cell wall (2). This mycobacterial lylation of arabinogalactan. Herein, we used biochemical and cell wall consists of a complex skeleton comprising covalently structural approaches to validate the Ag85 complex as a phar- linked macromolecules, such as peptidoglycan, arabinogalac- at CNRS on February 26, 2018 macological target of the CyC analogs. We found that CyC7␤, tan, and mycolic acids, in which non-covalently associated gly- 124 CyC8␤, and CyC17 bind covalently to the catalytic Ser residue colipids are interspersed (3). The mycolic acid portion of the in Ag85C; inhibit mycolyltransferase activity (i.e. the transfer of envelope is composed of very long fatty acids (C70–90) that are a fatty acid molecule onto trehalose); and reduce triacylglycerol either covalently attached to the arabinan moiety of the arabi- synthase activity, a property previously attributed to Ag85A. nogalactan (AG) polymer or found esterified to trehalose as Supporting these results, an X-ray structure of Ag85C in com- trehalose monomycolate (TMM) or trehalose dimycolate plex with CyC8␤ disclosed that this inhibitor occupies Ag85C’s (TDM). Because several key antitubercular drugs, such as iso- substrate-binding pocket. Importantly, metabolic labeling of niazid, SQ109, delamanid, or ethambutol, target different M. tuberculosis cultures revealed that the CyC compounds aspects of the biosynthetic steps responsible for the cell wall impair both trehalose dimycolate synthesis and mycolylation of attachment of mycolic acids (4–7), this pathway is of particular arabinogalactan. Overall, our study provides compelling evi- interest from a drug discovery perspective. dence that CyC analogs can inhibit the activity of the Ag85 com- The three functionally and structurally related members of plex in vitro and in mycobacteria, opening the door to a new the antigen 85 complex, designated Ag85A, -B, and -C, are strategy for inhibiting Ag85. The high-resolution crystal struc- among the most abundantly secreted proteins in M. tuberculo- ture obtained will further guide the rational optimization of new sis (8). These enzymes are responsible for the biosynthesis of CyC scaffolds with greater specificity and potency against TMM and TDM as well as the covalent attachment of mycolic M. tuberculosis. acids to AG (9–11). Deletion of fbpC2, encoding Ag85C, resulted in a 40% decrease in the AG-bound mycolic acids but This work was supported by Fondation pour la Recherche Médicale (FRM) failed to affect the production of non-covalently linked myco- Grants DEQ20150331719 (to L. K.) and ECO20160736031 (to M. R.) and by CNRS and INSERM. The authors declare that they have no conflicts of inter- est with the contents of this article. The atomic coordinates and structure factors (code 5OCJ) have been deposited in 5 The abbreviations used are: TB, tuberculosis; AG, arabinogalactan; CyC, cyc- the Protein Data Bank (http://wwpdb.org/). lipostins and cyclophostin; DGAT, diacylglycerol acyltransferase; FAME, 1 Both authors contributed equally to this work. fatty acid methyl ester; MAME, mycolic acid methyl ester(s); MIC, minimal 2 Supported by the Ph.D. Training program of the University of Science and inhibitory concentration; TAG, triacylglycerol; TDM, trehalose dimycolate; Technology of Hanoi. TLC, thin layer chromatography; TMM, trehalose monomycolate; xI50, 3 To whom correspondence may be addressed. Tel.: 33-4-34-35-94-47; E-mail: inhibitor molar excess leading to 50% inhibition; FP, fluorophosphonate; [email protected]. DTNB, 5,5Ј-dithio-bis-(2-nitrobenzoic acid); DEP, p-nitrophenyl phosphate; 4 To whom correspondence may be addressed. Tel.: 33-4-34-35-94-47; E-mail: TAMRA, carboxytetramethylrhodamine; ILI, intracellular lipid inclusion; [email protected]. r.m.s., root mean square; PDB, Protein Data Bank.

J. Biol. Chem. (2018) 293(8) 2755–2769 2755 © 2018 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A. Inhibition of Ag85C by cyclipostins and cyclophostin

lates (10), whereas deletion of fbpA or fbpB, encoding Ag85A strong activity against extracellular and intracellular mycobac- ␮ and Ag85B, respectively, led to reduced TDM levels (12–14), teria (MIC50 of 16.6 and 3.1 M, respectively), CyC8␤ was implying that although a level of functional redundancy exists mostly found to be active against intracellular bacteria (MIC50 ϳ ␮ in vivo between the three members, the contribution of each 11.7 M). In contrast, CyC17 was a potent inhibitor of in vitro ϳ ␮ member is significant. The lack of double and triple knockout growth (MIC50 0.5 M) but failed to show activity against mutants might indicate that the loss of two or more Ag85 intracellular bacilli (29). To identify the putative target(s) of enzymes is detrimental to M. tuberculosis viability. An addi- the CyC inhibitors, an activity-based protein profiling tional isoform, designated Ag85D or MPT51, has been charac- approach was used based on TAMRA-FP and desthiobio- terized but found to be inactive due to the lack of catalytic tin-FP probes and mass spectrometry analyses. This led to elements required for mycolyltransferase activity (11, 15, 16). the capture of several active serine/cysteine enzymes in a Ag85A/B/C share the same mycolic acid donor TMM, and their complex proteome before mass spectrometry identification, crystal structures present a highly conserved catalytic site, among which Ag85A (Rv3804c) and Ag85C (Rv0129c) were which further supports their similar enzymatic role (17–19). identified. Due to their importance in mycolic acid metabolism, the The present study was undertaken to further explore and Ag85 enzymes have often been proposed as attractive targets validate, through a combination of biochemical and struc- for future chemotherapeutic developments against TB (9, tural approaches, the specificity of inhibition of the Ag85 20–22). Because of their high structural conservation, it can be activity by the CyC analogs, to determine their mode of inferred that a single compound may inhibit all three enzymes action and to describe how they affect the mycolic acid pro-

of the complex at the same time and would make improbable file in M. tuberculosis. Downloaded from the development of resistance to inhibitors, because resistant Results mutants would require the simultaneous acquisition of muta- tions in at least two fbp genes. In addition, because these pro- CyC analogs inhibit TDM biosynthesis and transfer of mycolic teins are secreted, targeting the Ag85 complex will minimize acids to arabinogalactan in M. tuberculosis

the effect of efflux mechanisms that may result in resistance CyCs are a new class of compounds demonstrating potent http://www.jbc.org/ phenotypes. Early inhibitors, such as trehalose analogs, were antitubercular activity, presumably involving inhibition of the first designed as Ag85 inhibitors but were found to exhibit rel- Ag85 activity (29). The chemical structures of the cyclophostin

atively poor activity on whole mycobacterial cells (9, 23). analogs CyC7␤ and CyC8␤ and the cyclipostins CyC17 used in Another potentially selective fluorophosphonate ␣,␣-D-treha- this study are provided in Fig. 1A. To test whether treatment lose inhibitor of the three antigen 85 enzymes has been with these CyCs alters the mycolic acid composition of at CNRS on February 26, 2018 reported to form a stable, covalent complex with the Ag85 M. tuberculosis mc26230, cultures were exposed to increasing

enzyme following nucleophilic attack on the phosphorus atom concentrations of CyC17 or CyC7␤, the two inhibitors most 124 of the catalytic Ser (24). In the same manner, the 2-amino- active against extracellularly replicating M. tuberculosis (29), 6-propyl-4,5,6,7-tetrahydro-1-benzothiophene-3-carbonitrile, followed by metabolic labeling with sodium [2-14C]acetate and designated I3-AG85, inhibits Ag85C, and exposure of M. tuber- lipid analysis. Extraction and separation of the total mycolic culosis to this compound was associated with reduced survival acid methyl esters (MAME) by thin layer chromatography

rates in broth medium and in infected primary macrophages. (TLC) revealed that neither CyC17 nor CyC7␤ altered the de Moreover, I3-AG85 was active against a panel of multidrug- novo biosynthesis of mycolic acid (Fig. 1 (B and C), left). In resistant/extensively drug-resistant strains, although it exhib- contrast, separation of the apolar lipid fraction by TLC showed ited an MIC of 100 ␮M (25). By combining fragment-based drug a dose-dependent decrease in TDM levels associated with a discovery with early whole cell antibacterial screening, tetra- concomitant increase in the production of TMM, which is the hydro-1-benzothiophene analogs were discovered as potent natural substrate of the Ag85 proteins (Fig. 1 (B and C), middle). Ag85C inhibitory molecules against drug-susceptible and drug- To address whether CyC treatment also impacts the cell wall- resistant M. tuberculosis strains (26). The selenazole com- bound mycolic acids, radiolabeled mycolic acids were extracted pound ebselen (2-phenyl-1,2-benzisoselenazol-3(2H)-one) was from delipidated bacteria (30). The autoradiography/TLC anal- found to inhibit the activity of Ag85C through an original ysis confirmed a dose-dependent inhibition of [2-14C]acetate 209 mechanism by reacting with the conserved Cys residue incorporation into all three forms of the AG-attached mycolic located near the active site of the enzyme but not involved in the acids (␣, methoxy, and keto), suggesting that treatment with

catalytic activity (27, 28). Ebselen was shown to directly impede CyC17 or CyC7␤ inhibits AG mycolylation at low concentra- the production of TDM and mycolylation of AG (27). tions (Fig. 1 (B and C), right). A quantitative analysis of these Recently, cyclipostins and cyclophostin (CyC), representing effects is provided in the corresponding graphs (Fig. 1, B and C). a new class of monocyclic enolphosph(on)ate compounds, have Overall, this suggests that in vivo inhibition of the mycolyl- been discovered to act as powerful antitubercular agents affect- transferase activity by the CyC compounds results in decreased ing growth of M. tuberculosis both in vitro and in infected formation of the virulence-associated TDM and reduced trans- macrophages (29). Among the set of 27 CyC analogs previously fer of mycolic acids onto the essential cell wall AG. evaluated against M. tuberculosis H37Rv, eight compounds Covalent inhibition of the Ag85C mycolyltransferase activity exhibited potent anti-tubercular activities, particularly the

cyclophostin analogs CyC7␤ and CyC8␤ as well as the cycli- All three members of the Ag85 complex, sharing between 65 postins-related molecule CyC17. Whereas CyC7␤ exhibited a and 75% sequence identity, have been shown to possess a serine

2756 J. Biol. Chem. (2018) 293(8) 2755–2769 Inhibition of Ag85C by cyclipostins and cyclophostin Downloaded from http://www.jbc.org/ at CNRS on February 26, 2018

Figure 1. Ag85 complex mycolyltransferase activity is inhibited by CyC analogs in vivo. Exponentially growing M. tuberculosis mc26230 was incubated OADC/Tween 80 with increasing concentrations of CyC17 or CyC7␤ in 7H9 at 37 °C with agitation for 1 h. Subsequently, bacteria were labeled with sodium [2-14C]acetate for 6 h at 37 °C with agitation. The cultures were split, and from the first volume were extracted the total methyl esters of mycolates (MAME) and fatty acids (FAME). From the second volume, apolar and polar lipid fractions were obtained before derivatization of arabinogalactan MAME. A, chemical 2 structures of the CyC analogs used in this study. B and C, effect of CyC17 (B) or CyC7␤ (C) treatment on the mycolic acid profiles of M. tuberculosis mc 6230. Equal counts (50,000 cpm) of MAME ϩ FAME fraction were loaded on a TLC plate and resolved once using the solvent system hexane/ethyl acetate (95:5, v/v) run twice (far left). The apolar fraction was loaded (50,000 cpm), and TMM and TDM were visualized on a 1D TLC plate using the solvent system chloroform/ methanol/water (40:8:1, v/v/v) (middle left). Equal volumes of arabinogalactan MAME fraction were loaded, and ␣, methoxy, and keto mycolic acids were visualized on a 1D TLC plate using the solvent system hexane/ethyl acetate (95:5, v/v) run twice (middle right). Densitometric analysis (far right) was performed on the TLCs shown in the left panels. Histograms and error bars, means and S.D. values calculated from at least two independent experiments.

hydrolase mycolyl esterase/transferase activity (17–19). To test three compounds, with CyC8␤ being the most efficient inhibi- Ϯ ␮ Ϯ ␮ the hypothesis that CyC analogs inhibit the activity of the three tor (IC50 of 15 5 M), followed by CyC7␤ (IC50 of 43 3 M) Ϯ ␮ Ag85 members, we first cloned Ag85C (fbpcC2) into pET23b, and CyC17 (IC50 of 98 6 M)(Fig. 2A). Moreover, in terms of ϭ and the recombinant protein was produced in Escherichia coli. molar excess of inhibitor (xI50 IC50/[Ag85C]) (31), all three The protein was then purified from lysates of E. coli by succes- CyCs react almost in stoichiometry with pure Ag85C, as judged

sive nickel-affinity, anion-exchange, and size-exclusion chro- by their respective xI50 values of around 0.3, 0.8, and 1.8, matography steps, leading to 3 mg of pure protein/liter of cul- respectively. ture. Using a recently developed fluorescent assay based on To address the inhibitory effect on the mycolyltransferase resorufin butyrate as the acyl donor for Ag85C and trehalose as assay, Ag85C (55 ␮M) was next incubated for 30 min in its native ␮ the acyl acceptor (27), we investigated whether CyC7␤, CyC8␤, form with 500 M (i.e. enzyme/inhibitor molar ratio of 1:9) of and CyC17 inhibit the acyltransferase activity onto trehalose. In each CyC compound. As expected, the complete loss of activity each case, a dose-dependent inhibition was observed with all was confirmed by comparing the pretreated versus non-treated

J. Biol. Chem. (2018) 293(8) 2755–2769 2757 Inhibition of Ag85C by cyclipostins and cyclophostin Downloaded from

Figure 2. Inhibition of the Ag85C mycolyltransferase activity is mediated by the covalent binding of CyC analogs.A,theenzymaticactivityofAg85Cwastested http://www.jbc.org/ using a fluorescence-based assay in the presence of different concentrations of CyC7␤,CyC8␤,andCyC17.Theinhibitoryeffectwasdeterminedatthemaximumrateof the reaction. Error bars,S.D.calculatedfromthreeindependentexperiments.Curves for CyC7␤,CyC8␤,andCyC17 were fitted using the EC50 shift non-linear regression 2 S124A model in GraphPad Prism with R values of 0.9675, 0.9508, and 0.9415, respectively. B,equalamountsofeitherAg85CorAg85C were pretreated with CyC7␤, CyC8␤,andCyC17;incubatedwithTAMRA-FP,separatedbySDS-PAGE;andvisualizedbyCoomassieBluestaining(top)orin-gelfluorescencevisualization(middle). The merged image is shown at the bottom.TAMRAlabelingofAg85CispreventedbythecovalentbindingoftheCyCanalogstothecatalyticSer124.NoTAMRA-FPlabeling is seen for the Ag85CS124A variant, confirming Ser124 as the TAMRA-binding site. Cand D,globalmassmodificationofAg85C(C)andAg85CS124A (D)preincubatedwith CyC7␤,CyC8␤,andCyC17 as determined using an Ultraflex III mass spectrometer (Bruker Daltonics) in linear mode with the LP_66 kDa method. The mechanism of action at CNRS on February 26, 2018 of the phosphonates CyC7␤ and CyC8␤ and of the phosphate analog CyC17 based on mass spectrometry analyses is illustrated in C. a.u., arbitrary units.

124 Ag85C. All three Ag85C-CyC adducts were treated with 10 ␮M plex, as the reaction between the catalytic Ser and either ϩ TAMRA-FP fluorescent probe, known to bind to serine CyC7␤ or CyC8␤ is expected to yield mass increases of 374.2 enzymes (32), for 1 h, and equal amounts of proteins were sep- or ϩ402.25 Da, respectively. Moreover, such results are consis- arated by SDS-PAGE and visualized by Coomassie staining (Fig. tent with the known and irreversible classical mechanism of 2B, top) or in-gel fluorescence for TAMRA detection (Fig. 2B, action of phosphonate compounds, as demonstrated using middle). Pretreatment with either CyC7␤ or CyC8␤ resulted in a pure mycobacterial lipolytic enzymes (31). significant loss in fluorescence intensity (about 75%) as com- With respect to CyC17, the observed 322.1-Da mass shift pared with the non-treated protein, whereas incubation with increment was 124.18 Da lower than its expected theoretical CyC17 abrogated TAMRA labeling. This suggests that reaction molecular mass of 446.28 Da (Fig. 2C). This size difference may with the TAMRA probe is strongly impaired in the Ag85C-CyC arise from the specific chemical properties of phosphate (i.e. adducts, resulting in a decrease/loss of fluorescence emission. 124 CyC17) versus phosphonate (i.e. CyC7␤ and CyC8␤) chemical To determine the implication of the conserved catalytic Ser 124 groups. In all cases, the nucleophilic attack of catalytic Ser at in Ag85C in TAMRA labeling, this residue was replaced by an the phosphorus center induces ring opening. However, the Ala residue, and the mutated protein was purified (Fig. 2B, top). S124A reaction with CyC is very likely to form a new phosphate Exposure of TAMRA to Ag85C failed to produce a fluo- 17 triester, which in turn becomes susceptible to hydrolysis. From rescence signal (Fig. 2B, middle), indicating that the catalytic 124 Ser124 is required for binding of the probe. As expected, no these findings, it can be inferred that once the CyC17-Ser fluorescence emission was observed when pretreating the adduct is formed, it becomes rapidly hydrolyzed in the presence mutated protein with the CyC analogs (Fig. 2B, bottom). of water, resulting in the cleavage and release of the methyl MALDI-TOF mass spectrometry was further used to study 2-acetyl-4-hydroxybutyrate (i.e. 124.1 Da), accounting exactly the (covalent) nature of the inhibition. Mass increments of for the molecular mass discrepancy observed experimentally ϩ ϩ 383.5 and 402.3 Da in the presence of CyC7␤ and CyC8␤, (Fig. 2C). respectively, were observed within the global mass of treated Taken together, these findings conclusively indicate that Ag85C as compared with the global mass of untreated Ag85C Ag85C is covalently modified by CyC analogs, leading to the (Fig. 2C). In contrast, no changes in the global mass were inhibition of the mycolyltransferase activity and thus support- observed with the inactive Ag85CS124A protein (Fig. 2D). These ing the in vivo alteration of the mycolic acid pattern by these data thus support the formation of a covalent Ag85C-CyC com- compounds.

2758 J. Biol. Chem. (2018) 293(8) 2755–2769 Inhibition of Ag85C by cyclipostins and cyclophostin Downloaded from http://www.jbc.org/ at CNRS on February 26, 2018

Figure 3. DGAT activity of the antigen 85 complex and inhibition by CyC analogs. A, chemical reaction occurring while determining the DGAT activity. DTNB reacts with the free-thiol group coming from the release of SH-CoA during the formation of TAG from 1,2-dipalmitoylglycerol (DAG) and a molecule of acyl-CoA. B, comparison of the DGAT activity of Ag85A, Ag85B, Ag85C, and MPT51. Enzymatic activity was determined by the colorimetry-based assay illustrated in A. Inset, activity of the wildtype and S124A Ag85C proteins using palmitoyl-CoA (C16) as acyl donor molecule. Error bars, S.D. calculated from three independent experiments. C, inhibitory effect of CyC analogs on Ag85C DGAT activity. Inhibition was performed with increasing concentrations of CyC7␤, CyC8␤, and CyC17 using the colorimetry-based assay illustrated in A. The inhibitory effect was determined after 1 h of reaction. Error bars, S.D. calculated from 2 three independent experiments. Curves for CyC7␤, CyC8␤, and CyC17 were fitted using the EC50 shift non-linear regression model on GraphPad with R values of 0.9755, 0.9641, and 0.9422, respectively. D, comparison of the DGAT activity of Ag85C and Tgs1 in the absence or presence of CyC7␤, CyC8␤, and CyC17.

Ag85A, -85B, and -85C express DGAT activity and purified from lysates by successive nickel-affinity, anion- Although the mycolyltransferase activity of the Ag85 com- exchange, and size-exclusion chromatography steps. Because plex has been established for a long time (8, 9), more recent Ag85B was poorly expressed in E. coli, a synthetic gene was work suggested that Ag85A mediates the transesterification of produced by replacing low-usage codons with high-usage diacylglycerol using long-chain acyl-CoA to produce triglycer- codons, as reported previously (34) and subsequently cloned ides (TAG), which act as storage compounds for energy and into pET23a. All three proteins were assayed for DGAT activity carbon (33). Ag85A contains the same catalytic triad as Ag85C in the presence of acyl-CoA with various chain lengths (from or Ag85B, formed by residues Ser126, His262, and Glu230, and C4 to C18) as acyl donors and 1,2-dipalmitoyl-sn-glycerol (1,2- possesses a deep substrate-binding groove near the active-site dipalmitin) as the acyl acceptor, as illustrated in Fig. 3A. Trans- serine, suggesting that Ag85B and Ag85C, similarly to Ag85A, esterification in the presence of 5,5Ј-dithio-bis-(2-nitrobenzoic may also express diacylglycerol acyltransferase (DGAT) activ- acid) (DTNB) leads to the formation of TNB, which can readily ity. To test this hypothesis, all of the genes were cloned into be measured at 412 nm (33, 35). In agreement with previous pET23b, and the recombinant proteins were produced in E. coli findings, Ag85A was found to express DGAT activity, but such

J. Biol. Chem. (2018) 293(8) 2755–2769 2759 Inhibition of Ag85C by cyclipostins and cyclophostin

an activity was only detected with C12-C18 acyl-CoAs (Fig. 3B). whether overexpression of Ag85C in M. tuberculosis affects the Whereas Ag85B demonstrated lower activity than Ag85A, TAG content. M. tuberculosis was first transformed with either Ag85C showed the highest activity, which was optimal in the pMV261-Ag85C or pMV261-Ag85CS124A. Overexpression of presence of C16-CoA. No activity was detected with the C4- or either the wildtype or the catalytically dead proteins was C8-containing acyl chains. We also expressed and purified the checked by quantitative real-time PCR (Fig. 4A, left) and by Ag85 complex-related MPT51 (FbpC1) protein, which pos- immunoblotting using two different monoclonal antibodies sesses an overall structure similar to that of the Ag85 complex and purified Ag85A, -B, and -C as positive controls (Fig. 4A, members but is defective in the catalytic elements required for right). The 17/4 monoclonal antibody recognizes a well-con- mycolyltransferase activity (11, 16). As anticipated, MPT51 served epitope present in Ag85A and Ag85B but not in Ag85C failed to express any DGAT activity, suggesting that residues (37). In contrast, the 32/15 antibody revealed all three antigens important for mycolyltransferase activity are also key players in and the presence of more pronounced bands in the pMV261- the DGAT activity, as proposed earlier for the Ser126 in Ag85A Ag85C and pMV261-Ag85CS124A lysates, which, by compari- (33). Purified Ag85CS124A was next assayed with various acyl- son with the 17/4 blot, could clearly be attributed to Ag85C CoA substrates, and, as shown in Fig. 3B (inset), the DGAT (Fig. 4B, right). This indicates that both Ag85C variants were activity was abrogated in the mutant protein, implying that overproduced at comparable transcriptional and translational Ser124 plays a critical role in the enzymatic reaction. levels and allowed us to investigate whether this may affect the Overall, these data extend insights from previous findings intracellular TAG content of M. tuberculosis (38). TAGs are and indicate that all three members of the Ag85 complex often stored in the form of ILIs, which can be visualized by

express DGAT activity, with Ag85C exhibiting the most pro- staining with Nile Red (39, 40). As shown in Fig. 4B (left), Downloaded from nounced activity. This suggests that Ag85C may also make an although a punctiform labeling corresponding to ILIs is ob- important contribution in TAG synthesis in M. tuberculosis. served in the control strain carrying the empty pMV261, Nile Red straining was much more pronounced in the strain over- CyC analogs inhibit the in vitro DGAT activity of Ag85C but not producing Ag85C, and the effect returned to control levels in of Tgs1 S124A

the strain harboring pMV261-Ag85C . Quantification of http://www.jbc.org/ The above-mentioned results prompted us to investigate the fluorescence intensity over the entire length of the individ- whether CyC analogs alter the DGAT activity of Ag85C. This ual bacilli from each strain clearly indicates that large and was achieved by incubating the purified enzyme in the presence numerous ILIs were present in the Ag85C-overexpressing S124A of increasing concentrations of CyC7␤, CyC8␤, and CyC17 using strain, as compared with the control and Ag85C strains the colorimetric activity assay described in the legend to Fig. (Fig. 4B, right). 3A. As expected from the previous results on the inhibition of To check whether enhanced ILI formation coincided with at CNRS on February 26, 2018 mycolyltransferase activity, a dose-dependent inhibition of the increased de novo biosynthesis of TAG, metabolic labeling of DGAT activity with all three compounds was also observed M. tuberculosis cultures with sodium [2-14C]acetate was per-

(Fig. 3C). CyC7␤ appeared as the most potent inhibitor, with an formed, followed by extraction and separation of the apolar Ϯ ␮ ϭ IC50 value of 85 2 M (i.e. xI50 2.8), followed by CyC8␤ and lipid fraction by TLC. In the absence of CyC treatment, the Ϯ ␮ ϭ CyC17 exhibiting IC50 values of 121 4 M (i.e. xI50 4.0) and strain carrying pMV261-Ag85C produced moderately higher Ϯ ␮ ϭ 187 3 M (i.e. xI50 6.2), respectively. amounts of TAGs than the control strain containing the empty Because the TAG synthase tgs1 in M. tuberculosis and Myco- pMV261 or the strain overexpressing Ag85CS124A (Fig. 4C, left), bacterium abscessus has been reported as the major contributor supporting the in vivo contribution of Ag85C in TAG produc-

of TAG accumulation in the form of intracellular lipid inclu- tion. Importantly, exposure to CyC17 inhibited TAG biosynthe- sions (ILIs) in these two species (35, 36), we addressed whether sis in a dose-dependent manner in the control strain, thus the DGAT activity of Tgs1 may also be targeted by the CyC implying that the de novo biosynthesis of TAG is also targeted

analogs. Tgs1 from M. tuberculosis (Rv3130c) was expressed by CyC17 (Fig. 4C). In addition, a less pronounced decrease in and purified from E. coli and subsequently used in a DGAT TAG production occurred in the strain overexpressing Ag85C

assay in the absence or presence of either CyC7␤, CyC8␤, or as compared with the control strain, presumably because of the CyC17 (Fig. 3D). Whereas the activity of Tgs1 remained intact inherent capacity of this strain to synthesize more TAG that ␮ even in the presence of a 250 M concentration of each com- partially overcomes CyC17 inhibition (Fig. 4C). Collectively, pound, the DGAT activity of Ag85C assayed in the same con- these results suggest that TAG production in M. tuberculosis is

ditions was almost abrogated, suggesting that Tgs1 activity is inhibited by CyC17 and that this is dependent upon Ag85C not impacted by CyC treatment. DGAT activity. These results indicate that CyC analogs specifically inhibit Crystal structure of the CyC -bound Ag85C the DGAT activity of Ag85C but not of Tgs1 in vitro, in agree- 8␤ ment with the fact that members of the Tgs family were not To gain insight into the mode of action of the CyC com- identified in our original proteomic profiling study (29). pounds, crystallization studies of Ag85C were undertaken in the presence of the three CyC inhibitors. However, diffracting Overexpressing Ag85C in M. tuberculosis is associated with crystals were only obtained with CyC8␤ for which the X-ray reduced inhibition of TAG production by CyC17 structure of Ag85C bound to CyC8␤ was solved at a resolution That Ag85C expresses the highest DGAT activity among the of 1.8 Å (Table 1). The asymmetric unit contains two molecules three members of the Ag85 complex prompted us to address of Ag85C (Fig. 5A). Residues 6–282 and 8–282 for each subunit

2760 J. Biol. Chem. (2018) 293(8) 2755–2769 Inhibition of Ag85C by cyclipostins and cyclophostin Downloaded from http://www.jbc.org/ at CNRS on February 26, 2018

Figure 4. Biosynthesis of TAG in M. tuberculosis is inhibited by CyC17 and is dependent upon Ag85C expression. A, quantitative real-time PCR analysis showing the -fold increase in the Ag85C transcripts in M. tuberculosis mc26230 containing either pMV261 (ctrl), pMV261-Ag85C, or pMV261-Ag85CS124A (left). Western blotting using the 32/15 and 17/4 monoclonal antibodies probed against purified Ag85A/B/C and crude lysates of M. tuberculosis mc26230 containing either pMV261, pMV261-Ag85C, or pMV261-Ag85CS124A (right). B, Nile Red staining of M. tuberculosis strains growing exponentially (left) with the correspond- ing fluorescence quantification (right). Fluorescence quantification was performed on 30 bacilli of each group. Shown are the mean fluorescence and S.D. values. Means were compared by the two-tailed Mann–Whitney test. ns, non-significant; **, p Ͻ 0.01. Results shown are representative of two independent OADC/Tween 80 14 experiments. C, cultures were exposed to increasing concentrations of CyC17 in 7H9 and labeled with sodium [2- C]acetate for 4 h at 37 °C with agitation. The apolar fraction was extracted to analyze de novo synthesis of TAG. Equal counts (50,000 cpm) of apolar fraction were loaded, and TAG was visualized on a 1D TLC plate using the solvent system petroleum ether/diethyl ether (90:10, v/v) (left). Right, densitometric analysis of TLCs. Histograms and error bars, means and S.D. values calculated from four independent experiments. could be built, implying that the last 14 residues as well as the tions (data not shown), appears as a possible molecule interact- polyhistidine tag in the C terminus were not modeled. The ing with Phe150 and could be seen in both monomers. As Phe150 structure of Ag85C has been extensively reported (17). In brief, was shown to be involved in stacking of the lipid chain of octyl- the protein adopts a typical ␣/␤ hydrolase fold made of a central glucoside in the Ag85C-octylglucoside crystal structure (PDB ␤ ␣ -sheet surrounded by -helices. The two monomers are entry 1VA5 (19)), we tried to place the acyl chain of CyC8␤ in nearly identical, as their superposition over 274 residues gives this extra electron density, but refinement of this alternate con- an r.m.s. deviation of 0.24 Å. However, whereas a clear electron formation of CyC8␤ did not converge. Therefore, further mod- density could be seen for the entire structure of CyC8␤ in one eling of this electron density blob was not pursued. Although monomer, this was only the case for the headgroup of the sec- CyC8␤ has clearly reacted, as evidenced by the presence of an ond molecule (Fig. 5B). It is noteworthy that the extra but non- opened ring and the MALDI-TOF data, no covalent bond 124 interpretable electron density (Fig. 5B) in the vicinity of CyC8␤, between the catalytic Ser residue and the phosphonate group observed in all data sets collected from either co-crystallization of CyC8␤ was observed (Fig. 5, B and D). Therefore, CyC8␤ was or soaking experiments and in various crystallization condi- modeled in an opened conformation (Fig. 5, B and D). CyC8␤

J. Biol. Chem. (2018) 293(8) 2755–2769 2761 Inhibition of Ag85C by cyclipostins and cyclophostin

Table 1 Discussion Data collection and refinement statistics Toward the generation of new lead compounds with unex- Data collection statistics Beamline ESRF-ID23.1 plored modes of action in M. tuberculosis, the CyC analogs Wavelength (Å) 0.972 were initially designed to inhibit mycobacterial lipases (31). In Resolution range (Å) 48.09–1.8 (1.86–1.8)a Space group P 21 21 21 particular, by covalently binding to the catalytic serine, they Unit cell fully inactivated the monoacylglycerol lipase Rv0183 and the Å 67.39, 75.77, 137.32 Degrees 90, 90, 90 triacylglycerol lipase LipY from M. tuberculosis but not the Total reflections 528,844 (50,342) mammalian gastric and pancreatic lipases (31). Subsequent Unique reflections 65,871 (6486) Completeness (%) 99.92 (99.85) biochemical studies involving the selective labeling and enrich- Mean I/␴(I) 16.66 (2.24) ment of captured enzymes using appropriate fluorophospho- Wilson B-factor (Å2) 24.34 nate probes in combination with CyC17 resulted in the identi- Rmeas 0.0968 (1.049) Refinement statistics fication of 23 potential target lipolytic enzymes, all of which Reflections used in refinement 65,862 (6486) comprise catalytic serine or cysteine residues (29). Because they Rwork 0.151 (0.232) are multitarget-inhibitory compounds in mycobacteria, the use Rfree 0.175 (0.269) No. of non-hydrogen atoms 4895 of CyC analogs could prevent the selection of drug resistance Macromolecules 4308 Ligands 54 mechanisms. In addition, the lack of cytotoxicity in human cells Solvent 533 (29) makes them attractive hits to be further evaluated. No. of r.m.s. deviations Herein, we provide compelling evidence that at least some of

Bonds (Å) 0.006 Downloaded from Angles (degrees) 0.84 the CyC analogs primarily act by inhibition of the Ag85 com- Ramachandran favored (%) 96.7 Ramachandran allowed (%) 3.3 plex, resulting in decreased TDM formation and reduced Ramachandran outliers (%) 0.00 mycolylation of AG, an essential polymer of the mycobacterial Rotamer outliers (%) 1.12 Clashscore 1.07 cell wall. Although one cannot rule out the possibility that the Average B-factor 29.13 killing effect of the CyC on M. tuberculosis results from the Macromolecules 27.22 http://www.jbc.org/ Ligands 50.05 simultaneous and net effect on multiple physiological targets, Solvent 42.43 the inhibition of TMM and AG mycolylation is very likely to PDB accession number 5OCJ a represent the major cause of growth inhibition of M. tubercu- The values in parenthesis are for the highest-resolution shell. losis, at least in in vitro growing cultures. We demonstrate here that all three Ag85 members express DGAT activity in vitro, interacts through residues at the entrance of the Ag85C active with Ag85C being the most active, thereby extending previous at CNRS on February 26, 2018 work reporting the DGAT activity of Ag85A (33). Importantly, site (Fig. 5, C and D). The polar head of CyC8␤ is recognized through hydrogen bonds with the catalytic Ser124 side chain as the S124A site-directed mutation of the active site of Ag85C well as with the main chain of Leu40 and the Asp38 side chain via proved that this residue is involved in the DGAT activity of this two water molecules. The Arg41 side chain completes the inter- enzyme and TAG synthesis. Although the synthesis of TAG relies on the presence of multiple TAG synthases, such as the action with the headgroup of CyC8␤ by van der Waals interac- well-characterized Tgs1 (Rv3130c) (36), our work extends the tion (Fig. 5D). The long aliphatic chain of CyC8␤ is stabilized by 222 223 226 growing list of enzymes displaying DGAT activity in M. tuber- hydrophobic interactions involving the Ile , Pro , Phe , 227 culosis. The Ag85 proteins do not belong to the known DGAT and Leu side chains (Fig. 5D). The distance between the 124 families and do not possess the characteristic conserved hepta- phosphate of CyC ␤ and Ser of 3.6 Å clearly attests that in 8 peptide acyltransferase motif of the mycobacterial Tgs enzymes this crystal, the ligand is not covalently bound. Importantly, this involved in TAG biosynthesis (35, 41). Nevertheless, the DGAT loss of covalent binding was observed in multiple data sets col- activity of Ag85C, similarly to Ag85A (33), includes two con- lected, obtained either by soaking or co-crystallization experi- secutive reactions, the fatty acyl-CoA hydrolysis (thioesterifica- ments. However, the lack of covalent binding in the crystal tion) and the subsequent transfer of the acyl chain to the dia- structure does not rule out the well-known covalent inhibitory cylglycerol (transesterification). Overexpressing Ag85C in mechanism of the CyC analogs supported by MALDI-TOF M. tuberculosis was correlated with an increase in de novo TAG mass spectrometry analyses (Fig. 2, C and D). production and formation of lipid storage inclusions. These Furthermore, the polar headgroup of CyC8␤ is located where findings establish for the first time a connection between cell trehalose, the natural substrate of the Ag85 proteins, binds, as wall and TAG biosynthesis by Ag85C and expand our under- seen in the crystal of the trehalose-bound structure of Ag85B standing of this important enzyme in the physiology of (PDB entry 1F0P (18)) (Fig. 6A). Interestingly, the fatty acyl M. tuberculosis. However, a direct implication of the DGAT chain of CyC8␤ is placed in a very hydrophobic cavity that was activity of Ag85C in pathogenesis and persistence of M. tuber- proposed to be part of the TDM/TMM fatty chain recognition culosis requires further studies. In addition, under conditions site (17). In addition, structural comparison indicated that the where Ag85C is overexpressed, M. tuberculosis was more important residues in Ag85C interacting with CyC8␤ are fully refractory to TAG inhibition by CyC17, further emphasizing the conserved in Ag85B and Ag85A (Fig. 6B), strongly suggesting yet unexpected contribution of Ag85C as a player in TAG bio- that CyC8␤, and presumably all of the other CyC analogs, may synthesis. Inhibition of the DGAT activity of Ag85C, and there- inhibit the three members of the Ag85 complex. fore TAG inhibition, by the CyC compounds is very unlikely to

2762 J. Biol. Chem. (2018) 293(8) 2755–2769 Inhibition of Ag85C by cyclipostins and cyclophostin Downloaded from http://www.jbc.org/ at CNRS on February 26, 2018

Figure 5. Structural basis for Ag85C inhibition by CyC8␤. A, crystal structure of Ag85C in complex with CyC8␤. The figure displays the overall asymmetric unit with the two monomers represented as blue and magenta schematics. CyC8␤ is shown as sticks and colored in yellow. B, simulated annealing Fo Ϫ Fc OMIT map contoured at 3␴ attesting to the presence of two CyC8␤ that could be entirely modeled for one molecule and partially for the second one. The map also reveals the presence of an extra, but non-interpretable, electron density in the vicinity of the CyC8␤ molecule. C, surface representation of the Ag85C structure bound 124 to CyC8␤. The hydrophobic residues are colored in blue, and the catalytic Ser is shown in green. D, CyC8␤ binding site. Ag85C residues involved in CyC8␤ recognition are displayed as blue sticks for those involved in hydrogen bond (black dashes) formation. Residues in orange are involved in hydrophobic 41 124 228 interactions with the acyl chain of CyC8␤, and Arg in gray contributes to the recognition of the CyC8␤ headgroup by van der Waals interaction. Ser , Glu , and His260 form the catalytic triad. Red spheres, water molecules. participate in growth inhibition of M. tuberculosis in vitro, but molecules that restrict entry of M. tuberculosis into dormancy, it may have important consequences for in vivo survival and/or a strategy that would overcome mycobacterial persistence and for maintaining the bacilli in a non-replicating growth phase, prolonged chronic infections. such as in foamy macrophages in which M. tuberculosis is able Biochemical studies involving the TAMRA-FP probe that to hydrolyze the host-derived TAGs from lipid bodies to fatty binds to serine hydrolases along with mass spectrometry and acids, which are then reprocessed as TAGs and stored within structural analyses indicate that, in addition to covalently bind- ILIs (42, 43). In these subcellular structures, TAGs represent ing to the catalytic Ser124, the CyC analogs could also be com- the primary storage source of carbon and energy, allowing the peting with the binding of Ag85 substrate (i.e. the trehalose and bacteria to survive in a non-replicating state and to persist the acyl chain moieties of TMM). As the Ag85 complex mem- inside these foamy cells, which usually line the necrotic centers bers share similar substrate specificities, our results suggest of tubercle granulomas and have been proposed to be the intra- that CyC analogs could target not only Ag85C but also Ag85B cellular niche of M. tuberculosis during latent infection (42). and Ag85A, an assertion reinforced by the fact that Ag85A was Although this requires further exploration, inhibiting the also identified as a potential target in the original proteomic DGAT activity of Ag85C may help in designing new classes of screen approach (29). Comparison of the three structures

J. Biol. Chem. (2018) 293(8) 2755–2769 2763 Inhibition of Ag85C by cyclipostins and cyclophostin Downloaded from http://www.jbc.org/ at CNRS on February 26, 2018

Figure 6. Mode of inhibition of the Ag85 complex by CyC8␤. A, superposition of the Ag85B-trehalose (PDB code 1F0P, blue) and Ag85C-CyC8␤ (gray) crystal structures. The headgroup of CyC8␤ (yellow) occupies the same site as trehalose (green). B, Ag85C residues (gray) involved in the recognition of CyC8␤ are all strictly conserved in Ag85B (cyan) and Ag85A (magenta). C, superposition of the Ag85C-ebselen (PDB code 4QDU; blue) and Ag85C-CyC8␤ (gray) crystal structures. CyC8␤ binds far away from the ebselen-binding site and does not trigger structural rearrangement of helix ␣9. D, superposition of the Ag85C-DEP (PDB code 1DQY; cyan) and Ag85C-CyC8␤ (gray) crystal structures. CyC8␤ presents a similar mode of inhibition as DEP (cyan stick), a nonspecific ␴/␤ hydrolase inhibitor.

