The discovery of antibiotic drugs, which treat diseases caused by bacteria, has been a hugely valuable advance in modern medicine. They work by targeting specific cellular processes in bacteria, ultimately stopping them from multiplying or killing them outright. Antibiotics sometimes also affect their human hosts and can cause side-effects, such as gut problems or skin reactions.

Recent evidence suggests that antibiotics also have an impact on the human immune system. This may happen either indirectly, by affecting ‘friendly’ bacteria normally present in the body, or through direct effects on immune cells. In turn, this could change the effectiveness of drug treatments. For example, if an antibiotic weakens immune cells, the body could have difficulty fighting off the existing infection – or become more vulnerable to new ones.

However, even though new drugs are being introduced to combat the worldwide rise of antibiotic-resistant bacteria, their effects on immunity are still not well understood. For example, bedaquiline is an antibiotic recently developed to treat tuberculosis infections that are resistant to several drugs. Giraud-Gatineau et al. wanted to determine if bedaquiline altered the human immune response to bacterial infection independently from its direct anti-microbial effects.

Macrophages engulf foreign particles like bacteria and break them down using enzymes stored within small internal compartments, or ‘lysosomes’. Initial experiments using human macrophages, grown both with and without bedaquiline, showed that the drug did not harm the cells and that they grew normally. A combination of microscope imaging and genetic analysis revealed that exposure to bedaquiline not only increased the number of lysosomes within macrophage cells, but also the activity of genes and proteins that increase lysosomes’ ability to break down foreign particles.

These results suggested that bedaquiline treatment might make macrophages better at fighting infection, even if the drug itself had no direct effect on bacterial cells. Further studies, where macrophages were first treated with bedaquiline and then exposed to different types of bacteria known to be resistant to the drug, confirmed this hypothesis: in every case, the treated macrophages became efficient bacterial killers. In contrast, older anti-tuberculosis drugs did not have any such potentiating effect on the macrophages.

This work sheds new light on our how antibiotic drugs can interact with the cells of the human immune system, and can sometimes even boost our innate defences. Such immune-boosting effects could one day be exploited to make more effective treatments against bacterial infections.

Antibiotics are commonly used in the treatment of bacterial infections, and, in effectively combating such diseases, have substantially increased human life expectancy. As with most drugs, antibiotic treatment can also alter host metabolism, leading to adverse side-effects, including nausea, hepatotoxicity, skin reactions, and gastrointestinal and neurological disorders. Such side-effects can become critical when antibiotic treatment is long and involves several drugs, as in the treatment of tuberculosis (TB), where 2–28% of patients develop mild liver injury during treatment with first-line drugs (Agal et al., 2005).

Antibiotics can interfere with the immune system, indirectly through the disturbance of the body’s microbiota (Ubeda and Pamer, 2012), or directly by modulating the functions of immune cells. Such interactions may impact treatment efficacy or the susceptibility of the host to concomitant infection. For example, after treatment completion, TB patients are more vulnerable to reactivation and reinfection of the disease, suggesting therapy-related immune impairment (Cox et al., 2008). Drug-sensitive TB can be cured by combining up to four antibiotics in a 6-month treatment; specifically, isoniazid (INH), rifampicin (RIF), ethambutol (EMB) and pyrazinamide (PZA) for 2 months, and INH and RIF for additional 4 months. INH induces apoptosis of activated CD4⁺ T cells in Mycobacterium tuberculosis (MTB)-infected mice (Tousif et al., 2014) and leads to a decrease in Th1 cytokine production in household contacts with latent TB under preventive INH therapy (Biraro et al., 2015). RIF has immunomodulatory properties and acts as a mild immunosuppressive agent in psoriasis (Tsankov and Grozdev, 2011). RIF reduces inflammation by inhibiting IκBα degradation, mitogen-activated protein kinase (MAPK) phosphorylation (Bi et al., 2011), and Toll-like receptor 4 signaling (Wang et al., 2013). PZA treatment of MTB-infected human monocytes and mice significantly reduces the release of pro-inflammatory cytokines and chemokines (Manca et al., 2013). Recently, Puyskens et al. showed that several anti-TB drugs bind to the aryl hydrocarbon receptor and may impact host defense (Puyskens et al., 2020). It is therefore necessary to understand how antibiotic treatment modulates macrophage functions, and more generally, how it impacts the host immune response.

The worldwide rise in antibiotic resistance is a major threat to global health care. A growing number of bacterial infections, such as pneumonia, salmonellosis, and TB, are becoming harder to treat as the antibiotics used to treat them become less effective. While new antibiotics are being developed and brought to the clinic, their effects on the human immune system are not being studied in-depth. Here, we have investigated the impact of a recently approved anti-TB drug, bedaquiline (BDQ), on the transcriptional responses of human macrophages infected with MTB. Macrophages are the primary cell target of MTB, which has evolved several strategies to survive and multiply inside the macrophage phagosome, including prevention of phagosome acidification (Sturgill-Koszycki et al., 1994), inhibition of phagolysosomal fusion (Armstrong and Hart, 1975) and phagosomal rupture (Simeone et al., 2012; van der Wel et al., 2007). They play a central role in the host response to TB pathogenesis, by orchestrating the formation of granulomas, presenting mycobacterial antigens to T cells, and killing the bacillus upon IFN-γ activation (Cambier et al., 2014). BDQ is a diarylquinoline that specifically inhibits a subunit of the bacterial adenosine triphosphate (ATP) synthase, decreasing intracellular ATP levels (Andries et al., 2005; Koul et al., 2007). It has 20,000 times less affinity for human ATP synthase (Haagsma et al., 2009). The most common side effects of BDQ are nausea, joint and chest pain, headache, and arrhythmias (Diacon et al., 2012; TMC207-C208 Study Group et al., 2014). However, possible interactions between BDQ and the host immune response have not been studied in detail. Understanding the impact of BDQ on the host immune response may help to develop strategies aiming at improving drug efficacy and limiting side-effects, including cytotoxicity, alteration of cell metabolism, and immunomodulation.

The emergence of bacterial strains resistant to antibiotics requires the constant development of new antibiotics, which, beyond their bactericidal activity, may have a significant impact on cellular functions. Here, we have analyzed the effects of the new anti-TB drug BDQ on human macrophages. We found that in addition to its antibacterial activity, BDQ induces cell reprogramming, increasing macrophage bactericidal activity. Gene expression profiling revealed that 1495 genes were differentially expressed in MTB-infected macrophages incubated with BDQ, with over-representation of genes involved in metabolism, lysosome biogenesis, and acidification. Recent work has highlighted the role of metabolic reprogramming in controlling immunological effector functions, emphasizing the close connection between cell function and metabolism (Wang et al., 2019). In agreement with these results, we observed a substantial increase in both the number of acidic compartments and proteolytic activity of macrophages upon BDQ treatment.

BDQ is a cationic amphiphilic drug, consisting of a hydrophobic ring structure and a hydrophilic side chain with a charged cationic amine group (Diacon et al., 2012). Cationic amphiphilic drugs can accumulate in lysosomes through ion trapping (de Duve et al., 1974). At neutral pH, they passively diffuse across cell and organelle membranes but when they enter the luminal space of acidic compartments such as lysosomes, the amine group ionizes and becomes membrane-impermeable (MacIntyre and Cutler, 1988). Such lysosomotropic compounds usually increase the lysosomal pH and thus decrease lysosomal enzyme activity (Kazmi et al., 2013). However, our results reveal instead that BDQ triggers lysosomal activation, upregulating the expression of genes coding for hydrolases and for subunits of the lysosomal proton pump v-ATPase. Consistent with these observations, we observed that BDQ-treated cells significantly increase their ability to degrade DQ-BSA.

Pre-clinical studies have shown that BDQ may induce phospholipidosis, potentially explaining some of the drug’s observed toxicities (Diacon et al., 2012). Phospholipidosis, which is characterized by the accumulation of phospholipids in lysosomes, resulting in impaired lysosome function, is common upon treatment with cationic amphiphilic compounds (Shayman and Abe, 2013). Various phospholipid species have been described including phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, and lysobisphosphatidic acid (Reasor, 1984; Yamamoto et al., 1971a; Yamamoto et al., 1971b; Yoshikawa, 1991). In BDQ-treated macrophages, we only observed an increase in the amount of phosphatidylinositol and phosphatidylinositol-4-phosphate. The quantity of cardiolipin, phosphatidylethanolamine, and phosphatidylglycerol remained unchanged upon treatment. These observations do not indicate lysosomal dysfunction, but rather a targeted regulation of certain phospholipids by BDQ. In accordance with this idea, 28 genes involved in phospholipid metabolism were differentially expressed in BDQ-treated macrophages. Phosphatidylinositol phosphates regulate many cellular functions, including endosomal trafficking, endoplasmic reticulum (ER) export, autophagy, and phagosome-lysosome fusion (De Matteis et al., 2013; Levin et al., 2017). These phospholipids may thus be involved in the increase of autophagy and mycobacterial phagosome-lysosome fusion upon BDQ treatment.