strongly supports this hypothesis, as residues contacting CyC8␤ can be subjected to hydrolysis, rendering their covalent binding in Ag85C are strictly conserved in Ag85A and Ag85B. This is of potentially reversible, as shown here in the case of the CyC17- interest, as the inhibitor I3-AG85 binding to the active site of Ser124 adduct (29). Interestingly, using a chemical proteomic Ag85C exhibits only strict specificity toward Ag85C and does approach, the EZ120 ␤-lactone compound exhibiting strong not bind Ag85A and -B (25). Moreover, given their low xI50 antitubercular activity and resembling an electrophilic mimic values, the three CyC compounds are able to act in near stoi- of mycolic acids was recently found to block several serine chiometry and alter both the mycolyltransferase and DGAT hydrolases essential for the mycomembrane biosynthesis (44). activities of Ag85C. It is noteworthy that, among the three CyCs The polyketide synthase Pks13, whose ␤-keto mycolate is trans- investigated, the phosphate CyC17, which appears as the best ferred onto trehalose and reduced to yield TMM, as well as inhibitor against extracellular M. tuberculosis, was the least Ag85A were identified as primary targets of EZ120. However, efficient when assayed on pure recombinant enzyme. However, whether this ␤-lactone acts similarly to the CyC inhibitors in when assayed on living bacteria, CyC17 clearly affected TDM Ag85 awaits structural determination. synthesis and mycolylation of AG. The differences in activity Comparison of the Ag85C-CyC8␤ structure with that of with CyC7␤ and CyC8␤ may be related to the chemical proper- Ag85C-ebselen (PDB entry 4QDU (27)) shows that the mode of ties of the phosphate versus phosphonate chemical groups. On inhibition triggered by CyC8␤ is different. Ebselen indeed cova- the other hand, despite their high activity, phosphate inhibitors lently modifies Cys209, which is 13 Å away from the catalytic

2764 J. Biol. Chem. (2018) 293(8) 2755–2769 Inhibition of Ag85C by cyclipostins and cyclophostin

Ser124 (Fig. 6C). Inhibition by ebselen and its derivatives (azido codon-optimized version of the fbpB gene, encoding Ag85B, ebselen and adamantyl ebselen) is mediated by inducing struc- was synthesized (GenScript) and introduced within the pET23a tural rearrangements of helix ␣9 and the loop between helices 9 plasmid thanks to the NdeI and XhoI restriction sites, enabling and 10 that ends in destabilizing the hydrogen bond network of also the incorporation of a polyhistidine tag in the C terminus of the active site (27, 28, 45). Comparison of the crystal structures the Ag85B protein. The Ag85CS124A mutant was obtained by

of Ag85C-CyC8␤ and the Ag85C native structure (PDB entry using the PCR-driven overlap extension method (48). Briefly, 3HRH) possessing the same space group and crystallized in two separate PCRs were set up with the Phusion௡ DNA poly- similar conditions shows that the two structures are identical. merase (Thermo Fisher Scientific). The first one was set up with The superposition of the two structures yields an overall r.m.s. the forward primer used to amplify the wildtype fbpC2 gene and deviation over 251 residues of about 0.19 Å. Furthermore, no the reverse internal primer 5Ј-AAG ACC CAC CGC CGC local structural rearrangement was observed (data not shown). GTT-3Ј. The second one was set up with a forward internal Ј As expected, the mode of inhibition of CyC8␤ consists of block- primer, 5 -AAC GCG GCG GTG GGT CTT GCG ATG TCG ing the active site (31) and not of destabilizing the overall struc- GGC GGT TCC G-3Ј, overlapping the internal reverse primer ture and stability of the protein as reported for ebselen and its and containing the nucleotide substitution (changed nucleo-

analogs (27, 45). Furthermore, the mode of action of CyC8␤ is tide in boldface type) with the reverse primer used to amplify the more related to that of the diethyl p-nitrophenyl phosphate wildtype fbpC2 gene. The purified PCR products were het- (DEP), a nonspecific ␣/␤ hydrolase inhibitor that covalently erodimerized by heating to 95 °C for 1 min, followed by cooling modifies the Ser124 catalytic residue (17). Superposition of the to 60 °C for 10 min in the presence of Phusion௡ DNA polymer-

Ag85C-DEP (PDB entry 1DQY) and Ag85C-CyC8␤ structures ase and dNTPs to generate a double-stranded hybrid. A last Downloaded from highlights the similar positioning of the phosphonate groups of step of PCR was performed with the primers used to amplify the the two inhibitors (Fig. 6D). wildtype fbpC2 gene with the hybrid product obtained in the In summary, the data reported here offer a first look at the previous step as template. The mutated fbpC2 gene was finally potent inhibition of the M. tuberculosis Ag85C by cyclipostins cloned like the wildtype gene into pET23b and subjected to

and cyclophostin analogs, compounds that effectively inhibit DNA sequencing to confirm the proper introduction of the http://www.jbc.org/ growth of extracellularly and intracellularly replicating M. mutation. The coding sequence of the gene Rv3130c, which tuberculosis and their mechanism of action. Interestingly, a encodes Tgs1 from M. tuberculosis, was PCR-amplified using recent study indicated that these compounds were also effec- the forward primer 5Ј-GAG GAG CCA TGG aga atc tgta ctt cca tive against clinical isolates of the M. abscessus complex (46), ggg AAT GAA TCA CCT AAC GAC ACT TGA CGC-3Ј (NcoI mostly encountered in cystic fibrosis patients, and known to be site in boldface type, tobacco Etch virus protease cleavage site in at CNRS on February 26, 2018 intrinsically resistant to most antitubercular drugs. We antici- lowercase type) and the reverse primer 5Ј-ACG AGG AAG Ј pate that the high-resolution crystal structure of Ag85C-CyC8␤ CTT TCA CAC AAC CAG CGA TAG CGC T-3 (HindIII site will now open the way to the development, through structure- in boldface type). The PCR amplicon was treated with NcoI and based drug design, of improved inhibitors that target the Ag85 HindIII and ligated to NcoI-HindIII–linearized pET32a. This complex in various pathogenic mycobacteria. plasmid containing the polyhistidine and thioredoxin as fusion Experimental procedures tags in the N-terminal position was used to produce soluble recombinant Tgs1. Mycobacterial strains and growth conditions 2 Expression and purification of the individual Ag85 antigens M. tuberculosis mc 6230 (47) was grown on Middlebrook and MPT51 7H10 agar plates containing OADC (oleic acid, albumin, dex- trose, catalase) enrichment (Difco) and supplemented with 24 All four plasmids harboring the fbpA, fbpB, fbpC2, and fbpC1 ␮g/ml pantothenic acid. Liquid cultures were obtained by genes were used to transform the E. coli C41 (DE3) expression growing mycobacteria in Middlebrook 7H9 (Difco) supple- strain. Transformed bacteria were grown in Luria-Bertani ␮ mented with 10% OADC enrichment, 0.2% (v/v) glycerol, 0.05% medium containing ampicillin (200 g/ml) until the A600 (v/v), Tween 80 (Sigma), 24 ␮g/ml pantothenic acid, and 25 reached 0.6. Bacterial cultures were then placed on icy water for ␮g/ml kanamycin when required. 30 min before induction with 1 mM isopropyl ␤-D-1-thiogalac- topyranoside and further incubated at 16 °C for 20 h. Bacterial Plasmids and DNA manipulations pellets were collected by centrifugation (6,000 ϫ g, 4 °C, 1 h) The fbpC2 gene, encoding Ag85C, was amplified by PCR and resuspended in lysis buffer (50 mM Tris, pH 8.0, 200 mM from M. tuberculosis H37Rv genomic DNA using the forward NaCl, 20 mM imidazole, 5 mM ␤-mercaptoethanol, 1 mM benz- primer 5Ј-CTA CTT CAT ATG TTC TCT AGG CCC GGT amidine). Lysates were sonicated and clarified by centrifugation CTT CCA G-3Ј (NdeI site in boldface type) and the reverse (27,000 ϫ g, 4 °C, 45 min) before purification by nickel-affinity primer 5Ј-GAG ATT CTC GAG AGC AGC AGG CGC AGC chromatography with nickel-nitrilotriacetic acid–Sepharose AGG GG-3Ј (XhoI site in boldface type). The PCR product was beads and elution with lysis buffer containing 250 mM imidaz- cloned into pET23b cut with NdeI and XhoI (New England ole without benzamidine (GE Healthcare). Proteins were next Biolabs), enabling the incorporation of a polyhistidine tag in the dialyzed against 50 mM Tris-HCl, pH 8.0, and 5 mM ␤-mercap- C terminus of the Ag85C protein. The pET23b-fbpA and toethanol buffer and loaded on an anion-exchange HiTrap௡ Q pET23b-fbpC1 constructs carrying the genes encoding Ag85A Fast Flow column (GE Healthcare). The protein was eluted with and MPT51, respectively, were described previously (11). A a linear NaCl gradient. The final step of purification was by

J. Biol. Chem. (2018) 293(8) 2755–2769 2765 Inhibition of Ag85C by cyclipostins and cyclophostin

size-exclusion chromatography using a SuperdexTM 75 10/300 (butanoyl-CoA, octanoyl-CoA, lauroyl-CoA, palmitoyl-CoA, and GL column (GE Healthcare). Proteins were eluted in potassium oleoyl-CoA (Sigma-Aldrich)) in 50 mM potassium phosphate

phosphate buffer (50 mM KH2PO4/K2HPO4, pH 7.6) for DGAT buffer, pH 7.6, containing 2% DMSO. The enzyme concentration activity assessments. Ag85C was eluted in a sodium phosphate in the reaction was 3 ␮M (0.5 ␮M in the case of Tgs1). At the end of ␮ buffer (50 mM NaH2PO4/Na2HPO4, pH 6.0) for mycolyltrans- the assay, an equal volume of DTNB (360 g/ml) was added to the ferase activity assessments and in 50 mM Tris-HCl, pH 8.0, 200 reaction, and the absorbance was measured at 412 nm with a mM NaCl for crystallization experiments and stored at 4 °C. NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific), enabling the calculation of the specific activity of the enzymes Expression and purification of Tgs1 Ϫ1 Ϫ1 (nmol of TNB produced ϫ min ϫ mg of protein ). The M. tuberculosis Tgs1 was overproduced in E. coli and The mycolyltransferase activity assay was performed for 15 purified. Briefly, E. coli BL21 RosettaTM 2 was freshly trans- min at 35 °C based on a procedure described previously (27). formed with pET32a-tgs1. Exponentially growing bacteria cul- Measurements were taken every 15 s using a Multimode Micro- tured in 2 liters of NYZ Broth (BD Biosciences) were cooled on plate Reader POLARstar௡ Omega (BMG Labtech), and the icy water for 30 min, and 1 mM isopropyl ␤-D-1-thiogalactopy- activity of Ag85C was calculated at the maximum rate of the ranoside was added before incubation at 16 °C for 16 h with reaction. The reaction mixture was composed of 50 mM sodium agitation (200 rpm). Bacteria were then collected by centrifuga- phosphate (pH 6.0) containing 2% DMSO, 4 mM trehalose, and tion, the medium was discarded, and the pellet was resus- 12.5 ␮M resorufin butyrate (Sigma-Aldrich). The resorufin pended in lysis buffer containing 10% glycerol, which was main- butyrate was dissolved in DMSO and diluted 100-fold in the

tained for all subsequent buffers used. Lysates were produced reactions. The enzyme concentration in each reaction was 5.5 Downloaded from and subjected to purification via nickel affinity chromatogra- ␮M. Data presented were obtained from three independent phy. His-tagged tobacco etch virus protease was added to the experiments and analyzed by non-linear regression using eluted protein solution at a 1:50 (w/w) ratio, and the mixture GraphPad Prism version 5 software. was dialyzed overnight before again being subjected to nickel- Inhibition of the DGAT and mycolyltransferase activity

affinity chromatography. The fraction that flowed through the http://www.jbc.org/

nickel-nitrilotriacetic acid column, containing tagless Tgs1, CyC7␤, CyC8␤, and CyC17 were synthesized as described pre- was concentrated and subjected to size-exclusion chromatog- viously (31, 50). To study the inhibitory effect on DGAT activ- raphy using a Bio-rad ENrich SEC 650 (Bio-rad) and as buffer ity, a 30 ␮M concentration of either Ag85A, Ag85B, Ag85C, or

100 mM K2HPO4/KH2PO4, pH 7.5, supplemented with 400 mM MPT51 was co-incubated with increasing concentrations of NaCl and 10% glycerol. The fractions containing active Tgs1 CyC7␤, CyC8␤, and CyC17 for 1 h at room temperature in a at CNRS on February 26, 2018 were pooled and concentrated to 0.1 mg/ml. reaction mixture containing 50 mM potassium phosphate buffer (pH 7.6), 10% DMSO, and 0.5 times the critical micelle RNA extraction, cDNA production, and quantitative real-time concentration of n-dodecyl ␤-D-maltoside. Inhibition of the PCR mycolyltransferase activity was determined using 55 ␮M of

Mycobacterial RNA was purified using the Nucleospin RNA Ag85C co-incubated with increasing concentrations of CyC7␤, kit (Macherey Nagel) and assessed for purity on a NanoDrop CyC8␤, and CyC17 for 30 min at room temperature in 50 mM spectrometer and for integrity using a BioAnalyzer (Agilent). sodium phosphate buffer (pH 6.0), 10% DMSO, and 0.5 times the Subsequently, RNA was treated by DNase I (Life Technologies) critical micelle concentration of n-dodecyl ␤-D-maltoside. Ag85C and converted to cDNA using the SuperScript V reverse tran- and Ag85CS124A pretreated or not with the CyC analogs were fur- scriptase kit (Life Technologies). Quantitative real-time PCR ther incubated with 10 ␮M ActivX TAMRA-FP probe (Thermo was performed using the LightCycler 480 SYBR Green master Fisher Scientific) for1hatroom temperature in the darkness. The mix (Roche Applied Science) and primers specific to the house- reaction was stopped by adding 5ϫ Laemmli reducing buffer fol- keeping control gene sigA (forward, 5Ј-TGT ACT CGT GCG lowed by boiling, and proteins were separated by 12% SDS-PAGE. CAG TAA AG-3Ј; reverse, 5Ј-GTC GAA TGT CGG CGT TGA Subsequently, TAMRA FP-labeled proteins were detected by flu- Ј Ј ␭ ␭ TA-3 ) and fbpC2 (forward, 5 -CAG TTT CTA CAC CGA orescent gel scanning (TAMRA: ex 557 nm, em 583 nm) using CTG GTA TC-3Ј; reverse, 5Ј-TCT CTC TGG TAA GGA AGG the Cy௡3filterofaChemiDocMPImager(Bio-Rad)beforestain- TCT C-3Ј). Triplicate data were analyzed by the ⌬⌬Cp method ing of the gels with Coomassie Brilliant Blue dye. with correction for PCR efficiency. Overexpression of Ag85C variants in M. tuberculosis Western blotting The Rv0129c gene was amplified by PCR from M. tuberculo- Lysates of M. tuberculosis mc26230 wildtype or overexpress- sis H37Rv genomic DNA using the forward primer 5Ј-CCC ing Ag85C were prepared and subjected to Western blot anal- AGC TTG TTG ACA GGG TTC GTG-3Ј and the reverse ysis as described previously (49). primer 5Ј-ACC ATG GAT CCC TAG GCG CCC TGG GGC GCG-3Ј (BamHI site in boldface type). After amplification, the DGAT and mycolyltransferase assays PCR product was digested with BamHI (Promega) and cloned The DGAT activity assay was performed for 1 h at 37 °C into MscI/BamHI- digested pMV261, thus placing the Rv0129c using a protocol reported earlier (33). Briefly, the reaction mix- open reading frame under control of the hsp60 promoter to S124A ture was composed of 400 ␮M 1,2-dipalmitoyl-sn-glycerol and a yield pMV261-Ag85C. The pMV261-Ag85C mutant 500 ␮M concentration of the different acyl donor molecules tested plasmid was constructed by the QuikChange method using

2766 J. Biol. Chem. (2018) 293(8) 2755–2769 Inhibition of Ag85C by cyclipostins and cyclophostin pMV261-Ag85C as template, Phusion௡ DNA polymerase with Flexanalysis version 3.0 software (Bruker) with an adapted (Thermo Fisher Scientific), the forward primer 5Ј-GCG GCG analysis method. To eliminate salts from the samples, 10 ␮l of GTG GGT CTT GCG ATG TCG GGC GGT TCC-3Ј, and the each preparation was submitted to a desalting step on a C4 reverse primer 5Ј-GGA ACC GCC CGA CAT CGC AAG ACC Zip-Tip ␮column (Millipore). 1 ␮l of desalted sample was CAC CGC CG-3Ј (Ser 3 Ala mutation in boldface type). The mixed with 1 ␮l of ␣-cyano-4-hydroxycinnamic acid matrix in a DNA sequence was confirmed by DNA sequencing. M. tuber- 50% acetonitrile, 0.3% TFA mixture (1:1, v/v). 1 ␮l was spotted culosis mc26230 was subsequently electrotransformed with on the target, dried, and analyzed with the LP_66 kDa method. pMV261 as a control, pMV261-Ag85C, or pMV261-Ag85CS124A. Peak picking was performed with Flexanalysis version 3.0 soft- ware (Bruker) with an adapted analysis method. Parameters Whole-cell radiolabeling experiments and lipid analysis used were as follows: SNAP peak detection algorithm, S/N To investigate the CyC-induced changes in the lipid profile, threshold fixed to 6, and a quality factor threshold of 30. increasing drug concentrations were added to exponentially 2 Crystallization, data collection, structure determination, and growing M. tuberculosis mc 6230 cultures grown in Middle- refinement brook 7H9 supplemented with OADC enrichment and Tween 80 and 20 ␮g/ml pantothenate for 1 h. Subsequently, metabolic Crystals were grown in sitting drops at 18 °C by mixing 0.8 ␮l labeling of lipids was performed by adding 1 ␮Ci/ml sodium of protein (in 50 mM Tris-HCl, pH 8.0, and 200 mM NaCl) at a [2-14C]acetate (56 mCi/mmol; American Radio Chemicals) for concentration of 8 mg/ml with 0.8 ␮l of reservoir solution con- an additional 6 h at 37 °C. Cells were harvested and delipidated, sisting of 0.2 M magnesium chloride hexahydrate, 0.1 M sodium as described previously (51). The apolar lipid fraction contain- citrate tribasic dihydrate, pH 5.0, and 10% (w/v) polyethylene Downloaded from ing TMM and TDM was separated on a 1D TLC plate using the glycol 20,000. 1-month-old crystals were then soaked for 24 h solvent system chloroform/methanol/water (40:8:1, v/v/v) and with a final concentration in the drop of 5 mM CyC8␤. Crystals revealed after exposure to a film. Similarly, the apolar lipid frac- were fished with a litholoop and flash-cooled in liquid nitrogen tion, which also contains TAG, was separated on a 1D TLC without any cryoprotection. Data collection was performed at

plate using the solvent system petroleum ether/diethyl ether the ID-23.1 beamline at the ESRF synchrotron (Grenoble, http://www.jbc.org/ (90:10, v/v) and revealed after exposure to film. Delipidated France). Data were processed with XDS (53), and the structure cells were further processed to extract the arabinogalactan- was solved by molecular replacement with the structure of bound mycolic acids (52) and analyzed by TLC/autoradiogra- Ag85C as search model (PDB code 3HRH (54)) and using phy using hexane/ethyl acetate (95:5, v/v) run twice in the first Phaser from the PHENIX software suite (55). Manual adjust- dimension followed by exposure to a film to reveal 14C-labeled ments of the model were performed with Coot (56), and the at CNRS on February 26, 2018 mycolic acid methyl esters. structure was refined to 1.8 Å with PHENIX. PDB coordinates and structure factors were deposited in the Protein Data Bank Fluorescent microscopy experiments under accession number 5OCJ. Data collection and refinement Wildtype M. tuberculosis mc26230 or strains harboring statistics are displayed in Table 1. either the pMV261-Ag85C or its variant pMV261-Ag85CS124A were stained with Nile Red fluorescent probe (Interchim), as Author contributions—A.V., J.-F.C., S.C., M.B., and L.K. conceptual- described previously (40). Approximately 7.5 ϫ 107 cells (OD ization; A.V., M.R., P.F., and L.C. data curation; A.V., M.R., P.C.N., 1.5) were collected at 9,000 ϫ g for 3 min, washed twice with 500 P.F., L.C., R.R.P., G.R.G., J.-F.C., S.C., and M.B. investigation; A.V., ␮l of PBS-Tween 0.05%, and resuspended in 300 ␮l of PBS. Nile M.R., P.C.N., P.F., L.C., R.R.P., G.R.G., J.-F.C., S.C., and M.B. meth- Red (15 ␮l of a solution at 0.5 mg/ml solubilized in ethanol) was odology; A.V., M.R., P.F., L.C., C.D.S., J.-F.C., S.C., M.B., and L.K. added to the bacterial suspension, which was further incubated writing-review and editing; P.C.N. visualization; C.D.S. and S.C. for 30 min at 37 °C in the darkness. Cells were then centrifuged, resources; M.B. and L.K. supervision; L.K. funding acquisition; L.K. validation; L.K. writing-original draft; L.K. project administration. washed twice with PBS-Tween 0.05%, and resuspended in 300 ␮l of PBS. Bacteria were spotted between a 170-␮m-thick cov- Acknowledgments—We thank K. Huygen for kindly providing the erslip and a 1.5% agarose-PBS pad. Image acquisition was per- 17/4 and 32/15 monoclonal antibodies, W. R. Jacobs, Jr., for M. tuber- formed with an OLYMPUS FV1000 confocal microscope at culosis mc26230, and P. Santucci for help in fluorescent microscopy ␭ ϭ Ϯگ ␭ ex em 530/590 10 nm, and images were processed and experiments. This work benefited from the facilities and expertise of analyzed using ImageJ. the Platform for Microscopy of IMM. We thank the ESRF and SLS Mass spectrometry beamline staffs for support during data collection. Mass spectrometry analyses were done using the mass spectrometry facility of Marseille Mass analyses were performed on a MALDI-TOF-TOF Proteomics, supported by IBISA (Infrastructures Biologie Santé et Bruker Ultraflex III spectrometer (Bruker Daltonics, Wissem- Agronomie), the Cancéropôle PACA, the Provence-Alpes-Côte d’Azur bourg, France) controlled by the Flexcontrol version 3.0 pack- Region, the Institut Paoli-Calmettes, and the Centre de Recherche en age (Build 51). This instrument was used at a maximum accel- Cancérologie de Marseille. erating potential of 25 kV and was operated in linear mode using the m/z range from 20,000 to 100,000 (LP_66 kDa References method). Five external standards (Protein Calibration Standard 1. Dheda, K., Gumbo, T., Maartens, G., Dooley, K. E., McNerney, R., Murray, II, Bruker Daltonics) were used to calibrate each spectrum to a M., Furin, J., Nardell, E. A., London, L., Lessem, E., Theron, G., van Helden, mass accuracy within 200 ppm. Peak picking was performed P., Niemann, S., Merker, M., Dowdy, D., et al. (2017) The epidemiology,

J. Biol. Chem. (2018) 293(8) 2755–2769 2767 Inhibition of Ag85C by cyclipostins and cyclophostin

pathogenesis, transmission, diagnosis, and management of multidrug-re- 18. Anderson, D. H., Harth, G., Horwitz, M. A., and Eisenberg, D. (2001) An sistant, extensively drug-resistant, and incurable tuberculosis. Lancet Re- interfacial mechanism and a class of inhibitors inferred from two crystal spir. Med. 5, 291–360 CrossRef Medline structures of the Mycobacterium tuberculosis 30 kDa major secretory pro- 2. Barry, C. E., 3rd, and Mdluli, K. (1996) Drug sensitivity and environmental tein (Antigen 85B), a mycolyl transferase. J. Mol. Biol. 307, 671–681 adaptation of mycobacterial cell wall components. Trends Microbiol. 4, CrossRef Medline 275–281 CrossRef Medline 19. Ronning, D. R., Vissa, V., Besra, G. S., Belisle, J. T., and Sacchettini, J. C. 3. Brennan, P. J., and Nikaido, H. (1995) The envelope of mycobacteria. (2004) Mycobacterium tuberculosis antigen 85A and 85C structures con- Annu. Rev. Biochem. 64, 29–63 CrossRef Medline firm binding orientation and conserved substrate specificity. J. Biol. Chem. 4. Mikusová, K., Slayden, R. A., Besra, G. S., and Brennan, P. J. (1995) Bio- 279, 36771–36777 CrossRef Medline genesis of the mycobacterial cell wall and the site of action of ethambutol. 20. Daffé, M. (2000) The mycobacterial antigens 85 complex: from structure Antimicrob. Agents Chemother. 39, 2484–2489 CrossRef Medline to function and beyond. Trends Microbiol. 8, 438–440 CrossRef Medline 5. Vilchèze, C., Wang, F., Arai, M., Hazbón, M. H., Colangeli, R., Kremer, L., 21. Favrot, L., and Ronning, D. R. (2012) Targeting the mycobacterial enve- Weisbrod, T. R., Alland, D., Sacchettini, J. C., and Jacobs, W. R. (2006) lope for tuberculosis drug development. Expert Rev. Anti Infect. Ther. 10, Transfer of a point mutation in Mycobacterium tuberculosis inhA resolves 1023–1036 CrossRef Medline the target of isoniazid. Nat. Med. 12, 1027–1029 CrossRef Medline 22. Jackson, M., McNeil, M. R., and Brennan, P. J. (2013) Progress in targeting 6. Matsumoto, M., Hashizume, H., Tomishige, T., Kawasaki, M., Tsubouchi, cell envelope biogenesis in Mycobacterium tuberculosis. Future Microbiol. H., Sasaki, H., Shimokawa, Y., and Komatsu, M. (2006) OPC-67683, a 8, 855–875 CrossRef Medline nitro-dihydro-imidazooxazole derivative with promising action against 23. Rose, J. D., Maddry, J. A., Comber, R. N., Suling, W. J., Wilson, L. N., and tuberculosis in vitro and in mice. PLoS Med. 3, e466 CrossRef Medline Reynolds, R. C. (2002) Synthesis and biological evaluation of trehalose 7. Tahlan, K., Wilson, R., Kastrinsky, D. B., Arora, K., Nair, V., Fischer, E., analogs as potential inhibitors of mycobacterial cell wall biosynthesis. Car- Barnes, S. W., Walker, J. R., Alland, D., Barry, C. E., 3rd, and Boshoff, bohydr. Res. 337, 105–120 CrossRef Medline

H. I. (2012) SQ109 targets MmpL3, a membrane transporter of treha- 24. Barry, C. S., Backus, K. M., Barry, C. E., 3rd, and Davis, B. G. (2011) ESI-MS Downloaded from lose monomycolate involved in mycolic acid donation to the cell wall assay of M. tuberculosis cell wall antigen 85 enzymes permits substrate core of Mycobacterium tuberculosis. Antimicrob. Agents Chemother. profiling and design of a mechanism-based inhibitor. J. Am. Chem. Soc. 56, 1797–1809 CrossRef Medline 133, 13232–13235 CrossRef Medline 8. Harth, G., Lee, B. Y., Wang, J., Clemens, D. L., and Horwitz, M. A. (1996) 25. Warrier, T., Tropis, M., Werngren, J., Diehl, A., Gengenbacher, M., Schle- Novel insights into the genetics, biochemistry, and immunocytochemistry gel, B., Schade, M., Oschkinat, H., Daffe, M., Hoffner, S., Eddine, A. N., and

of the 30-kilodalton major extracellular protein of Mycobacterium tuber- Kaufmann, S. H. E. (2012) Antigen 85C inhibition restricts Mycobacte- http://www.jbc.org/ culosis. Infect. Immun. 64, 3038–3047 Medline rium tuberculosis growth through disruption of cord factor biosynthesis. 9. Belisle, J. T., Vissa, V. D., Sievert, T., Takayama, K., Brennan, P. J., and Antimicrob. Agents Chemother. 56, 1735–1743 CrossRef Medline Besra, G. S. (1997) Role of the major antigen of Mycobacterium tubercu- 26. Scheich, C., Puetter, V., and Schade, M. (2010) Novel small molecule in- losis in cell wall biogenesis. Science 276, 1420–1422 CrossRef Medline hibitors of MDR Mycobacterium tuberculosis by NMR fragment screening 10. Jackson, M., Raynaud, C., Lanéelle, M. A., Guilhot, C., Laurent-Winter, C., of antigen 85C. J. Med. Chem. 53, 8362–8367 CrossRef Medline

Ensergueix, D., Gicquel, B., and Daffé, M. (1999) Inactivation of the anti- 27. Favrot, L., Grzegorzewicz, A. E., Lajiness, D. H., Marvin, R. K., Boucau, J., at CNRS on February 26, 2018 gen 85C gene profoundly affects the mycolate content and alters the per- Isailovic, D., Jackson, M., and Ronning, D. R. (2013) Mechanism of inhi- meability of the Mycobacterium tuberculosis cell envelope. Mol. Micro- bition of Mycobacterium tuberculosis antigen 85 by ebselen. Nat. Com- biol. 31, 1573–1587 CrossRef Medline mun. 4, 2748 Medline 11. Kremer, L., Maughan, W. N., Wilson, R. A., Dover, L. G., and Besra, G. S. 28. Favrot, L., Lajiness, D. H., and Ronning, D. R. (2014) Inactivation of the (2002) The M. tuberculosis antigen 85 complex and mycolyltransferase Mycobacterium tuberculosis antigen 85 complex by covalent, allosteric activity. Lett. Appl. Microbiol. 34, 233–237 CrossRef Medline inhibitors. J. Biol. Chem. 289, 25031–25040 CrossRef Medline 12. Armitige, L. Y., Jagannath, C., Wanger, A. R., and Norris, S. J. (2000) 29. Nguyen, P. C., Delorme, V., Bénarouche, A., Martin, B. P., Paudel, R., Disruption of the genes encoding antigen 85A and antigen 85B of Myco- Gnawali, G. R., Madani, A., Puppo, R., Landry, V., Kremer, L., Brodin, P., bacterium tuberculosis H37Rv: effect on growth in culture and in macro- Spilling, C. D., Cavalier, J.-F., and Canaan, S. (2017) Cyclipostins and cy- phages. Infect. Immun. 68, 767–778 Medline clophostin analogs as promising compounds in the fight against tubercu- 13. Nguyen, L., Chinnapapagari, S., and Thompson, C. J. (2005) FbpA-depen- losis. Sci. Rep. 7, 11751 CrossRef Medline dent biosynthesis of trehalose dimycolate is required for the intrinsic mul- 30. Dupont, C., Viljoen, A., Dubar, F., Blaise, M., Bernut, A., Pawlik, A., tidrug resistance, cell wall structure, and colonial morphology of Myco- Bouchier, C., Brosch, R., Guérardel, Y., Lelièvre, J., Ballell, L., Herrmann, bacterium smegmatis. J. Bacteriol. 187, 6603–6611 CrossRef Medline J.-L., Biot, C., and Kremer, L. (2016) A new piperidinol derivative targeting 14. Puech, V., Guilhot, C., Perez, E., Tropis, M., Armitige, L. Y., Gicquel, B., mycolic acid transport in Mycobacterium abscessus. Mol. Microbiol. 101, and Daffé, M. (2002) Evidence for a partial redundancy of the fibronectin- 515–529 CrossRef Medline binding proteins for the transfer of mycoloyl residues onto the cell wall 31. Point, V., Malla, R. K., Diomande, S., Martin, B. P., Delorme, V., Carriere, arabinogalactan termini of Mycobacterium tuberculosis: redundancy of F., Canaan, S., Rath, N. P., Spilling, C. D., and Cavalier, J.-F. (2012) Syn- fibronectin-binding proteins in M. tuberculosis. Mol. Microbiol. 44, thesis and kinetic evaluation of cyclophostin and cyclipostins phospho- 1109–1122 CrossRef Medline nate analogs as selective and potent inhibitors of microbial lipases. J. Med. 15. Wilson, R. A., Rai, S., Maughan, W. N., Kremer, L., Kariuki, B. M., Harris, Chem. 55, 10204–10219 CrossRef Medline K. D. M., Wagner, T., Besra, G. S., and Fütterer, K. (2003) Crystallization 32. Liu, Y., Patricelli, M. P., and Cravatt, B. F. (1999) Activity-based protein and preliminary X-ray diffraction data of Mycobacterium tuberculosis profiling: the serine hydrolases. Proc. Natl. Acad. Sci. U.S.A. 96, FbpC1 (Rv3803c). Acta Crystallogr. D Biol. Crystallogr. 59, 2303–2305 14694–14699 CrossRef Medline CrossRef Medline 33. Elamin, A. A., Stehr, M., Spallek, R., Rohde, M., and Singh, M. (2011) The 16. Wilson, R. A., Maughan, W. N., Kremer, L., Besra, G. S., and Fütterer, K. Mycobacterium tuberculosis Ag85A is a novel diacylglycerol acyltrans- (2004) The structure of Mycobacterium tuberculosis MPT51 (FbpC1) de- ferase involved in lipid body formation. Mol. Microbiol. 81, 1577–1592 fines a new family of non-catalytic ␣/␤ hydrolases. J. Mol. Biol. 335, CrossRef Medline 519–530 CrossRef Medline 34. Lakey, D. L., Voladri, R. K., Edwards, K. M., Hager, C., Samten, B., Wallis, 17. Ronning, D. R., Klabunde, T., Besra, G. S., Vissa, V. D., Belisle, J. T., and R. S., Barnes, P. F., and Kernodle, D. S. (2000) Enhanced production of Sacchettini, J. C. (2000) Crystal structure of the secreted form of antigen recombinant Mycobacterium tuberculosis antigens in Escherichia coli by 85C reveals potential targets for mycobacterial drugs and vaccines. Nat. replacement of low-usage codons. Infect. Immun. 68, 233–238 CrossRef Struct. Biol. 7, 141–146 CrossRef Medline Medline

2768 J. Biol. Chem. (2018) 293(8) 2755–2769 Inhibition of Ag85C by cyclipostins and cyclophostin

35. Viljoen, A., Blaise, M., de Chastellier, C., and Kremer, L. (2016) 85C from Mycobacterium tuberculosis by ebselen derivatives. ACS Infect. MAB_3551c encodes the primary triacylglycerol synthase involved in lipid Dis. 3, 378–387 CrossRef Medline accumulation in Mycobacterium abscessus. Mol. Microbiol. 102, 611–627 46. Nguyen, P. C., Madani, A., Santucci, P., Martin, B. P., Paudel, R., Delattre, CrossRef Medline S., Herrmann, J.-L., Spilling, C. D., Kremer, L., Canaan, S., and Cavalier, 36. Daniel, J., Deb, C., Dubey, V. S., Sirakova, T. D., Abomoelak, B., Morbi- J.-F. (2017) Cyclophostin and cyclipostins analogs, new promising mole- doni, H. R., and Kolattukudy, P. E. (2004) Induction of a novel class of cules to treat mycobacterial-related diseases. Int. J. Antimicrob. Agents diacylglycerol acyltransferases and triacylglycerol accumulation in Myco- 10.1016/j.ijantimicag.2017.12.001 CrossRef Medline bacterium tuberculosis as it goes into a dormancy-like state in culture. J. 47. Sambandamurthy, V. K., Derrick, S. C., Hsu, T., Chen, B., Larsen, M. H., Bacteriol. 186, 5017–5030 CrossRef Medline Jalapathy, K. V., Chen, M., Kim, J., Porcelli, S. A., Chan, J., Morris, S. L., and ⌬ ⌬ 37. Huygen, K., Lozes, E., Gilles, B., Drowart, A., Palfliet, K., Jurion, F., Roland, Jacobs, W. R. (2006) Mycobacterium tuberculosis RD1 panCD: a safe I., Art, M., Dufaux, M., and Nyabenda, J. (1994) Mapping of TH1 helper and limited replicating mutant strain that protects immunocompetent T-cell epitopes on major secreted mycobacterial antigen 85A in mice in- and immunocompromised mice against experimental tuberculosis. Vac- fected with live Mycobacterium bovis BCG. Infect. Immun. 62, 363–370 cine 24, 6309–6320 CrossRef Medline Medline 48. Heckman, K. L., and Pease, L. R. (2007) Gene splicing and mutagenesis by 38. Daniel, J., Maamar, H., Deb, C., Sirakova, T. D., and Kolattukudy, P. E. PCR-driven overlap extension. Nat. Protoc. 2, 924–932 CrossRef Medline (2011) Mycobacterium tuberculosis uses host triacylglycerol to accumu- 49. Kremer, L., Dover, L. G., Morbidoni, H. R., Vilchèze, C., Maughan, W. N., late lipid droplets and acquires a dormancy-like phenotype in lipid-loaded Baulard, A., Tu, S.-C., Honoré, N., Deretic, V., Sacchettini, J. C., Locht, C., Jacobs, W. R., Jr., and Besra, G. S. (2003) Inhibition of InhA activity, but macrophages. PLoS Pathog. 7, e1002093 CrossRef Medline not KasA activity, induces formation of a KasA-containing complex in 39. Garton, N. J., Christensen, H., Minnikin, D. E., Adegbola, R. A., and Barer, mycobacteria. J. Biol. Chem. 20547–20554 CrossRef Medline M. R. (2002) Intracellular lipophilic inclusions of mycobacteria in vitro 278, 50. Vasilieva, E., Dutta, S., Malla, R. K., Martin, B. P., Spilling, C. D., and and in sputum. Microbiol. Read. Engl. 148, 2951–2958 CrossRef Medline

Dupureur, C. M. (2015) Rat hormone sensitive lipase inhibition by cycli- Downloaded from 40. Dhouib, R., Ducret, A., Hubert, P., Carrière, F., Dukan, S., and Canaan, postins and their analogs. Bioorg. Med. Chem. 23, 944–952 CrossRef S. (2011) Watching intracellular lipolysis in mycobacteria using time Medline lapse fluorescence microscopy. Biochim. Biophys. Acta 1811, 234–241 51. Kremer, L., Guérardel, Y., Gurcha, S. S., Locht, C., and Besra, G. S. (2002) CrossRef Medline Temperature-induced changes in the cell-wall components of Mycobac- 41. Kalscheuer, R., and Steinbüchel, A. (2003) A novel bifunctional wax ester terium thermoresistibile. Microbiol. Read. Engl. 148, 3145–3154 CrossRef synthase/acyl-CoA:diacylglycerol acyltransferase mediates wax ester and Medline http://www.jbc.org/ triacylglycerol biosynthesis in Acinetobacter calcoaceticus ADP1. J. Biol. 52. Coxon, G. D., Craig, D., Corrales, R. M., Vialla, E., Gannoun-Zaki, L., and Chem. 278, 8075–8082 CrossRef Medline Kremer, L. (2013) Synthesis, antitubercular activity and mechanism of 42. Peyron, P., Vaubourgeix, J., Poquet, Y., Levillain, F., Botanch, C., Bardou, resistance of highly effective thiacetazone analogues. PLoS One 8, e53162 F., Daffé, M., Emile, J.-F., Marchou, B., Cardona, P.-J., de Chastellier, C., CrossRef Medline and Altare, F. (2008) Foamy macrophages from tuberculous patients’ 53. Kabsch, W. (2010) Integration, scaling, space-group assignment and post- granulomas constitute a nutrient-rich reservoir for M. tuberculosis persis- refinement. Acta Crystallogr. D Biol. Crystallogr. 66, 133–144 CrossRef tence. PLoS Pathog. 4, e1000204 CrossRef Medline Medline at CNRS on February 26, 2018 43. Santucci, P., Bouzid, F., Smichi, N., Poncin, I., Kremer, L., De Chastellier, 54. Sanki, A. K., Boucau, J., Umesiri, F. E., Ronning, D. R., and Sucheck, S. J. C., Drancourt, M., and Canaan, S. (2016) Experimental models of foamy (2009) Design, synthesis and biological evaluation of sugar-derived esters, macrophages and approaches for dissecting the mechanisms of lipid ac- ␣-ketoesters and ␣-ketoamides as inhibitors for Mycobacterium tubercu- cumulation and consumption during dormancy and reactivation of tuber- losis antigen 85C. Mol. Biosyst. 5, 945–956 CrossRef Medline culosis. Front. Cell. Infect. Microbiol. 6, 122 Medline 55. Adams, P. D., Afonine, P. V., Bunkóczi, G., Chen, V. B., Echols, N., Headd, 44. Lehmann, J., Cheng, T.-Y., Aggarwal, A., Park, A. S., Zeiler, E., Raju, R. M., J. J., Hung, L.-W., Jain, S., Kapral, G. J., Grosse Kunstleve, R. W., McCoy, Akopian, T., Kandror, O., Bach, N. C., Sacchettini, J. C., Moody, D. B., A. J., Moriarty, N. W., Oeffner, R. D., Read, R. J., Richardson, D. C., et al. Rubin, E. J., and Sieber, S. A. (2018) An antibacterial ␤-lactone kills My- (2011) The Phenix software for automated determination of macromolec- cobacterium tuberculosis by infiltrating mycolic acid biosynthesis. Angew. ular structures. Methods 55, 94–106 CrossRef Medline Chem. Int. Ed. Engl. 57, 348–353 Medline 56. Emsley, P., Lohkamp, B., Scott, W. G., and Cowtan, K. (2010) Features and 45. Goins, C. M., Dajnowicz, S., Thanna, S., Sucheck, S. J., Parks, J. M., and development of Coot. Acta Crystallogr. D Biol. Crystallogr. 66, 486–501 Ronning, D. R. (2017) Exploring covalent allosteric inhibition of antigen CrossRef Medline

J. Biol. Chem. (2018) 293(8) 2755–2769 2769 Results – Article 4

Article 4: A biochemical and structural characterization of TesA a major thioesterase requires for PDIM and PGL syntheses in M. tuberculosis

TesA a putative thioesterase annotated Rv2928 and proposed to be involved in the biosynthesis pathway of both PDIM and PGL (Alibaud et al., 2011; Chavadi et al., 2011) has been identified as a potential target of CyC17 (Scheme 3).