Lysosomes are both digestive organelles of the endocytic and autophagic pathways and signaling hubs involved in nutrient sensing, cell growth and differentiation, transcriptional regulation, and metabolic homeostasis (Lamming and Bar-Peled, 2019; Lawrence and Zoncu, 2019). In response to nutrients and growth factors, the mechanistic target of the rapamycin complex 1 (mTORC1) is recruited and activated at the lysosomal surface, where it promotes ribosomal biogenesis, translation, and biosynthesis of lipids (Lamming and Bar-Peled, 2019; Lawrence and Zoncu, 2019). mTORC1 binds to and phosphorylates TFEB, resulting in its cytosolic sequestration (Roczniak-Ferguson et al., 2012; Settembre et al., 2012). Upon starvation or lysosomal stress, mTORC1 is released from the lysosomal membrane and becomes inactive (Lamming and Bar-Peled, 2019; Lawrence and Zoncu, 2019). The release of lysosomal Ca²⁺ activates the phosphatase calcineurin, which de-phosphorylates TFEB and promotes its nuclear translocation (Medina et al., 2015). TFEB then binds to CLEAR (coordinated lysosomal expression and regulation) elements within the promoters of genes involved in autophagy and lysosomal biogenesis and activates their expression (Lamming and Bar-Peled, 2019; Lawrence and Zoncu, 2019). We found that TFEB translocates from the cytoplasm to the nucleus in BDQ-treated cells, with the concomitant up-regulation of 85 genes containing CLEAR elements 18 hr after incubation with the drug.

Finding the direct host-cell target of BDQ would have been a major breakthrough, but unfortunately, we were unable to identify the target. However, it is possible that BDQ could modulate the host response by interfering with ion homeostasis inside the lysosome. Greenwood et al. recently showed that BDQ accumulated primarily in host cell lipid droplets (Greenwood et al., 2019). It is tempting to speculate that these droplets are actually lysosomes. Due to its high hydrophobicity, BDQ might accumulate in lysosomal membranes and thereby change the transmembrane ion permeability. In agreement with this hypothesis, it has been shown that BDQ can accumulate at the lipid membrane of liposomes and act as a H⁺/K⁺ ionophore (Hards et al., 2018). The ensuing lysosomal stress could then facilitate the dissociation of mTOR from the lysosomal membrane (Plescher et al., 2015) and the activation of TFEB. Interestingly, BDQ increases the expression of mucolipin 1 gene (mcoln1, Figure 3A, Figure 3—figure supplement 2A), which in turn can activate TFEB. Medina et al. showed indeed that lysosomal Ca²⁺ is released through MCOLN1 and activates calcineurin, which binds and dephosphorylates TFEB (Medina et al., 2015).

A striking feature of BDQ-treated macrophages is their capacity to control pathogenic bacterial infection. BDQ enhances macrophage innate defense mechanisms, including induction of antimicrobial effectors such as nitric oxide, phagosome-lysosome fusion, and autophagy. Other anti-TB drugs have been described to regulate autophagy. INH and PZA promote autophagy activation and phagosomal maturation in MTB-infected murine macrophages (Kim et al., 2012). This process was dependent of bacterial factors and was suggested to be essential for antimycobacterial drug action and for dampening proinflammatory cytokines (Kim et al., 2012). In our system, we did not detect increased autophagy in cells treated with INH, which may also be due to differences in the autophagy response in murine and human macrophages. Altogether, we demonstrate that BDQ is able to boost the innate defenses of human cells.

A growing number of pathogenic bacteria are becoming resistant to antibiotics, making their use less effective. In addition to the development of ‘classical’ drugs targeting key factors in bacterial physiology, host-directed therapy (HDT) has emerged as approach that could be used in adjunct with existing or future antibiotics (Machelart et al., 2017). For example, metformin, an FDA-approved drug for type II diabetes, increases the production of mitochondrial reactive oxygen species and stimulates phagosome-lysosome fusion by activating the 5′-adenosine monophosphate-activated protein kinase (AMPK) (Singhal et al., 2014), and recent studies suggest that metformin provides better outcomes in TB patients, especially those with diabetes mellitus (Yew et al., 2019). Pathogens manipulate host-signaling pathways to subvert innate and adaptive immunity. It might thus be possible to reprogram the host immune system to better control or even kill bacteria. For instance, MTB has developed several strategies to counteract autophagy, including the product of the enhanced intracellular survival (Eis) gene, which limits ROS generation (Shin et al., 2010). Our results clearly show that BDQ can bypass these escape mechanisms and allow more effective control of bacterial infection. We also showed that BDQ potentiates the activity of other anti-TB drugs, independently of its bactericidal activity on MTB. Hence, our work opens new avenues for downstream evaluation of the potential use of BDQ as a potent drug in HDT.

The raw fastq files of BDQ-treated cells have been deposited in NCBI's Gene Expression Omnibus (Edgar et al., 2002) and are accessible through GEO Series accession number GSE133145. The raw fastq files of cells stimulated with heat-killed MTB or treated with different antibiotics are accessible through GEO Series accession numbers GSE143627 and GSE143731.

1. 1. Giraud-Gatineau A
   2. Tailleux L

   (2019) NCBI Gene Expression Omnibus

   ID GSE133145. Bedaquiline remodels the macrophage response.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE133145
2. 1. Giraud-Gatineau A
   2. Tailleux L

   (2020) NCBI Gene Expression Omnibus

   ID GSE143627. Inactivated M. tuberculosis and M. tuberculosis Infection remodels the macrophage response.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE143627
3. 1. Giraud-Gatineau A
   2. Tailleux L

   (2020) NCBI Gene Expression Omnibus

   ID GSE143731. Genome-wide gene expression profiling of anti-tuberculosis drugs-treated macrophages.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE143731

1. 1. Agal S
   2. Baijal R
   3. Pramanik S
   4. Patel N
   5. Gupte P
   6. Kamani P
   7. Amarapurkar D

   (2005) Monitoring and management of antituberculosis drug induced hepatotoxicity

   Journal of Gastroenterology and Hepatology 20:1745–1752.

   https://doi.org/10.1111/j.1440-1746.2005.04048.x

   • PubMed
   • Google Scholar
2. 1. Andries K
   2. Verhasselt P
   3. Guillemont J
   4. Göhlmann HW
   5. Neefs JM
   6. Winkler H
   7. Van Gestel J
   8. Timmerman P
   9. Zhu M
   10. Lee E
   11. Williams P
   12. de Chaffoy D
   13. Huitric E
   14. Hoffner S
   15. Cambau E
   16. Truffot-Pernot C
   17. Lounis N
   18. Jarlier V

   (2005) A diarylquinoline drug active on the ATP synthase of Mycobacterium tuberculosis

   Science 307:223–227.

   https://doi.org/10.1126/science.1106753

   • PubMed
   • Google Scholar
3. 1. Armstrong JA
   2. Hart PD

   (1975) Phagosome-lysosome interactions in cultured macrophages infected with virulent tubercle bacilli reversal of the usual nonfusion pattern and observations on bacterial survival

   The Journal of Experimental Medicine 142:1–16.

   https://doi.org/10.1084/jem.142.1.1

   • PubMed
   • Google Scholar
4. 1. Barreiro LB
   2. Tailleux L
   3. Pai AA
   4. Gicquel B
   5. Marioni JC
   6. Gilad Y

   (2012) Deciphering the genetic architecture of variation in the immune response to Mycobacterium tuberculosis infection

   PNAS 109:1204–1209.

   https://doi.org/10.1073/pnas.1115761109

   • PubMed
   • Google Scholar
5. 1. Bi W
   2. Zhu L
   3. Wang C
   4. Liang Y
   5. Liu J
   6. Shi Q
   7. Tao E

   (2011) Rifampicin inhibits microglial inflammation and improves neuron survival against inflammation

   Brain Research 1395:12–20.

   https://doi.org/10.1016/j.brainres.2011.04.019

   • PubMed
   • Google Scholar
6. 1. Bindea G
   2. Mlecnik B
   3. Hackl H
   4. Charoentong P
   5. Tosolini M
   6. Kirilovsky A
   7. Fridman WH
   8. Pagès F
   9. Trajanoski Z
   10. Galon J

   (2009) ClueGO: a cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks

   Bioinformatics 25:1091–1093.

   https://doi.org/10.1093/bioinformatics/btp101

   • PubMed
   • Google Scholar
7. 1. Biraro IA
   2. Egesa M
   3. Kimuda S
   4. Smith SG
   5. Toulza F
   6. Levin J
   7. Joloba M
   8. Katamba A
   9. Cose S
   10. Dockrell HM
   11. Elliott AM

   (2015) Effect of isoniazid preventive therapy on immune responses to Mycobacterium tuberculosis: an open label randomised, controlled, exploratory study

   BMC Infectious Diseases 15:438.

   https://doi.org/10.1186/s12879-015-1201-8

   • PubMed
   • Google Scholar
8. 1. Blecher-Gonen R
   2. Barnett-Itzhaki Z
   3. Jaitin D
   4. Amann-Zalcenstein D
   5. Lara-Astiaso D
   6. Amit I

   (2013) High-throughput chromatin immunoprecipitation for genome-wide mapping of in vivo protein-DNA interactions and epigenomic states

   Nature Protocols 8:539–554.