Scheme 3. (A) The DIM gene cluster in M. tb. The tesA gene is in red arrow. (B) Scheme of phthiocerol dimycocerosate (DIM) and PGL biosynthesis in M. tb. The role of TesA interacting with the polyketide synthase PpsE is emphasized. This interaction allows the release of both phthiocerol and phenolphthiocerol that form the skeleton of DIM and PGL respectively.

Basically, gene encoding tesA is located in the gene cluster responsible for the biosynthesis and export of phthiocerol dimycocerosate (PDIM) and a second group of glycosylated compounds known as phenolphthiocerol dimycocerosates also called phenolic glycolipids (PGLs) (Rao and Ranganathan, 2004; Alibaud et al., 2011). These two complex lipids are non-covalently bound to other cell wall components, and have been shown to be important virulence factors and involved in host-pathogen interaction via multiple mechanisms in various mycobacterial pathogens (Camacho et al., 1999; Cox et al., 1999). The presence of PDIM is necessary for the optimal intracellular growth and contributes to the ability of M. tb to prevent phagosome

110 Results – Article 4 acidification (Astarie-Dequeker et al., 2009). Similarly to PDIM, PGLs are thought to be involved in virulence. However, as PGL are missing from many clinical isolates of M. tb, these lipids are unlikely to be essential for virulence but may strengthen the virulence of various strains such as some M. tb isolates of the East Asian/Beijing lineage (Daffe et al., 1987; Constant et al., 2002; Reed et al., 2004; Huet et al., 2009). TesA from M. tb is thus supposed to play an important role in vivo by increasing the virulence of pathogenic mycobacteria. Nevertheless although many ex vivo experiments have already been reported to decipher the physiological role of TesA, none or few information are however available regarding the biochemical and structural characterization of this enzyme. In this article, we successfully produced recombinant TesA in E. coli and realized its biochemical characterization as well as its structural determination by solving its X-ray tridimensional structure at 2.6 Å. We found that TesA, which exhibit both esterase and thioesterase activities, is belonging to the α/β-hydrolase fold family. TesA also possesses a mobile lid domain that is supposed to be stabilized in open conformation when a substrate is present inside the enzyme catalytic site. In addition, we demonstrated that in the TesA-CyC17 complex, the inhibitor was covalently linked to catalytic Ser104 residue, as confirmed by MALDI-TOF mass spectrometry.

111 Biochemical and structural characterization of TesA, a major thioesterase required for outer-envelope lipid biosynthesis in M. tuberculosis Phuong Chi Nguyen#, Van Son Nguyen¦, Benjamin P. Martin±, Patrick Fourquet¤, Luc Camoin¤, Chistopher D. Spilling±, Jean-François Cavalier#, Christian Cambillau¦ and Stéphane Canaan#,*

#Aix Marseille Univ, CNRS, LISM, IMM FR3479, Marseille, France ¦Aix Marseille Univ, CNRS, AFMB, Marseille, France ¤Aix Marseille Univ, CNRS, INSERM, Institut Paoli-Calmettes, CRCM, Marseille Protéomique, Marseille, France ±Department of Chemistry and Biochemistry, University of Missouri, One University Boulevard, St. Louis, MO 63121, USA *Corresponding author: [email protected], phone +33 491164093

Short title: M. tuberculosis TesA 3D structure Abstract With the high number of patients infected by tuberculosis (TB) and the sharp increase of drug resistant TB cases, developing new drugs to fight this disease has become increasingly urgent. In this context, analogs of the naturally occurring enolphosphates Cyclipostins and Cyclophostin (CyC analogs) offer new therapeutic opportunities. The CyC analogs display potent activity both in vitro and in infected macrophages against several pathogenic mycobacteria including Mycobacterium tuberculosis and Mycobacterium abscessus. Interestingly, these CyC inhibitors target several enzymes with active site serine or cysteine residues that play key roles in mycobacterial lipid and cell wall metabolism. Among them, TesA, a putative thioesterase involved in the synthesis of phthiocerol dimycocerosates (PDIMs) and phenolic glycolipids (PGLs), has been identified. These two lipids (PDIM and PGL) are non-covalently bound to the outer cell wall in several human pathogenic mycobacteria and are important virulence factors. Herein, we used biochemical and structural approaches to validate TesA as an effective pharmacological target of the CyC analogs. We confirmed both thioesterase and esterase activities of TesA, and showed that the most active inhibitor CyC17 binds covalently to the catalytic Ser104 residue leading to a total loss of enzyme activity. These data were supported by the X-ray structure, obtained at a 2.6 Å resolution, of a complex in which CyC17 is bound to TesA. Our study provides evidence that CyC17 inhibits the activity of TesA, thus paving the way to a new strategy for impairing the PDIM and PGL biosynthesis, potentially decreasing the virulence of associated mycobacterial species.

Keywords: lipids, virulence, Cyclipostins, Cyclophostin, PDIM, PGL, thioesterase inhibitor bacilli to hide from the immune system Introduction inside granulomas, it is challenging to Mycobacterium tuberculosis (M. efficiently treat the disease. Indeed, the M. tuberculosis) the causative agent of tuberculosis cell wall is composed of a large tuberculosis (TB) is responsible for an number of hydrophobic molecules having estimated 10.4 million new cases and 1.7 up to 90 carbon atoms chain length. This million deaths across the globe, as reported thick lipid rich structure is responsible for by WHO in 2017 [1]. With the emergence the high impermeability of the cell wall and of multi and extensive drug resistance thus to the inherent resistance of M. strains, TB remains the leading cause of tuberculosis to numerous macromolecules death from an infectious disease. Due to the including many antibiotics [2]. Among very complex composition of M. these outer membrane/surface-exposed tuberculosis cell wall and the capacity of the hydrophobic lipids, are the phthiocerol

dimycocerosate (PDIM), and a second enzymes in a complex proteome [19]. group of glycosylated compounds known as Among the 23 proteins identified via mass phenolic glycolipids (PGLs) and named spectrometry experiments [19], TesA a phenolphthiocerol dimycocerosates. These putative thioesterase (annotated Rv2928) two complex lipids, esterified with proposed to be involved in the biosynthesis multimethyl-branched long-chain fatty pathway of both PDIM and PGL [10, 20] acids, are non-covalently bound to other has been identified. cell wall components and have been shown TesA, which is localized in the to be important virulence factors [3]. membrane fraction of the bacteria [21, 22], Consistent with their main role in the is non-essential for the in vitro growth of M. permeability of the cell envelope and tuberculosis. However, the virulence of a pathogenicity [4], PDIMs are apparently ΔtesA deletion mutant strain of M. produced by all virulent clinical isolates of tuberculosis was strongly attenuated in M. tuberculosis. In contrast, the ability to infected cells, becoming more susceptible synthesize PGL has only been retained by to antibiotics [9, 10, 20, 23]. This result has M. canettii and some M. tuberculosis been attributed to the fact that the ΔtesA isolates of the East Asian/Beijing lineage mutant in M. tuberculosis H37Rv was [5-8]. As an example, it has been shown that defective in PDIMs [24]. More recently, PGLs could be in part responsible for the TesA has been proposed to release resistance to antibiotics of M. tuberculosis phtiocerol and phenolpthiocerol molecules strains of the W-Beijing family [8, 9]. that further condense with mycoserosic acid Moreover, the virulence of PGL/PDIM by the action of the PapA5 enzyme to form deficient-bacteria was strongly attenuated PDIM as well as PGL skeletons [9, 10]. in a mouse model suggesting a major impact Thus M. tuberculosis TesA represents a of these lipids on the viability of the major contributing partner in the synthesis mycobacteria in in vivo conditions [8-12]. of both PDIM and PGL by interacting with Hence, deciphering the PDIM and the PGL PpsE inside the polyketide synthase cluster biosynthesis pathway could be helpful to [10, 25] (Figure 1). Therefore, this pathway learn more about the cell wall assembly of may represent an interesting source of new clinical pathogenic bacteria, and to find new targets for the development of novel TB specific therapeutic targets. drugs, with particular relevance to the New antibiotics targeting one or emergence of drug resistant strains. more proteins involved in the synthesis of Consequently, M. tuberculosis TesA mycobacterial lipids would thus represent a is believed to play an important role in vivo promising way to fight and control TB [9, by increasing the virulence of pathogenic 13]. In this context, monocyclic mycobacteria. However, although many ex enolphosphate and phosphonate analogs vivo experiments have been reported to (CyC) of natural Cyclipostins and decipher the physiological role of TesA, Cyclophostin [14-17] represent a new little or no information is available family of potent and selective inhibitors regarding its biochemical and structural against mycobacteria only [18]. These CyC characterization. analogs act as powerful antitubercular The present study was undertaken to agents affecting growth of M. tuberculosis decipher, the selective inhibition of TesA both in vitro and in infected macrophages activity by the CyC17 inhibitor using a with very low toxicity towards mammalian combination of biochemical and structural host cells [19]. To identify the putative approaches. Remarkably, we report here the target(s) of the best inhibitor CyC17, an first crystal structure of this M. tuberculosis activity-based protein profiling (ABPP) enzyme in complex with the CyC17 approach was used, leading to the capture of covalently bound to the active site Ser104 several active Ser- and Cys-containing residue. This crystal structure may help

optimizing new CyC scaffolds with higher of 257 ±3.3 mU/mg. However, with pNP specificity and potency against M. esters carrying more than 8 carbon atoms, tuberculosis. this specific activity dropped sharply to 43 ±1.9, 23 ±0.3 and 15 ±0.8 mU/mg with Results pNP-C12, pNP-C14 and pNP-C16, Expression and purification of respectively (Figure 2B). Since similar recombinant TesA and mutant specific activities were obtained with pNP- Both TesA and TesAS104A were C5 (194 ±2.0 mU/mg) and pNP-C8 produced in recombinant forms and (162 ±7.7 mU/mg), this latter medium chain ester was chosen for additional purified to affinity, leading to 5-10 mg of pure recombinant proteins per liter of experiments, including inhibition tests with culture. Their purity as well as the expected the CyCs (see further) and Michaelis- Menten kinetic constants determination. In molecular weight (~29 kDa) was confirmed app this case, TesA showed apparent m and by 12% SDS-PAGE (Figure 2A). TesA as app well as TesAS104A were concentrated up to max values of around 1.5 ±0.057 μM and 3823 ±102 µM.s -1, respectively.� The 15 mg/mL, the highest concentration used app apparent� turnover number ( cat ) and the for crystallographic experiments (Figure app app 2A). specificity constant ( cat / m ) were then TesA has been previously predicted evaluated to be 445 ±11.5 s-1� and 288 ±3.3 to be a thioesterase [20]. However, it has μM.s-1, respectively (Figure� � S1). been reported that this family of enzymes High variations in the enzyme can also act as esterases, phospholipases, activity induced by the presence of lysophospholipases as well as lipases [26]. surfactant (i.e., Triton X-100) are usually To investigate whether TesA exhibits such associated with a conformational change of a diverse activity, a broad range of the enzyme in response to the new substrates have been tested including environment but without losing its catalytic Palmitoyl-coenzyme A, para-nitrophenyl properties. More precisely, such (pNP) esters carrying various carbon chain modifications often happen when a lid lengths, phospholipids and triacylglycerols. covering the active site is present. It is now As a control for phospholipase activity we well established in the case of lipolytic used the cutinase-like protein Cut6 enzymes, that surfactants can both help in (Rv3802c), a Pks13-associated thioesterase solubilizing medium to long chains proposed to be involved in mycolic acid substrates (here pNP-C8 to pNP-C16) by biosynthesis [27, 28]. TesA did not express forming mixed micelles, but can also any phospholipase, lysophospholipase, nor promote and stabilize the lid opening triacylglycerol lipase activities. However, process making the active site accessible to TesA displayed hydrolytic activity on ester solvent [29]. As a consequence of such lid substrates (Figure 2B) being more active opening, an enhanced enzyme activity will on pNP esters with short carbon chain be obtained whatever the length of the pNP lengths (C2-C5) than with the those bearing ester chain used, as observed here with medium and long carbon chain lengths (i.e., TesA for the soluble pNP-C2 to the poorly C8-C18). Indeed, the hydrolysis capacity of soluble pNP-C16. TesA decreased with increasing alkyl chain Notably, this latter pNP assay also lengths. Moreover in all cases, the presence allows the discrimination of esterases from of Triton X-100 led to a significant increase true lipases, which hydrolyze short-chains in the lipolytic activity of TesA, from 2-fold esters, and medium- to long-chain esters, with pNP-C2 and up to 127-fold when respectively [30]. Here, the absence of using pNP-C16. Under these conditions, lipase activity indicates that TesA behaves pNP-C2 having the shortest carbon chain as an esterase, in contrast to LipY or cfp21, length showed a maximum specific activity

the only M. tuberculosis true triacylglycerol To further characterize the observed lipases characterized so far [31, 32]. inhibitory effect, native TesA (15 µM) was As expected from the genome incubated for 30 min with 300 µM of CyC17 annotation and previous in vivo experiment (i.e., inhibitor molar excess xI =20). As [10, 20], TesA also exhibited thioesterase expected a complete loss of activity was activity with a specific activity of 6.82 observed when comparing the pre-treated vs mU/mg on Palmitoyl-CoA. It is noteworthy non-treated TesA. The resulting TesA- that this value is nearly 19 times lower than CyC17 adduct was further incubated for 1 h that of Cut6 used as positive control on the with 10 μM TAMRA-FP, a fluorescent same substrate (Figure 2C). probe known to bind to all enzymes with Finally, no enzyme activity was catalytic serine or cysteine [34, 35]. Equal obtained with any of the substrates for the amounts of adduct were separated by SDS- inactive TesAS104A mutant protein, where PAGE and visualized by Coomassie the catalytic Ser104 contained in the staining (Figure 4B, upper panel) or in-gel consensus sequence GHSMG was replaced fluorescence for TAMRA detection (Figure by an Ala. 4B, middle panel). Pre-treatment with Effects of Cyclipostins and Cyclophostin CyC17 (Figure 4B) resulted in a total loss of fluorescence intensity suggesting that the inhibitors CyC17, CyC6β, CyC7β, CyC8β and Orlistat on TesA activity reaction with the TAMRA probe was strongly impaired in the TesA-CyC17 The CyCs are a new class of potent adduct. To prove the role of catalytic anti-tubercular compounds exhibiting an Ser104 in TAMRA labelling, this residue activity against enzymes with serine or was mutated to an Ala residue. The cysteine in their active site [18, 19]. We resulting TesAS104A mutant was exposed to investigated on the ability of CyC17 to TAMRA-FP (Figure 4B, middle panel) efficiently inhibit the enzyme activity of and as expected, no fluorescence was TesA to confirm this enzyme an effective detected confirming that the Ser104 is the target of the inhibitor. In addition, CyC7β catalytic nucleophile required for the which has been found active against both binding of the probe. extracellular and intramacrophagic M. The TesA-CyC17 complex was tuberculosis, and CyC6β & CyC8β active on further purified on a gel filtration column infected macrophages [18, 19], were also for crystallization. Size exclusion tested to assess their ability to inhibit TesA chromatography coupled with online multi- activity. Orlistat which has been previously angle laser light scattering / quasi-elastic reported as an inhibitor of TesA was used as light scattering / refractive index (SEC- positive control [33]. The chemical UV/MALS/QELS/RI) demonstrated that structures of the four CyCs analogs used in TesA has a molecular mass of 29 kDa this study are provided in Figure 3. (Figure 4C – bright orange curve), which is In contrast to CyC6β, CyC7β and in agreement with the theoretical molecular CyC8β, which exhibited very weak or no weight of the monomer of 29,077 Da. In inhibitory effect up to a high molar excess contrast, the peak of the TesA-CyC17 xI = 200, a dose-dependent inhibition was complex was observed at 60 kDa (Figure observed with CyC17 and Orlistat (Figure 4C – grey curve). This increase in 4A). Both compounds were found to react molecular weight cannot result from almost stoichiometrically with pure TesA, binding of the CyC17 alone (Mw 446 Da). It as confirmed by their respective inhibitor is reasonable to attribute the molecular molar excess leading to 50% enzyme weight increase to a dimeric form of the inhibition, i.e. xI50 values of around 4.5 and enzyme (Figure 2). This finding suggests 12.4. that some conformational changes might occur which lead to a dimerization. Such

behavior has already been observed in the catalytic serine several lipolytic enzymes [36] and thanks to (I94DDPVAFFGHS104MGGMLAFEVALR the resolution of TesA 3D-structure a clear 116) with a mass of 2480.27 Da (Figure 5C, explanation can be proposed. upper panel). Similarly, TesAS104A It is noteworthy, however, that the produced a 2464.34 Da fragment (Figure traces indicating the molar mass across the 5D, upper panel). peaks were found to vary significantly; There was no peptide modification from around 24-38 kDa and 44-67 kDa for observed when comparing TesAS104A and S104A the monomer and the dimer, respectively the TesA -CyC17 mixture, confirming (Figure 4C). Seeing that each peak once again the role of Ser104 in the catalytic represents purified enzymes and is eluted process (Figure 5D, lower panel). from the SEC-column with a Gaussian In contrast with native TesA, the distribution, such mass range should not peptide fragment incorporating Ser104 was occur if the proteins are present simply as a absent in the TesA-CyC17 modified protein monomer or dimer in solution. From these suggesting mass modification (Figure 5D, findings, one can hypothesized that this lower panel). After digestion, the Ser104 observed molecular weight variation across incorporating peptide is expected to be the curve might result from transient shifted by a mass increment of +316.9 Da interaction between TesA due to the addition of the CyC17 inhibitor. monomers/dimers with the column. However, no isotopic peptides with the Mass spectrometry analysis desire mass shift were detected in the Mass spectrometry MALDI-TOF digested TesA-CyC17 complex mass analyses were performed on TesA and spectra. S104A TesA in presence or absence of CyC17 It is noteworthy, however, that the to confirm the covalent binding of this observed 316.9 Da mass shift increment in phosphate inhibitor to the active site Ser104 global mass was 129.38 Da lower than the residue (Figure 5). A mass increment of expected CyC17 theoretical molecular mass +316.9 Da was observed in the presence of of 446.28 Da (Figure 5). This size CyC17 within the global mass of untreated difference may arise from the specific TesA (Figure 5A). Conversely, no change chemical properties of phosphate inhibitor, in global mass was observed with the as already described in a previous study inactive TesAS104A protein, as expected with the Antigen 85 (Ag85) complex (i.e., (Figure 5B). Ag85A, Ag85B, and Ag85C) [38]. These Peptide mass fingerprinting (PMF) is secreted enzymes are responsible for the a useful tool to identify proteins and probe biosynthesis of trehalose dimycolate mass modifications within specific peptides (TDM) and mycolylation of [37]. This technique was used to show that arabinogalactan, two essential lipid TesA catalytic Ser104 covalently binds to components of the mycobacterial cell wall CyC17 (Figure 5C-D), excluding any non- [39, 40]. In all cases, the new phosphate covalent inhibition (i.e. fixation of the triester generated upon the nucleophilic inhibitor near the active site, blocking its attack of the catalytic serine at the S104A access). Both TesA and TesA exposed phosphorus center of CyC17, can be rapidly to CyC17 were digested (in-gel) using hydrolyzed in the presence of water trypsin to generate peptide mixtures which molecules, resulting in the cleavage and the were then analyzed using MALDI-TOF release of the methyl 2-acetyl- mass spectrometry. Trypsin cuts 4-hydroxybutyrate group (i.e., 124.1 Da), preferentially after positively charged accounting for the molecular mass residues like Arg or Lys. In absence of discrepancy observed experimentally inhibitor, tryptic digest of native TesA (Figure 5E). Finally, this transiently produced a 23 amino acid peptide harboring obtained penta-coordinate phosphorus

adduct may slowly rearrange to reach its involving the segment 135-180. In stable thermodynamic state. Such particular, the loop 135-145 exhibited two mechanism is not only perfectly in line with different tracks (Figure S3). the data generated by the proteomic The structure of TesA consists of a experiments, but above all, it was further core of 7 α-helices around 6 stranded confirmed by the 3D structure determined parallel β-sheets, upon which protrudes a below. domain formed of an extended stretch Overall, these data confirm the followed by 2 α-helices, with connectivity formation of a covalent complex between β1-α1-β2-α2-β3-α3-β4-(ext-α4-α5)-β5-α6- the TesA catalytic Ser104 and CyC17. β7-α7 (Figure 6A). Several structurally Moreover, our results are consistent with similar folds haves been previously the known and classical mechanism of reported and belong to the thioesterase fold action of phosphate based inhibitors [19, of the α/β-hydrolase family [43]. TesA 38]. structure is characterized by a central parallel β-sheet and a Ser-His-Asp catalytic Crystal structure of TesA-CyC17 triad involving here Ser104, His236 and Crystallization trials of TesA alone Asp208. However, the true hallmark of a or in the presence of α/β-hydrolase are the catalytic serine loop phosphonate/phosphate inhibitors were signature G-X-S-X-G and its special performed. Only the TesA-CyC17 complex conformation, with φ and ψ angles values of led to diffracting crystals. Datasets were ~60¡ and ‒120¡, respectively [43]. These collected at the European Synchrotron characteristics are indeed observed in TesA. Radiation facility (ESRF, Grenoble, Serine 104 is located in a loop with the France) or at SOLEIL (St Aubin, France). sequence G-H-S-M-G and has the expected The structure was solved by molecular φ and ψ angles (Figure 6B). replacement using the structure of the core TesA was crystallized with the domain of type II thioesterase RifR from CyC17 phosphonate inhibitor, which was Amycolatopsis mediterranei [41] as found to be covalently bound to Ser104. template model (PDB id: 3FLB), and The hydrocarbon tail of CyC17 could be refined to 2.6 Å resolution (Table 1). only traced to the first 4 carbon atoms of 16. Crystals belong to space group C2221 and The rest of the chain was not visible in the contain 8 molecules per asymmetric unit. electron density map, probably due to Analysis of the contacts between monomers disordering (Figure S4). In many with the PISA server [42] indicated that four complexes of lipases or esterases with groups of dimers can be identified (AC, BD, phosphate, an oxygen atom of the EG, FH). Moreover, the TesA dimer phosphate moiety is involved in an observed in solution seems to result from hydrogen bond to the so-called "oxyanion the hydrophobic association of two - hole" of the enzyme, one component being helices (i.e., ext-α4-α5) and the active site the NH group of the residue following the area of two symmetry related monomers catalytic serine [44]. In the present (Figure S2). This association buries an area complex, the phosphonate P-O1 establishes of 1120 Å2, amounting to ~10% of the total hydrogen bonds to the main NH groups of surface (11930 Å2). In the eight copies, the Ala37 and Met105 residues (Figure 6A). amino-acid chains could be traced from Within the α/β-hydrolase family, residues 24/28 to 258/260, depending upon many lipolytic enzymes have their active the chain. However, two segments at site controlled by a so-called (mobile) lid, positions 64 to 70/74 and 155/158 to formed by a surface loop [29, 36, 45, 46]. In 168/174 could not be traced in the electron all the cases studied, the lid domain was map density. A positional variation was found to undergo a conformational change observed between the eight copies in the presence of an inhibitor, making the

active site accessible to solvent [36]. In one observed with some lipases like Dog brief, the lid is a short domain that exhibits and Human Gastric Lipases [48, 49]. dramatic conformational changes between We then scanned the DALI server the apo- and the holo-forms. From the hits to find a structure well superposed with enzyme activity measurements obtained in TesA lid. The recently determined structure presence of detergent (Figure 2B), and of the thioesterase domain of M. since the structure has been only obtained as tuberculosis Polyketide Synthase 13 a dimer in complex with an inhibitor, we (Pks13; PDB id: 5V3X; Z= 20; rmsd=2.2 Å) suspect that the protruding domain formed [50] returned as an excellent hit by DALI. from an extended stretch of the peptide This structure consisting of a core domain backbone, followed by two α-helices (α4 and a lid domain, is in complex with an and α5) might be the TesA lid in an open inhibitor (TAM1). The superposition with conformation. Although, the TesA apo- TesA not only shows that both lid domains form could not be crystallized to confirm coincide well (Figure 8A), but also that the the presence/identification of the lid, we Pks13 thioesterase inhibitor TAM1 is performed a DALI server search [47] to located close to CyC17 (Figure 8). TAM1 is detect structurally similar proteins having indeed sandwiched between these two α- possibly closed and open lids, that could be helices and the lid, resulting in a catalytic used in a comparison with TesA structure. crevice of Pks13 thioesterase being much The DALI server returned several more restricted and deeper than that of TesA tens of structures with Z>15 and root mean (Figure 8A and B). square deviations (rmsd) lower than 3 Å. Among them, and of special interest, were Discussion the two crystal structures of RifR; i.e, PDB Although M. tuberculosis, the main ids 3FLB (Z=22; rmsd=1.2 Å; used for the cause of human TB, can be traced back to molecular replacement) and 3FLA (Z=22; 70,000 years ago [51], it currently remains rmsd=1.2 Å) [41]. This type II thioesterase the ninth cause of death worldwide and indeed possesses a helical lid that controls represents the primary cause of death by access of substrates to the active site [41]. single infectious pathogen, even ranking When the 3FLB and 3FLA structures were above HIV/AIDS [1]. The difficulties in superimposed to TesA, the core of the two treating the disease mainly result from the enzymes coincided within 1.4 Å rmsd. We fact that M. tuberculosis encounters diverse noticed, however, that the three helices microenvironments and can be found in a consisting of the lid subdomain of 3FLA variety of metabolic states during the covers the active site of the protein and do infection of the human host [52]. Available not superpose with their TesA counterparts, antituberculosis drugs cannot act against all i.e., the ext-α4-α5 helices proposed as a metabolic states of the bacteria, e.g. putative lid (Figure 7). The fact that most replicating vs non-replicating M. of thioesterases have an -helical insertion tuberculosis, or against multi- and that forms a lid over the active site, which extensively-drug resistant M. tuberculosis can differ in the number and disposition of (M/XDR-TB) vs. classical strain. helices and in its flexibility [41], prompted Moreover, despite the use of promising new us to suggest that the ext-α4-α5 domain drugs; such as bedaquiline which inhibits conformation observed in the TesA-CyC17 ATP synthase, and the nitroimidazoles complex structure is an open lid resulting delamanid and pretomanid which inhibit from a ~90¡ rotation of the two helices, as mycolic acid synthesis as well as energy compared to the 3FLA corresponding production [53, 54]; there is an urgent need helices in a closed conformation (Figure to discover new drugs to fight several 7A). Notably, the α5 helix may result from mycobacterial infections such as a smaller conformational change than the Tuberculosis, Leprosis and Buruli Ulcer.

The majority of commercially available mycobacteria (MIC50 of 12.6, 3.1 and 11.7 drugs target a single molecule/process µM, respectively). In contrast, CyC17 was a involved in the synthesis pathways essential potent inhibitor of in vitro growth (MIC50 ~ for the bacterial survival during the 0.5 µM) but failed to show activity against infection process. Unfortunately, impairing intracellular bacilli. These findings suggest only one protein may lead to the rapid that there are several modes of action of emergence of resistant strains. To overcome these related compounds (extracellular vs. these drawbacks, TB treatment is a 6-month intracellular) and that they probably target regimen of antibiotics: isoniazid, several enzymes. Using an activity-based rifampicin, ethambutol and pyrazinamide. protein profiling (ABPP) approach, 23 This treatment regimen is often extended to putative protein targets for CyC17 were 9-12 months to treat difficult cases. The identified, including the thioesterase TesA inherent difficulty to be compliant with [19]. such a long treatment is in part responsible Overall, these first results support for the appearance of resistant strains, the assumption that CyC inhibitors are representing a new challenge to control the multi-target compounds leading to the disease. As reported by Zumla et al. in 2013 inhibition of M. tuberculosis growth “there is growing awareness of the need for through the inhibition of various drugs that can kill M. tuberculosis in its mycobacterial Ser- or Cys-containing different physiological states” [55]. The enzymes involved in important priority for new drug development against physiological processes. In addition, the TB should thus be focused on the discovery lack of cytotoxicity towards host cells [19], of new candidates that would be able to i) makes these CyC analogs exciting reduce the treatment duration and cost; ii) molecules to be further evaluated as new impact several target enzymes; and iii) act lead candidates for treating the disease. on several physiological states of the Herein we provide clear evidence bacteria. that TesA is strongly inhibited by CyC17 but In this context, we have not by CyC6β, CyC7β or CyC8β (Figure demonstrated that the CyC analogs 4A). As showed in the crystal structure of represent powerful and selective inhibitors TesA in complex with the CyC17, the of mycobacterial enzymes [14, 15], with no inhibition of the enzyme results from the effect on the mammalian enzymes initially phosphorylation of the catalytic Ser104 targeted by natural parent molecules [14- (Figure 6). The inhibitory effect of CyC17 17]. The selectivity of the CyC derivatives on TesA allows the validation, a posteriori, toward the mycobacterial but not the human of this enzyme as an effective target of this enzymes, is therefore highly valuable and molecule, as initially postponed via prompted us to consider them as potential previous ABPP experiments [19]. However, anti-tubercular agents. Accordingly, among the absence of in vitro inhibition of TesA by the set of 27 CyC analogs previously CyC6β, CyC7β or CyC8β strengthens the fact evaluated against M. tuberculosis H37Rv, that M. tuberculosis bacterial clearance eight compounds were able to efficiently inside infected macrophages, observed inhibit M. tuberculosis growth either during ex vivo experiments in presence of extracellularly or within macrophages at these molecules [19], does not result from similar concentrations as isoniazid and the inhibition of this enzyme. In contrast the ethambutol [19]. Importantly, the best CyC Antigen 85 complex is targeted by CyC7β, compounds showed absence of toxicity in CyC8β as well as CyC17 [38]. From a mammalian cells at concentrations up to structure-activity relationship (SAR) point 100 µM [19]. More particularly, CyC6β, of view, CyC6β, CyC7β and CyC8β are CyC7β and CyC8β were mainly active phosphonate analogs while CyC17 is a against intracellularly-replicating phosphate (Figure 3). However, such a

chemical change (i.e., phosphonate vs. suggests the presence of a lid protecting the phosphate) is not a sufficient criteria to catalytic site and composed by 2 α-helices explain the observed difference in their [41, 50, 56]. These structural observations inhibitory potency, since any catalytic are in line with the results of pNP ester serine/cysteine is able to react similarly hydrolysis showing a clear increase in TesA with both chemical species [14, 19]. A more activity in presence of surfactant (i.e., plausible explanation relies on the core Triton X-100) even with the soluble short structure of these compounds and more chain pNP-C2 ester (Figure 2B). In the precisely on the positioning of the alkyl structure depicted in Figure 6, the lid would chain. If phosphonates CyC6β-8β are best be in open conformation allowing the described by the relationship between the interaction between CyC17 and the active OMe on phosphorus and the H-substituent Ser104. Moreover, in this crystal structure on the C-5 carbon atom, being here in a cis the polar head of CyC17 is also clearly (β-isomer) conformation; with CyC17, the covalently bound to Ser104 and its lipophilic C16-alkyl chain is directly hydrocarbon tail seems to be floating carried by the phosphorus atom. towards the hydrophobic crevice located Accordingly and as reported previously between the core and the lid. Only a few of [14], the presence of such lipophilic alkyl the carbons close to the phosphate group chain on the γ-carbon of the enol- could be modelled in the electron density phosphonate ring may significantly affect map (Figure S2). Despite many attempts, and modify the biological activity of the no suitable crystals were obtained in corresponding monocyclic analogs. The absence of inhibitor suggesting that TesA is exquisite chemoselectivity exerted by TesA difficult to crystallize in an assumed so- for CyC17 phosphate over the phosphonates called closed conformation. The presence of may thus result from a steric hindrance the inhibitor could, however, promote the provided by lateral alkyl chain located at the lid opening leading to the formation of C5-carbone atom in CyC6β-8β dimers (Figure 4C and Figure S1) that phosphonates, making reaction between the allows good quality of crystals to form. phosphorus atom with the catalytic Ser104 To conclude, in the present study we extremely difficult, as already observed confirm that the CyC analogs impair with mammalian lipases [14]. various target enzymes with different In a molecular point of view, the activities, including the thioesterase TesA. biochemical and mass spectrometry Consequently, a way to enhance the experiments involving CyC17, coupled with efficiency of this new family of anti- site directed mutagenesis conducted on tubercular molecules would be the use of Ser104 have confirmed the covalent these CyC as a cocktail mixture, or together binding of CyC17 to the catalytic serine of with known antibiotics to treat intra and/or TesA. Remarkably, after phosphorylation extracellular bacteria. of the serine by CyC17, a rearrangement of the inner structure of the bound inhibitor Material and Methods occurs with the release of methyl 2-acetyl 4- Bacterial strains and growth condition hydroxybutyrate group (Figure 5E). Such chemical rearrangement that has been Escherichia coli DH10β cells previously observed in the inhibited (Invitrogen) used in cloning experiments ¡ Ag85C-CyC17 complex [38], and can were grown at 37 C in Luria Bertani (LB) therefore be considered as a signature of the or on LB agar plates. In the expression CyC17 reactivity with Ser- and Cys- experiments, Escherichia coli T7 Iq pLysS containing enzymes. strain was used to carry the expression As found in other thioesterases and vector and express protein. Terrific Broth many lipolytic enzymes, TesA 3D structure (TB) broth (Difco) was used to grow the

strain with shaking at 250 rpm and 37 ¡C. washed with buffer A (20 mM Tris-HCl, pH All the media were supplemented with 100 8, 150 mM NaCl) containing 50 mM µg/mL ampicillin. imidazole and eluted with buffer A containing 250 mM imidazole. The Cloning, expression and purification of proteins hexahistidine tag was cleaved by incubation of the protein with the histidine-tagged TEV From genomic DNA of M. protease (ratio 40:1, protein: TEV, w/w) tuberculosis H37Rv, full-length of Rv2928 overnight, coupled with dialysis against gene (786 pb) encoding TesA enzyme (261 buffer A to eliminate imidazole. The AA, MW 29,077 Da, pI 5.11) was amplified untagged TesA was obtained in the flow using the Phusion¨ DNA polymerase through of a second Ni2+ affinity (ThermoFisher Scientific). After chromatography and further purified by purification the PCR product was cloned preparative Superdex 75 (GE Healthcare) into the Gateway pDEST14 expression gel filtration in buffer A. TesA was vector using Gateway cloning technology concentrated to 15 mg/mL for as described previously. The forward crystallization studies and other primers used is TesA-Fwd 5’ GGGG ACA experiments. Complexes of TesA with AGT TTG TAC AAA AAA GCA GGC different CyC inhibitors (CyC17, CyC7β, TTC GAA GGA GAT AGA ACC ATG CyC8β) were prepared by mixing pure TesA CAT CAC CAT CAC CAT CAC GAA with each CyC, previously dissolved in AAC CTG TAC TTC CAG GGT CTG DMSO, at a molar excess (xI) of 100. The GCC CGT CAC GGA CCA CGC TAT G mixture was left at 4 ¡C for at least 1 h 3’, and the reverse primer is TesA-Rev 5’ before submitted to an additional CTA AGC TCG ATC ATG CCA TTG preparative Superdex 75 gel filtration in GAG TG 3’. The final construct encodes the buffer A. TesA-inhibitor complexes were TesA gene fused to a N-terminal concentrated to 15 mg/mL for further hexahistidine tag followed by a TEV crystallization studies. cleavage site (bold). After checking the TesA mutant construction integrity of the DNA sequence by DNA sequencing (GATC-BIOTECH, Germany), The pDEST14-TesAS104A mutant the recombinant vector was transformed plasmid was constructed by the into Escherichia coli T7 Iq pLysS cells QuickChange method using pDEST14- (New England Biolabs) and then expressed TesA as matrix, the Phusion¨ DNA and purified. Briefly, the cells were grown polymerase (ThermoFisher Scientific), the S104A 5’ in Terrific Broth (TB) at 37¡C until OD600nm forward primer TesA -Fwd G GTG reached 0.6-1.0. Expression of TesA was GCA TTC TTT GGG CAC GCT ATG induced by 0.5 mM isopropyl-β-thio- GGC GGA ATG CTA GCC TTC 3’, and the galactoside (IPTG) for overnight at 17¡C. reverse primer TesAS104A-Rev 5’ GAA GGC Cells were harvested by centrifugation at TAG CAT TCC GCC CAT AGC GTG 4,000 × g for 15 min at 4¡C, resuspended in CCC AAA GAA TGC CAC C 3’ (Ser-to-Ala lysis buffer (50 mM Tris-HCl, pH 8, 300 mutation in bold). After DNA sequencing, mM NaCl, 10 mM imidazole, 0.25 mg/mL the recombinant plasmid was transformed lysozyme) and lysed by sonication at 50% into Escherichia coli T7 Iq pLysS cells and power during 4 min, with 45 short burst of the recombinant enzyme was produced and 45 s followed by intervals of 15 s for purified as described above. cooling. The lysate was cleared by MALS/QELS/UV/RI-coupled size × centrifugation at 18,000 g for 45 minutes exclusion chromatography and the supernatant was loaded onto a Ni2+ affinity chromatography column (HisTrap 5 Size exclusion chromatography was mL, GE Healthcare). The protein was performed on an Alliance 2695 HPLC

system (Waters) using a precalibrated released per min. Specific activities were KW802.5 column (Shodex) run in 20 mM expressed as mU/mg of pure enzyme. The Tris-HCl, pH 8, 150 mM NaCl at 0.5 same conditions were also applied to the mL/min. MALS, UV spectrophotometry, inactive TesAS104A catalytic mutant. QELS and RI were achieved with Negative controls included denaturated MiniDawn Treos (Wyatt Technology), a enzyme (10 min boiling) and protein Photo Diode Array 2996 (Waters), a filtrates (obtained during concentration DynaPro (Wyatt Technology) and an steps). All experiments were performed at Optilab rEX (Wyatt Technology), least in triplicate. respectively, as described [57]. Mass and Thioesterase activity assay hydrodynamic radius calculation was done Hydrolysis of Palmitoyl-CoEnzyme with ASTRA software (Wyatt Technology) A (Palmitoyl-CoA) (Sigma-Aldrich) was using a dn/dc value of 0.185 mL/g. used to measure thioesterase activity. After Enzymatic activity hydrolysis of the thioester bond, the free sulfur on CoA is attacked by DTNB (5-5'- Esterase activity Dithio-bis (2-nitrobenzoic acid)) (Sigma- The esterase enzyme activity was Aldrich) which releases a measurable performed as previously described with nitrophenyl group, TNB2-(5-Thio-2- minor modifications [58]. In brief, para- nitrobenzoate) ion detectable at 415 nm. nitrophenyl (pNP) esters (Sigma-Aldrich, Fresh stock solution of Palmitoyl-CoA (1 Saint-Quentin Fallavier, France) with mM) and DTNB (0.4 mM) were prepared in different carbon chain lengths were used as water. Palmitoyl-CoA and DTNB were substrates, including pNP acetate (pNP- added in each well of a 96-well microplate C2), valerate (pNP-C5), caprylate (pNP- at final concentration of 10 µM. The C8), laurate (pNP-C12), myristate (pNP- thioesterase activity of TesA (23.5 µg – 4.0 C14) palmitate (pNP-C16) and stearate µM final concentration) was compared to (pNP-C18). Stock solution (from 20 to 100 that of Cut6 (Rv3802c) (7.5 µg – 1.1 µM mM) of each substrate was freshly prepared final concentration) [28] which was used as in acetonitrile. Release of pNP was positive control. Triplicate assays were monitored at 410 nm and pH 7.5 using a 96- done at pH 8.0 in 10 mM Tris, 300 mM well plate spectrophotometer NaCl buffer. The absorbance at 415 nm was (PowerWaveTM, Bio-Tek Instruments) and continuously measured and the release of quantified using a pNP calibration curve (10 TNB2- was quantified using a calibration µM to 0.5 mM) with apparent Ɛ(λ=410nm) = curve. 8.4 mM-1 and 6.0 mM-1 when using 0.5% (w/v) Triton X-100. Enzymatic reactions Inhibition by CyC17, CyC7β, CyC8β and were performed at 37 ¡C over a period of 20 Orlistat min in a 96-well microplate filled with 100 Inhibition experiments were carried mM Tris-HCl buffer (pH 7.5) containing out using a classic lipase inhibitor pre- 100 mM NaCl to a final volume of 200 µL. incubation method, as previously described In each well, 2 mM of substrate and 50 µg [14, 15]. Mother solution of each inhibitor of enzyme (8.6 µM final concentration) CyC17, CyC6β, CyC7β, CyC8β and Orlistat were added. For each pNP ester, the assay were prepared in DMSO at a concentration was performed in presence or absence of of 4.6 mg/mL, 5.7 mg/mL, 5.9 mg/mL and 0.5% (w/v) Triton X-100 in the buffer. 5.0 mg/mL, respectively. TesA was pre- When Triton X-100 was present, the pNP incubated for 30 min at 37 ¡C with each esters were first solubilized in the buffer by CyC at various inhibitor molar excess (xI) sonication in a water bath for 1 min. ranging from 1 to 100 related to 1 mol of Activities were expressed in international enzyme. In each case, control experiments units (U), corresponding to 1 µmol of pNP were performed with the same volume of

solvent, without inhibitor. Residual analysis method. To eliminate salts from the activities were assayed using the samples solutions, 10 µL of each colorimetric assays with the substrate pNP- preparation was submitted to a desalting C8 as described above. The variation in the step on a C4 Zip-Tip µcolumn (Millipore). residual enzyme activity allowed One µL of desalted sample was mixed with determination of the inhibitor molar excess 1 µL sinapinic acid matrix in a 50% which reduced the enzyme activity to 50% acetonitrile/0.3% trifluoroacetic acid (TFA) of its initial value (xI50). Results are mixture (1:1, v/v). Then 1 µL was spotted expressed as mean values of at least two on the target, dried and analysed with the independent assays (CV%, 5.0%). LP_66Kda method. Peak picking was In view of the specific mechanism of performed with Flexanalysis 3.0 software action of lipolytic enzymes [36], the (Bruker) with an adapted analysis method. Michaelis-Menten–Henri model no longer Parameters used were as follows: SNAP applies [59] and the Km, Ki and IC50 values peak detection algorithm, S/N threshold often estimated for lipolytic enzymes and fixed to 6 and a quality factor threshold of expressed in terms of volume 30. concentrations are irrelevant when Tryptic digestions and peptide mass insoluble substrates and/or inhibitors are analyses involved, or when conformational changes Protein denaturating SDS-PAGE related to lid opening are occurring [29]. was performed on Nu-Page 4-12% Bis-Tris The use of more appropriate kinetic gels (Invitrogen, Life Technologies, constants such as the inhibitor molar excess Carlsbad, CA). The gel was stained with leading to 50% enzyme inhibition (i.e., xI50 Coomassie blue (Imperial Protein stain, value) [14, 15, 60] is therefore Pierce). Protein spots (~ 10 µg) excised recommended when assessing the from Coomassie blue stained gels were inhibitory potency of insoluble inhibitors, subjected to in-gel digestion with trypsin (2 such as these cyclic enol-phosphorus µL – 12.5 ng/µL) (Sequencing grade derivatives. Thereby, a xI50 value of 0.5 is modified porcine trypsin; Promega, synonymous with a 1:1 stoichiometric ratio Madison, WI, USA) according to a between the inhibitor and the lipolytic modified protocol from Shevchenko et al. enzyme, and is therefore the highest level of [61]. Tryptic peptides were then extracted inhibitory activity that can be achieved. from the gel by successive treatment with Mass spectrometry analyses 5% formic acid and 60% acetonitrile/5% formic acid, each treatment followed by a Total mass analyses 10 min sonication. Extracts were pooled Total mass analyses were performed and dried in a Speedvac evaporator. on a MALDI-TOF-TOF Bruker Ultraflex Peptide mass analyses were III spectrometer (Bruker Daltonics, performed on a MALDI-TOF-TOF Bruker Wissembourg, France) controlled by the (as described above) operating in reflectron Flexcontrol 3.0 package (Build 51). This mode using the m/z range from 600 to 3700 instrument was used at a maximum (RP Proteomics_2017 Method). The laser accelerating potential of 25 kV and was frequency was fixed to 200 Hz and operated in linear mode using the m/z range approximately 500 shots by sample were from 20,000 to 100,000 (LP_66 KDa cumulated. Five external standards (Peptide Method). Five external standards (Protein Calibration Standard, Bruker Daltonics) Calibration Standard II, Bruker Daltonics) were used to calibrate each spectrum to a were used to calibrate each spectrum to a mass accuracy within 50 ppm. One µL of mass accuracy within 200 ppm. Peak sample was mixed with 1 µL of a saturated picking was performed with Flexanalysis HCCA (α-cyano-4-hydroxycinnamic acid) 3.0 software (Bruker) with an adapted solution in acetonitrile/0.3% TFA (1:1, v/v).