   https://doi.org/10.1038/nprot.2013.023

   • PubMed
   • Google Scholar
9. 1. Cambier CJ
   2. Falkow S
   3. Ramakrishnan L

   (2014) Host evasion and exploitation schemes of Mycobacterium tuberculosis

   Cell 159:1497–1509.

   https://doi.org/10.1016/j.cell.2014.11.024

   • PubMed
   • Google Scholar
10. 1. Cox HS
    2. Morrow M
    3. Deutschmann PW

    (2008) Long term efficacy of DOTS regimens for tuberculosis: systematic review

    BMJ 336:484–487.

    https://doi.org/10.1136/bmj.39463.640787.BE

    • PubMed
    • Google Scholar
11. 1. de Duve C
    2. de Barsy T
    3. Poole B
    4. Trouet A
    5. Tulkens P
    6. Van Hoof F

    (1974) Commentary lysosomotropic agents

    Biochemical Pharmacology 23:2495–2531.

    https://doi.org/10.1016/0006-2952(74)90174-9

    • PubMed
    • Google Scholar
12. 1. De Matteis MA
    2. Wilson C
    3. D'Angelo G

    (2013) Phosphatidylinositol-4-phosphate: the golgi and beyond

    BioEssays 35:612–622.

    https://doi.org/10.1002/bies.201200180

    • PubMed
    • Google Scholar
13. 1. Diacon AH
    2. Donald PR
    3. Pym A
    4. Grobusch M
    5. Patientia RF
    6. Mahanyele R
    7. Bantubani N
    8. Narasimooloo R
    9. De Marez T
    10. van Heeswijk R
    11. Lounis N
    12. Meyvisch P
    13. Andries K
    14. McNeeley DF

    (2012) Randomized pilot trial of eight weeks of bedaquiline (TMC207) Treatment for Multidrug-Resistant tuberculosis: long-term outcome, tolerability, and effect on emergence of drug resistance

    Antimicrobial Agents and Chemotherapy 56:3271–3276.

    https://doi.org/10.1128/AAC.06126-11

    • Google Scholar
14. 1. Dobin A
    2. Davis CA
    3. Schlesinger F
    4. Drenkow J
    5. Zaleski C
    6. Jha S
    7. Batut P
    8. Chaisson M
    9. Gingeras TR

    (2013) STAR: ultrafast universal RNA-seq aligner

    Bioinformatics 29:15–21.

    https://doi.org/10.1093/bioinformatics/bts635

    • PubMed
    • Google Scholar
15. 1. Edgar R
    2. Domrachev M
    3. Lash AE

    (2002) Gene expression omnibus: ncbi gene expression and hybridization array data repository

    Nucleic Acids Research 30:207–210.

    https://doi.org/10.1093/nar/30.1.207

    • PubMed
    • Google Scholar
16. 1. Fiorillo M
    2. Lamb R
    3. Tanowitz HB
    4. Cappello AR
    5. Martinez-Outschoorn UE
    6. Sotgia F
    7. Lisanti MP

    (2016) Bedaquiline, an FDA-approved antibiotic, inhibits mitochondrial function and potently blocks the proliferative expansion of stem-like Cancer cells (CSCs)

    Aging 8:1593–1607.

    https://doi.org/10.18632/aging.100983

    • PubMed
    • Google Scholar
17. 1. Germic N
    2. Frangez Z
    3. Yousefi S
    4. Simon HU

    (2019) Regulation of the innate immune system by autophagy: monocytes, macrophages, dendritic cells and antigen presentation

    Cell Death & Differentiation 26:715–727.

    https://doi.org/10.1038/s41418-019-0297-6

    • PubMed
    • Google Scholar
18. 1. Greenwood DJ
    2. Dos Santos MS
    3. Huang S
    4. Russell MRG
    5. Collinson LM
    6. MacRae JI
    7. West A
    8. Jiang H
    9. Gutierrez MG

    (2019) Subcellular antibiotic visualization reveals a dynamic drug reservoir in infected macrophages

    Science 364:1279–1282.

    https://doi.org/10.1126/science.aat9689

    • PubMed
    • Google Scholar
19. 1. Gutierrez MG
    2. Master SS
    3. Singh SB
    4. Taylor GA
    5. Colombo MI
    6. Deretic V

    (2004) Autophagy is a defense mechanism inhibiting BCG and Mycobacterium tuberculosis survival in infected macrophages

    Cell 119:753–766.

    https://doi.org/10.1016/j.cell.2004.11.038

    • PubMed
    • Google Scholar
20. 1. Haagsma AC
    2. Abdillahi-Ibrahim R
    3. Wagner MJ
    4. Krab K
    5. Vergauwen K
    6. Guillemont J
    7. Andries K
    8. Lill H
    9. Koul A
    10. Bald D

    (2009) Selectivity of TMC207 towards mycobacterial ATP synthase compared with that towards the eukaryotic homologue

    Antimicrobial Agents and Chemotherapy 53:1290–1292.

    https://doi.org/10.1128/AAC.01393-08

    • Google Scholar
21. 1. Hards K
    2. McMillan DGG
    3. Schurig-Briccio LA
    4. Gennis RB
    5. Lill H
    6. Bald D
    7. Cook GM

    (2018) Ionophoric effects of the antitubercular drug bedaquiline

    PNAS 115:7326–7331.

    https://doi.org/10.1073/pnas.1803723115

    • PubMed
    • Google Scholar
22. 1. Ibrahim M
    2. Andries K
    3. Lounis N
    4. Chauffour A
    5. Truffot-Pernot C
    6. Jarlier V
    7. Veziris N

    (2007) Synergistic Activity of R207910 Combined with Pyrazinamide against Murine Tuberculosis

    Antimicrobial Agents and Chemotherapy 51:1011–1015.

    https://doi.org/10.1128/AAC.00898-06

    • Google Scholar
23. Website

    1. Janssen Therapeutics

    (2012) Sirturo (bedaquiline) US food and drug administration, center for drug evaluation and research

    Accessed June 29, 2012.

    https://www.accessdata.fda.gov/drugsatfda_docs/nda/2012/204384Orig1s000PharmR.pdf
24. 1. Kazmi F
    2. Hensley T
    3. Pope C
    4. Funk RS
    5. Loewen GJ
    6. Buckley DB
    7. Parkinson A

    (2013) Lysosomal sequestration (trapping) of lipophilic amine (cationic amphiphilic) drugs in immortalized human hepatocytes (Fa2N-4 cells)

    Drug Metabolism and Disposition 41:897–905.

    https://doi.org/10.1124/dmd.112.050054

    • PubMed
    • Google Scholar
25. 1. Kim JJ
    2. Lee HM
    3. Shin DM
    4. Kim W
    5. Yuk JM
    6. Jin HS
    7. Lee SH
    8. Cha GH
    9. Kim JM
    10. Lee ZW
    11. Shin SJ
    12. Yoo H
    13. Park YK
    14. Park JB
    15. Chung J
    16. Yoshimori T
    17. Jo EK

    (2012) Host cell autophagy activated by antibiotics is required for their effective antimycobacterial drug action

    Cell Host & Microbe 11:457–468.

    https://doi.org/10.1016/j.chom.2012.03.008

    • PubMed
    • Google Scholar
26. 1. Koul A
    2. Dendouga N
    3. Vergauwen K
    4. Molenberghs B
    5. Vranckx L
    6. Willebrords R
    7. Ristic Z
    8. Lill H
    9. Dorange I
    10. Guillemont J
    11. Bald D
    12. Andries K

    (2007) Diarylquinolines target subunit c of mycobacterial ATP synthase

    Nature Chemical Biology 3:323–324.

    https://doi.org/10.1038/nchembio884

    • PubMed
    • Google Scholar
27. 1. Lamming DW
    2. Bar-Peled L

    (2019) Lysosome: the metabolic signaling hub

    Traffic 20:27–38.

    https://doi.org/10.1111/tra.12617

    • PubMed
    • Google Scholar
28. 1. Lawrence RE
    2. Zoncu R

    (2019) The lysosome as a cellular centre for signalling, metabolism and quality control

    Nature Cell Biology 21:133–142.

    https://doi.org/10.1038/s41556-018-0244-7

    • PubMed
    • Google Scholar
29. 1. Levin R
    2. Hammond GR
    3. Balla T
    4. De Camilli P
    5. Fairn GD
    6. Grinstein S

    (2017) Multiphasic dynamics of phosphatidylinositol 4-phosphate during phagocytosis

    Molecular Biology of the Cell 28:128–140.

    https://doi.org/10.1091/mbc.e16-06-0451

    • PubMed
    • Google Scholar
30. 1. Liu WJ
    2. Ye L
    3. Huang WF
    4. Guo LJ
    5. Xu ZG
    6. Wu HL
    7. Yang C
    8. Liu HF

    (2016) p62 links the autophagy pathway and the ubiqutin-proteasome system upon ubiquitinated protein degradation

    Cellular & Molecular Biology Letters 21:29.

    https://doi.org/10.1186/s11658-016-0031-z

    • PubMed
    • Google Scholar
31. 1. Love MI
    2. Huber W
    3. Anders S

    (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2

    Genome Biology 15:550.

    https://doi.org/10.1186/s13059-014-0550-8

    • PubMed
    • Google Scholar
32. 1. Machelart A
    2. Song OR
    3. Hoffmann E
    4. Brodin P

    (2017) Host-directed therapies offer novel opportunities for the fight against tuberculosis

    Drug Discovery Today 22:1250–1257.

    https://doi.org/10.1016/j.drudis.2017.05.005

    • PubMed
    • Google Scholar
33. 1. MacIntyre AC
    2. Cutler DJ

    (1988) Role of lysosomes in hepatic accumulation of chloroquine

    Journal of Pharmaceutical Sciences 77:196–199.

    https://doi.org/10.1002/jps.2600770303

    • PubMed
    • Google Scholar
34. 1. Manca C
    2. Koo M-S
    3. Peixoto B
    4. Fallows D
    5. Kaplan G
    6. Subbian S

    (2013) Host targeted activity of pyrazinamide in Mycobacterium tuberculosis infection

    PLOS ONE 8:e74082.

    https://doi.org/10.1371/journal.pone.0074082

    • Google Scholar
35. 1. Medina DL
    2. Di Paola S
    3. Peluso I
    4. Armani A
    5. De Stefani D
    6. Venditti R
    7. Montefusco S
    8. Scotto-Rosato A
    9. Prezioso C
    10. Forrester A
    11. Settembre C
    12. Wang W
    13. Gao Q
    14. Xu H
    15. Sandri M
    16. Rizzuto R
    17. De Matteis MA
    18. Ballabio A

    (2015) Lysosomal calcium signalling regulates autophagy through calcineurin and TFEB

    Nature Cell Biology 17:288–299.

    https://doi.org/10.1038/ncb3114

    • Google Scholar
36. 1. Meilang Q
    2. Zhang Y
    3. Zhang J
    4. Zhao Y
    5. Tian C
    6. Huang J
    7. Fan H

    (2012) Polymorphisms in the SLC11A1 gene and tuberculosis risk: a meta-analysis update

    The International Journal of Tuberculosis and Lung Disease 16:437–446.

    https://doi.org/10.5588/ijtld.10.0743

    • PubMed
    • Google Scholar
37. 1. Palmieri M
    2. Impey S
    3. Kang H
    4. di Ronza A
    5. Pelz C
    6. Sardiello M
    7. Ballabio A

    (2011) Characterization of the CLEAR network reveals an integrated control of cellular clearance pathways

    Human Molecular Genetics 20:3852–3866.

    https://doi.org/10.1093/hmg/ddr306

    • PubMed
    • Google Scholar
38. 1. Palomino J-C
    2. Martin A
    3. Camacho M
    4. Guerra H
    5. Swings J
    6. Portaels F

    (2002) Resazurin microtiter assay plate: simple and inexpensive method for detection of drug resistance in Mycobacterium tuberculosis

    Antimicrobial Agents and Chemotherapy 46:2720–2722.

    https://doi.org/10.1128/AAC.46.8.2720-2722.2002

    • Google Scholar
39. 1. Pfaffl MW

    (2001) A new mathematical model for relative quantification in real-time RT-PCR

    Nucleic Acids Research 29:45e–45.

    https://doi.org/10.1093/nar/29.9.e45

    • Google Scholar
40. 1. Plescher M
    2. Teleman AA
    3. Demetriades C

    (2015) TSC2 mediates hyperosmotic stress-induced inactivation of mTORC1

    Scientific Reports 5:13828.

    https://doi.org/10.1038/srep13828

    • PubMed
    • Google Scholar
41. 1. Puyskens A
    2. Stinn A
    3. van der Vaart M
    4. Kreuchwig A
    5. Protze J
    6. Pei G
    7. Klemm M
    8. Guhlich-Bornhof U
    9. Hurwitz R
    10. Krishnamoorthy G
    11. Schaaf M
    12. Krause G
    13. Meijer AH
    14. Kaufmann SHE
    15. Moura-Alves P

    (2020) Aryl hydrocarbon receptor modulation by tuberculosis drugs impairs host defense and treatment outcomes

    Cell Host & Microbe 27:238–248.

    https://doi.org/10.1016/j.chom.2019.12.005

    • PubMed
    • Google Scholar
42. 1. Reasor MJ

    (1984)

    Phospholipidosis in the alveolar macrophage induced by cationic amphiphilic drugs

    Federation Proceedings 43:2578–2581.

    • PubMed
    • Google Scholar
43. 1. Remmerie A
    2. Scott CL

    (2018) Macrophages and lipid metabolism

    Cellular Immunology 330:27–42.

    https://doi.org/10.1016/j.cellimm.2018.01.020

    • Google Scholar
44. 1. Roczniak-Ferguson A
    2. Petit CS
    3. Froehlich F
    4. Qian S
    5. Ky J
    6. Angarola B
    7. Walther TC
    8. Ferguson SM

    (2012) The transcription factor TFEB links mTORC1 signaling to transcriptional control of lysosome homeostasis

    Science Signaling 5:ra42.

    https://doi.org/10.1126/scisignal.2002790

    • PubMed
    • Google Scholar
45. 1. Sena LA
    2. Chandel NS

    (2012) Physiological roles of mitochondrial reactive oxygen species

    Molecular Cell 48:158–167.

    https://doi.org/10.1016/j.molcel.2012.09.025

    • PubMed
    • Google Scholar
46. 1. Settembre C
    2. Di Malta C
    3. Polito VA
    4. Garcia Arencibia M
    5. Vetrini F
    6. Erdin S
    7. Erdin SU
    8. Huynh T
    9. Medina D
    10. Colella P
    11. Sardiello M
    12. Rubinsztein DC
    13. Ballabio A

    (2011) TFEB links autophagy to lysosomal biogenesis

    Science 332:1429–1433.

    https://doi.org/10.1126/science.1204592

    • PubMed
    • Google Scholar
47. 1. Settembre C
    2. Zoncu R
    3. Medina DL
    4. Vetrini F
    5. Erdin S
    6. Erdin S
    7. Huynh T
    8. Ferron M
    9. Karsenty G
    10. Vellard MC
    11. Facchinetti V
    12. Sabatini DM
    13. Ballabio A

    (2012) A lysosome-to-nucleus signalling mechanism senses and regulates the lysosome via mTOR and TFEB

    The EMBO Journal 31:1095–1108.

    https://doi.org/10.1038/emboj.2012.32

    • PubMed
    • Google Scholar
48. 1. Shayman JA
    2. Abe A

    (2013) Drug induced phospholipidosis: an acquired lysosomal storage disorder

    Biochimica Et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids 1831:602–611.

    https://doi.org/10.1016/j.bbalip.2012.08.013

    • Google Scholar
49. 1. Shin DM
    2. Jeon BY
    3. Lee HM
    4. Jin HS
    5. Yuk JM
    6. Song CH
    7. Lee SH
    8. Lee ZW
    9. Cho SN
    10. Kim JM
    11. Friedman RL
    12. Jo EK

    (2010) Mycobacterium tuberculosis eis regulates autophagy, inflammation, and cell death through redox-dependent signaling

    PLOS Pathogens 6:e1001230.

    https://doi.org/10.1371/journal.ppat.1001230

    • PubMed
    • Google Scholar
50. 1. Simeone R
    2. Bobard A
    3. Lippmann J
    4. Bitter W
    5. Majlessi L
    6. Brosch R
    7. Enninga J

    (2012) Phagosomal rupture by Mycobacterium tuberculosis results in toxicity and host cell death

    PLOS Pathogens 8:e1002507.

    https://doi.org/10.1371/journal.ppat.1002507

    • PubMed
    • Google Scholar
51. 1. Singhal A
    2. Jie L
    3. Kumar P
    4. Hong GS
    5. Leow MK
    6. Paleja B
    7. Tsenova L
    8. Kurepina N
    9. Chen J
    10. Zolezzi F
    11. Kreiswirth B
    12. Poidinger M
    13. Chee C
    14. Kaplan G
    15. Wang YT
    16. De Libero G

    (2014) Metformin as adjunct antituberculosis therapy

    Science Translational Medicine 6:263ra159.

    https://doi.org/10.1126/scitranslmed.3009885

    • PubMed
    • Google Scholar
52. 1. Sturgill-Koszycki S
    2. Schlesinger P
    3. Chakraborty P
    4. Haddix P
    5. Collins H
    6. Fok A
    7. Allen R
    8. Gluck S
    9. Heuser J
    10. Russell D

    (1994) Lack of acidification in Mycobacterium phagosomes produced by exclusion of the vesicular proton-ATPase

    Science 263:678–681.

    https://doi.org/10.1126/science.8303277

    • Google Scholar
53. 1. Tailleux L
    2. Neyrolles O
    3. Honoré-Bouakline S
    4. Perret E
    5. Sanchez F
    6. Abastado JP
    7. Lagrange PH
    8. Gluckman JC
    9. Rosenzwajg M
    10. Herrmann JL

    (2003) Constrained intracellular survival of Mycobacterium tuberculosis in human dendritic cells

    The Journal of Immunology 170:1939–1948.