Then 1 µL was spotted on the target, dried Manual adjustments of the model were and analysed with the RP Proteomics_2015 performed with Coot [65] and the structure method. Peak picking was performed with was refined to 2.75 Å for crystal form 1 and Flexanalysis 3.0 software (Bruker), using 2.6 Å for crystal form 2 with BUSTER [66]. the following parameters: SNAP peak Covalent link between CyC17 and catalytic detection algorithm, S/N threshold fixed to serine and restraints for the inhibitor were 6 and a quality factor threshold of 30. generated by Jligand [67]. Data collection Crystallization and data processing of and refinement statistics are displayed in TesA Table 1.

Crystallization trials were carried Funding/Acknowledgments out using the sitting-drop vapour diffusion This work was supported by the method [62]. The reservoirs of the Greiner CNRS and Aix Marseille University. P.C.N plates were filled with a TECAN pipetting was supported by the PhD Training robot, and nanodrops were dispensed by a program from the University of Science and Mosquito robot (TTP Labtech). No crystal Technology of Hanoi (837267E). V.S.N hits were found when using apo-form of was supported by a PhD grant from the TesA, although many different screening French Embassy in Vietnam (792803C). conditions and protein concentrations were Proteomics analyses were supported by the used. All three complexes of inhibitor- Institut Paoli-Calmettes and the Centre de bound TesA (TesA-CyC17, TesA-CyC7β, Recherche en Cancérologie de Marseille. TesA-CyC8β, at a molar excess (xI) of 100) Proteomic analyses were done using the resulted in different crystal forms, however mass spectrometry facility of Marseille only crystals of TesA-CyC17 appeared in Proteomics (marseille-proteomique.univ- 45% PEG 600, 0.1 M HEPES pH 7.5 amu.fr) supported by IBISA diffracted well. The resolution of these (Infrastructures Biologie Santé et crystals was improved by adding of 10% Agronomie), the Cancéropôle PACA, the (v/v) 2-methyl-2,4-pentanediol (MPD) and Provence-Alpes-Côte d'Azur Region, the 5 mM β-Octyl glucoside into the well Institut Paoli-Calmettes and the Centre de solution before drop preparation. Crystals Recherche en Cancérologie de Marseille. were directly cryo-cooled in well solution The atomic coordinates and structure without any cryo-protection. Datasets were factors (codes 6FW5 and 6FVJ) have been collected at the SOLEIL synchrotron (St deposited in the Protein Data Bank Aubin, France). Data were processed with (http://wwpdb.org/). XDSME [63] and the structure was solved by molecular replacement with the structure References of the core domain of RifR a thioesterase [1] WHO. from the rifamycin biosynthetic pathway http://www.who.int/tb/publications/global_ from Amycolatopsis mediterranei as search report/en/. 2017. model (PDB id: 3FLB) [41] and using [2] Hett EC, Rubin EJ. Bacterial growth and Phaser from the PHENIX software suite cell division: a mycobacterial perspective. [64]. Microbiol Mol Biol Rev. 2008;72:126-56, Crystal form 1 had the F222 space group table of contents. with unit cell parameters: a=77.58, [3] Jackson M. The Mycobacterial Cell b=224.65, c=226.62 Å and α = β = γ = 90¡, Envelope—Lipids. Cold Spring Harb 4 molecule of TesA in 1 asymmetric unit. Perspect Med. 2014;4. Crystal form 2 had the C2221 space group [4] Camacho LR, Constant P, Raynaud C, with unit cell parameters: a=79.07, Lanéelle M-A, Triccas JA, Gicquel B, et al. b=224.58, c=222.21 Å and α = β = γ = 90¡, Analysis of the Phthiocerol Dimycocerosate 8 molecules of TesA in 1 asymmetric unit. Locus of Mycobacterium tuberculosis :

Evidence that this lipid is involved in the invasion by inducing changes in the cell wall permiability barrier. J Biol Chem. organization of plasma membrane lipids. 2001;276:19845-54. PLoS Pathog. 2009;5:e1000289. [5] Constant P, Perez E, Malaga W, [12] Dhungel S, Ranjit C, Sapkota BR, Laneelle MA, Saurel O, Daffe M, et al. Role Macdonald M. Role of PGL-I of M. leprae of the pks15/1 gene in the biosynthesis of in TNF-alpha production by in vitro whole phenolglycolipids in the Mycobacterium blood assay. Nepal Med Coll J. 2008;10:1- tuberculosis complex. Evidence that all 3. strains synthesize glycosylated p- [13] Ferreras JA, Stirrett KL, Lu X, Ryu JS, hydroxybenzoic methyl esters and that Soll CE, Tan DS, et al. Mycobacterial strains devoid of phenolglycolipids harbor a phenolic glycolipid virulence factor frameshift mutation in the pks15/1 gene. J biosynthesis: mechanism and small- Biol Chem. 2002;277:38148-58. molecule inhibition of polyketide chain [6] Daffé M, Lacave C, Laneelle MA, initiation. Chem Biol. 2008;15:51-61. Laneelle G. Structure of the major [14] Point V, Malla RK, Diomande S, triglycosyl phenol-phthiocerol of Martin BP, Delorme V, Carriere F, et al. Mycobacterium tuberculosis (strain Synthesis and kinetic evaluation of Canetti). Eur J Biochem. 1987;167:155-60. cyclophostin and cyclipostins phosphonate [7] Huet G, Constant P, Malaga W, Laneelle analogs as selective and potent inhibitors of MA, Kremer K, van Soolingen D, et al. A microbial lipases. J Med Chem. lipid profile typifies the Beijing strains of 2012;55:10204-19. Mycobacterium tuberculosis: identification [15] Point V, Malla RK, Carriere F, Canaan of a mutation responsible for a modification S, Spilling CD, Cavalier JF. of the structures of phthiocerol Enantioselective inhibition of microbial dimycocerosates and phenolic glycolipids. J lipolytic enzymes by nonracemic Biol Chem. 2009;284:27101-13. monocyclic enolphosphonate analogues of [8] Reed MB, Domenech P, Manca C, Su H, cyclophostin. J Med Chem. 2013;56:4393- Barczak AK, Kreiswirth BN, et al. A 401. glycolipid of hypervirulent tuberculosis [16] Martin BP, Vasilieva E, Dupureur CM, strains that inhibits the innate immune Spilling CD. Synthesis and comparison of response. Nature. 2004;431:84-7. the biological activity of monocyclic [9] Mohandas P, Budell WC, Mueller E, Au phosphonate, difluorophosphonate and A, Bythrow GV, Quadri LE. Pleiotropic phosphate analogs of the natural AChE consequences of gene knockouts in the inhibitor cyclophostin. Bioorg Med Chem. phthiocerol dimycocerosate and phenolic 2015;23:7529–34. glycolipid biosynthetic gene cluster of the [17] Vasilieva E, Dutta S, Malla RK, Martin opportunistic human pathogen BP, Spilling CD, Dupureur CM. Rat Mycobacterium marinum. FEMS Microbiol hormone sensitive lipase inhibition by Lett. 2016;363:fnw016. cyclipostins and their analogs. Bioorg Med [10] Alibaud L, Rombouts Y, Trivelli X, Chem. 2015;23:944-52. Burguiere A, Cirillo SL, Cirillo JD, et al. A [18] Nguyen PC, Madani A, Santucci P, Mycobacterium marinum TesA mutant Martin BP, Paudel RR, Delattre S, et al. defective for major cell wall-associated Cyclophostin and cyclipostins analogs, new lipids is highly attenuated in Dictyostelium promising molecules to treat mycobacterial- discoideum and zebrafish embryos. Mol related diseases. Int J Antimicrob Agents. Microbiol. 2011;80:919-34. 2018;51:651-4. [11] Astarie-Dequeker C, Le Guyader L, [19] Nguyen PC, Delorme V, Benarouche Malaga W, Seaphanh FK, Chalut C, Lopez A, Martin BP, Paudel R, Gnawali GR, et al. A, et al. Phthiocerol dimycocerosates of M. Cyclipostins and Cyclophostin analogs as tuberculosis participate in macrophage

promising compounds in the fight against phospholipase/thioesterase and is inhibited tuberculosis. Sci Rep. 2017;7:11751. by the antimycobacterial agent [20] Chavadi SS, Edupuganti UR, tetrahydrolipstatin. PLoS One. Vergnolle O, Fatima I, Singh SM, Soll CE, 2009;4:e4281. et al. Inactivation of tesA reduces cell wall [29] Delorme V, Dhouib R, Canaan S, lipid production and increases drug Fotiadu F, Carrière F, Cavalier J-F. Effects susceptibility in mycobacteria. J Biol of Surfactants on Lipase Structure, Activity Chem. 2011;286:24616-25. and Inhibition. Pharmaceutical Research. [21] de Souza GA, Wiker HG. A proteomic 2011;28:1831-42. view of mycobacteria. Proteomics. [30] Chahinian H, Nini L, Boitard E, Dubes 2011;11:3118-27. JP, Comeau LC, Sarda L. Distinction [22] Gu S, Chen J, Dobos KM, Bradbury between esterases and lipases: a kinetic EM, Belisle JT, Chen X. Comprehensive study with vinyl esters and TAG. Lipids. proteomic profiling of the membrane 2002;37:653-62. constituents of a Mycobacterium [31] Ulker S, Placidi C, Point V, Gadenne tuberculosis strain. Mol Cell Proteomics. B, Serveau-Avesque C, Canaan S, et al. 2003;2:1284-96. New lipase assay using Pomegranate oil [23] Griffin JE, Gawronski JD, Dejesus coating in microtiter plates. Biochimie. MA, Ioerger TR, Akerley BJ, Sassetti CM. 2015. High-resolution phenotypic profiling [32] Schué M, Maurin D, Dhouib R, defines genes essential for mycobacterial N'Goma JC, Delorme V, Lambeau G, et al. growth and cholesterol catabolism. PLoS Two cutinase-like proteins secreted by Pathog. 2011;7:e1002251. Mycobacterium tuberculosis show very [24] Waddell SJ, Chung GA, Gibson KJ, different lipolytic activities reflecting their Everett MJ, Minnikin DE, Besra GS, et al. physiological function. Faseb J. Inactivation of polyketide synthase and 2010;24:1893-903. related genes results in the loss of complex [33] Ravindran MS, Rao SP, Cheng X, lipids in Mycobacterium tuberculosis Shukla A, Cazenave-Gassiot A, Yao SQ, et H37Rv. Lett Appl Microbiol. 2005;40:201- al. Targeting Lipid Esterases in 6. Mycobacteria Grown Under Different [25] Rao A, Ranganathan A. Interaction Physiological Conditions Using Activity- studies on proteins encoded by the based Profiling with Tetrahydrolipstatin phthiocerol dimycocerosate locus of (THL). Molecular & Cellular Proteomics. Mycobacterium tuberculosis. Mol Genet 2014;13:435-48. Genomics. 2004;272:571-9. [34] Liu Y, Patricelli MP, Cravatt BF. [26] Kovacic F, Granzin J, Wilhelm S, Activity-based protein profiling: The serine Kojic-Prodic B, Batra-Safferling R, Jaeger hydrolases. Proceedings of the National KE. Structural and functional Academy of Sciences. 1999;96:14694-9. characterisation of TesA - a novel [35] Ortega C, Anderson LN, Frando A, lysophospholipase A from Pseudomonas Sadler NC, Brown RW, Smith RD, et al. aeruginosa. PLoS One. 2013;8:e69125. Systematic Survey of Serine Hydrolase [27] Crellin PK, Vivian JP, Scoble J, Chow Activity in Mycobacterium tuberculosis FM, West NP, Brammananth R, et al. Defines Changes Associated with Tetrahydrolipstatin inhibition, functional Persistence. Cell Chem Biol. 2016;23:290- analyses, and three-dimensional structure of 8. a lipase essential for mycobacterial [36] Aloulou A, Rodriguez JA, Fernandez viability. J Biol Chem. 2010;285:30050-60. S, van Oosterhout D, Puccinelli D, Carrière [28] Parker SK, Barkley RM, Rino JG, F. Exploring the specific features of Vasil ML. Mycobacterium tuberculosis interfacial enzymology based on lipase Rv3802c encodes a studies. Biochimica et Biophysica Acta

(BBA) - Molecular and Cell Biology of revealed by X-Ray crystallography. Nature. Lipids. 2006;1761:995-1013. 1993;362:814-20. [37] Pappin DJC, Hojrup P, Bleasby AJ. [47] Holm L, Rosenstrom P. Dali server: Rapid identification of proteins by peptide- conservation mapping in 3D. Nucleic Acids mass fingerprinting. Curr Biol. 1993;3:327- Res. 2010;38:W545-9. 32. [48] Roussel A, Canaan S, Egloff MP, [38] Viljoen A, Richard M, Nguyen PC, Riviere M, Dupuis L, Verger R, et al. Fourquet P, Camoin L, Paudal RR, et al. Crystal structure of human gastric lipase Cyclipostins and Cyclophostin analogs and model of lysosomal acid lipase, two inhibit the antigen 85C from lipolytic enzymes of medical interest. J Biol Mycobacterium tuberculosis both in vitro Chem. 1999;274:16995-7002. and in vivo. J Biol Chem. 2018. [49] Roussel A, Miled N, Berti-Dupuis L, [39] Belisle JT, Vissa VD, Sievert T, Riviere M, Spinelli S, Berna P, et al. Crystal Takayama K, Brennan PJ, Besra GS. Role structure of the open form of dog gastric of the major antigen of Mycobacterium lipase in complex with a phosphonate tuberculosis in cell wall biogenesis. inhibitor. J Biol Chem. 2002;277:2266-74. Science. 1997;276:1420-2. [50] Aggarwal A, Parai MK, Shetty N, [40] Jackson M, Raynaud C, Laneelle MA, Wallis D, Woolhiser L, Hastings C, et al. Guilhot C, Laurent-Winter C, Ensergueix Development of a Novel Lead that Targets D, et al. Inactivation of the antigen 85C M. tuberculosis Polyketide Synthase 13. gene profoundly affects the mycolate Cell. 2017;170:249-59 e25. content and alters the permeability of the [51] Comas I, Coscolla M, Luo T, Borrell S, Mycobacterium tuberculosis cell envelope. Holt KE, Kato-Maeda M, et al. Out-of- Mol Microbiol. 1999;31:1573-87. Africa migration and Neolithic coexpansion [41] Claxton HB, Akey DL, Silver MK, of Mycobacterium tuberculosis with Admiraal SJ, Smith JL. Structure and modern humans. Nature Genetics. functional analysis of RifR, the type II 2013;45:1176. thioesterase from the rifamycin biosynthetic [52] Ferraris D, Miggiano R, Rossi F, Rizzi pathway. J Biol Chem. 2009;284:5021-9. M. Mycobacterium tuberculosis Molecular [42] Krissinel E, Henrick K. Inference of Determinants of Infection, Survival Macromolecular Assemblies from Strategies, and Vulnerable Targets. Crystalline State. J Mol Biol. Pathogens. 2018;7:17. 2007;372:774-97. [53] Wellington S, Hung DT. The [43] Carr PD, Ollis DL. Alpha/beta expanding diversity of Mycobacterium hydrolase fold: an update. Protein Pept Lett. tuberculosis drug targets. ACS Infect Dis. 2009;16:1137-48. 2018:DOI: 10.1021/acsinfecdis.7b00255. [44] Martinez C, Nicolas A, van Tilbeurgh [54] Yuan T, Sampson NS. Hit Generation H, Egloff M-P, Cudrey C, Verger R, et al. in TB Drug Discovery: From Genome to Cutinase, a lipolytic enzyme with a Granuloma. Chem Rev. 2018. preformed oxyanion hole. Biochemistry. [55] Zumla A, Nahid P, Cole ST. Advances 1994;33:83-9. in the development of new tuberculosis [45] Khan FI, Lan D, Durrani R, Huan W, drugs and treatment regimens. Nat Rev Zhao Z, Wang Y. The Lid Domain in Drug Discov. 2013;12:388-404. Lipases: Structural and Functional [56] Miled N, Roussel A, Bussetta C, Berti- Determinant of Enzymatic Properties. Front Dupuis L, Riviere M, Buono G, et al. Bioeng Biotechnol. 2017;5:16. Inhibition of dog and human gastric lipases [46] van Tilbeurgh H, Egloff M-P, Martinez by enantiomeric phosphonate inhibitors: a C, Rugani N, Verger R, Cambillau C. structure-activity study. Biochemistry. Interfacial activation of the lipase- 2003;42:11587-93. procolipase complex by mixed micelles

[57] Sciara G, Blangy S, Siponen M, Mc Development of a technology for Grath S, van Sinderen D, Tegoni M, et al. A automation and miniaturization of protein topological model of the baseplate of crystallization. J Biotechnol. 2001;85:7-14. lactococcal phage Tuc2009. J Biol Chem. [63] Kabsch W. Xds. Acta Crystallogr D 2008;283:2716-23. Biol Crystallogr. 2010;66:125-32. [58] Delorme V, Diomande SV, Dedieu L, [64] Adams PD, Afonine PV, Bunkoczi G, Cavalier JF, Carriere F, Kremer L, et al. Chen VB, Davis IW, Echols N, et al. MmPPOX Inhibits Mycobacterium PHENIX: a comprehensive Python-based tuberculosis Lipolytic Enzymes Belonging system for macromolecular structure to the Hormone-Sensitive Lipase Family solution. Acta Crystallogr D Biol and Alters Mycobacterial Growth. PLoS Crystallogr. 2010;66:213-21. One. 2012;7:e46493. [65] Emsley P, Cowtan K. Coot: model- [59] Verger R, de Haas GH. Interfacial building tools for molecular graphics. Acta enzyme kinetics of lipolysis. Annual Crystallogr D Biol Crystallogr. Review Biophys Bioeng. 1976;5:77-117. 2004;60:2126-32. [60] Ransac S, Riviere C, Soulie JM, [66] Blanc E, Roversi P, Vonrhein C, Gancet C, Verger R, de Haas GH. Flensburg C, Lea SM, Bricogne G. Competitive inhibition of lipolytic Refinement of severely incomplete enzymes. I. A kinetic model applicable to structures with maximum likelihood in water-insoluble competitive inhibitors. BUSTER-TNT. Acta Crystallogr D Biol Biochimica et Biophysica Acta (BBA) - Crystallogr. 2004;60:2210-21. Molecular and Cell Biology of Lipids. [67] Lebedev AA, Young P, Isupov MN, 1990;1043:57-66. Moroz OV, Vagin AA, Murshudov GN. [61] Shevchenko A, Wilm M, Vorm O, JLigand: a graphical tool for the CCP4 Mann M. Mass spectrometric sequencing of template-restraint library. Acta Crystallogr proteins silver-stained polyacrylamide gels. D Biol Crystallogr. 2012;68:431-40. Anal Chem. 1996;68:850-8. [62] Mueller U, Nyarsik L, Horn M, Rauth H, Przewieslik T, Saenger W, et al.

Table 1. Data collection and refinement statistics (numbers in brackets refer to the highest resolution bin)

DATA Form1 Form2 COLLECTION* PDB 6FW5 6FVJ Source SOLEIL, PROXIMA 1 SOLEIL, PROXIMA 1 Detector PILATUS 6M PILATUS 6M

Wavelength (Å) 0.978570 0.97934

Space group F222 C2221 cell (Å) a = 77.58, b = 224.65, c = 226.62 a = 79.07, b = 222.58, c = 222.21

Angles (¡) α=β=γ=90 α=β=γ=90

Nr. of monomers 4 8 Resolution limits (Å) 43.87-2.75 (2.91-2.75) 43.80-2.60 (2.75-2.6) Rmerge 0.106 (1.001) 0.116 (0.907) CC1/2 0.997 (0.743) 0.999 (0.773) Unique reflections 24307 (4055) 61244 (9662)

Mean((I)/sd(I)) 10.00 (1.38) 15.94 (2.43) Completeness (%) 93.1 (97.1) 99.6 (98.8) Multiplicity 7.18 (7.19) 13.4 (13.5) REFINEMENT*

Resolution (Å) 39.89-2.75 (2.87-2.75) 43.8-2.6 (2.67-2.6) Number of reflections 24306 (3026) 61125 (4464) Number of protein / 6766/80/20 12774/327/157 water / ligand (atoms) Test set reflections 1216 (152) 3056 (223)

Rwork/Rfree (%) 27.2/28.1 (29.4/31.2) 23.4/25.4 (25.6/25.3) r.m.s.d.bonds (Å) / 0.011/1.23 0.008/1.04 angles (¡) B-Wilson / B-mean Å 145.61/133.02 85.66/85.27

Ramachandran: preferred / allowed / 92.17/6.56/1.27 94.44/3.89/1.67 outliers (%) Average B-factor per chain: 119.2/128.2/142.7/145.7 72.2/67.6/74.8/67.2/87.9/91.2/113.8/112.3 A/B/C/D/E/F/G/H

Figure 1

Figure 1: (A) The PDIM gene cluster in M. tuberculosis. The tesA genes is shown in red. The black arrows represent predicted open reading. (B) Schematic representation of phthiocerol dimycocerosate (PDIM) and phenolphthiocerol (PGL) biosynthesis in M. tuberculosis. The interaction between TesA and PpsE is represented. The hydrolyzed thioester bond is indicated by a green arrow in PDIM synthesis pathway, and by a red arrow in PGL biosynthesis. After condensation with mycoseric acid, the products released by TesA lead to the synthesis of both PDIM and PGL.

Figure 2

Figure 2: Biochemical characterization: (A) TesA and TesAS104A purification. Lane 1, standard from Euromedex; lane 2 and 3, 3 µg of purified TesA and TesA S104A mutant after Tev proteolysis. (B) Esterase activity on pNitrophenyl esters with chain lengths varying from C2 to C18 in presence or in absence of 0.5% Triton X-100. (C) Thioesterase activity measurements corresponding to Palmitoyl-CoEnzyme A hydrolysis. Cut6 (Rv3802c) was used as positive control for the thioesterase activity. Each experiment is the mean of three independents assays. NA, no activity.

Figure 3

Figure 3: Chemical structures of CyC6β, CyC7β, CyC8β cis-monocyclic enol phosphonate analogs to Cyclophostin and CyC17 a monocyclic enol phosphate analog to Cyclipostins. Adapted from [19].

Figure 4

Figure 4: Inhibition of the TesA esterase activity is mediated by the covalent binding of CyC analogs. (A) The enzymatic activity of TesA was tested using pNP ester assay in the presence of different concentrations of CyC7β, CyC8β, CyC17 and Orlistat as control. The inhibitory effect was determined at the maximum rate of the reaction. Error bars represent the standard deviation calculated from three independent experiments. Dose-response curves for CyC7β, CyC8β, CyC17 and Orlistat were fitted in Kaleidagraph 4.2 Software (Synergy S104A Software). (B) Equal amounts of either TesA or TesA were pre-treated with CyC17, incubated with TAMRA-FP, separated by SDS-PAGE and visualized by Coomassie blue staining (upper panel) or in-gel fluorescence visualization (middle panel). The merged image is shown in the lower panel. TAMRA labelling of TesA is blocked by the covalent binding of the CyC analogs to the catalytic Ser104 as evidenced with the TesAS104A variant. (C) MALS/QELS/UV/RI analysis. The corresponding curve depicting the variation in UV absorbance at 280 nm as a function of time (min. after sample injection in the High Performance Liquid Chromatography system) for TesA (orange brown chromatogram) and TesA-CyC17 complex (grey chromatogram) have been superimposed and reported on the same chromatogram. The traces indicating the molar mass (indicated on the left, in Da) are shown on each peak by an arrow. Molecular masses of 31,800 Da (CV% <5%) for the monomer and 56,000 Da (CV% <4%) for the dimer have been calculated by the ASTRA software (Wyatt Technology). Each curve is representative of two different experiments.

Figure 5

Figure 5: Mass spectrometry analyses. Global mass modification of TesA (A) and TesAS104A (C) without (blue and pink spectra) or pre-incubated with CyC17 (green and blue spectra), as determined using an Ultraflex III mass spectrometer (Brucker Daltonics) in linear mode with the LP_66kDa_method. In order to localize the peptide containing the catalytic Ser104, TesA and TesAS104A were digested by trypsin. (B) and (D), peptide mass fingerprint on the digested TesA/TesAS104A (upper spectra, blue and pink), and peptide mass fingerprint on the digested S104A TesA/TesA pre-treated by CyC17 (lower spectra, green and blue). (E) Mechanism of action of the phosphate analog CyC17 based on mass spectrometry analyses.

Figure 6

Figure 6: (A) Ribbon view of overall structure of TesA, face view and 90¡ rotation around vertical axis. The α-helices are colored cyan blue and the β-strands magenta. (B) The catalytic machinery of TesA. The catalytic triad Ser104-His236-Asp208 side chains are colored by atom type (N: blue, O: red, C: grey) with black labels. The hydrogen bonds are colored green. The oxyanion hole residues (Ala37 and Met105) are labelled in white with yellow colored hydrogen bonds.

Figure 7

Figure 7: Identification of TesA lid. (A) Ribbon view of the superposition of TesA with RifR (PDB id: 3FLA; Type II thioesterase from Rifamycin NRPS/PKS biosynthetic pathway). TesA and RifR are colored in orange and blue, respectively. The hypothetical rotation from closed to open form is identified by a blue arrow. (B) Left: surface representation of TesA (orange) with the inhibitor CyC17 (stick representation) in the active site. Right: same as on the left, but with the ribbon view of RifR (PDB id: 3FLA) superimposed. The helices of RifR lid (blue color) cover the catalytic site of TesA.

Figure 8

Figure 8: Comparison of TesA with the thioesterase domain of M. tuberculosis Polyketide Synthase 13 (Pks13). (A) Ribbon view of the superposition of TesA with the thioesterase domain of Pks13 (PDB id: 5V3X) colored orange and cyan blue, respectively. TesA and Pks13 inhibitors are represented as sticks and spheres, respectively. Both lids superimposed well and are in open conformation. The grey arrow identifies the two helices insertion in Pks13 thioesterase domain. (B) Left: Surface representation of Pks13 thioesterase domain (cyan blue) and its inhibitor. Right: Surface representation of TesA (orange) and inhibitor CyC17. The grey arrow identifies the position of the two helices insertion in Pks13 thioesterase domain.

Figure S1

Figure S1: Apparent Michaelis-Menten kinetic constants determination using pNP-C8 as substrate

Figure S2

Figure S2: Dimer formation of TesA in presence of inhibitor CyC17

Figure S3

Figure S3: Superposition of 8 monomers of TesA identified in the asymmetric unit

Figure S4

Figure S4: 2Fo-Fc electron density map contoured at 1 σ of the ligand (CyC17) and residues within a 8 Å radius from ligand to each atom

Results – Article 5

Article 5: Oxadiazolone derivatives against M. tuberculosis

In parallel, the second series of 19 new compounds so-called Oxadiazolone (OX) derivatives, which have the core-structure of the human hormone sensitive lipase inhibitor MmPPOX, have been tested for their antibacterial activity against M. tb, M. marinum and M. bovis BCG. Susceptibility testing were performed using the resazurin microtiter assay (REMA). Nearly all 19 OXs were active against M. bovis BCG and M. marinum growth. However, with

MIC50 values in the range 1.9-53 µM, M. marinum was nearly 2-times more sensitive to OX 2 compounds than M. bovis BCG (MIC50 from 3.5 to >120 µM). In the case of M. tb mc 6230, 14 OXs were active against this bacteria; the two best growth inhibitors obtained being iBpPPOX and BePOX which displayed similar MIC50 value (mean 32.1 ±1.0 µM). Each of the 19 OX derivatives have been further evaluated for their antitubercular activity on virulent M. tb H37Rv-GFP using the previously described high-content screening assay based on the fluorescence measurement of GFP-expressing bacteria. Among tested compounds, 6 potential OX candidates exhibited interesting antitubercular properties with a very low toxicity towards host macrophages (CC50 > 100 µM). BePOX and HPOX impaired exclusively M. tb growth in culture broth medium with the same moderate 2 MIC50 (30.8 and 44.6 µM, respectively) than obtained previously on M. tb mc 6230. In contrast, iBPOX, HpPPOX and BepPPOX showed a clear preference against intracellularly-replicating mycobacteria with similar MIC50 values (3.5-17.1 µM) than the first line antibiotics. Of interest, only iBpPPOX exhibited moderate (32.0 µM) to quite good (8.5 µM) activity against extracellular and intramacrophagic M. tb, respectively. We further used the ABPP approach to identify potential enzymes targeted by our OX compounds. Contrary to CyC inhibitor, we incubated directly iBpPPOX and HPOX inhibitor with living bacteria for 2-3 h. before cell lysis and addition of Desthiobiotin-FP probe. In the downstream mass analysis, we will perform directly the digestion with trypsin and the resulting peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) for subsequent protein identification.

142 Oxadiazolone derivatives, new promising multi-target inhibitors against M. tuberculosis Phuong Chi Nguyen1¤, Vincent Delorme2¤ #, Anaïs Bénarouche1, Alexandre Guy3, Valérie Landry2, Stéphane Audebert4, Matthieu Pophillat4, Luc Camoin4, Céline Crauste3, Jean-Marie Galano3, Thierry Durand3, Priscille Brodin2, Stéphane Canaan1* and Jean-François Cavalier1*

1 LISM, Institut de Microbiologie de la Méditerranée, CNRS and Aix-Marseille Univ., Marseille, France 2 Univ. Lille, CNRS, INSERM, CHU Lille, Institut Pasteur de Lille, U1019 - UMR 8204 - CIIL - Center for Infection and Immunity of Lille, Lille, France 3 Institut des Biomolécules Max Mousseron (IBMM), UMR 5247, Université de Montpellier, CNRS, ENSCM, 15 Avenue Charles Flahault, 34093 Montpellier Cedex 5, France 4 Aix-Marseille Univ, Inserm, CNRS, Institut Paoli-Calmettes, CRCM, Marseille Protéomique, Marseille, France

¤ Authors have contributed equally to this work Corresponding authors: [email protected] (J.-F. Cavalier); [email protected] (S. Canaan)

# Current Address: Tuberculosis Research Laboratory, Institut Pasteur Korea, Seongnam-si, Gyeonggi-do, 13488 Republic of Korea.

ABSTRACT A set of 19 oxadiazolone (OX) derivatives have been investigated for their antimycobacterial activity against three pathogenic slow-growing mycobacteria: Mycobacterium marinum, Mycobacterium bovis BCG and the avirulent Mycobacterium tuberculosis (M. tb) mc26230. The encouraging minimal inhibitory concentrations (MIC) values obtained prompted us to test them against virulent M. tb H37Rv growth either in broth medium or inside macrophages. The OX compounds displayed a diversity of action and were found to act either on extracellular M. tb growth only with moderated MIC50, or both intracellularly on infected macrophages as well as extracellularly on bacterial growth. Of interest, all OX derivatives exhibited very low toxicity towards host macrophages. Among the six potential OXs identified, HPOX, a selective inhibitor of extracellular M. tb growth, was selected and further used in a competitive labelling/enrichment assay against the activity-based probe Desthiobiotin-FP, in order to identify its putative target(s). This approach, combined with mass spectrometry, identified 18 potential candidates, all being serine or cysteine enzymes involved in M. tb lipid metabolism and/or in cell wall biosynthesis. Among them, Ag85A, CaeA, TesA, KasA and MetA have been reported as essential for in vitro growth of M. tb and/or its survival and persistence inside macrophages. Overall, our findings support the assumption that OX derivatives may represent a novel class of multi-target inhibitors leading to the arrest of M. tb growth through a cumulative inhibition of a large number of Ser- and Cys-containing enzymes involved in various important physiological processes. Keywords Tuberculosis, Oxadiazolone, Lipolytic enzyme inhibitors, Activity-based probe (ABP) Abbreviations TB, tuberculosis; M. tb, Mycobacterium tuberculosis; INH, isoniazid; PZA, pyrazinamide; RIF, rifampicin; EMB, ethambutol; ETO, ethionamide; MDR, multidrug-resistant strains; XDR, extensively drug-resistant strains; TDR, totally drug-resistant strains; MIC50, minimal inhibitory concentrations leading to 50% of growth inhibition; CC50, compound concentration leading to 50% of cell cytotoxicity; HSL, hormone-sensitive lipase; CyC, Cyclipostins & Cyclophostin analogs; ABPP, activity-based protein profiling; REMA, resazurin microtiter assay.