    https://doi.org/10.4049/jimmunol.170.4.1939

    • PubMed
    • Google Scholar
54. 1. Tailleux L
    2. Waddell SJ
    3. Pelizzola M
    4. Mortellaro A
    5. Withers M
    6. Tanne A
    7. Castagnoli PR
    8. Gicquel B
    9. Stoker NG
    10. Butcher PD
    11. Foti M
    12. Neyrolles O

    (2008) Probing host pathogen cross-talk by transcriptional profiling of both Mycobacterium tuberculosis and infected human dendritic cells and macrophages

    PLOS ONE 3:e1403.

    https://doi.org/10.1371/journal.pone.0001403

    • PubMed
    • Google Scholar
55. 1. TMC207-C208 Study Group
    2. Diacon AH
    3. Pym A
    4. Grobusch MP
    5. de los Rios JM
    6. Gotuzzo E
    7. Vasilyeva I
    8. Leimane V
    9. Andries K
    10. Bakare N
    11. De Marez T
    12. Haxaire-Theeuwes M
    13. Lounis N
    14. Meyvisch P
    15. De Paepe E
    16. van Heeswijk RP
    17. Dannemann B

    (2014) Multidrug-resistant tuberculosis and culture conversion with bedaquiline

    New England Journal of Medicine 371:723–732.

    https://doi.org/10.1056/NEJMoa1313865

    • PubMed
    • Google Scholar
56. 1. Tousif S
    2. Singh DK
    3. Ahmad S
    4. Moodley P
    5. Bhattacharyya M
    6. Van Kaer L
    7. Das G

    (2014) Isoniazid induces apoptosis of activated CD4+ T cells: implications for post-therapy tuberculosis reactivation and reinfection

    The Journal of Biological Chemistry 289:30190–30195.

    https://doi.org/10.1074/jbc.C114.598946

    • PubMed
    • Google Scholar
57. 1. Troegeler A
    2. Lastrucci C
    3. Duval C
    4. Tanne A
    5. Cougoule C
    6. Maridonneau-Parini I
    7. Neyrolles O
    8. Lugo-Villarino G

    (2014) An efficient siRNA-mediated gene silencing in primary human monocytes, dendritic cells and macrophages

    Immunology and Cell Biology 92:699–708.

    https://doi.org/10.1038/icb.2014.39

    • PubMed
    • Google Scholar
58. 1. Tsankov N
    2. Grozdev I

    (2011) Rifampicin – A mild immunosuppressive agent for psoriasis

    Journal of Dermatological Treatment 22:62–64.

    https://doi.org/10.3109/09546630903496975

    • Google Scholar
59. 1. Ubeda C
    2. Pamer EG

    (2012) Antibiotics, microbiota, and immune defense

    Trends in Immunology 33:459–466.

    https://doi.org/10.1016/j.it.2012.05.003

    • Google Scholar
60. 1. van der Wel N
    2. Hava D
    3. Houben D
    4. Fluitsma D
    5. van Zon M
    6. Pierson J
    7. Brenner M
    8. Peters PJ

    (2007) M. tuberculosis and M. leprae translocate from the phagolysosome to the cytosol in myeloid cells

    Cell 129:1287–1298.

    https://doi.org/10.1016/j.cell.2007.05.059

    • PubMed
    • Google Scholar
61. 1. Wang X
    2. Grace PM
    3. Pham MN
    4. Cheng K
    5. Strand KA
    6. Smith C
    7. Li J
    8. Watkins LR
    9. Yin H

    (2013) Rifampin inhibits Toll‐like receptor 4 signaling by targeting myeloid differentiation protein 2 and attenuates neuropathic pain

    The FASEB Journal 27:2713–2722.

    https://doi.org/10.1096/fj.12-222992

    • Google Scholar
62. 1. Wang A
    2. Luan HH
    3. Medzhitov R

    (2019) An evolutionary perspective on immunometabolism

    Science 363:eaar3932.

    https://doi.org/10.1126/science.aar3932

    • Google Scholar
63. 1. Weiss G
    2. Schaible UE

    (2015) Macrophage defense mechanisms against intracellular bacteria

    Immunological Reviews 264:182–203.

    https://doi.org/10.1111/imr.12266

    • Google Scholar
64. 1. Wynn TA
    2. Chawla A
    3. Pollard JW

    (2013) Macrophage biology in development, homeostasis and disease

    Nature 496:445–455.

    https://doi.org/10.1038/nature12034

    • Google Scholar
65. 1. Yamamoto A
    2. Adachi S
    3. Ishikawa K
    4. Yokomura T
    5. Kitani T

    (1971a) Studies on drug-induced lipidosis 3. lipid composition of the liver and some other tissues in clinical cases of "Niemann-Pick-like syndrome" induced by 4,4'-diethylaminoethoxyhexestrol

    The Journal of Biochemistry 70:775–784.

    https://doi.org/10.1093/oxfordjournals.jbchem.a129695

    • PubMed
    • Google Scholar
66. 1. Yamamoto A
    2. Adachi S
    3. Kitani T
    4. Shinji Y
    5. Seki K

    (1971b)

    Drug-induced lipidosis in human cases and in animal experiments. Accumulation of an acidic glycerophospholipid

    Journal of Biochemistry 69:613–615.

    • PubMed
    • Google Scholar
67. 1. Yew WW
    2. Chang KC
    3. Chan DP
    4. Zhang Y

    (2019) Metformin as a host-directed therapeutic in tuberculosis: is there a promise?

    Tuberculosis 115:76–80.

    https://doi.org/10.1016/j.tube.2019.02.004

    • PubMed
    • Google Scholar
68. 1. Yoshikawa H

    (1991) Effects of drugs on cholesterol esterification in normal and Niemann-Pick type C fibroblasts: ay-9944, other cationic amphiphilic drugs and DMSO

    Brain and Development 13:115–120.

    https://doi.org/10.1016/S0387-7604(12)80118-5

    • PubMed
    • Google Scholar
69. 1. Zhang Y
    2. Mitchison D

    (2003)

    The curious characteristics of pyrazinamide: a review

    The International Journal of Tuberculosis and Lung Disease 7:6–21.

    • PubMed
    • Google Scholar

Acceptance summary:

This paper provides evidence that the antibiotic bedaquiline not only has direct anti-microbial effects but may also act on host cells to enhance their capacity for killing bacteria. Bedaquiline is an important antibiotic for the treatment of tuberculosis, and a better understanding of its activities is important for developing next-generation therapeutics for this globally significant disease. More-broadly, understanding the host-directed effects of antibiotics is also significant for other bacterial infections for which drug resistance is increasing.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "The antibiotic bedaquiline activates host macrophage innate immune resistance to bacterial infection" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor and a Senior Editor.

Our decision has been reached after consultation between the reviewers. We found that the work addresses an important topic and is interesting. However, the reviewers have raised substantial technical concerns. Therefore, the decision is to reject your manuscript without requesting a revision. Nevertheless, in addition to the individual reviews, we are providing a consolidated review (below) that captures the consensus of the three reviewers in case you feel that the concerns are easily addressable within a reasonable timeframe.

1) A major concern is that the manuscript does not cleanly demonstrate that the effects of BDQ observed on MTB growing in cells is in fact due to the effects of BDQ on the host cells vs. effects of BDQ on MTB. The use of a BDQ resistant mutant is a nice attempt to address this concern, but it is not entirely satisfactory. First, the data shown in Figure 1—figure supplement 2C demonstrate that BDQ (5µg/mL) does appear to have an effect on growth of the resistant mutant. Second, even if growth of the bacteria is unimpeded in broth, the antibiotic may still have an effect on the bacteria in cells. Even a minor effect on MTB (e.g., lysis of a few bacteria) might have a large effect on the host response. Unless this central concern can be satisfactorily addressed (which we fear it may not be) then we do not feel the paper would be a strong candidate for eLife.

2) Another major concern relates to the presentation of the RNAseq data in Figure 1A-B and Figure supplement 3A-B. Figure 1AB shows the effect of BDQ on MTB-infected macrophages, whereas Figure supplement 3A-B shows the effect of BDQ on uninfected macrophages. If BDQ acts on host cells independent of an effect on bacteria, then we would expect similar genes to be induced in both conditions. However, the data are presented only at the level of GO-term annotations. We require gene-level expression data comparing gene induction in all four conditions (+/- BDQ and +/- MTB) in a single analysis so that we can assess whether in fact the effects of BDQ are similar +/- MTB. It may well be that MTB modulates the host response to BDQ, e.g., via activation of TLRs. In such a case, it might be useful to compare the response of cells infected with live TB +/- BDQ with the response to cells treated with TLR agonists or killed bacteria +/- BDQ. If, however, the main effects of BDQ on host gene expression only appear to occur in conjunction with infection with live MTB, then we would be concerned that the effects of BDQ are via their effects on MTB and not direct host-directed effects.

3) Another major concern is that there is no direct evidence that the apparent effects of BDQ on MTB growing in cells is due to induction of TFEB. This should be shown with TFEB mutant cells. These cells could also be used to show that the induction of lysosomal-related genes and increased lysotracker staining by BDQ is due to TFEB activation.

4) As raised by reviewer 2, it is imperative to show that the synergy between BDQ and PZA on the BDQ-resistant mutant does not occur in broth.

Other comments:

- A MIC for BDQ on the resistant mutant should be reported.