1. Introduction RIF (1.0 µg/mL) [16, 17]. Molecular modeling With 10.4 million new cases and 1.7 million indicated that these compounds possessed all deaths in 2016, tuberculosis (TB) caused by necessary features to block the enzymatic the pathogenic strain Mycobacterium activity of the mycobacterial cytochrome tuberculosis (M. tb) remains the most P450-dependent 14a-sterol demethylase contagious and deadly disease worldwide [1]. (P45014DMs) [18], also a target for antifungal Despite the quadritherapy treatment involving drug design [19]. These findings thus suggest isoniazid (INH), pyrazinamide (PZA), the potential use of such OX derivatives as rifampicin (RIF) and ethambutol (EMB), the alternative therapeutic agents for TB [20]. introduction of new molecules on the market to Few years ago, we reported that the OX strengthen or replace this first-line antibiotics compound MmPPOX (Figure 1B), a regimen is a slow and tedious process [2, 3]. reversible inhibitor of the hormone-sensitive Only a few drugs were able to pass the lipase (HSL) family of proteins [21, 22], was selection stages (e.g., bedaquiline [4], able to inhibit the growth of M. tb with MIC delamanide [5] and PA-824 [6]). However, the values of around 15-25 µg/mL, as determined appearance and spread of multidrug-resistant on solid medium [23]. Keeping in mind its (MDR), extensively drug-resistant (XDR) and strong affinity toward the HSL family member totally drug-resistant (TDR) strains [7-10] proteins (Lip-HSL), we further investigated highlights the urgent need for finding new the in vitro inhibition of recombinant M. tb therapeutic targets to fight against M. tb. Lip-HSL. As expected, all purified Lip-HSL In this context, oxadiazolone-core (OX) were strongly inhibited by MmPPOX, which compounds represent attractive tools. reacted with the catalytic serine residue by Compound S57 (Figure 1A), initially forming a covalent but (slowly) reversible described in 1954 [11, 12] as being active bond (i.e., carbamate or thiocarbamate bound, against TB [13-15], was used as template for Figure 1B). Such an inhibitor could then be the synthesis of 3,5-substituted 1,3,4- considered as a long-life substrate rather than a oxadiazole-2-one derivatives [16, 17]. These true inhibitor as already observed with Orlistat molecules were found to exhibit interesting (also named Tetrahydrolipstatine or THL), a anti-mycobacterial activity against M. tb representative member of reversible serine H37Rv, with minimal inhibitory hydrolase inhibitors [24-26]. concentrations (MIC) of 1.25Ð8 µg/mL, comparable to those of INH (0.5 µg/mL) and

Figure 1. Chemical structure of (A) 5-(pyridin-4-yl)-1,3,4-oxadiazol-2(3H)-one S57, and (B) 5-methoxy-3- (3-phenoxyphenyl)-1,3,4-oxadiazol-2(3H)-one MmPPOX and its mode of action. (C) General procedure for the one step preparation of 5-alkoxy-3-phenyl substituted-1,3,4-oxadiazol-2(3H)-one compounds from either (3-phenoxyphenyl)hydrazine hydrochloride (1), (4-phenoxyphenyl)hydrazine hydrochloride (2) or phenylhydrazine hydrochloride (3), giving OX derivatives 5a-k (i.e., RmPPOX), 6a,d,e,k (i.e., RpPPOX) or 7a,d,e,k (i.e., RPOX), respectively. Reagents and conditions: i) Alkyl chloroformate 4a-k, Pyridine, 0 ¡C to

RT; ii) ClCO2CCl3, CHCl2, Pyridine, 0 ¡C to RT, 40-85%. Adapted from [27]. From these first results, a new series of 18 slow-growing mycobacteria: M. marinum, M. lipophilic OX derivatives based on MmPPOX bovis BCG and M. tb mc26230, a H37Rv strain core-structure have been synthesized (Figure with its RD1 region and panCD genes deleted, 1C) and tested for their anti-mycobacterial resulting in an avirulent pan(−) phenotype activity. More precisely, each oxadiazolone [28]. The corresponding MIC50 values, as molecule has been tested against M. tb H37Rv determined by the REMA assay [29-31], are for i) its capacity to inhibit in vitro growth; ii) reported in Table 1. First, it is noteworthy that its antitubercular activity on M. tb-infected the concentrations needed to inhibit 50% of the macrophages, and iii) its cytotoxicity towards bacterial growth (MIC50) obtained for macrophages. Interestingly, some analogs rifampicin (RIF) and isoniazid (INH), used were found to inhibit M. tb growth in vitro here as reference antibiotics, were in and/or inside macrophages without any agreement with literature data [32-34]. significant toxicity to host cells. In addition, Nearly all 19 OXs were active against M. bovis using an activity-based protein profiling BCG and M. marinum growth (Table 1). (ABPP) assay, the potential target enzymes of However, with MIC50 values in the range 1.9- HPOX, acting only on M. tb extracellular 53 µM, M. marinum was nearly 2-times more growth, were further identified. sensitive to OX compounds than M. bovis BCG (MIC50 from 3.5 to >120 µM). BePOX, HPOX and iBPOX, for which the phenoxy 2. Results and Discussion substituent is absent, exhibited the most potent 2.1 Synthesis of oxadiazolone-core (OX) antibacterial activity towards M. marinum, derivatives. with mean MIC50 of 2.3 ±0.33 µM, The set of new 18 lipophilic OX derivatives comparable to that of RIF (1.4 µM). based on MmPPOX core-structure was Interestingly, these three OX compounds were designed by varying the nature of the R chain also among the best inhibitors of M. bovis BCG and/or the positioning of the phenoxy group (in growth; HPOX being the best one (MIC50 = meta or para position) when present (Figure 3.5 µM). 1C), and synthesized as previously reported From these encouraging data obtained using [27]. The (phenoxy)phenyl group, proposed to two slow-growing mycobacteria, it was be responsible for strong hydrophobic tempting to extrapolate that these OXs would interactions and structural stiffening [27], was behave in a similar way against M. tb growth conserved in most of the new candidate and conclude that the same three compounds, inhibitors. In addition, modifying the R chain i.e. BePOX, HPOX and iBPOX, would then born by the oxadiazolone ring allow an be promising antiTB molecules. investigation of the influence of the Drug susceptibility testing of the 19 OXs was lipophilicity on the antibacterial activity thus further assessed using the non-virulent M. exerted by these molecules. To remain tb mc26230 strain. Among all tested consistent with the names of previous compounds, 14 OXs were active against M. tb compounds [23, 27], we used a specific mc26230. The two best growth inhibitors nomenclature for these derivatives noted obtained were iBpPPOX and BePOX which Rm(or p)PPOX; where m(or p)P represents displayed similar MIC50 value (mean 32.1 ±1.0 the meta (or para)-Phenoxy group when µM). In all other cases, MIC50 values were present; P the phenyl group; OX the indicative either of a weak (mean MIC50 = Oxadiazolone core; and R the alkyl chain (i.e., 84.1 ±9.1 µM for MmPPOX, MPOX and Be, benzyloxyethyl; M; methyl, E, ethyl; B, MemPPOX), or a moderate (mean MIC50 = butyl; iB, isobutyl; H, hexyl; O, octyl; Eh, 2- 46.9 ±5.1 µM for BmPPOX, iBmPPOX, ethylhexyl; D, decyl; Do, dodecyl; Me, HmPPOX, HpPPOX, HPOX, OmPPOX, methoxyethyl). EhmPPOX, BemPPOX and BepPPOX) antibacterial activity (Table 1). From these 2.2 Susceptibility testing on selected data, M. tb mc26230 was found nearly 13- and mycobacteria. 5-times less sensitive to the OX compounds The antibacterial properties of the OX compounds were first evaluated towards three than M. marinum and M. bovis BCG, 2.3 High-content screening assay on virulent respectively. M. tb H37Rv. Surprisingly, iBPOX, which differs from As recently observed for another class of BePOX and HPOX by the length of its R potent M. tb growth inhibitors, the substituent, was not active against this Cyclipostins & Cyclophostin analogs (CyC) mycobacteria. Moreover, no clear trends or [31], MIC50 values determined against M. tb rules in terms of structure-activity mc26230 do not necessarily correlate with the relationships (SAR) have emerged regarding activity against M. tb H37Rv, particularly the the potency of these oxadiazolone-core activity against intracellular bacteria. compounds against M. marinum and M. bovis Consequently, each of the 19 OX derivatives BCG. Indeed, increasing the lipophilicity by have been further evaluated for their specific varying the nature of the R chain on the antitubercular activity against extracellularly- oxadiazolone ring and/or the positioning of the and intracellularly-growing virulent M. tb phenoxy group when present had no real H37Rv-GFP strain, using a high-content impact on the anti-mycobacterial activity. screening assay based on the fluorescence Interestingly, with M. tb mc26230, some SAR measurement of GFP-expressing bacteria. In tendencies can however be set up. First, and as vitro growth (i.e., extracellular assay) of M. tb mentioned above, the positioning of the H37Rv-GFP was first monitored after 5 days phenoxy group in meta or para position has no at 37¡C in presence of increasing real impact on the antibacterial activity of the concentrations of candidate inhibitors [31, 35- corresponding compounds (i.e., MmPPOX vs. 37]. Intracellular growth of M. tb H37Rv-GFP MpPPOX; iBmPPOX vs. iBpPPOX; was also assessed following a 5-day exposure HmPPOX vs. HpPPOX; BemPPOX vs. of infected Raw264.7 murine macrophage cell BepPPOX). Remarkably, iBPOX bearing the line to the various OX compounds. In this case, short chain isobutyl has no activity as the percentage of infected cells and the number compared to the phenoxyphenyl derivatives of living host cells allowed to determine both iBpPPOX and iBmPPOX. This is however not the MIC50 and the compound concentration the case with the medium chains hexyl and leading to 50% of host cell cytotoxicity, i.e. benzyloxyethyl OXs, for which the respective CC50 [35, 38, 39]. activity of HPOX and BePOX is retained and Among all tested molecules, 6 potential OX even slightly better than HmPPOX & candidates exhibited interesting antitubercular HpPPOX on the one hand, and BemPPOX & properties (Table 2 and Figure 2). BePOX BepPPOX on the other hand. Finally, in and HPOX impaired exclusively M. tb growth absence of the bulky phenoxy group, the best in culture broth medium with the same 2 MIC50s against M. tb mc 6230 were also moderate MIC50 (30.8 and 44.6 µM, obtained with HPOX and BePOX vs. MPOX respectively) than obtained previously on M. tb and iBPOX. In brief, the R chain length thus mc26230. In contrast, iBPOX, HpPPOX and seems to affect the potency of the tested BepPPOX showed a clear preference against compounds. More globally, the best growth intracellularly-replicating mycobacteria with inhibitors were found to carry a middle chain similar MIC50 values (3.5-17.1 µM) than the length of around 6 to 9 carbon atoms (i.e., first line antibiotics. Of interest, only hexyl, 2-ethylhexyl, octyl or benzyloxyethyl iBpPPOX exhibited moderate (32.0 µM) to chains). Longer or shorter chain’s OXs (i.e., quite good (8.5 µM) activity against methyl, methoxyethyl, ethyl, decyl or dodecyl) extracellular and intramacrophagic M. tb, exhibited no or only very poor activity. respectively (Figure 2). Taken together, with the fact that iBpPPOX Beside antibacterial activity, significantly, all displayed the best (but moderate) antibacterial these 6 OX inhibitors exhibited very low activity, all these findings imply that the OXs toxicity towards host macrophages with CC50 may exhibit a different range of selectivity > 100 µM, similarly to INH (CC50 > 150 µM) 2 against M. tb mc 6230 mycobacterial enzymes, and ethionamide (CC50 ≥ 120 µM), two potent leading to various target enzymes. anti-TB drugs (Table 2). Their respective selectivity index (SI = CC50/MIC50) on intramacrophagic M. tb vs. raw264.7 cells was the phenoxy group in para position clearly thus found to be in a range from 5.8 and up to shifted the activity of the corresponding OX 28. from extracellular- (i.e., HPOX & BePOX) to Interestingly, these observed differences in the intracellular-replicating bacilli (i.e., HpPPOX behavior of OX compounds; in particular the & BepPPOX). On the other hand, with the higher activity against intracellular bacteria short isobutyl chain, both iBpPPOX and than against extracellular ones, have already iBPOX are found most active against been reported in the case of the CyC intramacrophagic-replicating M. tb. compounds family [31]. As previously From these findings, it is tempting to assume discussed above, inhibition of M. tb H37Rv that these OX compounds thus lead to the may result from several (and probably inhibition of specific but most likely distinct different) mechanisms of action or penetration mycobacterial target enzymes between of the OX derivatives. With the hexyl and intramacrophagic- vs. extracellularly- benzyloxyethyl chain, the absence/presence of replicating bacilli.

Figure 2. In vitro and ex vivo dose-response activity of the OX derivatives against M. tb H37Rv. (A) Activity of BePOX, iBpPPOX and HPOX against GFP-expressing M. tb replicating in broth medium, expressed as normalized relative fluorescence units (RFU%). (B) Activity of iBpPPOX, BepPPOX, HpPPOX and iBPOX against M. tb replicating inside Raw264.7 macrophages. Results are expressed as the percentage of infected macrophages after 5 days post-infection. For each concentration, data are means ± SD of at least two independent assays performed in duplicate. The MIC50 of BePOX, iBpPPOX and HPOX replicating in culture broth medium were 30.8 µM, 32.0 µM and 44.6 µM, respectively. The MIC50 of BepPPOX, iBpPPOX, HpPPOX and iBPOX replicating inside macrophages were 3.5 µM, 8.5 µM, 9.5 µM and 17.1 µM, respectively. Values are means ± SD of three independent assays performed in triplicate (CV% < 5%). 2.4 Targets identification by Activity-based (Figure 3). In parallel, the remaining lysate protein profiling. was incubated with TAMRA-FP, also non- Based on the aforementioned results, and specifically targeting (Ser/Cys)-based taking into account their known mechanism of enzymes, to reveal the candidates presumably action as well as their strong affinity for reacting with HPOX on SDS-PAGE gel, using serine/cysteine enzymes (Figure 1B), one can fluorescence scanning [31]. Around 9 distinct hypothesize that OX inhibitors might target bands labelled by TAMRA-FP were visible in and impair the activity of various enzymes the fluorescence readout (Figure 3B Ðlane E) involved in several processes of M. tb and could also be detected by silver staining pathogenic life cycle, thus resulting in bacterial after release of the enzymes captured by death without any cytotoxicity towards host Desthiobiotin-FP (Figure 3B Ð lane B). In cells. Accordingly, target(s) identification contrast, pre-treatment with HPOX (Figure experiments were conducted by applying an 3A) resulted in a decrease in fluorescence activity-based protein profiling (ABPP) intensity of all visible bands, as exemplified by approach [40-43]. In order to take into account the black arrows in Figure 3B Ð lane D. the penetration/diffusion of the inhibitor Indeed, the enzymes previously inactivated by through the mycobacterial cell wall, all HPOX inhibitor will thus be unable to further experiments have been performed on living react with the probes. The respective enriched bacterial cells and not with a lysate, as mixtures (Figure 3B Ð lanes A-B) were previously described [31]. digested with trypsin and the resulting peptides Here, HPOX, that selectively inhibits M. tb were analyzed by liquid chromatography- growth only in culture broth medium, was tandem mass spectrometry (LC-MS/MS) selected for such experiments. M. tb mc26230 followed by subsequent label free cells were grown to log phase and then quantification analysis. The proteins that were incubated with HPOX compound or DMSO as also found in the control experiment (i.e., a control. After cell lysis, part of the lysate was Figure 3B - lane A: DMSO alone for used for competitive probe unspecific binding to streptavidin-magnetic labelling/enrichment assay using the non- beads) were not taken into account. specific Desthiobiotin-FP probe, targeting Serine/Cysteine (Ser/Cys)-based enzymes

Figure 3. Activity based protein profiling (ABPP) workflow for the identification of the proteins covalently bound to OX inhibitors. (A) Cell culture of M. tb mc26230 was pre-treated with selected HPOX inhibitor prior cell lysis and further incubation with Desthiobiotin-FP probe. Samples were then treated with streptavidin-magnetic beads for the capture and enrichment of labelled proteins. (B) Equal amounts of proteins obtained in A were separated by SDS-PAGE and visualized by silver staining (right panel Ð lanes A-C) or in- gel fluorescence (left panel - lanes D-E: TAMRA detection). Enzymes whose TAMRA-FP labelling is impeded because of the presence of HPOX in their active-site are shown by arrowheads. (C) Tryptic digestion followed by tandem mass spectrometry analyses and subsequent differential proteomics analysis allowed identifying the target proteins.

The resulting mycobacterial targets of HPOX ranged in their functional category from were displayed as volcano plot (Figure 4). intermediary metabolism/respiration (6 Only proteins identified with a permutation proteins), lipid metabolism (5 proteins), cell false discovery rate (pFDR) of 5% and a score wall/cell processes (6 proteins), and virulence threshold value ≥ 60 were selected, therefore / detoxification / adaptation (1 protein) (Table leading to a panel of 18 distinct proteins 3). (Table 3). These identified enzyme candidates

Figure 4. Volcano plot of proteomic analysis of HPOX in M. tb mc26230 culture. Volcano plot plots significance two-sample t-test (-Log p-value) versus fold-change (Log2 (LFQ normalized intensity in HPOX versus Desthiobiotin-FP condition) on the y and x axes, respectively. Here the plot is zoomed on the positive difference between the two conditions. The full line is indicative of protein hits obtained at a permutation false discovery rate of 1% (pFDR). Only protein hits obtained at a pFDR of 5% (dashed line) and with a score threshold value ≥ 60 have been selected, and are highlighted in black (non-essential) or in red (essential). Data results from two different experiments processed three times.

More interestingly, 5 out of 18 identified As expected, the identified proteins targeted by proteins have been annotated as essential HPOX were all Ser/Cys-based enzymes. enzymes [48] (Table 3). These include the Among them, a variety of Ser/Cys hydrolases antigen 85 complex, Ag85A (Rv3804c), were detected. These included the putative Ag85B (Rv1886c) and Ag85C (Rv0129c) β-lactamase Rv1367c possibly involved in cell [49]; the thioesterase TesA (Rv2928) [50]; the wall biosynthesis; two amidases AmiC carboxylesterase CaeA (Rv2224c) [51]; the (Rv2888c) and AmiD (Rv2888c); BpoC beta-ketoacyl synthase KasA (Rv2245) [52]; (Rv0554) a putative serine hydrolase [42]; two and the sole putative α/β-hydrolase MetA members of the lipase family Lip (LipH [44] (Rv3341) belonging to the homoserine O- and LipV [45]); three Cutinase-like proteins acetyltransferase family proteins in M. tb [53]. (Cfp21, Cut2 and Cut3) [46]; and the Moreover, these OXs derivatives behave monoacylglycerol lipase Rv0183 [47]. against M. tb extracellular growth as two well- known non-specific Ser-/Cys-enzyme inhibitors, namely the Orlistat (MIC ~ 25 µM) The first 13 oxadiazolone derivatives 5a-k, 6k [40, 54] and the human lysosomal acid lipase and 7k were synthesized as described inhibitor lalistat (MIC ~ 25Ð50 µM) [41]. previously [27, 55]. The new six derivatives Similarly than these latter compounds, HPOX, 6a,d,e and 7a,d,e were prepared from and certainly all the other active OX commercial (4-phenoxyphenyl)hydrazine compounds, may act as multi-target inhibitors hydrochloride (2) and phenylhydrazine by impairing the activities of multiple non- hydrochloride (3), respectively, by performing essential lipolytic enzymes as well as essential both the coupling reaction with alkyl proteins involved in various important chloroformate 2a-k (step i) and the cyclization physiological pathways of M. tb life cycle. reaction with diphosgene (step ii) in a one-pot two-steps reaction [27]. All compound were at 3. Conclusion least 98% pure [27]. Stock solutions (4 In conclusion, we have developed a new series mg/mL) in which the oxadiazolone compounds of promising oxadiazolone-core compounds, were found to be completely soluble in active against three pathogenic mycobacteria. dimethyl sulfoxide (DMSO), were prepared Although they exhibited moderate MIC50 prior to drug susceptibility testing. See values against M. tb H37Rv, as compared to Supplementary Material for NMR and HRMS classic antibiotics or the more recent CyC spectra of the new six OX derivatives. analogs [31], OX probes would however represent attractive chemical tools, similarly 4.1.1 General procedure for the one step than Orlistat [40, 54] and lalistat [41], for preparation of 5-alkoxy-3-aryl-1,3,4- identifying Ser-/Cys-containing enzymes in oxadiazol-2(3H)-one compounds. living mycobacteria. Indeed, by blocking 4.1.1.1. 5-(2-(benzyloxy)ethoxy)-3-(3- extracellular and/or intracellular M. tb growth, phenoxyphenyl)-1,3,4-oxadiazol-2(3H)-one we anticipate that the OX probes could thus (6a = BepPPOX). (4- provide interesting insights in the mechanisms phenoxyphenyl)hydrazine hydrochloride [55] operating during mycobacterial replication, 2 (8.2 g, 34.6 mmol, 1 equiv.) and 1-methyl persistence and/or reactivation that are major pyrrolidone (2.41 ml, 31.1 mmol, 0.9 equiv.) issues for deciphering the susceptibility and were dissolved in dry pyridine (700 mL). The the general development of TB. In addition, solution was cooled in an ice bath to 0 ¡C. since OX derivatives can be more easily Then, 2-benzyloxyethyl chloroformate 4a synthesized compared to CyC analogs [31], (6.87 mL, 38.6 mmol, 1.1 equiv.) was added chemical modulations may be further dropwise over a period of 30 min at 0-5 ¡C and investigated to discover more selective and allowed to stir for 1 h at 0 ¡C and 1 h at room potent chemical structure, in the near future. temperature. The reaction mixture was diluted Moreover, as these compounds target bacterial by addition of methylene chloride (300 mL) pathway yet unexploited by anti-TB and dry pyridine (70 mL) and the mixture was compounds, the identification of the cooled at -10 ¡C using an ice-salt bath. A mycobacterial enzymes inhibited by these OX solution of diphosgene (6.26 mL, 51.8 mmol, compounds in vivo will contribute to 1.5 equiv.) in methylene chloride (30 mL) was background information for the development added dropwise using a syringe pump over a of new therapeutic strategies for elimination of period of 1 h while maintaining -10¡C with an either actively replicating or latent bacilli from ice-salt bath. After the addition is complete the infected individuals. Accordingly, given the reaction mixture stirred 1 h at -10 ¡C and 2 h at importance of such enzymes for M. tb viability room temperature. The reaction mixture was during infection, they should represent new diluted with water (1 L) and extracted with attractive drug targets. Such experiments are diethyl ether (3× 250 mL). The combined currently underway, and will be reported in organic layers were washed with water (2× 250 due course. mL) and brine (3× 100 mL), dried over MgSO4, and filtered. Purification by column 4. Experimental Section chromatography using cyclohexane/ethyl 4.1. Chemistry. acetate (98/2 to 95/5, v/v) as eluent gave the title compound 6a (BepPPOX) as a yellow oil similar method as described above for 6a. (9.94 g, 71%). Analytical data for BepPPOX: Analytical data for BePOX: pale yellow oil Rf (AcOEt/Cyclohexane 1:3, v/v) 0.36. HRMS (79%). Rf (AcOEt/Cyclohexane 1:3, v/v) 0.36. + + (ESI) m/z [M+H] calcd. for C23H21N2O5: HRMS (ESI) m/z [M+H] calcd. for 1 405.1445 Da ; found : 405.1446 Da. H NMR C7H17N2O4: 313.1183 Da ; found : 313.1183 δ 7.71 (dd, J = 9.1 Hz, J = 2.2 Hz, 2H), 7.31 Da. 1H NMR δ 7.75 (m, 2H), 7.18-7.73 (m, (m, 7H), 6.98-7.13 (m, 5H), 4.60 (s, 2H), 4.54 8H), 4.60 (s, 2H), 4.54 (m, 2H), 3.83 (m, 2H). (m, 2H), 3.83 (m, 2H). 13C NMR δ 157.14 (s), 13C NMR δ 155.23 (s), 148.23 (s), 137.44 (s), 155.23 (s), 154.77 (s), 148.29 (s), 137.43 (s), 136.17 (s), 129.12 (2 × d), 128.55 (2 × d), 131.56 (s), 129.85 (2 × d), 128.55 (2 × d), 127.98 (d), 127.79 (2 × d), 125.56 (d), 117.95 127.99 (d), 127.79 (2 × d), 123.47 (d), 119.83 (2 × d), 73.41 (t), 70.55 (t), 66.98 (t). (2 × d), 119.47 (2 × d), 118.74 (2 × d), 73.42 (t), 70.56 (t), 66.99 (t). 4.1.1.5. 5-isoButyloxy-3-phenyl-1,3,4- oxadiazol-2(3H)-one (7d = iBPOX). Prepared 4.1.1.2. 5-isoButyloxy-3-(4-phenoxyphenyl)- using Isobutyl chloroformate 4d and 1,3,4-oxadiazol-2(3H)-one (6d = iBpPPOX). phenylhydrazine hydrochloride 3 applying Prepared using Isobutyl chloroformate 4d similar method as described above for 6a. applying similar method as described above Analytical data for iBPOX: white crystals for 6a. Analytical data for iBpPPOX: pale (77%). mp: 76-77¡C. Rf (AcOEt/Cyclohexane + yellow oil (87%). Rf (AcOEt/Cyclohexane 1:3, 1:3, v/v) 0.55. HRMS (ESI) m/z [M+Na] + v/v) 0.55. HRMS (ESI) m/z [M+H] calcd. for calcd. for C12H14N2O3Na: 257.0897 Da ; 1 C18H19N2O4: 327.1339 Da ; found : 327.1342 found : 257.0895 Da. H NMR δ 7.80 (m, 2H), Da. 1H NMR δ 7.72 (dd, J = 9.2 Hz, J = 2.3 Hz, 7.42 (m, 2H), 7.25 (m, 1H), 4.18 (d, J = 6.6 Hz, 2H), 7.30 (m, 2H), 6.97-7.12 (m, 5H), 4.14 (t, 2H), 1.85 (m, 1H), 1.05 (d, 6H). 13C NMR δ J = 6.6 Hz, 2H), 2.15 (m, 1H), 1.0 (d, 6H). 13C 155.63 (s), 148.51 (s), 136.44 (s), 129.29 NMR δ 157.17 (s), 155.42 (s), 154.67 (s), (2 × d), 125.67 (d), 118.14 (2 × d), 77.57 (t), 148.36 (s), 131.66 (s), 129.83 (2 × d), 123.42 27.94 (d), 18.89 (2 × q). (d), 119.79 (2 × d), 119.49 (2 × d), 118.69 (2 × d), 71.39 (t), 22.75 (t), 18.70 (2 × q). 4.1.1.6. 5-Hexyloxy-3-phenyl-1,3,4-oxadiazol- 2(3H)-one (7e = HPOX). Prepared using 4.1.1.3. 5-Hexyloxy-3-(4-phenoxyphenyl)- Hexyl chloroformate 4e and phenylhydrazine 1,3,4-oxadiazol-2(3H)-one (6e = HpPPOX). hydrochloride 3 applying similar method as Prepared using Hexyl chloroformate 4e described above for 6a. Analytical data for applying similar method as described above HPOX: white crystals (71%). mp: 34-35¡C. Rf for 6a. Analytical data for HpPPOX: yellow (AcOEt/Cyclohexane 1:3, v/v) 0.59. HRMS + oil (85%). Rf (AcOEt/Cyclohexane 1:3, v/v) (ESI) m/z [M+H] calcd. for C14H19N2O3: 0.95. HRMS (ESI) m/z [M+H]+ calcd. for 263.1390 Da ; found : 263.1390 Da. 1H NMR C20H23N2O4: 355.1652 Da ; found : 355.1652 δ 7.80 (ddd, J = 8.4 Hz, J = 2.3 Hz, J = 0.75 Da. 1H NMR δ 7.72 (dd, J = 9.0 Hz, J = 2.0 Hz, Hz, 2H), 7.42 (dd, J = 7.6 Hz, J = 1.7 Hz, 2H), 2H), 7.30 (m, 2H), 6.97-7.12 (m, 5H), 4.37 (t, 7.22 (t, J = 7.4 Hz, 1H), 4.40 (t, J = 6.6 Hz, J = 6.6 Hz, 2H), 1.80 (m, 2H), 1.30-1.58 (m, 2H), 1.85 (m, 2H), 1.33-1.49 (m, 6H), 0.93 (t, 6H), 0.90 (t, J = 6.8 Hz, 3H). 13C NMR δ 3H). 13C NMR δ 155.54 (s), 148.53 (s), 136.45 157.16 (s), 155.33 (s), 154.67 (s), 148.38 (s), (s), 129.30 (2 × d), 125.67 (d), 118.13 (2 × d), 131.65 (s), 129.83 (2 × d), 123.42 (d), 119.78 71.97 (t), 31.45 (t), 28.55 (t), 25.36 (t), 22.68 (2 × d), 119.49 (2 × d), 118.68 (2 × d), 71.79 (t), 14.17 (q). (t), 31.36 (t), 28.35 (t), 25.17 (t), 22.50 (t), 13.99 (q). 4.2. Biological evaluation 4.2.1. Bacterial strains and growth condition. 4.1.1.4. 5-Benzyloxyethoxy-3-phenyl-1,3,4- M. marinum ATCC BAA-535/M, M. bovis oxadiazol-2(3H)-one (7a = BePOX). Prepared BCG Pasteur 1173P2 and M. tb mc26230 using 2-benzyloxyethyl chloroformate 4a and (H37Rv DRD1 DpanCD [28]) strains were phenylhydrazine hydrochloride 3 applying routinely grown in Middlebrook 7H9 broth (BD Difco) supplemented with 0.2% glycerol, 4.2.4. High-content screening assay in infected 0.05% Tween 80 (Sigma-Aldrich), 10% oleic macrophages - Intracellular assay. acid, albumin, dextrose, catalase (OADC The growth of M. tb H37Rv-GFP strain in enrichment; BD Difco) (7H9-S) and 24 µg/mL macrophages was monitored by automated D-panthothenate (M. tb mc26230). For further fluorescence confocal microscope (Opera, intra and extracellular assays, M. tb H37Rv Perkin-Elmer) as already described [35, 56]. expressing GFP [35] was grown for 14 days in Briefly, bacteria were washed twice with PBS 7H9-S supplemented with 50 µg/mL and resuspended in RPMI 1640 medium hygromycin B (Euromedex). All cultures were (Invitrogen) supplemented with 10% heat- kept at 37 ¡C without shaking, except M. inactivated fetal bovine serum (FBS, marinum which was grown at 32 ¡C. Invitrogen). Murine (Raw264.7) macrophages (American Type Culture Collection TIB-71) 4.2.2. Susceptibility testing on M. marinum, M. were infected at a multiplicity of infection bovis BCG and M. tb mc26230. (MOI) of 2:1 and incubated 2 hours at 37 ¡C in The concentrations of compound leading to RPMI 1640 medium containing 10% FBS. 50% of bacterial growth (MIC50) were first Cells were then washed, treated with 50 µg/mL determined using the resazurin microtiter assay amikacin (Euromedex) for 1 hour at 37 ¡C to (REMA) [29, 30]. Briefly, log-phase bacteria kill all extra-cellular bacteria, washed again were diluted to a cell density of 5×106 cells/mL and finally seeded in 384-well plates (5 × 105 and 100 µL of this inoculum was grown in a cells/mL), containing 2-fold dilutions of 96-well plate in the presence of serial dilutions compounds in DMSO. The final volume of of compounds. After 7-14 days incubation, 20 DMSO was kept under 0.3%. Plates were µL of a 0.025% (w/v) resazurin solution was incubated for 5 days at 37 ¡C, 5% CO2. added to each well (200 µL) and incubation Infected cells were stained for 30 min using was continued until the appearance of a color Syto60 dye (Invitrogen) at a final change (from blue to pink) in the control well concentration of 5 µM before reading using (bacteria without antibiotics). Fluorescence of fluorescence confocal microscope (20X water the resazurin metabolite resorufin (lexcitation, objective; GFP: lex 488 nm, lem 520 nm; 530 nm; lemission, 590 nm) was then measured Syto60: lex 640 nm, lem 690 nm). Sigmoidal [30] and the concentration leading to 50% dose-response curves were fitted using Prism growth inhibition was defined as the MIC50. software (sigmoidal dose-response, variable See Supplementary Material for detailed slope model). The MIC50 was determined protocol. using ten-point dose-response curves as an average of the MIC50 of all parameters, the 4.2.3. High-content screening assay - ratio of infected cells and the bacterial area per Extracellular assay. infected cell. A 14 days old culture of M. tb H37Rv-GFP was washed twice with PBS and resuspended in 4.3. HPOX target enzymes identification 7H9 medium containing 10% OADC, 0.5% 4.3.1. Activity-based protein profiling (ABPP). glycerol, 0.05% Tween 80 and 50 µg/mL Homogeneous bacterial suspension of M. tb 2 hygromycin B. Bacteria were seeded in 384 mc 6230 in 7H9-S was adjusted at an OD600 of well plates (7 × 105 bacteria/mL) containing 2- 60 and then incubated with the selected HPOX fold dilutions of the compounds in DMSO. The inhibitor (400 µM final concentration) or final volume of DMSO was kept under 0.3%. DMSO (control) at 37¡C for 2-3 h. under Plates were incubated at 37 ¡C, 5% CO2 for 5 gentle shaking at 75 rpm. Bacteria were then days. Bacterial fluorescence levels (RFU) were washed 3 times with PBS containing 0.05% recorded using a fluorescent microplate reader Tween 80, resuspended in PBS buffer at a 1:1 (Victor X3, Perkin-Elmer). The MIC50 of all (w/v) ratio and then lysed by mechanical tested compounds were determined using ten- disruption on a BioSpec Beadbeater. Both point dose-response curves. HPOX-treated M. tb mc26230 and DMSO- control lysate samples (750 µL Ð 0.75 mg total proteins) were labeled with 2 µM Desthiobiotin-FP probe for 90 min at room statistical analysis was done with Perseus temperature. Samples were enriched for program (version 1.5.6.0). Differential biotinylated proteins using Nanolink proteins were detected using a two-sample t- streptavidin magnetic beads 0.8 µm (Solulink), test at 0.01 and 0.05 permutation based FDR. according to the manufacturer’s instructions. The mass spectrometry proteomics data, The resulting captured biotinylated proteins including search results, have been deposited solution was mixed with 5X Laemmli reducing to the ProteomeXchange Consortium sample buffer, and heated at 95 ¡C for 5 min. (www.proteomexchange.org) [59] via the The released denatured proteins were PRIDE partner repository with the dataset subjected to tryptic digestion, peptide identifier PXD010255. Detailed Materials and extraction, and LC-MS/MS analysis as Methods is given in Supplementary Material. described below. Alternatively, the HPOX-treated M. tb Acknowledgements 2 mc 6230 and DMSO-control lysate samples This work was supported by the CNRS and (100 µL Ð 100 µg total proteins) were Aix-Marseille University. PCN was supported incubated with 2 µM ActivX TAMRA-FP by a PhD Training program from the probe (Thermo Fisher Scientific) for 90 min at University of Science and Technology of room temperature and in absence of light. The Hanoi (837267E). Financial support to PB, VD reaction was stopped by adding 5X Laemmli and VL was provided by the European reducing sample buffer and boiling at 95 ¡C for Community (ERC-STG INTRACELLTB n¡ 5 min. The labeled proteins were further 260901, MM4TB n¡260872, CycloNHit n¡ separated by SDS-PAGE electrophoresis. 608407), the Agence Nationale de la TAMRA fluorescence (TAMRA: λex 557 nm, Recherche (ANR-10-EQPX-04-01, ANR-14- λem 583 nm) was detected using a ChemiDoc CE08-0017), the Feder (12001407 (D-AL) MP Imager (Bio-Rad). Detailed Materials and Equipex Imaginex BioMed), the Région Nord Methods is given in Supplementary Material. Pas de Calais (convention n¡ 12000080). Proteomics analysis was supported by the 4.3.2. Protein identification and Institut Paoli-Calmettes and the Centre de quantification. Protein extract were loaded and Recherche en Cancérologie de Marseille. stacked on a NuPAGE gel (Life Technologies). Proteomic analyses were done using the mass Stained bands were submitted to an in-gel spectrometry facility of Marseille Proteomics trypsin digestion [57]. Peptides extracts were (marseille-proteomique.univ-amu.fr) reconstituted with 0.1% trifluoroacetic acid in supported by IBISA (Infrastructures Biologie 4% acetonitrile and analyzed by liquid Santé et Agronomie), Plateforme chromatography (LC)-tandem mass Technologique Aix-Marseille, the spectrometry (MS/MS) using an Orbitrap Cancéropôle PACA, the Provence-Alpes-Côte Fusion Lumos Tribrid Mass Spectrometer d'Azur Région, the Institut Paoli-Calmettes (Thermo Electron, Bremen, Germany) online and the Centre de Recherche en Cancérologie with a an Ultimate 3000RSLC nano de Marseille. chromatography system (Thermo Fisher Scientific, Sunnyvale, CA). Protein Supplementary Material identification and quantification were Supplementary data associated with this article processed using the MaxQuant computational can be found, in the online version. proteomics platform, version 1.5.3.8 [58] using a UniProt M. tuberculosis ATCC 25618 database (date 2018.01; 2164 entries). The Tables Table 1. Antibacterial activities of the oxadiazolone derivatives against M. marinum, M. Bovis BCG and M. tb mc26230 using the REMA method a

MIC (µM) Compounds 50 M. marinum M. bovis BCG M. tb mc26230 INH 20.5 0.71 1.5 RIF 1.4 0.027 0.015 MmPPOX 7.0 15.0 71.3 MpPPOX 11.5 39.9 >120 MPOX 52.9 >120 89.8 EmPPOX 3.4 7.1 >120 MemPPOX 2.7 9.4 91.2

BmPPOX 4.1 6.6 52.2 iBmPPOX 13.6 4.4 57.2 iBpPPOX 6.9 8.5 31.1 iBPOX 2.5 5.7 >120 HmPPOX 11.1 5.8 48.8 HpPPOX 3.4 14.0 41.3 HPOX 2.6 3.5 40.5 BemPPOX 5.6 10.5 49.5 BepPPOX 10.1 18.5 44.5 BePOX 1.9 6.6 33.2 OmPPOX 8.0 11.0 44.7

EhmPPOX 8.3 8.1 43.3

DmPPOX 10.5 80.8 >120

DomPPOX 19.5 22.6 >120

a Experiments were performed as described in Materials and Methods. MIC50: compound minimal concentration leading to 50% of growth inhibition, as determined by the REMA assay. The best MIC50 obtained for each strain are highlighted in bold. Values are mean of at least two independent assays performed in triplicate. INH, isoniazid; RIF, rifampicin.

Table 2. Antitubercular activities of the most active OX derivatives against M. tb H37Rv a

Extracellular Intracellular

Compounds growth macrophage growth b MIC50 (µM) MIC50 (µM) CC50 (µM) INH c 1.2 1.2 >150 RIF c 0.01 2.9 24 ETO c 6.0 6.0 120

iBpPPOX 32.0 8.5 >100

iBPOX >50 17.1 >100

HpPPOX >50 9.5 >100

HPOX 44.6 No effect >100

BepPPOX >50 3.5 >100

BePOX 30.8 No effect >100

a Experiments were performed as described in Materials and Methods. MIC50: compound minimal concentration leading to 50% of growth inhibition. CC50: compound concentration leading to 50% of cell cytotoxicity. Values are means of three independent assays performed in triplicate (CV% < 5%). b Raw264.7 macrophages were infected by M. tb H37Rv-GFP at a MOI of 2:1. c Data from [35]. INH, isoniazid; RIF, rifampicin; ETO, ethionamide.

Table 3. HPOX target proteins identified in M. tb mc26230 culture by LC-ESI-MS/MS analysis. a

Mol. weight Functional Protein name Rv number Essentiality Location b Function [kDa] Category c Cutinase Cfp21 Rv1984c 21.8 CF; CW; M Lipase/esterase CW/CP Probable cutinase Cut2 Rv2301 23.9 M, WCL, CF hydrolase CW/CP Probable cutinase Cut3 Rv3451 26.5 M, CF hydrolase CW/CP Lipase LipV Rv3203 27.9 WCL Lipase/esterase IM/R Non-heme bromoperoxidase BpoC Rv0554 28.4 M hydrolase V/D/A Thioesterase TesA Rv2928 29.2 in vitro growth M Lipase/esterase LM Monoglyceride lipase Rv0183 Rv0183 30.3 M Lipase/esterase IM/R Lipase LipH Rv1399c 33.9 M Lipase/esterase IM/R Secreted antigen 85-B FbpB (Ag85B) Rv1886c 34.6 CF; CW; M Lipase/esterase LM Secreted antigen 85-A FbpA (Ag85A) Rv3804c 35.7 in vitro CF; CW; M Lipase/esterase LM Secreted antigen 85-C FbpC (Ag85C) Rv0129c 36.8 CF; CW Lipase/esterase LM Probable homoserine O-acetyltransferase MetA Rv3341 39.8 essential gene M Acyltransferase IM/R β-ketoacyl-ACP synthase KasA Rv2245 43.3 essential gene CF; CW; M Lipase/esterase LM Hypothetical protein LH57_07490 Rv1367c 43.7 M β-lactamase CW/CP Trigger factor Tig Rv2462c 50.6 M, WCL, CF Protein export CW/CP Amidase AmiD Rv3375 50.6 CW; M Amidase IM/R Amidase AmiC Rv2888c 50.9 CW; M Amidase IM/R Putative Carboxylesterase A CaeA Rv2224c 55.9 Macrophage and in vitro M, WCL, CF Lipase/esterase CW/CP

a Only positive hits with a pFDR of 5% and a score threshold value ≥ 60 were selected. b CF: Culture filtrate; CW: Cell wall; M: Membrane fraction; WCL: Whole cell lysate. c IM/R: Intermediary metabolism/respiration; CW/CP: cell wall/cell processes; LM: Lipid metabolism; V/D/A: Virulence, detoxification, adaptation.

References review of current concepts and future [1] WHO, Global tuberculosis report, challenges, Clinical medicine, 14 (2014) 279- http://www.who.int/tb/publications/global_re 285. port/en/ (2017). [10] G. Gunther, F. van Leth, N. Altet, M. [2] A.I. Zumla, S.H. Gillespie, M. Hoelscher, Dedicoat, R. Duarte, G. Gualano, H. Kunst, I. P.P. Philips, S.T. Cole, I. Abubakar, T.D. Muylle, V. Spinu, S. Tiberi, P. Viiklepp, C. McHugh, M. Schito, M. Maeurer, A.J. Nunn, Lange, Beyond multidrug-resistant New antituberculosis drugs, regimens, and tuberculosis in Europe: a TBNET study, Int J adjunct therapies: needs, advances, and future Tuberc Lung Dis, 19 (2015) 1524-1527. prospects, The Lancet. Infectious diseases, 14 [11] J.A. Aeschlimann, A. Stempel. (2014) 327-340. Heterocyclic antituberculous compounds. [3] M. Pai, M.A. Behr, D. Dowdy, K. Dheda, US2665279, 1954. M. Divangahi, C.C. Boehme, A. Ginsberg, S. [12] A.E. Wilder Smith, The Action of Swaminathan, M. Spigelman, H. Getahun, D. Phosgene on Acid Hydrazides to Give 1, 3, 4- Menzies, M. Raviglione, Tuberculosis, Nature Oxdiazolones of Interest in the Treatment of Reviews Disease Primers, 2 (2016) 16076. Tuberculosis, Science, 119 (1954) 514. [4] K. Andries, P. Verhasselt, J. Guillemont, [13] A.E. Wilder Smith, H. Brodhage, H.W. Gohlmann, J.M. Neefs, H. Winkler, J. Biological spectrum of some new Van Gestel, P. Timmerman, M. Zhu, E. Lee, P. tuberculostatic 1,3,4-oxadiazolones with Williams, D. de Chaffoy, E. Huitric, S. special reference to cross-resistance and rates Hoffner, E. Cambau, C. Truffot-Pernot, N. of emergence of resistance, Nature, 192 (1961) Lounis, V. Jarlier, A diarylquinoline drug 1195. active on the ATP synthase of Mycobacterium [14] A.E. Wilder Smith, H. Brodhage, G. tuberculosis, Science, 307 (2005) 223-227. Haukenes, Some tuberculostatic 1,3,4- [5] M. Matsumoto, H. Hashizume, T. oxadiazolones(-5) and 1,3,4- Tomishige, M. Kawasaki, H. Tsubouchi, H. oxadiazolthiones(-5). II: Biological spectrum Sasaki, Y. Shimokawa, M. Komatsu, OPC- in vitro and activity in vivo in relation to 67683, a nitro-dihydro-imidazooxazole resistance emergence, Arzneimittel- derivative with promising action against Forschung, 12 (1962) 275-280. tuberculosis in vitro and in mice, PLoS Med, 3 [15] A.E. Wilder Smith, Some recently (2006) e466. synthesised tuberculostatic 4-substituted [6] A.H. Diacon, R. Dawson, F. von Groote- oxadiazolones and oxadiazol-thiones. VI, Bidlingmaier, G. Symons, A. Venter, P.R. Arzneimittel-Forschung, 16 (1966) 1034- Donald, C. van Niekerk, D. Everitt, H. Winter, 1038. P. Becker, C.M. Mendel, M.K. Spigelman, 14- [16] M.G. Mamolo, D. Zampieri, L. Vio, M. day bactericidal activity of PA-824, Fermeglia, M. Ferrone, S. Pricl, G. Scialino, E. bedaquiline, pyrazinamide, and moxifloxacin Banfi, Antimycobacterial activity of new 3- combinations: a randomised trial, Lancet, 380 substituted 5-(pyridin-4-yl)-3H-1,3,4- (2012) 986-993. oxadiazol-2-one and 2-thione derivatives. [7] A.A. Velayati, M.R. Masjedi, P. Farnia, P. Preliminary molecular modeling Tabarsi, J. Ghanavi, A.H. Ziazarifi, S.E. investigations, Bioorg Med Chem., 13 (2005) Hoffner, Emergence of new forms of totally 3797-3809. drug-resistant tuberculosis bacilli: super [17] D. Zampieri, M.G. Mamolo, E. Laurini, extensively drug-resistant tuberculosis or M. Fermeglia, P. Posocco, S. Pricl, E. Banfi, totally drug-resistant strains in iran, Chest, 136 G. Scialino, L. Vio, Antimycobacterial activity (2009) 420-425. of new 3,5-disubstituted 1,3,4-oxadiazol- [8] C.D. Acosta, A. Dadu, A. Ramsay, M. 2(3H)-one derivatives. Molecular modeling Dara, Drug-resistant tuberculosis in Eastern investigations, Bioorg Med Chem., 17 (2009) Europe: challenges and ways forward, Public 4693-4707. Health Action, 4 (2014) S3-S12. [18] A. Bellamine, A.T. Mangla, W.D. Nes, [9] G. Gunther, Multidrug-resistant and M.R. Waterman, Characterization and extensively drug-resistant tuberculosis: a catalytic properties of the sterol 14alpha- demethylase from Mycobacterium targeting selectively gastric lipolysis, tuberculosis, Proc Natl Acad Sci USA, 96 European Journal of Medicinal Chemistry, 123 (1999) 8937-8942. (2016) 834-848. [19] D.J. Sheehan, C.A. Hitchcock, C.M. [28] V.K. Sambandamurthy, S.C. Derrick, T. Sibley, Current and emerging azole antifungal Hsu, B. Chen, M.H. Larsen, K.V. Jalapathy, agents, Clin Microbiol Rev., 12 (1999) 40-79. M. Chen, J. Kim, S.A. Porcelli, J. Chan, S.L. [20] K.J. McLean, A.J. Dunford, R. Neeli, Morris, W.R. Jacobs, Jr., Mycobacterium M.D. Driscoll, A.W. Munro, Structure, tuberculosis DeltaRD1 DeltapanCD: a safe and function and drug targeting in Mycobacterium limited replicating mutant strain that protects tuberculosis cytochrome P450 systems, Arch immunocompetent and immunocompromised Biochem Biophys., 464 (2007) 228-240. mice against experimental tuberculosis, [21] Y. Ben Ali, H. Chahinian, S. Petry, G. Vaccine, 24 (2006) 6309-6320. Muller, R. Lebrun, R. Verger, F. Carrière, L. [29] J.C. Palomino, A. Martin, M. Camacho, Mandrich, M. Rossi, G. Manco, L. Sarda, A. H. Guerra, J. Swings, F. Portaels, Resazurin Abousalham, Use of an Inhibitor To Identify microtiter assay plate: simple and inexpensive Members of the Hormone-Sensitive Lipase method for detection of drug resistance in Family, Biochemistry, 45 (2006) 14183- Mycobacterium tuberculosis, Antimicrob 14191. Agents Chemother, 46 (2002) 2720-2722. [22] Y. Ben Ali, R. Verger, F. Carrière, S. [30] J. Rybniker, A. Vocat, C. Sala, P. Busso, Petry, G. Muller, A. Abousalham, The F. Pojer, A. Benjak, S.T. Cole, Lansoprazole is molecular mechanism of human hormone- an antituberculous prodrug targeting sensitive lipase inhibition by substituted 3- cytochrome bc1, Nat. Commun., 6 (2015) phenyl-5-alkoxy-1,3,4-oxadiazol-2-ones, 7659. Biochimie, 94 (2012) 137-145. [31] P.C. Nguyen, V. Delorme, A. [23] V. Delorme, S.V. Diomandé, L. Dedieu, Bénarouche, B.P. Martin, R. Paudel, G.R. J.-F. Cavalier, F. Carrière, L. Kremer, J. Gnawali, A. Madani, R. Puppo, V. Landry, L. Leclaire, F. Fotiadu, S. Canaan, MmPPOX Kremer, P. Brodin, C.D. Spilling, J.-F. Inhibits Mycobacterium tuberculosis Lipolytic Cavalier, S. Canaan, Cyclipostins and Enzymes Belonging to the Hormone-Sensitive Cyclophostin analogs as promising Lipase Family and Alters Mycobacterial compounds in the fight against tuberculosis, Growth, PLoS ONE, 7 (2012) e46493. Scientific Reports, 7 (2017) 11751. [24] B. Borgström, Mode of action of [32] R.J. Wallace, Jr., D.R. Nash, L.C. Steele, tetrahydrolipstatin: A derivative of the V. Steingrube, Susceptibility testing of slowly naturally occuring lipase inhibitor lipstatin, growing mycobacteria by a microdilution MIC Biochim. Biophys. Acta, 962 (1988) 308-316. method with 7H9 broth, J Clin Microbiol, 24 [25] P. Hadváry, W. Sidler, W. Meister, W. (1986) 976-981. Vetter, H. Wolfer, The lipase inhibitor [33] A. Aubry, V. Jarlier, S. Escolano, C. tetrahydrolipstatin binds covalently to the Truffot-Pernot, E. Cambau, Antibiotic putative active site serine of pancreatic lipase, susceptibility pattern of Mycobacterium J. Biol. Chem., 266 (1991) 2021-2027. marinum, Antimicrob Agents Chemother, 44 [26] A. Bénarouche, V. Point, F. Carrière, J.-F. (2000) 3133-3136. Cavalier, Using the reversible inhibition of [34] N. Ritz, M. Tebruegge, T.G. Connell, A. gastric lipase by Orlistat for investigating Sievers, R. Robins-Browne, N. Curtis, simultaneously lipase adsorption and substrate Susceptibility of Mycobacterium bovis BCG hydrolysis at the lipid-water interface, vaccine strains to antituberculous antibiotics, Biochimie, 101 (2014) 221-231. Antimicrob Agents Chemother, 53 (2009) 316- [27] V. Point, A. Bénarouche, J. Zarillo, A. 318. Guy, R. Magnez, L. Fonseca, B. Raux, J. [35] T. Christophe, M. Jackson, H.K. Jeon, D. Leclaire, G. Buono, F. Fotiadu, T. Durand, F. Fenistein, M. Contreras-Dominguez, J. Kim, Carrière, C. Vaysse, L. Couëdelo, J.-F. A. Genovesio, J.P. Carralot, F. Ewann, E.H. Cavalier, Slowing down fat digestion and Kim, S.Y. Lee, S. Kang, M.J. Seo, E.J. Park, absorption by an oxadiazolone inhibitor H. Skovierova, H. Pham, G. Riccardi, J.Y. Nam, L. Marsollier, M. Kempf, M.L. Joly- [42] C. Ortega, L.N. Anderson, A. Frando, Guillou, T. Oh, W.K. Shin, Z. No, U. N.C. Sadler, R.W. Brown, R.D. Smith, A.T. Nehrbass, R. Brosch, S.T. Cole, P. Brodin, Wright, C. Grundner, Systematic Survey of High content screening identifies decaprenyl- Serine Hydrolase Activity in Mycobacterium phosphoribose 2' epimerase as a target for tuberculosis Defines Changes Associated with intracellular antimycobacterial inhibitors, Persistence, Cell Chem Biol, 23 (2016) 290- PLoS Pathog, 5 (2009) e1000645. 298. [36] T. Christophe, F. Ewann, H.K. Jeon, J. [43] K.R. Tallman, S.R. Levine, K.E. Beatty, Cechetto, P. Brodin, High-content imaging of Small Molecule Probes Reveal Esterases with Mycobacterium tuberculosis-infected Persistent Activity in Dormant and macrophages: an in vitro model for Reactivating Mycobacterium tuberculosis, tuberculosis drug discovery, Future Med. ACS Infect. Dis., 2 (2016) 936-944. Chem., 2 (2010) 1283-1293. [44] S. Canaan, D. Maurin, H. Chahinian, B. [37] P. Brodin, T. Christophe, High-content Pouilly, C. Durousseau, F. Frassinetti, L. screening in infectious diseases, Curr Opin Scappuccini-Calvo, C. Cambillau, Y. Bourne, Chem Biol, 15 (2011) 534-539. Expression and characterization of the protein [38] M. Flipo, M. Desroses, N. Lecat-Guillet, Rv1399c from Mycobacterium tuberculosis. A B. Dirie, X. Carette, F. Leroux, C. Piveteau, F. novel carboxyl esterase structurally related to Demirkaya, Z. Lens, P. Rucktooa, V. Villeret, the HSL family, Eur J Biochem, 271 (2004) T. Christophe, H.K. Jeon, C. Locht, P. Brodin, 3953-3961. B. Deprez, A.R. Baulard, N. Willand, [45] G. Singh, S. Arya, D. Narang, D. Jadeja, Ethionamide boosters: synthesis, biological U.D. Gupta, K. Singh, J. Kaur, activity, and structure-activity relationships of Characterization of an acid inducible lipase a series of 1,2,4-oxadiazole EthR inhibitors, J Rv3203 from Mycobacterium tuberculosis Med Chem., 54 (2011) 2994-3010. H37Rv, Mol Biol Rep, 41 (2014) 285-296. [39] J. Neres, R.C. Hartkoorn, L.R. Chiarelli, [46] N.P. West, F.M. Chow, E.J. Randall, J. R. Gadupudi, M.R. Pasca, G. Mori, A. Wu, J. Chen, J.M. Ribeiro, W.J. Britton, Venturelli, S. Savina, V. Makarov, G.S. Kolly, Cutinase-like proteins of Mycobacterium E. Molteni, C. Binda, N. Dhar, S. Ferrari, P. tuberculosis: characterization of their variable Brodin, V. Delorme, V. Landry, A.L. de Jesus enzymatic functions and active site Lopes Ribeiro, D. Farina, P. Saxena, F. Pojer, identification, FASEB J., 23 (2009) 1694- A. Carta, R. Luciani, A. Porta, G. Zanoni, E. 1704. De Rossi, M.P. Costi, G. Riccardi, S.T. Cole, [47] K. Côtes, R. Dhouib, I. Douchet, H. 2-Carboxyquinoxalines Kill Mycobacterium Chahinian, A. De Caro, F. Carriere, S. Canaan, tuberculosis through Noncovalent Inhibition of Characterization of an exported DprE1, ACS Chemical Biology, 10 (2015) monoglyceride lipase from Mycobacterium 705-714. tuberculosis possibly involved in the [40] M.S. Ravindran, S.P. Rao, X. Cheng, A. metabolism of host cell membrane lipids, Shukla, A. Cazenave-Gassiot, S.Q. Yao, M.R. Biochem J., 408 (2007) 417-427. Wenk, Targeting Lipid Esterases in [48] J.C. Sacchettini, D.R. Ronning, The Mycobacteria Grown Under Different mycobacterial antigens 85 complex--from Physiological Conditions Using Activity- structure to function: response, Trends based Profiling with Tetrahydrolipstatin Microbiol, 8 (2000) 441. (THL), Molecular & Cellular Proteomics, 13 [49] J.T. Belisle, V.D. Vissa, T. Sievert, K. (2014) 435-448. Takayama, P.J. Brennan, G.S. Besra, Role of [41] J. Lehmann, J. Vomacka, K. Esser, M. the major antigen of Mycobacterium Nodwell, K. Kolbe, P. Ramer, U. Protzer, N. tuberculosis in cell wall biogenesis, Science, Reiling, S.A. Sieber, Human lysosomal acid 276 (1997) 1420-1422. lipase inhibitor lalistat impairs Mycobacterium [50] L. Alibaud, Y. Rombouts, X. Trivelli, A. tuberculosis growth by targeting bacterial Burguiere, S.L. Cirillo, J.D. Cirillo, J.F. hydrolases, MedChemComm, 7 (2016) 1797- Dubremetz, Y. Guerardel, G. Lutfalla, L. 1801. Kremer, A Mycobacterium marinum TesA mutant defective for major cell wall-associated M. Jackson, A.J. Lenaerts, H. Pham, V. Jones, lipids is highly attenuated in Dictyostelium M.J. Seo, Y.M. Kim, M. Seo, J.J. Seo, D. Park, discoideum and zebrafish embryos, Molecular Y. Ko, I. Choi, R. Kim, S.Y. Kim, S. Lim, S.- Microbiology, 80 (2011) 919-934. A. Yim, J. Nam, H. Kang, H. Kwon, C.-T. Oh, [51] S. Lun, W.R. Bishai, Characterization of Y. Cho, Y. Jang, J. Kim, A. Chua, B.H. Tan, a Novel Cell Wall-anchored Protein with M.B. Nanjundappa, S.P.S. Rao, W.S. Barnes, Carboxylesterase Activity Required for R. Wintjens, J.R. Walker, S. Alonso, S. Lee, J. Virulence in Mycobacterium tuberculosis, J Kim, S. Oh, T. Oh, U. Nehrbass, S.-J. Han, Z. Biol Chem., 282 (2007) 18348-18356. No, J. Lee, P. Brodin, S.-N. Cho, K. Nam, J. [52] R.A. Slayden, C.E. Barry, 3rd, The role of Kim, Discovery of Q203, a potent clinical KasA and KasB in the biosynthesis of candidate for the treatment of tuberculosis, Nat meromycolic acids and isoniazid resistance in Med, 19 (2013) 1157-1160. Mycobacterium tuberculosis, Tuberculosis [57] A. Shevchenko, M. Wilm, O. Vorm, M. (Edinb), 82 (2002) 149-160. Mann, Mass spectrometric sequencing of [53] G. Johnson, The alpha/beta Hydrolase proteins silver-stained polyacrylamide gels, Fold Proteins of Mycobacterium tuberculosis, Anal Chem., 68 (1996) 850-858. With Reference to their Contribution to [58] J. Cox, M. Mann, MaxQuant enables high Virulence, Curr Protein Pept Sci, 18 (2017) peptide identification rates, individualized 190-210. p.p.b.-range mass accuracies and proteome- [54] C. Rens, F. Laval, M. Daffe, O. Denis, R. wide protein quantification, Nat Biotechnol., Frita, A. Baulard, R. Wattiez, P. Lefevre, V. 26 (2008) 1367-1372. Fontaine, Effects of lipid-lowering drugs on [59] J.A. Vizcaino, E.W. Deutsch, R. Wang, A. vancomycin susceptibility of mycobacteria, Csordas, F. Reisinger, D. Rios, J.A. Dianes, Z. Antimicrobial Agents and Chemotherapy, 60 Sun, T. Farrah, N. Bandeira, P.A. Binz, I. (2016) 6193-6199. Xenarios, M. Eisenacher, G. Mayer, L. Gatto, [55] V. Point, K.V.P. Pavan Kumar, S. Marc, A. Campos, R.J. Chalkley, H.J. Kraus, J.P. V. Delorme, G. Parsiegla, S. Amara, F. Albar, S. Martinez-Bartolome, R. Apweiler, Carrière, G. Buono, F. Fotiadu, S. Canaan, J. G.S. Omenn, L. Martens, A.R. Jones, H. Leclaire, J.-F. Cavalier, Analysis of the Hermjakob, ProteomeXchange provides discriminative inhibition of mammalian globally coordinated proteomics data digestive lipases by 3-phenyl substituted 1,3,4- submission and dissemination, Nat oxadiazol-2(3H)-ones, European Journal of Biotechnol., 32 (2014) 223-226. Medicinal Chemistry, 58 (2012) 452-463. [56] K. Pethe, P. Bifani, J. Jang, S. Kang, S. Park, S. Ahn, J. Jiricek, J. Jung, H.K. Jeon, J. Cechetto, T. Christophe, H. Lee, M. Kempf,

Supplementary Material

Oxadiazolone derivatives, new promising multi-target inhibitors against M. tuberculosis Phuong Chi Nguyen1¤, Vincent Delorme2¤ #, Anaïs Bénarouche1, Alexandre Guy3, Valérie Landry2, Stéphane Audebert4, Matthieu Pophilat4, Luc Camoin4, Céline Crauste3, Jean-Marie Galano3, Thierry Durand3, Priscille Brodin2, Stéphane Canaan1* and Jean-François Cavalier1*

1 LISM, Institut de Microbiologie de la Méditerranée, CNRS and Aix-Marseille Univ., Marseille, France 2 Univ. Lille, CNRS, INSERM, CHU Lille, Institut Pasteur de Lille, U1019 - UMR 8204 - CIIL - Center for Infection and Immunity of Lille, Lille, France 3 Institut des Biomolécules Max Mousseron (IBMM), UMR 5247, Université de Montpellier, CNRS, ENSCM, 15 Avenue Charles Flahault, 34093 Montpellier Cedex 5, France 4 Aix-Marseille Univ, CNRS, INSERM, CNRS, Institut Paoli-Calmettes, CRCM, Marseille Protéomique, Marseille, France

¤ Authors have contributed equally to this work

Corresponding authors: [email protected] (J.-F. Cavalier); [email protected] (S. Canaan)

# Current Address: Tuberculosis Research Laboratory, Institut Pasteur Korea, Seongnam-si, Gyeonggi-do, 13488 Republic of Korea.

Contents

General methods of Synthesis S2 1H, 13C NMR and HRMS spectra (Figure S1-S6) S3

Detailed protocols for susceptibility testing and ABPP experiments S15

S1

General methods of Synthesis. Commercially available starting materials were purchased from Sigma-Aldrich (St Quentin Fallavier, France), TCI (Zwijndrecht, Belgium) or Alfa Aesar (Karlsruhe, Germany) and used without further purification. All reactions were carried out under strictly anhydrous conditions; and all solvents were purified according to usual methods

[1]. Analytical thin-layer chromatography (TLC) was carried out on Merck silica gel F254 (60 Å, 40–63 μm, 230–400 mesh) pre-coated aluminum sheets, and the following detection methods were used: UV lamp (254 nm) and PMA: dipped into a solution containing 5% phosphomolybdic acid in absolute ethanol, and heated on a hot plate. Flash chromatography separations were performed using Macherey Nagel silica gel 60 (230-400 mesh) according to Still et al. [2] 1H and 13C NMR spectra were recorded on BRUKER Avance 300 spectrometer operating at 300 MHz for 1H and 75 MHz for 13C. Chemical shifts are reported in ppm on the

1 scale from an internal standard of solvent (CDCl3, δ = 7.27 ppm for H NMR and 77.23 ppm in 13C NMR). Coupling constants (J) are reported in hertz. 1H spectral splitting patterns are designated as s, singlet; d, doublet; t, triplet; q, quartet; quint, quintet; sext, sextet; sept, septet; m, multiplet. 13C spectral splitting patterns are designated as s, singlet (quaternary elements); d, doublet (CH); t, triplet (CH2); q, quartet (CH3). Accurate mass measurements (HRMS) were performed with a SYNAPT G2 HDMS (Waters) mass spectrometer equipped with an electrospray ionization source (ESI), operated in the positive mode. In this hybrid instrument, ions were measured using an orthogonal acceleration time-of-flight (oa-TOF) mass analyzer. Infrared spectra were recorded on a Perkin Elmer Spectrum one spectrophotometer. Melting points were measured using a Stuart SMP3 without corrections. High-performance liquid chromatography analyses were used to confirm the purity of all compounds (≥ 98%), and were performed on a Perkin Elmer liquid chromatograph equipped with a Quaternary LC pump model 200Q/410 with series 200 autosampler and an LC 200 diode array detector (DAD) (all from Perkin Elmer, Bradford, CT, USA). Separation was performed on a C-18 reversed phase column Waters NovaPack (3.9 × 150 mm, 5.0 µm particle size) operating at 35 ¡C. The compound analyses were conducted at a flow rate of 1 mL/min using isocratic elution with a mobile phase of acetonitrile and water (50:50, v/v) and UV detection at 254 nm wavelength.

S2

S3

1 13 Fig. S1. H & C NMR spectra recorded on BRUKER Avance 300 spectrometer using CDCl3 as an internal standard of solvent, and HRMS spectrum of BepPPOX in ESI positive mode. Two internal standards (IS1 and IS2) are detected at m/z 388.2540 and m/z 432.2803 Da, respectively.

S4

S5

1 13 Fig. S2. H & C NMR spectra recorded on BRUKER Avance 300 spectrometer using CDCl3 as an internal standard of solvent, and HRMS spectrum of iBpPPOX in ESI positive mode. Two internal standards (IS1 and IS2) are detected at m/z 309.2271 and m/z 367.2690 Da, respectively.

S6

S7

1 13 Fig. S3. H & C NMR spectra recorded on BRUKER Avance 200 spectrometer using CDCl3 as an internal standard of solvent, and HRMS spectrum of HpPPOX in ESI positive mode. Two internal standards (IS1 and IS2) are detected at m/z 344.2278 and m/z 388.2540 Da, respectively.

S8

S9

1 13 Fig. S4. H & C NMR spectra recorded on BRUKER Avance 200 spectrometer using CDCl3 as an internal standard of solvent, and HRMS spectrum of BePOX in ESI positive mode. Two internal standards (IS1 and IS2) are detected at m/z 300.2016 and m/z 327.2013 Da, respectively.

S10

S11

1 13 Fig. S5. H & C NMR spectra recorded on BRUKER Avance 200 spectrometer using CDCl3 as an internal standard of solvent, and HRMS spectrum of iBPOX in ESI positive mode. Two internal standards (IS1 and IS2) are detected at m/z 239.1489 and m/z 262.1308 Da, respectively.

S12

S13

1 13 Fig. S6. H & C NMR spectra recorded on BRUKER Avance 200 spectrometer using CDCl3 as an internal standard of solvent, and HRMS spectrum of HPOX in ESI positive mode. Two internal standards (IS1 and IS2) are detected at m/z 239.1489 and m/z 283.1751 Da, respectively.

S14

Resazurin microtiter assay (REMA) for MIC determination. Susceptibility testing was first performed using the Middlebrook 7H9 broth microdilution method. All assays for each strain were carried out at least in triplicate. MICs of the OXs were determined in 96-well flat-bottom Nunclon Delta Surface microplates with lid (Thermo Fisher Scientific, ref. 167008) using the resazurin microtiter assay (REMA [3, 4]). Briefly, log-phase bacteria were diluted to a cell density of 5 × 106 cells/mL in 7H9-S (7H9 broth + 10% OADC + 0.5% glycerol + 0.05% Tween 80, and 24 µg/mL D-panthothenate when needed). Then 100 µL of the above inoculum (i.e., 5 × 105 cells per well) was added to each well containing 100 µL 7H9 -S medium, serial two- fold dilutions of the selected OX analog or controls to a final volume of 200 µL. Growth controls containing no inhibitor (i.e., bacteria only = B), inhibition controls containing 50 µg/mL kanamycin and sterility controls (i.e., medium only = M) without inoculation were also included. Plates were incubated at 37 ¡C (32 ¡C for M. marinum) in a humidity chamber [5] to prevent evaporation for 7-14 days. Then, 20 µL of a 0.02% (w/v) resazurin solution was added to each well, and the plates were incubated at 37¡C for color change from blue to pink or violet and for a reading of fluorescence units (FU). Fluorescence corresponding to the resazurin reduction to its metabolite resorufin was quantified using a Tecan Spark 10M multimode microplate reader (Tecan Group Ltd, France) with excitation at 530 nm and emission at 590 nm. For fluorometric MIC determinations, a background subtraction was performed on all wells with a mean of M wells. Relative fluorescence units were defined as: RFU% = (test well FU/mean FU of B wells) × 100. MIC values were determined by fitting the RFU% sigmoidal dose-response curves [4] in Kaleidagraph 4.2 software (Synergy Software). The lowest drug concentration inhibiting 50% of growth was defined as the MIC50.

Activity-based protein profiling (ABPP). From 1 L of culture at the logarithmically growth 2 stage (OD600 ~1.5), M. tb mc 6230 cells were harvested by centrifugation at 4,000 g for 15 min, and an homogeneous bacterial suspension in 7H9 S was adjusted at an OD600 of 60. Five mL of this suspension was incubated with the selected OX inhibitor (400 µM final concentration) or DMSO (control) at 37¡C for 2-3 h. under gentle shaking at 75 rpm. Bacteria were then washed 3 times with PBS containing 0.05% Tween 80, and resuspended in PBS buffer at a 1:1 (w/v) ratio. The bacterial cells (500 µL) were then mixed with 350 µL of 0.1 mm diameter glass beads (BioSpec) in a 2-mL Eppendorf tube and disrupted during 2 × 2.5 min of violent shaking, with ice cooling between each run, using Mini-Beadbeater-96 (BioSpec). The lysate was cooled down in ice for 5 min and then centrifuged at 4¡C and at 13,500 g for 15 min to remove the cell

S15 debris and unbroken cells. Supernatants were adjusted to a concentration of 1 mg/mL of total proteins, snap frozen in liquid nitrogen and stored at ‒80°C until further use. Both OX-treated M. tb mc26230 and DMSO-control lysate samples (750 µL – 0.75 mg total proteins) were incubated with 2 µM Desthiobiotin-FP probe for 90 min at room temperature. The reaction was next stopped by adding 0.45 g of urea (10 M final concentration) to denature proteins completely. Unreacted probes were removed using Zeba Spin desalting column (7K MWCO, Thermo Fisher Scientific) and labelled proteins were further captured by 200 µg Nanolink streptavidin magnetic beads 0.8 µm (Solulink), according to the manufacturer’s instructions. First, 20 µL of a 10 mg/mL NanoLink streptavidin magnetic beads was transferred into a 1.5 mL Eppendorf tube. The Wash Buffer (50 mM Tris-HCl, 150 mM NaCl, 0.05% Tween 20, pH 8.0) was then added to bring the final volume to 250 µL and the resulting mi xture was mixed gently to resuspend and wash the beads. The tube was placed on a magnetic stand for 2 min. and the supernatant was discarded. The tube was removed from the magnetic stand and the beads were washed two more times with the Wash Buffer (250 µL). Each M. tb mc26230 treated-lysate sample was enriched for labelled proteins by transfer to the previously washed beads (around 200 µg). The lysate/beads suspensions were incubated for 1 h . at room temperature with mild shaking. The tubes were then placed on the magnetic stand for 2 min to collect the beads, and the supernatant was removed. The beads containing bound, biotinylated proteins were washed three time carefully with the Wash Buffer, as described above, and resuspended in 50 µL PBS buffer pH 7.4 containing 50 mM free biotin. The resulting solution was mixed with 5X Laemmli reducing sample buffer, and heated at 95¡C for 5 min. This step allowed the recovery of the captured labelled proteins by exchanging the initially captured desthiobiotin/streptavidin complex to the greater affinity biotin/streptavidin complex. The denatured proteins were further placed on the magnetic stand for 2 min to collect the released beads. The supernatant was snap frozen in liquid nitrogen and stored at ‒80°C before mass spectrometry experiments. To check for unspecific binding, a DMSO-treated lysate sample was also incubated only with the streptavidin-magnetic beads in absence of Desthiobiotin-FP probe treatment, and processed as described above.

In-gel detection of total M. tb potential target enzymes using TAMRA-FP probe. Alternatively, the OX-treated M. tb mc26230 and DMSO-control lysate samples (100 µL – 100 µg total proteins) were incubated with 2 μM ActivX TAMRA-FP probe (Thermo Fisher Scientific) for 90 min at room temperature and in absence of light. The reaction was stopped

S16 by adding 5X Laemmli reducing sample buffer and boiling at 95¡C for 5 min. The labelled proteins were further analyzed by SDS-PAGE electrophoresis (12% Bis-Tris gel) followed by ¨ fluorescent gel scanning (TAMRA: λex 557 nm, λem 583 nm) and detection using the Cy 3 filter of a ChemiDoc MP Imager (Bio-Rad).

Mass spectrometry analysis. Protein extract were loaded on NuPAGE 4-12% Bis-Tris acrylamide gels (Life Technologies) to stack proteins in a single band that was stained with Imperial Blue (Pierce, Rockford, IL) and cut from the gel. Gels pieces were submitted to an in- gel trypsin digestion [6] with slight modifications. Briefly, gel pieces were washed and destained using 100 mM NH4HCO3. Destained gel pieces were shrunk with 100 mM ammonium bicarbonate in 50% acetonitrile and dried at room temperature. Protein spots were then rehydrated using 10 mM DTT in 100 mM ammonium bicarbonate pH 8.0 for 45 min at 56 ¡C. This solution was replaced by 55 mM iodoacetamide in 100 mM ammonium bicarbonate pH 8.0 and the gel pieces were incubated for 30 min at room temperature in the dark. They were then washed twice in 100 mM ammonium bicarbonate and finally shrunk by incubation for 5 min with 100 mM ammonium bicarbonate in 50% acetonitrile. The resulting alkylated gel pieces were dried at room temperature. The dried gel pieces were re-swollen by incubation in 100 mM ammonium bicarbonate pH 8.0 supplemented with trypsin (12.5 ng/µL; Promega) for 1 h at 4 ¡C and then incubated overnight at 37 ¡C. Peptides were harvested by collecting the initial digestion solution and carrying out two extractions; first in 5% formic acid and then in 5% formic acid in 60% acetonitrile. Pooled extracts were dried down in a centrifugal vacuum system. Samples were reconstituted with 0.1% trifluoroacetic acid in 4% acetonitrile and analyzed by liquid chromatography (LC)-tandem mass spectrometry (MS/MS) using an Orbitrap Fusion Lumos Tribrid Mass Spectrometer (Thermo Electron, Bremen, Germany) online with a an Ultimate 3000RSLCnano chromatography system (Thermo Fisher Scientific, Sunnyvale, CA). Peptides were separated on a Dionex Acclaim PepMap RSLC C18 column. First peptides were concentrated and purified on a pre-column from Dionex (C18 PepMap100, 2 cm × 100 µm I.D, 100 Å pore size, 5 µm particle size) in solvent A (0.1% formic acid in 2% acetonitrile). In the second step, peptides were separated on a reverse phase LC EASY-Spray C18 column from Dionex (PepMap RSLC C18, 15 cm × 75 µm I.D, 100 Å pore size, 2 µm particle size) at 300 nL/min flow rate. After column equilibration using 4% of solvent B (20% water - 80% acetonitrile - 0.1% formic acid), peptides were eluted from the analytical column by a two steps linear gradient (4-20% acetonitrile/H2O; 0.1 % formic acid for 90 min and 20-

45% acetonitrile/H2O; 0.1 % formic acid for 30 min). For peptide ionization in the EASY-Spray

S17 nanosource, spray voltage was set at 2.2 kV and the capillary temperature at 275 ¡C. The Orbitrap Lumos was used in data dependent mode to switch consistently between MS and MS/MS. Time between Masters Scans was set to 3 seconds. MS spectra were acquired with the Orbitrap in the range of m/z 400-1600 at a FWHM resolution of 120 000 measured at 400 m/z. AGC target was set at 4.0e5 with a 50 ms Maximum Injection Time. For internal mass calibration the 445.120025 ions was used as lock mass. The more abundant precursor ions were selected and collision-induced dissociation fragmentation was performed in the ion trap to have maximum sensitivity and yield a maximum amount of MS/MS data. Number of precursor ions was automatically defined along run in 3s windows using the “Inject Ions for All Available pararallelizable time option” with a maximum injection time of 300 ms. The signal threshold for an MS/MS event was set to 5000 counts. Charge state screening was enabled to exclude precursors with 0 and 1 charge states. Dynamic exclusion was enabled with a repeat count of 1 and a duration of 60 s.

Protein identification and quantification. Relative intensity-based label-free quantification (LFQ) was processed using the MaxLFQ algorithm [7] from the freely available MaxQuant computational proteomics platform, version 1.5.3.8 [8]. The acquired raw LC Orbitrap MS data were first processed using the integrated Andromeda search engine [9]. Spectra were searched against a UniProt M. tuberculosis ATCC 25618 database (date 2018.01; 2164 entries). This database was supplemented with a set of 245 frequently observed contaminants. The following parameters were used for searches: (i) trypsin allowing cleavage before proline; (ii) two missed cleavages were allowed; (ii) monoisotopic precursor tolerance of 20 ppm in the first search used for recalibration, followed by 4.5 ppm for the main search and 0.5 Da for fragment ions from MS/MS ; (iii) cysteine carbamidomethylation (+57.02146) as a fixed modification and methionine oxidation (+15.99491) and N-terminal acetylation (+42.0106) as variable modifications; (iv) a maximum of five modifications per peptide allowed; and (v) minimum peptide length was 7 amino acids and a maximum mass of 4,600 Da. The match between runs option was enabled to transfer identifications across different LC-MS/MS replicates based on their masses and retention time within a match time window of 0.7 min and using an alignment time window of 20 min. The quantification was performed using a minimum ratio count of 1 (unique+razor) and the second peptide option to allow identification of two co-fragmented co- eluting peptides with similar masses. The false discovery rate (FDR) at the peptide and protein levels were set to 1% and determined by searching a reverse database. For protein grouping, all proteins that cannot be distinguished based on their identified peptides were assembled into a

S18 single entry according to the MaxQuant rules. The statistical analysis was done with Perseus program (version 1.5.6.0) from the MaxQuant environment (www.maxquant.org). The LFQ normalised intensities were uploaded from the proteinGroups.txt file. First, proteins marked as contaminant, reverse hits, and “only identified by site” were discarded. Quantifiable proteins were defined as those detected in at least 100% of samples in at least one condition. Protein LFQ normalized intensities were base 2 logarithmized to obtain a normal distribution. Missing values were replaced using data imputation by randomly selecting from a normal distribution centred on the lower edge of the intensity values that simulates signals of low abundant proteins using default parameters (a downshift of 1.8 standard deviation and a width of 0.3 of the original distribution). In this way, imputation of missing values in the controls allows statistical comparison of protein abundances that are present only in the inhibitors samples. To determine whether a given detected protein was specifically differential a two-sample t-test were done using permutation based FDR-controlled at 0.01 and 0.05 and employing 250 permutations. The p value was adjusted using a scaling factor s0 with a value of 1 [10]. The mass spectrometry proteomics data, including search results, have been deposited to the ProteomeXchange Consortium (www.proteomexchange.org) [11] via the PRIDE partner repository with the dataset identifier PXD010255. Proteomic data are available at https://www.ebi.ac.uk/pride/archive/login with identifier PXD010255 Username: [email protected] / password: Hhpb9H0Y

S19

References

[1] D. Perrin, W. Armarego, D.R. Perrin, Purification of Laboratory Chemicals, 2nd Ed,

Pergamon Press, Oxford, 1980.

[2] W.C. Still, M. Kahn, A. Mitra, Rapid chromatographic technique for preparative separation with moderate resolution, J. Org. Chem., 43 (1978) 2923-2925.

[3] J.C. Palomino, A. Martin, M. Camacho, H. Guerra, J. Swings, F. Portaels, Resazurin microtiter assay plate: simple and inexpensive method for detection of drug resistance in

Mycobacterium tuberculosis, Antimicrob Agents Chemother, 46 (2002) 2720-2722.

[4] J. Rybniker, A. Vocat, C. Sala, P. Busso, F. Pojer, A. Benjak, S.T. Cole, Lansoprazole is an antituberculous prodrug targeting cytochrome bc1, Nature communications, 6 (2015) 7659.

[5] A. Walzl, N. Kramer, M.R. Mazza, D. Falkenhagen, M. Hengstschläger, D. Schwanzer-

Pfeiffer, H. Dolznig, A Simple and Cost Efficient Method to Avoid Unequal Evaporation in

Cellular Screening Assays, Which Restores Cellular Metabolic Activity, Int. J. Appl. Sci.

Technol., 2 (2012) 17–21.

[6] A. Shevchenko, M. Wilm, O. Vorm, M. Mann, Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels, Anal Chem., 68 (1996) 850-858.

[7] J. Cox, M.Y. Hein, C.A. Luber, I. Paron, N. Nagaraj, M. Mann, Accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed

MaxLFQ, Mol Cell Proteomics, 13 (2014) 2513-2526.

[8] J. Cox, M. Mann, MaxQuant enables high peptide identification rates, individualized p.p.b.- range mass accuracies and proteome-wide protein quantification, Nat Biotechnol., 26 (2008)

1367-1372.

[9] J. Cox, N. Neuhauser, A. Michalski, R.A. Scheltema, J.V. Olsen, M. Mann, Andromeda: a peptide search engine integrated into the MaxQuant environment, J Proteome Res., 10 (2011)

1794-1805.

S20

[10] V.G. Tusher, R. Tibshirani, G. Chu, Significance analysis of microarrays applied to the ionizing radiation response, Proc Natl Acad Sci U S A, 98 (2001) 5116-5121.

[11] J.A. Vizcaino, E.W. Deutsch, R. Wang, A. Csordas, F. Reisinger, D. Rios, J.A. Dianes, Z.

Sun, T. Farrah, N. Bandeira, P.A. Binz, I. Xenarios, M. Eisenacher, G. Mayer, L. Gatto, A.

Campos, R.J. Chalkley, H.J. Kraus, J.P. Albar, S. Martinez-Bartolome, R. Apweiler, G.S.

Omenn, L. Martens, A.R. Jones, H. Hermjakob, ProteomeXchange provides globally coordinated proteomics data submission and dissemination, Nat Biotechnol., 32 (2014) 223-

226.

S21

CONCLUSIONS AND PERSPECTIVES

Conclusions and perspectives

Conclusions M. tb is one of the most ancient microorganism as well as one of the deadliest pathogenic bacteria in human health. The emergency of drug resistant M. tb strains recently promoted the scientists to explore new lead compounds with novel mechanisms or targets, to eradicate the growth of the bacilli. We assume that our CyC and OX synthesized compounds are such molecules. In this work, we have implemented a large screening using this two series of potential mycobacterial inhibitors. We initially confirmed that several molecules exhibit antitubercular activities varying from very good to moderate levels, as judged by their MIC values, which are comparable to currently used anti-TB drugs. We determined here different mode of actions of the molecules: some of them are extracellular inhibitors only, e.g., CyC17, BePOX; intracellular inhibitors active exceptionally on intramacrophagic bacilli such as CyC6β, CyC8β, HpPPOX and BepPPOX; or

dual-inhibitors thus exhibiting antibacterial activity in both conditions, like CyC7β and iBpPPOX. The exact mechanisms responsible of such effects remain not clear yet and need to be extensively studied. However one can hypothesize that the intracellular bacterial target(s) of these compounds are different than those leading to extracellular bacteria killing; thus suggesting a more critical physiological role of these enzymatic target(s) in intracellular vs. extracellular bacteria. Alternatively, such inhibitors might also enhance the macrophage defense mechanism and induce a specific response of the host cells to kill intracellular bacteria. Another issue resides in the bioavailability of these molecules which may differ, thus being responsible for the diverse outcomes on antibacterial activities. The way they access/penetrate the bacterial membrane and/or diffuse through the host macrophage and selectively target the bacteria still remains unknown, and should deserve to be studied in details. Despite their different mode of action, all these potential compounds displayed a very low toxicity to human macrophages, up to >100 µM. However it should be interesting to study the reason why these compounds able to target serine and cysteine enzymes have not effect in macrophages. More globally, CyCs displayed a better bacterial clearance than that of OX molecules; and based on the aforementioned results, one can assume that both family of inhibitors would alter the outcome of the infection by impairing mycobacterial growth within host cells. In addition, they may also affect the entry of bacilli into the persistence phase and/or interfere with reactivation of dormant bacilli in macrophages.