- Figure 1—figure supplement 2B is missing information about what concentration of BDQ was used.

- Reviewer 2 raised a number of questions about the data in Figure 1—figure supplement 2C that should be addressed.

- As raised by reviewer 3, the proposed mechanism connecting BDQ to calcium and TFEB activation is only weakly supported by data. The authors should consider toning down their conclusions related to this part of the paper.

Reviewer #1:

Tuberculosis remains a major global problem and kills more people than any other single pathogen. This paper addresses the mechanism of action of one of the most important new tuberculosis drugs, bedaquiline (BDQ), and shows that in addition to its well-known direct anti-microbial effects, BDQ also has significant 'host-directed' effects that can contribute to its anti-microbial activity. In particular, BDQ is shown to enhance lysosomal activity in macrophages. This appears to be via activation of the host transcription factor, TFEB.

In general the paper is well-written, and the experiments are well-described and well-performed. There is significant current interest in the idea of host-directed therapies for TB, and the idea that a major new TB drug might act in part via its effects on host cells is interesting and potentially very important. I have only a few minor comments:

1) Transcriptional profiling is performed on both MTB-infected and in uninfected cells. It is claimed that the effects of BDQ are "similar" in both situations. However, the way the data are presented makes it hard to see how similar or different the effects of BDQ are +/- MTB. Could the authors present a heat-map (indicating key genes, particularly the lysosomal signature) with all four conditions (+/- MTB, +/- BDQ) in a single analysis/panel? If there are significant differences, it might be relevant to comment on these.

2) A weakness of the manuscript is that the direct host-cell target of BDQ is not identified. Thus, the mechanism leading to TFEB induction is unclear. However, I tend to feel that identification of the target of BDQ is beyond what could be reasonably asked of this study. However, perhaps the authors could at least briefly discuss what the host cell target of BDQ might be and acknowledge that they have not identified it. In addition, the evidence that the effects of BDQ require TFEB is only indirect. It would thus be desirable, if feasible, for the authors to show, e.g., with TFEB knock out cells, that the effects of BDQ on lysosome-related gene induction does require TFEB.

Reviewer #2:

The manuscript by Giraud-Gatineau et al. concludes that the newest anti-tuberculosis antibiotic, bedaquiline (BDQ), works by not only inhibiting ATP synthase in the bacterium, but also by synergistically working directly on host macrophages to boost their antimicrobial effects during infection. There have been many reports of similar kinds of effects of different antibiotic regimens on activating immune responses the host, but nearly all suffer from the inherent difficulty in conclusively determining whether the "host effects" of the antibiotic are due to direct action on macrophage pathways or whether the effects on the host are secondary to the effects of the antibiotic on the physiology of the bacteria. Indeed, each of the papers cited by the authors on this subject suffer from this problem, and the field remains murky.

It would be an exciting finding that BDQ activates host macrophage pathways (perhaps by partial inhibition of the host ATP synthase?), but unfortunately my main criticism of this paper is that I find that there are multiple issues that center around the main problem of whether BDQ actually increases the antibacterial capacity of macrophages by directly activating host pathways. I certainly appreciate the difficulty of the task, and the authors consider bacterial load as one potential factor by using a BDQ-resistant mutant, but many of my concerns relate to the data itself of these key experiments, as well as the possibility of effects on bacterial physiology induced by the antibiotic. For example, even a small amount of lysis of the bacteria could influence host transcriptional responses. Likewise, it seems that comparing resistance of a bacterium in culture vs. in a cell doesn't necessarily show that the effect is on the host. It seems likely that the antibiotic is still working solely on the bug, but the host environment sensitizes the bacterium to the antibiotic.

1) It seems likely to me that the transcriptional (Figure 1) and functional effects (Figure 2) of BDQ on infected macrophages is due to indirect effects on bacterial physiology. The use of the BDQ-resistant mutant is an important experiment, but I have doubts that the concentration used for the macrophage infection studies (5µg/mL) does not have any effect on the bacteria. This concentration apparently inhibits the growth of the resistant mutant Figure 1—figure supplement 2C inside macrophages, and this could be due to either activation of antimicrobial host effects or partial inhibition of bacterial growth. It is not clear what effects this concentration has on the bacteria, the mutant was selected on 0.3 µg/mL BDQ, and I could not find an MIC stated in the manuscript. The growth curve of bacteria grown in axenic culture +/- BDQ (panel B) is missing any information about the concentration of BDQ used, and there is no direct comparison of the growth kinetics between wild-type and mutant bacteria at 5 µg/mL. Moreover, the assumption is that treatment with 5µg/mL of BDQ doesn't change the bacteria at all, and thus any differences must be due to host modulation, yet I think there are multiple reasons to doubt this.

2) There are a number of other worrisome details about the experiment displayed in Figure 1—figure supplement 2C. First, it is curious that neither the wild-type nor resistant bacteria grew very well between the 18hour and 5day time points in the absence of BDQ. It appears they double at best, but much more would be expected. Second, the BDQ-resistant mutant didn't grow at all over this 4+ day time course in the presence of 5µg/mL BDQ. It seems unlikely to me that this effects of BDQ are solely due to host activation. Third, why are the CFU levels of the mutant bacteria so much higher than the wild-type at 18 hours?

3) It seems important to demonstrate in all of their bacterial infections, including Figure 4, that the macrophage monolayers are intact and of equal density amongst all the treatments. Indeed, while the authors test the effects of BDQ on macrophage viability in the absence of bacteria, it seems important to assess its effect during the heightened inflammatory response during infection.

4) I believe that the results with the host effects of BDQ would be more conclusive if they could show that treatment with different antibiotics gives rise to different results.

5) The authors also perform treatment of macrophages with BDQ in the absence of infection, and claim that similar pathways are activated, inferring that these effects are independent of the presence of bacteria. Unfortunately, these comparisons were evaluated at the level of GO pathway annotation, and I believe strongly that if BDQ is directly activating gene expression, perhaps via PTFB, that the comparison should be performed at the gene level, i.e. as done in Figure 2A.

6) I find the conclusion drawn from the apparent synergy btw PZA and BDQ is misleading for two reasons. First of all, given the residual sensitivity of the resistant mutant to high levels of BDQ, it seems imperative that this synergy be tested in culture. Second, given the model that BDQ activates host antimicrobial activities, I expect that BDQ would also synergize with other antibiotics besides PZA.

7) Much of the lysosome results in Figure 2 is complicated by the mild sensitivity of the resistant strain to BDQ. The one experiment that is interesting is the no-infection in panel G, but there is only a correlation, there is no evidence provided that this is the host effect of BDQ.

Reviewer #3:

In this manuscript, Giraud-Gatineau et al. describe how bedaquiline (BDQ), an antibiotic in clinical use for TB, activates bactericidal functions in human primary macrophages. Transcriptional analysis of such treated macrophages leads them to examine the autophagy-lysosomal pathway, which they find is upregulated and necessary for the effect of BDQ in reducing M. tuberculosis load in infected macrophages in vitro. Finally, they show that transcription factor TFEB is activated by BDQ, and that the activity of BDQ in macrophages is dependent on calcium, suggesting that the MCOLN1-calcineurin-TFEB pathway described by Medina et al., (2015) is implicated.

Overall, the concept of antibiotics serving dual functions as host-directed therapies is appealing and of great interest. However, there are several major concerns that must be addressed:

1) The proposed mechanism of enhancement of bactericidal activity is through TFEB, but the only evidence presented is correlative (TFEB immunofluorescence showing TFEB in the nucleus). This must be directly tested using loss-of-function approaches.

2) Similarly, the implication from the RNAseq, LysoTracker, and TFEB IF is that TFEB upregulates the genes responsible for lysosomal biogenesis and autophagy, thus increasing bacterial clearance. The inducible recruitment of TFEB to relevant promoters must be directly tested.

3) The mechanism connecting BDQ to calcium release and TFEB activation is not shown. Neither is the requirement for calcineurin.

4) Although the synergistic effect of BDQ with PZA seems to approach 1 log (~10% survival) killing of M. tuberculosis, the effect size of BDQ alone (Figure 3F) is quite small. This raises questions of how significant these observations are to TB. This is important, since the whole setup is to find new therapeutic agents to control MDR TB.

5) The authors use BAPTA (extracellular) to eliminate calcium and test the effect of loss of TFEB activation on bacterial CFU. This is done with S. aureus (not TB). Also, if they really are using BAPTA, they would not be inhibiting intracellular calcium release but import from extracellular stores, changing the interpretation of the experiment.

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "The antibiotic bedaquiline activates host macrophage innate immune resistance to bacterial infection" for further consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor and a Senior Editor.

The manuscript has been improved and the reviewers are generally enthusiastic about publication but there are some remaining issues that need to be addressed before acceptance. Please revise your manuscript to address these issues.

Reviewer #1:

The revised manuscript addresses the main concerns of this reviewer. Overall, I think the data strongly indicate that the antibiotic bedaquiline (BDQ) can induce antibacterial responses by direct action on host cells. The mechanism of action of BDQ is of great importance to the TB field, and the concept that antibiotics can act (in part) by targeting the host is of even broader importance.