184 Conclusions and perspectives

Subsequently, we first validated that CyC analogs react with α/β-hydrolase family, under the nucleophilic attack of the catalytic serine/cysteine residues to the phosph(on)ate ring, thus forming a covalent bond and generating inactive enzymes via phosphorylation of their catalytic serine/cysteine residues. In addition, the chemical modifications made on their core structure allow to improve their efficiency and activity towards microbial enzymes, as well as their specific antibacterial properties against mycobacteria only. In an attempt to elucidate the mechanism of action of CyCs, ABPP approach was successfully applied, allowing the identification of mycobacterial enzymes impaired by the inhibitors during mycobacterial growth. Selective labelling and enrichment of captured enzymes using appropriate fluorophosphonate probes in combination with CyC17 resulted in the identification of 23 potential target enzymes, including several lipolytic enzymes as well as non-lipolytic enzymes, among them the antigen 85 complex (Ag85A, Ag85C), the thioesterase TesA, the carboxylesterase CaeA, and the hydrolase HsaD appeared as promising target candidates; the latter two proteins being annotated as essential enzymes in M. tb. We have next validated two main targets identified: Ag85C and TesA. The important role of Ag85C has been recognized for decades and often been proposed as attractive target for prospective therapeutic treatment of TB. We demonstrated that the presence of CyC compounds impair TDM synthesis and mycolylation of arabinogalactan in vitro and in vivo. The crystal structure of Ag85C in complex with CyC8β was resolved. However, no covalent link between the catalytic serine and the phosphonate group of the molecules was observed in the crystal.

This can be explained by the possibility of hydrolysis of CyC8β from protein, in other words, the covalent serine – phosphorous is reversible. However the inhibition is still happened due to the occupation of the compound inside the active site. Indeed, CyC8β has clearly reacted as evidenced by the presence of an opened ring attribute to the nucleophilic attack of catalytic serine, combined with the MALDI-TOF data clearly shown a mass modification of 402.3 Da. Regarding TesA, we confirmed that this enzyme displays both esterase and thioesterase activities, but no phospholipase nor lipase activity. The inhibition assay performed with four representative CyC analogs (i.e., CyC6, CyC7β, CyC8β and CyC17) show that only CyC17 was able to efficiently inhibit TesA. Importantly, the crystal structure of TesA has also been solved at the resolution of 2.6 Å in complex with CyC17 covalently linked to the catalytic Ser104. These two studies, therefore provide clear evidence that Ag85C and TesA are effective targets of CyC17. This further strengthens the hypothesis that this inhibitor, and presumably the other CyC analogs, represent multi-target agents. Accordingly, by blocking at the same time the activities

185 Conclusions and perspectives

of various lipolytic enzymes CyC17 would strongly interfere with the acquisition and consumption of host cell-derived lipids by the mycobacteria, and also destabilize the cell envelope assembly. In such conditions, the large spectrum of inhibitory effects exerted by our CyC analogs cannot be considered as a weakness if only M. tb is impacted, and on the contrary can open new avenues for the treatment of TB. Above all, this work led to the identification of very promising anti-TB candidates that should be able to act against bacteria in various physiological stages, thus allowing a faster sterilization. Finally, CyCs also represent promising molecules to treat mycobacterial related diseases. Indeed, a screening test on a diverse range of bacteria help us to clarify that the CyC compounds selectively. Since mycobacteria have a similar but unique cell wall composition and structure compare to cell wall of gram negative or positive, it is not surprising since CyCs express an exclusive inhibitory activity on mycobacterial strains. More importantly, the best compound

CyC17 can effectively stop the growth of 26 clinical mycobacterial strains, many of them are multi-resistant species belonging to the M. abscessus-chelonae clade. These results thus open new opportunities in the treatment of mycobacterial related diseases including cystic fibrosis. Regarding OX derivatives, these compounds impair the in vitro growth of three mycobacterial strains, i.e., M. marinum, M. bovis BCG and M. tb mc26230 from good to moderate extents. Similarly to CyC compounds, OX derivatives are also found to be able to inhibit extracellular and/or intracellular M. tb growth. ABPP approach was also applied to identify the targets in M. tb mc26230. The experiments are in process of mass spectrometry analysis. The targets determination will give the ideas for further studies such as chemical modifications to improve to efficiency and specificity of the molecules to bacterial enzymes. To sum up, our study bring a very attractive suggestion of utilization these compounds in treatment of several M. tb bacterial types in active TB or in LTBI regardless they are drug resistant strains or not. Our compounds promise the development of multi-targets TB drugs which direct to not only new mechanism i.e., lipid metabolism but also novel potential targets. Although the synergistic activities have not been assayed yet, we anticipated the possibility of using our molecules in combination with other ancient anti-TB drugs in order to obtain the efficient therapeutic approach. Besides TB, the treatment of mycobacterial-related diseases e.g., cystic fibrosis, skin and soft tissue diseases, etc… using our molecules are also in prospect. Perspectives Results from our study can be considered as the fundamental to developing the new serial therapeutic drugs in TB treatment. Regarding to the first series of CyCs, a broad range of work is needed to be explored. Among potential CyC compounds, only extracellular molecule CyC17

186 Conclusions and perspectives was specifically identified its 23 possible targets. The two important targets i.e., Ag85C and TesA were extensively studied on their biochemical characteristic as well as their structures in the co-exist of CyC inhibitors were revealed. From there, their physiological roles to bacteria has been also confirmed. The other 21 targets of CyC17, however, still remain to be exploited. Several of them have not been biochemically characterized yet nor crystal structures such as AmiC, AmiB2, LipM, LipE, Rv1367, etc… The role of proteins which thought to be essential for bacterial pathogenesis including CaeA, hsaD, Ag85A are needed to be further elucidated.

In addition, the targets of intracellular inhibitors like CyC6β or CyC7β and CyC8β are necessary to be identified, in order to figure out the mode of action of those compounds as well as the behavior changes of intraphagosomal bacilli. The permeability (active or passive transport) of such molecules across the host cell membrane are also interesting to be revealed. The attachment of a fluorescent tag to the molecules without affecting their inhibition activity in order to observe the action of these molecules in situ can be considered. The primitive test indicating the low cytotoxicity of CyCs against Raw264.7 murine macrophages suggesting that they are positive to be applied in treatment of TB in human. Bioavalability of the molecules is needed to be assessed before testing on animal models such as zebrafish or mice. These experiments, in one hand, evaluate the efficiency of CyC compounds on bacterial clearance in vivo, and in the other hand, verify the possibility of using these molecules in phases of clinical research. The OX family with several optimistic molecules to be potential antimycobacterial agents, display however the issue of inhibition in parallel human hormone sensitive lipases. Thus the primary perspective could be related to chemical modifications in the purpose to improve the selectivity and efficiency against microbial enzymes. Both family of molecules can be considered as promising drugs candidates in the treatment of mycobacterial related diseases since they showed the antibacterial activity against several mycobacterial strains. M. abscessus, the causative agent of cystic fibrosis, arose recently as an antibiotic nightmare, are not excluded. The further experiments including targets identification, targets validation and characteristic as well as primary testing on animal models are awaiting to be implemented. Last but not least, since the quadruple cocktail of antibiotics was applied to eradicate M. tb in current standard TB treatment, suggesting that the combination of several drugs in order to give an appropriate regimen is necessary. Thus, the evaluation of the synergistic effect of our molecules with available antibiotics/drugs are critical to accomplish the process of application of these compound in TB treatment.

187

REFERENCES References

References

A

Abdallah, A.M., Gey van Pittius, N.C., Champion, P.A., Cox, J., Luirink, J., Vandenbroucke-Grauls, C.M., Appelmelk, B.J., Bitter, W., 2007. Type VII secretion--mycobacteria show the way. Nat Rev Microbiol 5, 883-891. Abdallah, A.M., Verboom, T., Hannes, F., Safi, M., Strong, M., Eisenberg, D., Musters, R.J., Vandenbroucke-Grauls, C.M., Appelmelk, B.J., Luirink, J., Bitter, W., 2006. A specific secretion system mediates PPE41 transport in pathogenic mycobacteria. Mol Microbiol 62, 667-679. Aeschlimann, J.A., Stempel, A., 1954. Heterocyclic antituberculous compounds. Hoffmann La Roche Inc., p. 2 pp. Ahmad, S., 2011. Pathogenesis, immunology, and diagnosis of latent Mycobacterium tuberculosis infection. Clin Dev Immunol 2011, 814943. Alangaden, G.J., Kreiswirth, B.N., Aouad, A., Khetarpal, M., Igno, F.R., Moghazeh, S.L., Manavathu, E.K., Lerner, S.A., 1998. Mechanism of resistance to amikacin and kanamycin in Mycobacterium tuberculosis. Antimicrob Agents Chemother 42, 1295-1297. Alibaud, L., Rombouts, Y., Trivelli, X., Burguiere, A., Cirillo, S.L., Cirillo, J.D., Dubremetz, J.F., Guerardel, Y., Lutfalla, G., Kremer, L., 2011. A Mycobacterium marinum TesA mutant defective for major cell wall-associated lipids is highly attenuated in Dictyostelium discoideum and zebrafish embryos. Mol Microbiol 80, 919-934. Almeida Da Silva, P.E., Palomino, J.C., 2011. Molecular basis and mechanisms of drug resistance in Mycobacterium tuberculosis: classical and new drugs. J Antimicrob Chemother 66, 1417-1430. Andersen, P., Doherty, T.M., 2005. The success and failure of BCG - implications for a novel tuberculosis vaccine. Nat Rev Microbiol 3, 656-662. Andrews, J.R., Noubary, F., Walensky, R.P., Cerda, R., Losina, E., Horsburgh, C.R., 2012. Risk of progression to active tuberculosis following reinfection with Mycobacterium tuberculosis. Clin Infect Dis 54, 784-791. Astarie-Dequeker, C., Le Guyader, L., Malaga, W., Seaphanh, F.K., Chalut, C., Lopez, A., Guilhot, C., 2009. Phthiocerol dimycocerosates of M. tuberculosis participate in macrophage invasion by inducing changes in the organization of plasma membrane lipids. PLoS pathogens 5, e1000289.

B

Backus, K.M., Dolan, M.A., Barry, C.S., Joe, M., McPhie, P., Boshoff, H.I., Lowary, T.L., Davis, B.G., Barry, C.E., 3rd, 2014. The three Mycobacterium tuberculosis antigen 85 isoforms have unique substrates and activities determined by non-active site regions. J Biol Chem 289, 25041-25053. Bailo, R., Bhatt, A., Ainsa, J.A., 2015. Lipid transport in Mycobacterium tuberculosis and its implications in virulence and drug development. Biochem Pharmacol 96, 159-167. Bakala N'goma J, C., Schue, M., Carriere, F., Geerlof, A., Canaan, S., 2010. Evidence for the cytotoxic effects of Mycobacterium tuberculosis phospholipase C towards macrophages. Biochimica et biophysica acta 1801, 1305-1313. Bakala N'Goma, J.C., Le Moigne, V., Soismier, N., Laencina, L., Le Chevalier, F., Roux, A.L., Poncin, I., Serveau-Avesque, C., Rottman, M., Gaillard, J.L., Etienne, G., Brosch, R., Herrmann, J.L., Canaan, S., Girard-Misguich, F., 2015. Mycobacterium abscessus phospholipase C expression is induced during coculture within amoebae and enhances M. abscessus virulence in mice. Infect Immun 83, 780-791. Balganesh, M., Dinesh, N., Sharma, S., Kuruppath, S., Nair, A.V., Sharma, U., 2012. Efflux pumps of Mycobacterium tuberculosis play a significant role in antituberculosis activity of potential drug candidates. Antimicrob Agents Chemother 56, 2643-2651. Bassam, H.M., Mayank, G.V. (Eds.), 2013. Tuberculosis - Current Issues in Diagnosis and Management. InTech. Beaulieu, A.M., Rath, P., Imhof, M., Siddall, M.E., Roberts, J., Schnappinger, D., Nathan, C.F., 2010. Genome-wide screen for Mycobacterium tuberculosis genes that regulate host immunity. PloS one 5, e15120.

189 References

Belisle, J.T., Vissa, V.D., Sievert, T., Takayama, K., Brennan, P.J., Besra, G.S., 1997. Role of the major antigen of Mycobacterium tuberculosis in cell wall biogenesis. Science 276, 1420-1422. Bellamine, A., Mangla, A.T., Nes, W.D., Waterman, M.R., 1999. Characterization and catalytic properties of the sterol 14alpha-demethylase from Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 96, 8937-8942. Ben Ali, Y., Chahinian, H., Petry, S., Muller, G., Lebrun, R., Verger, R., Carriere, F., Mandrich, L., Rossi, M., Manco, G., Sarda, L., Abousalham, A., 2006. Use of an inhibitor to identify members of the hormone-sensitive lipase family. Biochemistry 45, 14183-14191. Ben Ali, Y., Verger, R., Carrière, F., Petry, S., Muller, G., Abousalham, A., 2012. The molecular mechanism of human hormone-sensitive lipase inhibition by substituted 3-phenyl-5-alkoxy-1,3,4- oxadiazol-2-ones. Biochimie 94, 137-145. Benarouche, A., Point, V., Carriere, F., Cavalier, J.F., 2014. Using the reversible inhibition of gastric lipase by Orlistat for investigating simultaneously lipase adsorption and substrate hydrolysis at the lipid- water interface. Biochimie 101, 221-231. Bernstein, J., Lott, W.A., Steinberg, B.A., Yale, H.L., 1952. Chemotherapy of experimental tuberculosis. V. Isonicotinic acid hydrazide (nydrazid) and related compounds. Am Rev Tuberc 65, 357- 364. Borgström, B., 1988. Mode of action of tetrahydrolipstatin: A derivative of the naturally occuring lipase inhibitor lipstatin. Biochim. Biophys. Acta 962, 308-316. Bos, K.I., Harkins, K.M., Herbig, A., Coscolla, M., Weber, N., Comas, I., Forrest, S.A., Bryant, J.M., Harris, S.R., Schuenemann, V.J., Campbell, T.J., Majander, K., Wilbur, A.K., Guichon, R.A., Wolfe Steadman, D.L., Cook, D.C., Niemann, S., Behr, M.A., Zumarraga, M., Bastida, R., Huson, D., Nieselt, K., Young, D., Parkhill, J., Buikstra, J.E., Gagneux, S., Stone, A.C., Krause, J., 2014. Pre-Columbian mycobacterial genomes reveal seals as a source of New World human tuberculosis. Nature 514, 494- 497. Brennan, P.J., 2003. Structure, function, and biogenesis of the cell wall of Mycobacterium tuberculosis. Tuberculosis (Edinb) 83, 91-97. Brewer, T.F., Colditz, G.A., 1995. Bacille Calmette-Guerin vaccination for the prevention of tuberculosis in health care workers. Clin Infect Dis 20, 136-142. Brust, B., Lecoufle, M., Tuaillon, E., Dedieu, L., Canaan, S., Valverde, V., Kremer, L., 2011. Mycobacterium tuberculosis lipolytic enzymes as potential biomarkers for the diagnosis of active tuberculosis. PloS one 6, e25078.

C

Caire-Brandli, I., Papadopoulos, A., Malaga, W., Marais, D., Canaan, S., Thilo, L., de Chastellier, C., 2014. Reversible lipid accumulation and associated division arrest of Mycobacterium avium in lipoprotein-induced foamy macrophages may resemble key events during latency and reactivation of tuberculosis. Infection and immunity 82, 476-490. Camacho, L.R., Ensergueix, D., Perez, E., Gicquel, B., Guilhot, C., 1999. Identification of a virulence gene cluster of Mycobacterium tuberculosis by signature-tagged transposon mutagenesis. Mol Microbiol 34, 257-267. Cambau, E., Drancourt, M., 2014. Steps towards the discovery of Mycobacterium tuberculosis by Robert Koch, 1882. Clin Microbiol Infect 20, 196-201. Caminero, J.A., Sotgiu, G., Zumla, A., Migliori, G.B., 2010. Best drug treatment for multidrug-resistant and extensively drug-resistant tuberculosis. The Lancet. Infectious diseases 10, 621-629. Campbell, I.A., Bah-Sow, O., 2006. Pulmonary tuberculosis: diagnosis and treatment. BMJ 332, 1194- 1197. Camus, J.C., Pryor, M.J., Medigue, C., Cole, S.T., 2002. Re-annotation of the genome sequence of Mycobacterium tuberculosis H37Rv. Microbiology 148, 2967-2973. Canaan, S., Maurin, D., Chahinian, H., Pouilly, B., Durousseau, C., Frassinetti, F., Scappuccini-Calvo, L., Cambillau, C., Bourne, Y., 2004a. Expression and characterization of the protein Rv1399c from Mycobacterium tuberculosis. Eur. J. Biochem. 271, 3953-3961. Canaan, S., Maurin, D., Chahinian, H., Pouilly, B., Durousseau, C., Frassinetti, F., Scappuccini-Calvo, L., Cambillau, C., Bourne, Y., 2004b. Expression and characterization of the protein Rv1399c from

190 References

Mycobacterium tuberculosis. A novel carboxyl esterase structurally related to the HSL family. Eur J Biochem 271, 3953-3961. Candido, P.H., Nunes Lde, S., Marques, E.A., Folescu, T.W., Coelho, F.S., de Moura, V.C., da Silva, M.G., Gomes, K.M., Lourenco, M.C., Aguiar, F.S., Chitolina, F., Armstrong, D.T., Leao, S.C., Neves, F.P., Mello, F.C., Duarte, R.S., 2014. Multidrug-resistant nontuberculous mycobacteria isolated from cystic fibrosis patients. Journal of clinical microbiology 52, 2990-2997. Cao, J., Dang, G., Li, H., Li, T., Yue, Z., Li, N., Liu, Y., Liu, S., Chen, L., 2015. Identification and Characterization of Lipase Activity and Immunogenicity of LipL from Mycobacterium tuberculosis. PloS one 10, e0138151. Cardona, P.J. (Ed), 2012. Understanding Tuberculosis - Deciphering the Secret Life of the Bacilli. Cardona, P.J., Llatjos, R., Gordillo, S., Diaz, J., Ojanguren, I., Ariza, A., Ausina, V., 2000. Evolution of granulomas in lungs of mice infected aerogenically with Mycobacterium tuberculosis. Scand J Immunol 52, 156-163. Carr, P., D. , Ollis, D., L. , 2009. a/bHydrolase Fold: An Update. Protein Pept Lett. 16, 1137-1148. Carvalho, C., Aires-Barros, M., Cabral, J., 1998. Cutinase structure, function and biocatalytic applications. Electron. J. Biotechnol. 1, 160-173. Cavalier, J.-F., Buono, G., Verger, R., 2000. Covalent inhibition of digestive lipases by chiral phosphonates. Acc. Chem. Res. 33, 579-589. Caws, M., Duy, P.M., Tho, D.Q., Lan, N.T., Hoa, D.V., Farrar, J., 2006. Mutations prevalent among rifampin- and isoniazid-resistant Mycobacterium tuberculosis isolates from a hospital in Vietnam. Journal of clinical microbiology 44, 2333-2337. Chakhaiyar, P., Nagalakshmi, Y., Aruna, B., Murthy, K.J., Katoch, V.M., Hasnain, S.E., 2004. Regions of high antigenicity within the hypothetical PPE major polymorphic tandem repeat open-reading frame, Rv2608, show a differential humoral response and a low T cell response in various categories of patients with tuberculosis. The Journal of infectious diseases 190, 1237-1244. Chavadi, S.S., Edupuganti, U.R., Vergnolle, O., Fatima, I., Singh, S.M., Soll, C.E., Quadri, L.E., 2011. Inactivation of tesA reduces cell wall lipid production and increases drug susceptibility in mycobacteria. J Biol Chem 286, 24616-24625. Chiaradia, L., Lefebvre, C., Parra, J., Marcoux, J., Burlet-Schiltz, O., Etienne, G., Tropis, M., Daffe, M., 2017. Dissecting the mycobacterial cell envelope and defining the composition of the native mycomembrane. Sci Rep 7, 12807. Christophe, T., Ewann, F., Jeon, H.K., Cechetto, J., Brodin, P., 2010. High-content imaging of Mycobacterium tuberculosis-infected macrophages: an in vitro model for tuberculosis drug discovery. Future medicinal chemistry 2, 1283-1293. Christophe, T., Jackson, M., Jeon, H.K., Fenistein, D., Contreras-Dominguez, M., Kim, J., Genovesio, A., Carralot, J.P., Ewann, F., Kim, E.H., Lee, S.Y., Kang, S., Seo, M.J., Park, E.J., Skovierova, H., Pham, H., Riccardi, G., Nam, J.Y., Marsollier, L., Kempf, M., Joly-Guillou, M.L., Oh, T., Shin, W.K., No, Z., Nehrbass, U., Brosch, R., Cole, S.T., Brodin, P., 2009. High content screening identifies decaprenyl-phosphoribose 2' epimerase as a target for intracellular antimycobacterial inhibitors. PLoS Pathog 5, e1000645. Clay, H., Davis, J.M., Beery, D., Huttenlocher, A., Lyons, S.E., Ramakrishnan, L., 2007. Dichotomous role of the macrophage in early Mycobacterium marinum infection of the zebrafish. Cell host & microbe 2, 29-39. Cole, S.T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., Gordon, S.V., Eiglmeier, K., Gas, S., Barry, C.E., 3rd, Tekaia, F., Badcock, K., Basham, D., Brown, D., Chillingworth, T., Connor, R., Davies, R., Devlin, K., Feltwell, T., Gentles, S., Hamlin, N., Holroyd, S., Hornsby, T., Jagels, K., Krogh, A., McLean, J., Moule, S., Murphy, L., Oliver, K., Osborne, J., Quail, M.A., Rajandream, M.A., Rogers, J., Rutter, S., Seeger, K., Skelton, J., Squares, R., Squares, S., Sulston, J.E., Taylor, K., Whitehead, S., Barrell, B.G., 1998. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393, 537-544. Constant, P., Perez, E., Malaga, W., Laneelle, M.A., Saurel, O., Daffe, M., Guilhot, C., 2002. Role of the pks15/1 gene in the biosynthesis of phenolglycolipids in the Mycobacterium tuberculosis complex. Evidence that all strains synthesize glycosylated p-hydroxybenzoic methyl esters and that strains devoid of phenolglycolipids harbor a frameshift mutation in the pks15/1 gene. J Biol Chem 277, 38148-38158.

191 References

Cooper, A.M., 2009. Cell-mediated immune responses in tuberculosis. Annual review of immunology 27, 393-422. Corper, H.J., Cohn, M.L., 1949. The probable mechanism of streptomycin action in tuberculosis. Yale J Biol Med 21, 181-200. Cotes, K., Dhouib, R., Douchet, I., Chahinian, H., de Caro, A., Carriere, F., Canaan, S., 2007. Characterization of an exported monoglyceride lipase from Mycobacterium tuberculosis possibly involved in the metabolism of host cell membrane lipids. The Biochemical journal 408, 417-427. Cox, J.S., Chen, B., McNeil, M., Jacobs, W.R., Jr., 1999. Complex lipid determines tissue-specific replication of Mycobacterium tuberculosis in mice. Nature 402, 79-83. Cravatt, B.F., Wright, A.T., Kozarich, J.W., 2008. Activity-based protein profiling: from enzyme chemistry to proteomic chemistry. Annu Rev Biochem 77, 383-414. Crellin, P.K., Vivian, J.P., Scoble, J., Chow, F.M., West, N.P., Brammananth, R., Proellocks, N.I., Shahine, A., Le Nours, J., Wilce, M.C.J., Britton, W.J., Coppel, R.L., Rossjohn, J., Beddoe, T., 2010. Tetrahydrolipstatin Inhibition, Functional Analyses, and Three-dimensional Structure of a Lipase Essential for Mycobacterial Viability. J. Biol. Chem. 285, 30050-30060.

D

Daffe, M., Lacave, C., Laneelle, M.A., Laneelle, G., 1987. Structure of the major triglycosyl phenol- phthiocerol of Mycobacterium tuberculosis (strain Canetti). Eur J Biochem 167, 155-160. Daher Ede, F., da Silva, G.B., Jr., Barros, E.J., 2013. Renal tuberculosis in the modern era. Am J Trop Med Hyg 88, 54-64. Dahle, U.R., 2006. Extensively drug resistant tuberculosis: beware patients lost to follow-up. BMJ 333, 705. Daleke, M.H., Cascioferro, A., de Punder, K., Ummels, R., Abdallah, A.M., van der Wel, N., Peters, P.J., Luirink, J., Manganelli, R., Bitter, W., 2011. Conserved Pro-Glu (PE) and Pro-Pro-Glu (PPE) protein domains target LipY lipases of pathogenic mycobacteria to the cell surface via the ESX-5 pathway. J Biol Chem 286, 19024-19034. Daniel, J., Maamar, H., Deb, C., Sirakova, T.D., Kolattukudy, P.E., 2011. Mycobacterium tuberculosis uses host triacylglycerol to accumulate lipid droplets and acquires a dormancy-like phenotype in lipid- loaded macrophages. PLoS pathogens 7, e1002093. Daniel, T.M., 2015. Jean-Antoine Villemin and the infectious nature of tuberculosis. Int J Tuberc Lung Dis 19, 267-268. Daniel, T.M., Iversen, P.A., 2015. Hippocrates and tuberculosis. Int J Tuberc Lung Dis 19, 373-374. Daniel, V.S., Daniel, T.M., 1999. Old Testament biblical references to tuberculosis. Clin Infect Dis 29, 1557-1558. de Chastellier, C., 2009. The many niches and strategies used by pathogenic mycobacteria for survival within host macrophages. Immunobiology 214, 526-542. de Lima Corvino, D.F., Kosmin, A.R., 2017. Tuberculosis, Screening. Deb, C., Daniel, J., Sirakova, T.D., Abomoelak, B., Dubey, V.S., Kolattukudy, P.E., 2006. A novel lipase belonging to the hormone-sensitive lipase family induced under starvation to utilize stored triacylglycerol in Mycobacterium tuberculosis. J Biol Chem 281, 3866-3875. Debasu, D., Desalegn, W., Biniam, M., 2014. Review on Molecular mechanism of first line antibiotic resistance in Mycobacterium tuberculosis. Mycobacterial diseases 4. Dedieu, L., Serveau-Avesque, C., Kremer, L., Canaan, S., 2013. Mycobacterial lipolytic enzymes: a gold mine for tuberculosis research. Biochimie 95, 66-73. Delorme, V., Diomande, S.V., Dedieu, L., Cavalier, J.F., Carriere, F., Kremer, L., Leclaire, J., Fotiadu, F., Canaan, S., 2012. MmPPOX inhibits Mycobacterium tuberculosis lipolytic enzymes belonging to the hormone-sensitive lipase family and alters mycobacterial growth. PloS one 7, e46493. Desmoras, J., Ambrosi, D., Boesch, R., 1974. Laboratory and greenhouse study of a new herbicide, the 23 465 RP derived of oxadiazon. Phytiatrie Phytopharmacie (France) 23, 153-164. Detjen, A.K., Keil, T., Roll, S., Hauer, B., Mauch, H., Wahn, U., Magdorf, K., 2007. Interferon-gamma release assays improve the diagnosis of tuberculosis and nontuberculous mycobacterial disease in children in a country with a low incidence of tuberculosis. Clin Infect Dis 45, 322-328.

192 References

Dhouib, R., Laval, F., Carriere, F., Daffe, M., Canaan, S., 2010. A monoacylglycerol lipase from Mycobacterium smegmatis Involved in bacterial cell interaction. Journal of bacteriology 192, 4776- 4785. Donald, P.R., Marais, B.J., Barry, C.E., 3rd, 2010. Age and the epidemiology and pathogenesis of tuberculosis. Lancet 375, 1852-1854. Drain, P.K., Losina, E., Coleman, S.M., Giddy, J., Ross, D., Katz, J.N., Bassett, I.V., 2016. Rapid urine lipoarabinomannan assay as a clinic-based screening test for active tuberculosis at HIV diagnosis. BMC Pulm Med 16, 147.

E

EA/BMRC, 1972. Controlled clinical trial of short-course (6-month) regimens of chemotherapy for treatment of pulmonary tuberculosis. Lancet 1, 1079-1085. Edlin, B.R., Tokars, J.I., Grieco, M.H., Crawford, J.T., Williams, J., Sordillo, E.M., Ong, K.R., Kilburn, J.O., Dooley, S.W., Castro, K.G., et al., 1992. An outbreak of multidrug-resistant tuberculosis among hospitalized patients with the acquired immunodeficiency syndrome. N Engl J Med 326, 1514-1521. Egmond, M.R., de Vlieg, J., 2000. Fusarium solani pisi cutinase. Biochimie 82, 1015-1021. Eisenhuber, E., 2002. The tree-in-bud sign. Radiology 222, 771-772.

F

Fattorini, L., Piccaro, G., Mustazzolu, A., Giannoni, F., 2013. Targeting Dormant Bacilli to Fight Tuberculosis. Mediterr J Hematol Infect Dis. 5, e2013072. Fine, P.E., 1995. Variation in protection by BCG: implications of and for heterologous immunity. Lancet 346, 1339-1345. Fischl, M.A., Uttamchandani, R.B., Daikos, G.L., Poblete, R.B., Moreno, J.N., Reyes, R.R., Boota, A.M., Thompson, L.M., Cleary, T.J., Lai, S., 1992. An outbreak of tuberculosis caused by multiple- drug-resistant tubercle bacilli among patients with HIV infection. Ann Intern Med 117, 177-183. Flammersfeld, A., Lang, C., Flieger, A., Pradel, G., 2017. Phospholipases during membrane dynamics in malaria parasites. Int J Med Microbiol. Fletcher, H.A., Schrager, L., 2016. TB vaccine development and the End TB Strategy: importance and current status. Trans R Soc Trop Med Hyg 110, 212-218. Flipo, M., Desroses, M., Lecat-Guillet, N., Dirie, B., Carette, X., Leroux, F., Piveteau, C., Demirkaya, F., Lens, Z., Rucktooa, P., Villeret, V., Christophe, T., Jeon, H.K., Locht, C., Brodin, P., Deprez, B., Baulard, A.R., Willand, N., 2011. Ethionamide boosters: synthesis, biological activity, and structure- activity relationships of a series of 1,2,4-oxadiazole EthR inhibitors. J Med Chem 54, 2994-3010. Flynn, J.L., Chan, J., 2001. Tuberculosis: latency and reactivation. Infection and immunity 69, 4195- 4201. Frieden, T.R., Fujiwara, P.I., Washko, R.M., Hamburg, M.A., 1995. Tuberculosis in New York City-- turning the tide. N Engl J Med 333, 229-233.

G

Galagan, J.E., 2014. Genomic insights into tuberculosis. Nat Rev Genet 15, 307-320. Gengenbacher, M., Kaufmann, S.H., 2012. Mycobacterium tuberculosis: success through dormancy. FEMS Microbiol Rev 36, 514-532. Ginsburg, A.S., Grosset, J.H., Bishai, W.R., 2003. Fluoroquinolones, tuberculosis, and resistance. The Lancet. Infectious diseases 3, 432-442. Gordon, S.V., Brosch, R., Billault, A., Garnier, T., Eiglmeier, K., Cole, S.T., 1999. Identification of variable regions in the genomes of tubercle bacilli using bacterial artificial chromosome arrays. Mol Microbiol 32, 643-655. Goude, R., Amin, A.G., Chatterjee, D., Parish, T., 2009. The arabinosyltransferase EmbC is inhibited by ethambutol in Mycobacterium tuberculosis. Antimicrob Agents Chemother 53, 4138-4146. Grzegorzewicz, A.E., Pham, H., Gundi, V.A., Scherman, M.S., North, E.J., Hess, T., Jones, V., Gruppo, V., Born, S.E., Kordulakova, J., Chavadi, S.S., Morisseau, C., Lenaerts, A.J., Lee, R.E., McNeil, M.R.,

193 References

Jackson, M., 2012. Inhibition of mycolic acid transport across the Mycobacterium tuberculosis plasma membrane. Nat Chem Biol 8, 334-341. Guenin-Mace, L., Simeone, R., Demangel, C., 2009. Lipids of pathogenic Mycobacteria: contributions to virulence and host immune suppression. Transbound Emerg Dis 56, 255-268. Gunther, G., 2014. Multidrug-resistant and extensively drug-resistant tuberculosis: a review of current concepts and future challenges. Clinical medicine 14, 279-285. Gunther, G., van Leth, F., Altet, N., Dedicoat, M., Duarte, R., Gualano, G., Kunst, H., Muylle, I., Spinu, V., Tiberi, S., Viiklepp, P., Lange, C., 2015. Beyond multidrug-resistant tuberculosis in Europe: a TBNET study. Int J Tuberc Lung Dis 19, 1524-1527. Guo, J., Zheng, X., Xu, L., Liu, Z., Xu, K., Li, S., Wen, T., Liu, S., Pang, H., 2010. Characterization of a Novel Esterase Rv0045c from Mycobacterium tuberculosis. PLoS One 5, e13143.

H

Haalck, L., Spener, F., 1997. On the inhibition of microbial lipases by tetrahydrolipstatin. Methods in enzymology 286, 252-263. Hadvary, P., Sidler, W., Meister, W., Vetter, W., Wolfer, H., 1991. The lipase inhibitor tetrahydrolipstatin binds covalently to the putative active site serine of pancreatic lipase. J Biol Chem 266, 2021-2027. Hards, K., Robson, J.R., Berney, M., Shaw, L., Bald, D., Koul, A., Andries, K., Cook, G.M., 2015. Bactericidal mode of action of bedaquiline. J Antimicrob Chemother 70, 2028-2037. Heck, A.M., Yanovski, J.A., Calis, K.A., 2000. Orlistat, a new lipase inhibitor for the management of obesity. Pharmacotherapy 20, 270-279. Heep, M., Brandstatter, B., Rieger, U., Lehn, N., Richter, E., Rusch-Gerdes, S., Niemann, S., 2001. Frequency of rpoB mutations inside and outside the cluster I region in rifampin-resistant clinical Mycobacterium tuberculosis isolates. Journal of clinical microbiology 39, 107-110. Hershkovitz, I., Donoghue, H.D., Minnikin, D.E., Besra, G.S., Lee, O.Y., Gernaey, A.M., Galili, E., Eshed, V., Greenblatt, C.L., Lemma, E., Bar-Gal, G.K., Spigelman, M., 2008. Detection and molecular characterization of 9,000-year-old Mycobacterium tuberculosis from a Neolithic settlement in the Eastern Mediterranean. PloS one 3, e3426. Hett, E.C., Rubin, E.J., 2008. Bacterial growth and cell division: a mycobacterial perspective. Microbiol Mol Biol Rev 72, 126-156, table of contents. Hingley-Wilson, S.M., Sambandamurthy, V.K., Jacobs, W.R., Jr., 2003. Survival perspectives from the world's most successful pathogen, Mycobacterium tuberculosis. Nature immunology 4, 949-955. Holmberg, A., Blomstergren, A., Nord, O., Lukacs, M., Lundeberg, J., Uhlen, M., 2005. The biotin- streptavidin interaction can be reversibly broken using water at elevated temperatures. Electrophoresis 26, 501-510. Holmquist, M., 2000. Alpha/Beta-hydrolase fold enzymes: structures, functions and mechanisms. Curr Protein Pept Sci 1, 209-235. Huang, J., Bushey, D.F., 1987. Novel 1,3,4-oxadiazol-2(3H)-ones as potential pest control agents. J. Agric. Food Chem. 35, 368-373. Huet, G., Constant, P., Malaga, W., Laneelle, M.A., Kremer, K., van Soolingen, D., Daffe, M., Guilhot, C., 2009. A lipid profile typifies the Beijing strains of Mycobacterium tuberculosis: identification of a mutation responsible for a modification of the structures of phthiocerol dimycocerosates and phenolic glycolipids. J Biol Chem 284, 27101-27113. Hunter, R.L., Armitige, L., Jagannath, C., Actor, J.K., 2009. TB research at UT-Houston--a review of cord factor: new approaches to drugs, vaccines and the pathogenesis of tuberculosis. Tuberculosis (Edinb) 89 Suppl 1, S18-25.

194 References

I

Im, J.G., Itoh, H., Shim, Y.S., Lee, J.H., Ahn, J., Han, M.C., Noma, S., 1993. Pulmonary tuberculosis: CT findings--early active disease and sequential change with antituberculous therapy. Radiology 186, 653-660. Indrigo, J., Hunter, R.L., Jr., Actor, J.K., 2003. Cord factor trehalose 6,6'-dimycolate (TDM) mediates trafficking events during mycobacterial infection of murine macrophages. Microbiology 149, 2049- 2059. Istivan, T.S., Coloe, P.J., 2006. Phospholipase A in Gram-negative bacteria and its role in pathogenesis. Microbiology 152, 1263-1274.

J

Jackson, M., Raynaud, C., Laneelle, M.A., Guilhot, C., Laurent-Winter, C., Ensergueix, D., Gicquel, B., Daffe, M., 1999. Inactivation of the antigen 85C gene profoundly affects the mycolate content and alters the permeability of the Mycobacterium tuberculosis cell envelope. Mol Microbiol 31, 1573-1587. Jarlier, V., Nikaido, H., 1990. Permeability barrier to hydrophilic solutes in Mycobacterium chelonei. Journal of bacteriology 172, 1418-1423. Jia, X., Yang, L., Dong, M., Chen, S., Lv, L., Cao, D., Fu, J., Yang, T., Zhang, J., Zhang, X., Shang, Y., Wang, G., Sheng, Y., Huang, H., Chen, F., 2017. The Bioinformatics Analysis of Comparative Genomics of Mycobacterium tuberculosis Complex (MTBC) Provides Insight into Dissimilarities between Intraspecific Groups Differing in Host Association, Virulence, and Epitope Diversity. Front Cell Infect Microbiol 7, 88. Johansen, K.A., Gill, R.E., Vasil, M.L., 1996. Biochemical and molecular analysis of phospholipase C and phospholipase D activity in mycobacteria. Infection and immunity 64, 3259-3266. Johnson, G., 2017. The alpha/beta Hydrolase Fold Proteins of Mycobacterium tuberculosis, With Reference to their Contribution to Virulence. Curr Protein Pept Sci. 18, 190-210.

K

Kaur, D., Guerin, M.E., Skovierova, H., Brennan, P.J., Jackson, M., 2009. Chapter 2: Biogenesis of the cell wall and other glycoconjugates of Mycobacterium tuberculosis. Adv Appl Microbiol 69, 23-78. Khan, F.I., Lan, D., Durrani, R., Huan, W., Zhao, Z., Wang, Y., 2017. The Lid Domain in Lipases: Structural and Functional Determinant of Enzymatic Properties. Front Bioeng Biotechnol 5, 16. Kidd, D., Liu, Y., Cravatt, B.F., 2001. Profiling Serine Hydrolase Activities in Complex Proteomes. Biochemistry 40, 4005-4015. Kieser, K.J., Rubin, E.J., 2014. How sisters grow apart: mycobacterial growth and division. Nat Rev Microbiol 12, 550-562. Kiss, L.E., Ferreira, H.S., Beliaev, A., Torrao, L., Bonifacio, M.J., Learmonth, D.A., 2011. Design, synthesis, and structure-activity relationships of 1,3,4-oxadiazol-2(3H)-ones as novel FAAH inhibitors. MedChemComm 2, 889-894. Koch, R., 1882. Die Ätiologie der Tuberkulose. Berliner klinische Wochenschrift 15, 10. Kremer, L., de Chastellier, C., Dobson, G., Gibson, K.J., Bifani, P., Balor, S., Gorvel, J.P., Locht, C., Minnikin, D.E., Besra, G.S., 2005. Identification and structural characterization of an unusual mycobacterial monomeromycolyl-diacylglycerol. Mol Microbiol 57, 1113-1126. Kremer, L., Dover, L.G., Morbidoni, H.R., Vilcheze, C., Maughan, W.N., Baulard, A., Tu, S.C., Honore, N., Deretic, V., Sacchettini, J.C., Locht, C., Jacobs, W.R., Jr., Besra, G.S., 2003. Inhibition of InhA activity, but not KasA activity, induces formation of a KasA-containing complex in mycobacteria. J Biol Chem 278, 20547-20554. Kremer, L., Maughan, W.N., Wilson, R.A., Dover, L.G., Besra, G.S., 2002. The M. tuberculosis antigen 85 complex and mycolyltransferase activity. Letters in applied microbiology 34, 233-237. Kurokawa, T., Suzuki, K., Hayaoka, T., Nakagawa, T., Izawa, T., Kobayashi, M., Harada, N., 1993. Cyclophostin, acetylcholinesterase inhibitor from Streptomyces lavendulae. J Antibiot (Tokyo) 46, 1315-1318.

195 References

L

L.M.Fu, C.S.F.-L., 2005. Is Mycobacterium tuberculosis a closer relative to Gram-positive or Gram- negative bacterial pathogens? Tuberculosis (Edinb) 82(2/3), 6. Lack, N.A., Yam, K.C., Lowe, E.D., Horsman, G.P., Owen, R.L., Sim, E., Eltis, L.D., 2010. Characterization of a carbon-carbon hydrolase from Mycobacterium tuberculosis involved in cholesterol metabolism. J Biol Chem 285, 434-443. Lambert, D.M., Fowler, C.J., 2005. The Endocannabinoid System: Drug Targets, Lead Compounds, and Potential Therapeutic Applications. J. Med. Chem. 48, 5059-5087. Lamichhane, G., Tyagi, S., Bishai, W.R., 2005. Designer arrays for defined mutant analysis to detect genes essential for survival of Mycobacterium tuberculosis in mouse lungs. Infection and immunity 73, 2533-2540. Lanckriet, C., Levy-Bruhl, D., Bingono, E., Siopathis, R.M., Guerin, N., 1995. Efficacy of BCG vaccination of the newborn: evaluation by a follow-up study of contacts in Bangui. Int J Epidemiol 24, 1042-1049. Lawn, S.D., Wilkinson, R., 2006. Extensively drug resistant tuberculosis. BMJ 333, 559-560. Layre, E., Paepe, D.C., Larrouy-Maumus, G., Vaubourgeix, J., Mundayoor, S., Lindner, B., Puzo, G., Gilleron, M., 2011. Deciphering sulfoglycolipids of Mycobacterium tuberculosis. J Lipid Res 52, 1098- 1110. Lee, J.Y., 2015. Diagnosis and Treatment of Extrapulmonary Tuberculosis. Tuberc Respir Dis. (Seoul) 78, 47-55. Lehmann, J.V., J.; Esser, K.; Nodwell, M.; Kolbe, K.; Ramer, P.; Protzer, U.; Reiling, N.; Sieber, S. A., 2016. Human lysosomal acid lipase inhibitor lalistat impairs Mycobacterium tuberculosis growth by targeting bacterial hydrolases. MedChemComm Lenfant, N., Hotelier, T., Bourne, Y., Marchot, P., Chatonnet, A., 2013a. Proteins with an alpha/beta hydrolase fold: Relationships between subfamilies in an ever-growing superfamily. Chem Biol Interact 203, 266-268. Lenfant, N., Hotelier, T., Velluet, E., Bourne, Y., Marchot, P., Chatonnet, A., 2013b. ESTHER, the database of the alpha/beta-hydrolase fold superfamily of proteins: tools to explore diversity of functions. Nucleic Acids Res 41, D423-429. Leung, D., Hardouin, C., Boger, D.L., Cravatt, B.F., 2003. Discovering potent and selective reversible inhibitors of enzymes in complex proteomes. Nat Biotech 21, 687-691. Leyten, E.M., Lin, M.Y., Franken, K.L., Friggen, A.H., Prins, C., van Meijgaarden, K.E., Voskuil, M.I., Weldingh, K., Andersen, P., Schoolnik, G.K., Arend, S.M., Ottenhoff, T.H., Klein, M.R., 2006. Human T-cell responses to 25 novel antigens encoded by genes of the dormancy regulon of Mycobacterium tuberculosis. Microbes Infect 8, 2052-2060. Lin, P.L., Flynn, J.L., 2010. Understanding latent tuberculosis: a moving target. J Immunol 185, 15-22. Liu, Y., Patricelli, M.P., Cravatt, B.F., 1999. Activity-based protein profiling: the serine hydrolases. Proceedings of the National Academy of Sciences of the United States of America 96, 14694-14699. Longhi, S., Cambillau, C., 1999. Structure-activity of cutinase, a small lipolytic enzyme. Biochim Biophys Acta 1441, 185-196. Lovewell, R.R., Sassetti, C.M., VanderVen, B.C., 2016. Chewing the fat: lipid metabolism and homeostasis during M. tuberculosis infection. Curr Opin Microbiol 29, 30-36. Low, K.L., Rao, P.S., Shui, G., Bendt, A.K., Pethe, K., Dick, T., Wenk, M.R., 2009. Triacylglycerol utilization is required for regrowth of in vitro hypoxic nonreplicating Mycobacterium bovis bacillus Calmette-Guerin. Journal of bacteriology 191, 5037-5043. Lu, J.J., Pan, W., Hu, Y.J., Wang, Y.T., 2012. Multi-target drugs: the trend of drug research and development. PloS one 7, e40262. Luca, S., Mihaescu, T., 2013. History of BCG Vaccine. Maedica (Buchar) 8, 53-58. Lun, S., Bishai, W.R., 2007. Characterization of a novel cell wall-anchored protein with carboxylesterase activity required for virulence in Mycobacterium tuberculosis. J Biol Chem 282, 18348-18356. Lun, S., Miranda, D., Kubler, A., Guo, H., Maiga, M.C., Winglee, K., Pelly, S., Bishai, W.R., 2014. Synthetic Lethality Reveals Mechanisms of Mycobacterium tuberculosis Resistance to β-Lactams. mBio 5, e01767.

196 References

Luthi-Peng, Q., Marki, H.P., Hadvary, P., 1992. Identification of the active-site serine in human pancreatic lipase by chemical modification with tetrahydrolipstatin. FEBS Lett 299, 111-115.

M

Mamadou Daffé, J.-M.R., 2008. The mycobacterial cell envelop. Mamolo, M.G., Zampieri, D., Vio, L., Fermeglia, M., Ferrone, M., Pricl, S., Scialino, G., Banfi, E., 2005. Antimycobacterial activity of new 3-substituted 5-(pyridin-4-yl)-3H-1,3,4-oxadiazol-2-one and 2-thione derivatives. Preliminary molecular modeling investigations. Bioorg. Med. Chem. 13, 3797- 3809. Marino, S., Cilfone, N.A., Mattila, J.T., Linderman, J.J., Flynn, J.L., Kirschner, D.E., 2015. Macrophage polarization drives granuloma outcome during Mycobacterium tuberculosis infection. Infect Immun 83, 324-338. Marrakchi, H., Laneelle, M.A., Daffe, M., 2014. Mycolic acids: structures, biosynthesis, and beyond. Chem Biol 21, 67-85. Martin, B.P., Vasilieva, E., Dupureur, C.M., Spilling, C.D., 2015. Synthesis and comparison of the biological activity of monocyclic phosphonate, difluorophosphonate and phosphate analogs of the natural AChE inhibitor cyclophostin. Bioorg Med Chem 23, 7529-7534. Matsui, T., Carneiro, C.R., Leao, S.C., 2000. Evidence for the expression of native Mycobacterium tuberculosis phospholipase C: recognition by immune sera and detection of promoter activity. Braz J Med Biol Res 33, 1275-1282. McKinney, J.D., Honer zu Bentrup, K., Munoz-Elias, E.J., Miczak, A., Chen, B., Chan, W.T., Swenson, D., Sacchettini, J.C., Jacobs, W.R., Jr., Russell, D.G., 2000. Persistence of Mycobacterium tuberculosis in macrophages and mice requires the glyoxylate shunt enzyme isocitrate lyase. Nature 406, 735-738. McLean, K.J., Dunford, A.J., Neeli, R., Driscoll, M.D., Munro, A.W., 2007. Structure, function and drug targeting in Mycobacterium tuberculosis cytochrome P450 systems. Arch. Biochem. Biophys. 464, 228-240. Medina-Franco, J.L., Giulianotti, M.A., Welmaker, G.S., Houghten, R.A., 2013. Shifting from the single to the multitarget paradigm in drug discovery. Drug Discov Today 18, 495-501. Meniche, X., Labarre, C., de Sousa-d'Auria, C., Huc, E., Laval, F., Tropis, M., Bayan, N., Portevin, D., Guilhot, C., Daffe, M., Houssin, C., 2009. Identification of a stress-induced factor of Corynebacterineae that is involved in the regulation of the outer membrane lipid composition. Journal of bacteriology 191, 7323-7332. Migliori, G.B., De Iaco, G., Besozzi, G., Centis, R., Cirillo, D.M., 2007. First tuberculosis cases in Italy resistant to all tested drugs. Euro Surveill 12, E070517 070511. Minkkila, A., Savinainen, J.R., Kasnanen, H., Xhaard, H., Nevalainen, T., Laitinen, J.T., Poso, A., Leppanen, J., Saario, S.M., 2009. Screening of various hormone-sensitive lipase inhibitors as endocannabinoid-hydrolyzing enzyme inhibitors. ChemMedChem 4, 1253-1259. Mishra, K.C., de Chastellier, C., Narayana, Y., Bifani, P., Brown, A.K., Besra, G.S., Katoch, V.M., Joshi, B., Balaji, K.N., Kremer, L., 2008. Functional role of the PE domain and immunogenicity of the Mycobacterium tuberculosis triacylglycerol hydrolase LipY. Infection and immunity 76, 127-140. Mitchison, D.A., 1985. The action of antituberculosis drugs in short-course chemotherapy. Tubercle 66, 219-225. Miyoshi-Akiyama, T., Matsumura, K., Iwai, H., Funatogawa, K., Kirikae, T., 2012. Complete annotated genome sequence of Mycobacterium tuberculosis Erdman. Journal of bacteriology 194, 2770. Moodley, R., Godec, T.R., 2016. Short-course treatment for multidrug-resistant tuberculosis: the STREAM trials. Eur Respir Rev 25, 29-35. Muccioli, G.G., Labar, G., Lambert, D.M., 2008. CAY10499, a novel monoglyceride lipase inhibitor evidenced by an expeditious MGL assay. Chembiochem 9, 2704-2710. Mukherjee, A., Lodha, R., Kabra, S.K., 2017. Current therapies for the treatment of multidrug-resistant tuberculosis in children in India. Expert Opin Pharmacother 18, 1595-1606.

197 References

N

Narasimhan, P., Wood, J., Macintyre, C.R., Mathai, D., 2013. Risk factors for tuberculosis. Pulm Med 2013, 828939. Nardini, M., Dijkstra, B.W., 1999. Alpha/beta hydrolase fold enzymes: the family keeps growing. Curr Opin Struct Biol 9, 732-737. Neres, J., Hartkoorn, R.C., Chiarelli, L.R., Gadupudi, R., Pasca, M.R., Mori, G., Venturelli, A., Savina, S., Makarov, V., Kolly, G.S., Molteni, E., Binda, C., Dhar, N., Ferrari, S., Brodin, P., Delorme, V., Landry, V., de Jesus Lopes Ribeiro, A.L., Farina, D., Saxena, P., Pojer, F., Carta, A., Luciani, R., Porta, A., Zanoni, G., De Rossi, E., Costi, M.P., Riccardi, G., Cole, S.T., 2015. 2-Carboxyquinoxalines kill mycobacterium tuberculosis through noncovalent inhibition of DprE1. ACS Chem Biol 10, 705-714. Nessar, R., Cambau, E., Reyrat, J.M., Murray, A., Gicquel, B., 2012. Mycobacterium abscessus: a new antibiotic nightmare. J Antimicrob Chemother 67, 810-818. Newman, J.W., Morisseau, C., Hammock, B.D., 2005. Epoxide hydrolases: their roles and interactions with lipid metabolism. Prog Lipid Res 44, 1-51. Nguyen, L., Chinnapapagari, S., Thompson, C.J., 2005. FbpA-Dependent biosynthesis of trehalose dimycolate is required for the intrinsic multidrug resistance, cell wall structure, and colonial morphology of Mycobacterium smegmatis. J Bacteriol 187, 6603-6611. Nguyen, P.C., Delorme, V., Bénarouche, A., Martin, B.P., Paudel, R., Gnawali, G.R., Madani, A., Puppo, R., Landry, V., Kremer, L., Brodin, P., Spilling, C.D., Cavalier, J.-F., Canaan, S., 2017. Cyclipostins and Cyclophostin analogs as promising compounds in the fight against tuberculosis. Scientific Reports 7, 11751. Nunes-Alves, C., Booty, M.G., Carpenter, S.M., Jayaraman, P., Rothchild, A.C., Behar, S.M., 2014. In search of a new paradigm for protective immunity to TB. Nat Rev Microbiol 12, 289-299.

O

Okerberg, E.S., Wu, J., Zhang, B., Samii, B., Blackford, K., Winn, D.T., Shreder, K.R., Burbaum, J.J., Patricelli, M.P., 2005. High-resolution functional proteomics by active-site peptide profiling. Proceedings of the National Academy of Sciences of the United States of America 102, 4996-5001. Ollis, D.L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F., Franken, S.M., Harel, M., Remington, S.J., Silman, I., Schrag, J., et al., 1992. The alpha/beta hydrolase fold. Protein Eng 5, 197-211. Onyango, R.O., 2011. State of the Globe: Tracking Tuberculosis is the Test of Time. J Glob Infect Dis 3, 1-3. Orme, I.M., Basaraba, R.J., 2014. The formation of the granuloma in tuberculosis infection. Semin Immunol 26, 601-609. Ortega, C., Anderson, L.N., Frando, A., Sadler, N.C., Brown, R.W., Smith, R.D., Wright, A.T., Grundner, C., 2016a. Systematic Survey of Serine Hydrolase Activity in Mycobacterium tuberculosis Defines Changes Associated with Persistence. Cell chemical biology 23, 290-298. Ortega, C., Anderson, L.N., Frando, A., Sadler, N.C., Brown, R.W., Smith, R.D., Wright, A.T., Grundner, C., 2016b. Systematic Survey of Serine Hydrolase Activity in Mycobacterium tuberculosis Defines Changes Associated with Persistence. Cell Chem Biol 23, 290-298.

P

Pai, M., Behr, M.A., Dowdy, D., Dheda, K., Divangahi, M., Boehme, C.C., Ginsberg, A., Swaminathan, S., Spigelman, M., Getahun, H., Menzies, D., Raviglione, M., 2016a. Tuberculosis. Nat Rev Dis Primers 2, 16076. Pai, M., Nicol, M.P., Boehme, C.C., 2016b. Tuberculosis Diagnostics: State of the Art and Future Directions. Microbiol Spectr 4. Pai, M., Rodrigues, C., 2015. Management of latent tuberculosis infection: An evidence-based approach. Lung India 32, 205-207. Palomino, J.C., Martin, A., 2014. Drug Resistance Mechanisms in Mycobacterium tuberculosis. Antibiotics (Basel) 3, 317-340.

198 References

Park, M.M., Davis, A.L., Schluger, N.W., Cohen, H., Rom, W.N., 1996. Outcome of MDR-TB patients, 1983-1993. Prolonged survival with appropriate therapy. Am J Respir Crit Care Med 153, 317-324. Parker, S.K., Barkley, R.M., Rino, J.G., Vasil, M.L., 2009. Mycobacterium tuberculosis Rv3802c encodes a phospholipase/thioesterase and is inhibited by the antimycobacterial agent tetrahydrolipstatin. PloS one 4, e4281. Parker, S.K., Curtin, K.M., Vasil, M.L., 2007. Purification and characterization of mycobacterial phospholipase A: an activity associated with mycobacterial cutinase. Journal of bacteriology 189, 4153- 4160. Patricelli, M.P., Giang, D.K., Stamp, L.M., Burbaum, J.J., 2001. Direct visualization of serine hydrolase activities in complex proteomes using fluorescent active site-directed probes. Proteomics 1, 1067-1071. Pethe, K., Swenson, D.L., Alonso, S., Anderson, J., Wang, C., Russell, D.G., 2004. Isolation of Mycobacterium tuberculosis mutants defective in the arrest of phagosome maturation. Proceedings of the National Academy of Sciences of the United States of America 101, 13642-13647. Petry, S., Baringhaus, K.H., Schoenafinger, K., Jung, C., Kleine, H., Müller, G., 2004. High-throughput screening of hormone-sensitive lipase and subsequent computer-assisted compound optimisation. In: Müller, G., and Petry, S. (Ed.), Lipases and phospholipases in drug development. Wiley-VCH, Weinheim, pp. 121-137. Peyron, P., Vaubourgeix, J., Poquet, Y., Levillain, F., Botanch, C., Bardou, F., Daffe, M., Emile, J.F., Marchou, B., Cardona, P.J., de Chastellier, C., Altare, F., 2008. Foamy macrophages from tuberculous patients' granulomas constitute a nutrient-rich reservoir for M. tuberculosis persistence. PLoS pathogens 4, e1000204. Phuong Chi Nguyen, V.D., Anaïs Bénarouche1, Benjamin P. Martin2, Abdeldjalil Madani, Remy Puppo3, Valérie Landry4, Laurent Kremer5,6, Priscille Brodin4, Christopher D. Spilling2, Jean- François Cavalier1* and Stéphane Canaan1*, 2017. Cyclipostins & Cyclophostin analogs: new promising drugs for fighting TB. Cell Chemical Biology. Podinovskaia, M., Lee, W., Caldwell, S., Russell, D.G., 2013. Infection of macrophages with Mycobacterium tuberculosis induces global modifications to phagosomal function. Cellular microbiology 15, 843-859. Point, V., Benarouche, A., Zarrillo, J., Guy, A., Magnez, R., Fonseca, L., Raux, B., Leclaire, J., Buono, G., Fotiadu, F., Durand, T., Carriere, F., Vaysse, C., Couedelo, L., Cavalier, J.F., 2016. Slowing down fat digestion and absorption by an oxadiazolone inhibitor targeting selectively gastric lipolysis. European journal of medicinal chemistry 123, 834-848. Point, V., Malla, R.K., Carriere, F., Canaan, S., Spilling, C.D., Cavalier, J.F., 2013. Enantioselective inhibition of microbial lipolytic enzymes by nonracemic monocyclic enolphosphonate analogues of cyclophostin. Journal of medicinal chemistry 56, 4393-4401. Point, V., Malla, R.K., Diomande, S., Martin, B.P., Delorme, V., Carriere, F., Canaan, S., Rath, N.P., Spilling, C.D., Cavalier, J.F., 2012a. Synthesis and kinetic evaluation of cyclophostin and cyclipostins phosphonate analogs as selective and potent inhibitors of microbial lipases. Journal of medicinal chemistry 55, 10204-10219. Point, V., Pavan Kumar, K.V.P., Marc, S., Delorme, V., Parsiegla, G., Amara, S., Carrière, F., Buono, G., Fotiadu, F., Canaan, S., Leclaire, J., Cavalier, J.-F., 2012b. Analysis of the discriminative inhibition of mammalian digestive lipases by 3-phenyl substituted 1,3,4-oxadiazol-2(3H)-ones. Eur J Med Chem. 58, 452-463. Prasad, R., Singh, A., Balasubramanian, V., Gupta, N., 2017. Extensively drug-resistant tuberculosis in India: Current evidence on diagnosis & management. Indian J Med Res 145, 271-293. Pyle, M.M., 1947. Relative numbers of resistant tubercle bacilli in sputa of patients before and during treatment with streptomycin. Proc Staff Meet Mayo Clin 22, 465-473.

R

Ramakrishnan, L., 2012. Revisiting the role of the granuloma in tuberculosis. Nature reviews. Immunology 12, 352-366. Rao, A., Ranganathan, A., 2004. Interaction studies on proteins encoded by the phthiocerol dimycocerosate locus of Mycobacterium tuberculosis. Mol Genet Genomics 272, 571-579.

199 References

Rauwerdink, A., Kazlauskas, R.J., 2015. How the Same Core Catalytic Machinery Catalyzes 17 Different Reactions: the Serine-Histidine-Aspartate Catalytic Triad of alpha/beta-Hydrolase Fold Enzymes. ACS Catal 5, 6153-6176. Ravindran, M.S., Rao, S.P., Cheng, X., Shukla, A., Cazenave-Gassiot, A., Yao, S.Q., Wenk, M.R., 2014. Targeting lipid esterases in mycobacteria grown under different physiological conditions using activity- based profiling with tetrahydrolipstatin (THL). Molecular & cellular proteomics : MCP 13, 435-448. Raza, S., Fransson, L., Hult, K., 2001. Enantioselectivity in Candida antarctica lipase B: a molecular dynamics study. Protein Sci 10, 329-338. Reed, M.B., Domenech, P., Manca, C., Su, H., Barczak, A.K., Kreiswirth, B.N., Kaplan, G., Barry, C.E., 3rd, 2004. A glycolipid of hypervirulent tuberculosis strains that inhibits the innate immune response. Nature 431, 84-87. Rengarajan, J., Bloom, B.R., Rubin, E.J., 2005. Genome-wide requirements for Mycobacterium tuberculosis adaptation and survival in macrophages. Proceedings of the National Academy of Sciences of the United States of America 102, 8327-8332. Rengarajan, J., Murphy, E., Park, A., Krone, C.L., Hett, E.C., Bloom, B.R., Glimcher, L.H., Rubin, E.J., 2008. Mycobacterium tuberculosis Rv2224c modulates innate immune responses. Proceedings of the National Academy of Sciences of the United States of America 105, 264-269. Rens, C., Laval, F., Daffe, M., Denis, O., Frita, R., Baulard, A., Wattiez, R., Lefevre, P., Fontaine, V., 2016. Effects of lipid-lowering drugs on vancomycin susceptibility of mycobacteria. Antimicrob Agents Chemother. Richter, L., Saviola, B., 2009. The lipF promoter of Mycobacterium tuberculosis is upregulated specifically by acidic pH but not by other stress conditions. Microbiol Res 164, 228-232. Rosenbaum, A.I., Cosner, C.C., Mariani, C.J., Maxfield, F.R., Wiest, O., Helquist, P., 2010. Thiadiazole carbamates: potent inhibitors of lysosomal acid lipase and potential Niemann-Pick type C disease therapeutics. Journal of medicinal chemistry 53, 5281-5289. Rothschild, B.M., Martin, L.D., 2003. Frequency of pathology in a large natural sample from Natural Trap Cave with special remarks on erosive disease in the Pleistocene. Reumatismo 55, 58-65. Rousseau, C., Neyrolles, O., Bordat, Y., Giroux, S., Sirakova, T.D., Prevost, M.C., Kolattukudy, P.E., Gicquel, B., Jackson, M., 2003. Deficiency in mycolipenate- and mycosanoate-derived acyltrehaloses enhances early interactions of Mycobacterium tuberculosis with host cells. Cellular microbiology 5, 405-415. Roux, A.L., Catherinot, E., Ripoll, F., Soismier, N., Macheras, E., Ravilly, S., Bellis, G., Vibet, M.A., Le Roux, E., Lemonnier, L., Gutierrez, C., Vincent, V., Fauroux, B., Rottman, M., Guillemot, D., Gaillard, J.L., 2009. Multicenter study of prevalence of nontuberculous mycobacteria in patients with cystic fibrosis in france. J Clin Microbiol 47, 4124-4128. Rubin, E.J., 2009. The granuloma in tuberculosis--friend or foe? N Engl J Med 360, 2471-2473. Russell, D.G., 2001. Mycobacterium tuberculosis: here today, and here tomorrow. Nature reviews. Molecular cell biology 2, 569-577. Russell, D.G., Cardona, P.J., Kim, M.J., Allain, S., Altare, F., 2009a. Foamy macrophages and the progression of the human tuberculosis granuloma. Nature immunology 10, 943-948. Russell, D.G., Vanderven, B.C., Glennie, S., Mwandumba, H., Heyderman, R.S., 2009b. The macrophage marches on its phagosome: dynamic assays of phagosome function. Nature reviews. Immunology 9, 594-600.

S

Sambandamurthy, V.K., Derrick, S.C., Hsu, T., Chen, B., Larsen, M.H., Jalapathy, K.V., Chen, M., Kim, J., Porcelli, S.A., Chan, J., Morris, S.L., Jacobs, W.R., Jr., 2006. Mycobacterium tuberculosis DeltaRD1 DeltapanCD: a safe and limited replicating mutant strain that protects immunocompetent and immunocompromised mice against experimental tuberculosis. Vaccine 24, 6309-6320. Saravanan, P., Dubey, V.K., Patra, S., 2012. Potential selective inhibitors against Rv0183 of Mycobacterium tuberculosis targeting host lipid metabolism. Chem Biol Drug Des 79, 1056-1062. Sasindran, S.J., Torrelles, J.B., 2011. Mycobacterium Tuberculosis Infection and Inflammation: what is Beneficial for the Host and for the Bacterium? Front Microbiol 2, 2.

200 References

Sassetti, C.M., Boyd, D.H., Rubin, E.J., 2003. Genes required for mycobacterial growth defined by high density mutagenesis. Mol Microbiol 48, 77-84. Saunders, B.M., Cooper, A.M., 2000. Restraining mycobacteria: role of granulomas in mycobacterial infections. Immunol Cell Biol 78, 334-341. Savas, C.P., Gehring, A., Johnson, R.J., Hoops, G., 2013. Substrate specificity of Rv0045c, a bacterial esterase from Mycobacterium tuberculosis. The FASEB Journal 27, 559.552-559.552. Saviola, B., Woolwine, S.C., Bishai, W.R., 2003. Isolation of Acid-Inducible Genes of Mycobacterium tuberculosis with the Use of Recombinase-Based In Vivo Expression Technology. Infect Immun. 71, 1379-1388. Schilke, K., Weyer, K., Bretzel, G., Amthor, B., Brandt, J., Sticht-Groh, V., Fourie, P.B., Haas, W.H., 1999. Universal pattern of RpoB gene mutations among multidrug-resistant isolates of Mycobacterium tuberculosis complex from Africa. Int J Tuberc Lung Dis 3, 620-626. Schoenafinger, K., Petry, S., Müller, G., Baringhaus, K.-H., 2001. Substituted 3-phenyl-5-alkoxy-1,3,4- oxadiazol-2-one and use thereof for inhibiting hormone-sensitive lipase. Aventis Pharma GMBH, p. 44. Schue, M., Maurin, D., Dhouib, R., Bakala N'Goma, J.C., Delorme, V., Lambeau, G., Carriere, F., Canaan, S., 2010. Two cutinase-like proteins secreted by Mycobacterium tuberculosis show very different lipolytic activities reflecting their physiological function. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 24, 1893-1903. Seibert, G., Toti, L., Wink, J., 2008. Treating mycobacterial infections with cyclipostins. World patent Application, Sanofi-Aventis Deutschland GmbH. Sensi, P., 1983. History of the development of rifampin. Rev Infect Dis 5 Suppl 3, S402-406. Shanahan, E.R., Pinto, R., Triccas, J.A., Britton, W.J., West, N.P., 2010. Cutinase-like protein-6 of Mycobacterium tuberculosis is recognised in tuberculosis patients and protects mice against pulmonary infection as a single and fusion protein vaccine. Vaccine 28, 1341-1346. Sheehan, D.J., Hitchcock, C.A., Sibley, C.M., 1999. Current and emerging azole antifungal agents. Clin. Microbiol. Rev. 12, 40-79. Shen, G., Singh, K., Chandra, D., Serveau-Avesque, C., Maurin, D., Canaan, S., Singla, R., Behera, D., Laal, S., 2012. LipC (Rv0220) is an immunogenic cell surface esterase of Mycobacterium tuberculosis. Infection and immunity 80, 243-253. Shi, W., Zhang, X., Jiang, X., Yuan, H., Lee, J.S., Barry, C.E., 3rd, Wang, H., Zhang, W., Zhang, Y., 2011. Pyrazinamide inhibits trans-translation in Mycobacterium tuberculosis. Science 333, 1630-1632. Silva Miranda, M., Breiman, A., Allain, S., Deknuydt, F., Altare, F., 2012. The tuberculous granuloma: an unsuccessful host defence mechanism providing a safety shelter for the bacteria? Clin Dev Immunol 2012, 139127. Simon, G.M., Cravatt, B.F., 2010. Activity-based proteomics of enzyme superfamilies: serine hydrolases as a case study. J Biol Chem 285, 11051-11055. Singh, G., Kumar, A., Arya, S., Gupta, U.D., Singh, K., Kaur, J., 2016. Characterization of a novel esterase Rv1497 of Mycobacterium tuberculosisH37Rv demonstrating beta-lactamase activity. Enzyme Microb Technol 82, 180-190. Singh, G., Singh, G., Jadeja, D., Kaur, J., 2010. Lipid hydrolizing enzymes in virulence: Mycobacterium tuberculosis as a model system. Crit Rev Microbiol 36, 259-269. Sinsimer, D., Huet, G., Manca, C., Tsenova, L., Koo, M.S., Kurepina, N., Kana, B., Mathema, B., Marras, S.A., Kreiswirth, B.N., Guilhot, C., Kaplan, G., 2008. The phenolic glycolipid of Mycobacterium tuberculosis differentially modulates the early host cytokine response but does not in itself confer hypervirulence. Infection and immunity 76, 3027-3036. Sirgel, F.A., Tait, M., Warren, R.M., Streicher, E.M., Bottger, E.C., van Helden, P.D., Gey van Pittius, N.C., Coetzee, G., Hoosain, E.Y., Chabula-Nxiweni, M., Hayes, C., Victor, T.C., Trollip, A., 2012. Mutations in the rrs A1401G gene and phenotypic resistance to amikacin and capreomycin in Mycobacterium tuberculosis. Microb Drug Resist 18, 193-197. Slomski, A., 2013. South Africa warns of emergence of "totally" drug-resistant tuberculosis. JAMA 309, 1097-1098. Smith, I., 2003. Mycobacterium tuberculosis pathogenesis and molecular determinants of virulence. Clinical microbiology reviews 16, 463-496.

201 References

Stackebrandt, E., Rainey, F.A., Ward-Rainey, N.L., 1997. Proposal for a New Hierarchic Classification System, Actinobacteria classis nov. International Journal of Systematic and Evolutionary Microbiology 47, 12. Stalder, H., Oesterhelt, Borgström, 1992. Tetrahydrolipstatin: degradation products produced by human carboxyl-ester lipase. Helvetica Chimica Acta 75, 11. Stalder, H., Schneider, P.R., Oesterhelt, G., 1990. Tetrahydrolipstatin: Thermal and Hydrolytic Degradation Medicinal and biological Chemistry 73, 15. Stanley, S.A., Raghavan, S., Hwang, W.W., Cox, J.S., 2003. Acute infection and macrophage subversion by Mycobacterium tuberculosis require a specialized secretion system. Proceedings of the National Academy of Sciences of the United States of America 100, 13001-13006. Steele, M.A., Des Prez, R.M., 1988. The role of pyrazinamide in tuberculosis chemotherapy. Chest 94, 845-850. Sultana, R., Tanneeru, K., Guruprasad, L., 2011. The PE-PPE domain in mycobacterium reveals a serine alpha/beta hydrolase fold and function: an in-silico analysis. PloS one 6, e16745. Takayama, K., Armstrong, E.L., Kunugi, K.A., Kilburn, J.O., 1979. Inhibition by ethambutol of mycolic acid transfer into the cell wall of Mycobacterium smegmatis. Antimicrob Agents Chemother 16, 240- 242.

T

Tallman, K.R., Levine, S.R., Beatty, K.E., 2016. Small Molecule Probes Reveal Esterases with Persistent Activity in Dormant and Reactivating Mycobacterium tuberculosis. ACS Infect Dis. Tang, P., Johnston, J., 2017. Treatment of Latent Tuberculosis Infection. Curr Treat Options Infect Dis 9, 371-379. Telenti, A., Imboden, P., Marchesi, F., Lowrie, D., Cole, S., Colston, M.J., Matter, L., Schopfer, K., Bodmer, T., 1993. Detection of rifampicin-resistance mutations in Mycobacterium tuberculosis. Lancet 341, 647-650. Thakare, R., Soni, I., Dasgupta, A., Chopra, S., 2015. Delamanid for the treatment of pulmonary multidrug-resistant tuberculosis. Drugs Today (Barc) 51, 117-123. Tieman, C., 1981. Pesticidal 3-(2,3-dihydrobenzofuran-7-yl)-5-methoxy-1,3,4-oxadiazol-2 (3H)-one. Shell Oil Company. Tortoli, E., 2014. Microbiological features and clinical relevance of new species of the genus Mycobacterium. Clinical microbiology reviews 27, 727-752. Trivedi, O.A., Arora, P., Vats, A., Ansari, M.Z., Tickoo, R., Sridharan, V., Mohanty, D., Gokhale, R.S., 2005. Dissecting the mechanism and assembly of a complex virulence mycobacterial lipid. Mol Cell 17, 631-643. Tseng, S.T., Tai, C.H., Li, C.R., Lin, C.F., Shi, Z.Y., 2015. The mutations of katG and inhA genes of isoniazid-resistant Mycobacterium tuberculosis isolates in Taiwan. J Microbiol Immunol Infect 48, 249- 255. Tuaillon, E., Rubbo, P.-A., Canaan, S., Kremer, L., Nagot, N., Van, D.P.P., Vendrell, J.-P., 2011. Diagnosis method of active tuberculosis. In: Techniques, U.M.S.E., I, U.M., Montpellier, C.H.U.D., Scientifique, C.N.D.L.R. (Eds.), France.

U

Udwadia, Z.F., Amale, R.A., Ajbani, K.K., Rodrigues, C., 2012. Totally drug-resistant tuberculosis in India. Clin Infect Dis 54, 579-581. Ulrichs, T., Kaufmann, S.H., 2006. New insights into the function of granulomas in human tuberculosis. J Pathol 208, 261-269. Unissa, A.N., Subbian, S., Hanna, L.E., Selvakumar, N., 2016. Overview on mechanisms of isoniazid action and resistance in Mycobacterium tuberculosis. Infect Genet Evol 45, 474-492.

202 References

V

Vandal, O.H., Nathan, C.F., Ehrt, S., 2009. Acid resistance in Mycobacterium tuberculosis. Journal of bacteriology 191, 4714-4721. Vasilieva, E., Dutta, S., Malla, R.K., Martin, B.P., Spilling, C.D., Dupureur, C.M., 2015. Rat hormone sensitive lipase inhibition by cyclipostins and their analogs. Bioorg Med Chem 23, 944-952. Velayati, A.A., Farnia, P., Masjedi, M.R., 2013. The totally drug resistant tuberculosis (TDR-TB). Int J Clin Exp Med 6, 307-309. Velayati, A.A., Masjedi, M.R., Farnia, P., Tabarsi, P., Ghanavi, J., ZiaZarifi, A.H., Hoffner, S.E., 2009. Emergence of new forms of totally drug-resistant tuberculosis bacilli: super extensively drug-resistant tuberculosis or totally drug-resistant strains in iran. Chest 136, 420-425. Vertesy, L., Beck, B., Bronstrup, M., Ehrlich, K., Kurz, M., Muller, G., Schummer, D., Seibert, G., 2002. Cyclipostins, novel hormone-sensitive lipase inhibitors from Streptomyces sp. DSM 13381. II. Isolation, structure elucidation and biological properties. J Antibiot (Tokyo) 55, 480-494. Vilcheze, C., Jacobs, W.R., Jr., 2014. Resistance to Isoniazid and Ethionamide in Mycobacterium tuberculosis: Genes, Mutations, and Causalities. Microbiol Spectr 2, MGM2-0014-2013.

W

Waddell, S.J., Chung, G.A., Gibson, K.J., Everett, M.J., Minnikin, D.E., Besra, G.S., Butcher, P.D., 2005. Inactivation of polyketide synthase and related genes results in the loss of complex lipids in Mycobacterium tuberculosis H37Rv. Lett Appl Microbiol 40, 201-206. Weldingh, K., Rosenkrands, I., Jacobsen, S., Rasmussen, P.B., Elhay, M.J., Andersen, P., 1998. Two- dimensional electrophoresis for analysis of Mycobacterium tuberculosis culture filtrate and purification and characterization of six novel proteins. Infect Immun 66, 3492-3500. West, N.P., Cergol, K.M., Xue, M., Randall, E.J., Britton, W.J., Payne, R.J., 2011. Inhibitors of an essential mycobacterial cell wall lipase (Rv3802c) as tuberculosis drug leads. Chem Commun (Camb) 47, 5166-5168. West, N.P., Chow, F.M., Randall, E.J., Wu, J., Chen, J., Ribeiro, J.M., Britton, W.J., 2009. Cutinase- like proteins of Mycobacterium tuberculosis: characterization of their variable enzymatic functions and active site identification. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 23, 1694-1704. Wheeler, P.R., Ratledge, C., 1988. Use of carbon sources for lipid biosynthesis in Mycobacterium leprae: a comparison with other pathogenic mycobacteria. J Gen Microbiol 134, 2111-2121. WHO, 2015. WHO Global Tuberculosis Report 2015. http://apps.who.int/iris/bitstream/handle/10665/191102/9789241565059_eng.pdf?sequence=1. WHO, 2016. WHO Global Tuberculosis Report 2016. http://apps.who.int/medicinedocs/documents/s23098en/s23098en.pdf. WHO, 2017. Global tuberculosis report. http://www.who.int/tb/publications/global_report/en/. Wilder Smith, A.E., 1954. The Action of Phosgene on Acid Hydrazides to Give 1, 3, 4-Oxdiazolones of Interest in the Treatment of Tuberculosis. Science 119, 514. Wilder Smith, A.E., 1966. Some recently synthesised tuberculostatic 4-substituted oxadiazolones and oxadiazol-thiones. VI. Arzneimittelforschung 16, 1034-1038. Wilder Smith, A.E., Brodhage, H., 1961. Biological spectrum of some new tuberculostatic 1,3,4- oxadiazolones with special reference to cross-resistance and rates of emergence of resistance. Nature 192, 1195. Wilder Smith, A.E., Brodhage, H., Haukenes, G., 1962. Some tuberculostatic 1,3,4-oxadiazolones(-5) and 1,3,4-oxadiazolthiones(-5). II: Biological spectrum in vitro and activity in vivo in relation to resistance emergence. Arzneimittelforschung 12, 275-280. Wink, J., Schmidt, F.R., Seibert, G., Aretz, W., 2002. Cyclipostins: novel hormone-sensitive lipase inhibitors from Streptomyces sp. DSM 13381. I. Taxonomic studies of the producer microorganism and fermentation results. J Antibiot (Tokyo) 55, 472-479.

203 References

Y

Yang, P.Y., Liu, K., Ngai, M.H., Lear, M.J., Wenk, M.R., Yao, S.Q., 2010. Activity-based proteome profiling of potential cellular targets of Orlistat--an FDA-approved drug with anti-tumor activities. J Am Chem Soc 132, 656-666. Yedgar, S., Lichtenberg, D., Schnitzer, E., 2000. Inhibition of phospholipase A(2) as a therapeutic target. Biochimica et Biophysica Acta 1488.

Z

Zampieri, D., Mamolo, M.G., Laurini, E., Fermeglia, M., Posocco, P., Pricl, S., Banfi, E., Scialino, G., Vio, L., 2009. Antimycobacterial activity of new 3,5-disubstituted 1,3,4-oxadiazol-2(3H)-one derivatives. Molecular modeling investigations. Bioorg. Med. Chem. 17, 4693-4707. Zhang, M., Wang, J.D., Li, Z.F., Xie, J., Yang, Y.P., Zhong, Y., Wang, H.H., 2005. Expression and characterization of the carboxyl esterase Rv3487c from Mycobacterium tuberculosis. Protein Expr Purif 42, 59-66. Zhang, Y., Yew, W.W., 2015. Mechanisms of drug resistance in Mycobacterium tuberculosis: update 2015. Int J Tuberc Lung Dis 19, 1276-1289. Zhang, Y., Yew, W.W., Barer, M.R., 2012. Targeting Persisters for Tuberculosis Control. Antimicrob Agents Chemother. 56, 2223-2230. Zheng, X., Guo, J., Xu, L., Li, H., Zhang, D., Zhang, K., Sun, F., Wen, T., Liu, S., Pang, H., 2011. Crystal Structure of a Novel Esterase Rv0045c from Mycobacterium tuberculosis. PLoS One 6, e20506. Zuber, B., Chami, M., Houssin, C., Dubochet, J., Griffiths, G., Daffe, M., 2008. Direct visualization of the outer membrane of mycobacteria and corynebacteria in their native state. Journal of bacteriology 190, 5672-5680. Zumla, A., Nahid, P., Cole, S.T., 2013. Advances in the development of new tuberculosis drugs and treatment regimens. Nat Rev Drug Discov 12, 388-404. Zumla, A.I., Gillespie, S.H., Hoelscher, M., Philips, P.P., Cole, S.T., Abubakar, I., McHugh, T.D., Schito, M., Maeurer, M., Nunn, A.J., 2014. New antituberculosis drugs, regimens, and adjunct therapies: needs, advances, and future prospects. The Lancet. Infectious diseases 14, 327-340.

204