My only query relates to new data (Figure 7) showing that human macrophages induce nitrite, which the authors claim is derived from host NO. As far as I am aware, there has been considerable difficulty demonstrating induction of NO by human macrophages in culture. In some cases, the NO2- observed has actually been produced by the bacteria themselves (see Bussel, Zhang and Nathan, 2013). The authors may want to take this into account.

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

[…]

1) A major concern is that the manuscript does not cleanly demonstrate that the effects of BDQ observed on MTB growing in cells is in fact due to the effects of BDQ on the host cells vs. effects of BDQ on MTB. The use of a BDQ resistant mutant is a nice attempt to address this concern, but it is not entirely satisfactory. First, the data shown in Figure 1—figure supplement 2C demonstrate that BDQ (5µg/mL) does appear to have an effect on growth of the resistant mutant. Second, even if growth of the bacteria is unimpeded in broth, the antibiotic may still have an effect on the bacteria in cells. Even a minor effect on MTB (e.g., lysis of a few bacteria) might have a large effect on the host response. Unless this central concern can be satisfactorily addressed (which we fear it may not be) then we do not feel the paper would be a strong candidate for eLife.

We thank the reviewer 2 for raising this point, acknowledging that this may be due to problems in our writing. Our main goal was to show that BDQ can activate macrophages, making them resistant to any bacterial infections, and not just to infection with MTB. In the revised manuscript, we present new data, which clearly show that BDQ induces bactericidal functions by directly targeting the host.

The first question is to what extent is the MTB mutant resistant to BDQ (BDQr-MTB). The MIC₉₉ of the mutant is 36 µg/mL, which is much higher than the concentration that we used. In our paper, BDQ is used at a maximum concentration of 5 µg/mL during 48 hours and 1 µg/mL during 5 days. At these concentrations, we never observed bacterial killing nor decrease of bacterial growth in liquid medium (Figure 1—figure supplement 2B). We also showed that the bacillary load of resistant MTB inside macrophages is the same in untreated cells as in cells after 18 hours of BDQ treatment (the time point chosen for the RNAseq experiment). In contrast, in the same experiment using BDQ-susceptible MTB there is a 70% decrease in the bacillary load (Figure 1—figure supplement 2C).

We agree that BDQ might kill a few intracellular bacteria (even if we did not detect any changes by CFUs), but it is very unlikely that this effect could explain the macrophage resistance to bacterial infection. First, the effect of BDQ can also be inhibited by 3-MA and BAF, whereas these two autophagy inhibitors have no effect on untreated cells (Figure 6F). These results suggest a host-targeted effect of BDQ. The regulation of the lysosomal pathway by BDQ is also observed in uninfected macrophages and in cells stimulated with heat-inactivated MTB or with TLR agonists such as LPS and Pam3CSK4 (Figure 3—figure supplement 2). Such changes were not observed in cells treated with the first line anti-TB drugs ethambutol (EMB), isoniazid (INH), pyrazinamide (PZA) and rifampicin (RIF) (new transcriptome analysis presented in Figure 5A-C and Figure 5—figure supplement 2, and LysoTracker staining analyzed by flow cytometry Figure 5D).

If the lysis of a few bacteria by BDQ induced macrophage activation, then we would expect a greater response when the cells are stimulated by heat-killed bacteria or drug-susceptible MTB than with BDQr-MTB. Again, as shown in Figure 3—figure supplement 2, we did not observe any differences in the intensity of LysoTracker staining. Similar results have also been obtained when comparing the expression level of lysosomal genes. Overall, our results strongly argue for a host-dependent effect of BDQ.

2) Another major concern relates to the presentation of the RNAseq data in Figure 1A-B and Figure supplement 3A-B. Figure 1AB shows the effect of BDQ on MTB-infected macrophages, whereas Figure supplement 3A-B shows the effect of BDQ on uninfected macrophages. If BDQ acts on host cells independent of an effect on bacteria, then we would expect similar genes to be induced in both conditions. However, the data are presented only at the level of GO-term annotations. We require gene-level expression data comparing gene induction in all four conditions (+/- BDQ and +/- MTB) in a single analysis so that we can assess whether in fact the effects of BDQ are similar +/- MTB. It may well be that MTB modulates the host response to BDQ, e.g., via activation of TLRs. In such a case, it might be useful to compare the response of cells infected with live TB +/- BDQ with the response to cells treated with TLR agonists or killed bacteria +/- BDQ. If, however, the main effects of BDQ on host gene expression only appear to occur in conjunction with infection with live MTB, then we would be concerned that the effects of BDQ are via their effects on MTB and not direct host-directed effects.

Here we present new analysis, comparing gene expression level in all four conditions (+/- BDQ and +/- MTB). Upon BDQ treatment, 127 and 1043 genes, respectively, were differentially expressed in naïve cells and in MTB-infected macrophages, and the expression of 452 genes was modulated in both conditions. It is not surprising that the expression of more genes was affected in MTB-infected cells as MTB infection induces extensive remodeling of the transcriptome (Barreiro et al., 2012; Tailleux et al., 2008). We then classified all these genes by performing gene-set enrichment analysis. We showed that the expression of 70% of the lysosomal gene differentially expressed in MTB-infected macrophages, were also up-regulated in naïve cells upon BDQ treatment. These results are now included in Figure 1D and Figure 3A. We also performed additional experiments to determine whether the effects of BDQ on host gene expression were dependent of infection with live MTB. Briefly, cells were untreated or stimulated with LPS (TLR4 agonist), Pam3CSK4 (TLR1/2 agonist), heat-killed bacteria, drug-susceptible MTB or BDQr-MTB, and treated with BDQ. After 18 hours, RNA was collected and we performed RT-qPCR on a panel of lysosomal genes. We also analyzed the intensity of the LysoTracker staining using flow cytometry (Figure 3—figure supplement 2). Our results clearly show that the main effects on lysosome biogenesis/activation occurred with BDQ treatment and were not exclusively seen after infection with live MTB.

3) Another major concern is that there is no direct evidence that the apparent effects of BDQ on MTB growing in cells is due to induction of TFEB. This should be shown with TFEB mutant cells. These cells could also be used to show that the induction of lysosomal-related genes and increased lysotracker staining by BDQ is due to TFEB activation.

Using siRNA-mediated gene silencing (Troegeler et al., 2014), we inactivated the expression of TFEB in macrophages and confirmed its major role in the resistance to bacterial infection upon BDQ treatment. We also present a chromatin immunoprecipitation assay with anti-TFEB antibody. As suggested by reviewers 1 and 3, we showed that TFEB activation is required for (i) the increased LysoTracker staining (Figure 9G), (ii) the induction of a panel of lysosomal-related genes (Figure 9H), and (iii) a better control of S. aureus and MTB infection upon BDQ treatment (Figure 9I-J).

4) As raised by reviewer 2, it is imperative to show that the synergy between BDQ and PZA on the BDQ-resistant mutant does not occur in broth.

We performed the suggested experiment. As shown in Figure 4B, we found no synergy between BDQ and PZA on the BDQ-resistant mutant cultivated in Middlebrook 7H9 liquid medium.

- A MIC for BDQ on the resistant mutant should be reported.

This has been added in the revised version. We determined the MIC₉₉ (defined as the concentration required to prevent 99% growth) of both wild-type and BDQ-resistant MTB by the colorimetric resazurin microtiter assay (Palomino et al., 2002). The MIC for susceptible MTB was 0.07 µg/mL, a value similar to previously published studies (Andries et al., 2005), and the MIC₉₉ of the BDQr-MTB was 36 µg/mL.

- Figure 1—figure supplement 2B is missing information about what concentration of BDQ was used.

This has been corrected in the revised version.

- Reviewer 2 raised a number of questions about the data in Figure 1—figure supplement 2C that should be addressed.

We thank the reviewer for the points raised about the experiment displayed in Figure 1—figure supplement 2C, which could question the virulence of the both wild-type and BDQ-resistant MTB. In the former version of our paper, we only quantified the number of intracellular bacteria, excluding the extracellular bacilli released from necrotic cells. This gave the wrong impression that bacteria do not multiply within macrophages. In the revised manuscript we provide new data from CFUs and confocal microscopy showing that wild-type and BDQ-resistant MTB are both virulent and replicate inside macrophages (Figure supplement 2C-E).

- As raised by reviewer 3, the proposed mechanism connecting BDQ to calcium and TFEB activation is only weakly supported by data. The authors should consider toning down their conclusions related to this part of the paper.

The calcium chelation experiments were performed with BAPTA-AM, not BAPTA (please see our comments to reviewer #3). We would like to apologize for this error and have corrected it in the revised manuscript.

Reviewer #1:

[…]

1) Transcriptional profiling is performed on both MTB-infected and in uninfected cells. It is claimed that the effects of BDQ are "similar" in both situations. However, the way the data are presented makes it hard to see how similar or different the effects of BDQ are +/- MTB. Could the authors present a heat-map (indicating key genes, particularly the lysosomal signature) with all four conditions (+/- MTB, +/- BDQ) in a single analysis/panel? If there are significant differences, it might be relevant to comment on these.

We thank the reviewer for his positive feedback and for fair suggestions. In the revised manuscript, we present a new analysis (Figure 1). MTB infection induces important transcriptome remodeling (7,093 genes were differentially expressed upon infection with BDQr-MTB) so it is not surprising that the expression of more genes were altered after BDQ treatment of MTB-infected macrophages than uninfected macrophages. In the new version, we comment on these and present functional annotation clustering of the corresponding genes (Figure 1A-B). In the former version of our manuscript, we wrote that “We observed similar results with uninfected Mφs-treated with BDQ […] indicating that the effect of BDQ is not dependent on MTB infection”. We were principally referring here to the lysosome pathway and lipid metabolism. 452 genes were differentially expressed in both uninfected- and MTB infected cells. Classification of these genes using gene-set enrichment analysis shows an enrichment for genes associated with lysosome (Figure 1C-D). The expression of 70% of the lysosomal genes found to be differentially expressed in BDQr-MTB-infected cells, were also regulated in naïve cells upon BDQ treatment (Figure 3A). As suggested, we have added a first heat-map of the genes most up- and down-regulated in naïve- and MTB-infected cells treated with BDQ (Figure 1D), and a second heat-map focusing on the expression of lysosomal genes in all four conditions +/- MTB +/- BDQ (Figure 3A).

2) A weakness of the manuscript is that the direct host-cell target of BDQ is not identified. Thus, the mechanism leading to TFEB induction is unclear. However, I tend to feel that identification of the target of BDQ is beyond what could be reasonably asked of this study. However, perhaps the authors could at least briefly discuss what the host cell target of BDQ might be and acknowledge that they have not identified it. In addition, the evidence that the effects of BDQ require TFEB is only indirect. It would thus be desirable, if feasible, for the authors to show, e.g., with TFEB knock out cells, that the effects of BDQ on lysosome-related gene induction does require TFEB.

Finding the direct host-cell target of BDQ would have been a major breakthrough, but unfortunately, we were unable to identify the target. However, it is possible that BDQ could modulate the host response by interfering with ion homeostasis inside the lysosome. Greenwood et al., recently showed that BDQ accumulated primarily in host cell lipid droplets (Greenwood et al., 2019). It is tempting to speculate that these droplets are actually lysosomes. Due to its high hydrophobicity, BDQ might accumulate in lysosomal membranes and thereby change the transmembrane ion permeability. In agreement with this hypothesis, it has been shown that BDQ can accumulate at the lipid membrane of liposomes and act as a H⁺/K⁺ ionophore (Hards et al., 2018). The ensuing lysosomal stress could then facilitate the dissociation of mTOR from the lysosomal membrane (Plescher et al., 2015) and the activation of TFEB. Interestingly, BDQ increases the expression of mucolipin 1 gene (mcoln1, Figure 3A and Figure 3—figure supplement 2A). Recently, Medina et al. showed that lysosomal Ca²⁺ is released through MCOLN1 and activates calcineurin, which binds and dephosphorylates TFEB (Medina et al., 2015) (Medina et al., 2015). These two hypotheses are now mentioned in the Discussion section of our paper.

In the revised manuscript, we present additional experiments, including silencing of TFEB expression using siRNA and chromatin purification with anti-TFEB antibody. The results show that the effects of BDQ on the induction of genes linked to lysosomes indeed require TFEB (Figure 9 and Figure 9—figure supplement 1).

Reviewer #2:

[…]

1) It seems likely to me that the transcriptional (Figure 1) and functional effects (Figures 2- of BDQ on infected macrophages is due to indirect effects on bacterial physiology. The use of the BDQ-resistant mutant is an important experiment, but I have doubts that the concentration used for the macrophage infection studies (5µg/mL) does not have any effect on the bacteria. This concentration apparently inhibits the growth of the resistant mutant Figure S1C inside macrophages, and this could be due to either activation of antimicrobial host effects or partial inhibition of bacterial growth. It is not clear what effects this concentration has on the bacteria, the mutant was selected on 0.3 µg/mL BDQ, and I could not find an MIC stated in the manuscript. The growth curve of bacteria grown in axenic culture +/- BDQ (panel B) is missing any information about the concentration of BDQ used, and there is no direct comparison of the growth kinetics between wild-type and mutant bacteria at 5 µg/mL. Moreover, the assumption is that treatment with 5µg/mL of BDQ doesn't change the bacteria at all, and thus any differences must be due to host modulation, yet I think there are multiple reasons to doubt this.

We thank the reviewer 2 for raising the question to what extent is the MTB mutant resistant to BDQ (BDQr-MTB). We first determined the MIC₉₉ (defined as the concentration required to prevent 99% growth) of both wild-type- and BDQ-resistant MTB by the colorimetric resazurin microtiter assay (Palomino et al., 2002). The MIC for susceptible MTB was 0.07 µg/mL, a value similar to previously published study (Andries et al., 2005), while the MIC₉₉ of the BDQr-MTB was 36 µg/mL. The CFUs and the growth curves of both WT and BDQr-MTB, with and without different concentrations of BDQ are now included in the manuscript (Figure 1—figure supplement 2B-C). Unlike drug-susceptible MTB, BDQrMTB continues to growth in the presence of 5 µg/mL BDQ, and we observed a slight decrease in growth only after 7 days of treatment. In our paper, BDQ is used at a maximum concentration of 5 µg/mL during 48 h and 1 µg/mL during 5 days. At these concentrations, we observed neither bacterial killing nor a decrease of bacterial growth in liquid medium (Figure 1—Figure supplement 2B-C). We also show that the bacillary load of resistant MTB inside macrophages is the same in untreated cells as in cells after 18 h of BDQ treatment (the time point chosen for the RNAseq experiment). In contrast, in the same experiment using BDQ-susceptible MTB there is a 70% decrease in the bacillary load (Figure 1—figure supplement 2C).

We agree that we cannot exclude the possibility that BDQ may affect the metabolism of BDQr-MTB, but it seems unlikely that this would explain the macrophage resistance to bacterial infection upon BDQ treatment. First, the effect of BDQ can be inhibited by 3-MA and BAF, whereas these two autophagy inhibitors have no effect on untreated cells (Figure 6F). These results suggest a host-targeted effect of BDQ. The regulation of the lysosomal pathway by BDQ seen in MTB infected macrophages was also observed in naïve macrophages and in cells stimulated with heat-inactivated MTB or with TLR agonists LPS and Pam3CSK4 (Figure 3—figure supplement 2). These changes were not observed in cells treated with the first line anti-TB drugs ethambutol (EMB), isoniazid (INH), pyrazinamide (PZA) and rifampin (RIF) (new transcriptome analysis presented in Figure 5A-B-C and Figure 6—figure supplement 2, and LysoTracker staining analyzed by flow cytometry Figure 5D).

If the lysis of a few bacteria by BDQ induces macrophage activation, we would expect a greater response when the cells are stimulated by heat-killed bacteria or drug-susceptible MTB than was observed with BDQr-MTB. Again, as shown in Figure 3—figure supplement 2B, we did not observe any differences in the intensity of LysoTracker staining. Similar results have also been obtained when comparing the expression level of lysosomal genes.

Overall, our results strongly argue for a host-dependent effect of BDQ.

2) There are a number of other worrisome details about the experiment displayed in Figure S1C. First, it is curious that neither the wild-type nor resistant bacteria grew very well between the 18hour and 5day time points in the absence of BDQ. It appears they double at best, but much more would be expected. Second, the BDQ-resistant mutant didn't grow at all over this 4+ day time course in the presence of 5µg/mL BDQ. It seems unlikely to me that this effects of BDQ are solely due to host activation. Third, why are the CFU levels of the mutant bacteria so much higher than the wild-type at 18hours?

We thank the reviewer for the points raised about the experiment displayed in the former Figure 1—figure supplement 1C, which could question the virulence of both wild-type and BDQ-resistant MTB. In the former version of our paper, we only quantified the number of intracellular bacteria, excluding the extracellular bacillus released from necrotic cells. This gave the erroneous impression that the bacteria did not multiply inside macrophages. In the revised manuscript, we provide new data from CFUs and confocal microscopy showing that wild-type and BDQ-resistant MTB are virulent and replicate within macrophages.

3) It seems important to demonstrate in all of their bacterial infections, including Figure 4, that the macrophage monolayers are intact and of equal density amongst all the treatments. Indeed, while the authors test the effects of BDQ on macrophage viability in the absence of bacteria, it seems important to assess its effect during the heightened inflammatory response during infection.

The viability of naïve and infected macrophages, treated with the different molecules used in this study, were assessed using the MTT or LDH assay. As shown in Figure 1—figure supplement 1, we noted no toxicity of the compounds. As expected, we only observe a decrease in cell viability of MTB-infected macrophages after 48 h of infection.

4) I believe that the results with the host effects of BDQ would be more conclusive if they could show that treatment with different antibiotics gives rise to different results.

In the initial version of our manuscript, we only tested whether first-line anti-TB drugs could induce an increase in LysoTracker staining. We agree that further analysis of the effects of other drugs on the macrophages response would be more conclusive. We therefore used RNAseq to characterize the genome-wide gene expression profiles of naïve cells and cells stimulated with heat-killed MTB, and treated with ethambutol (EMB), isoniazid (INH), pyrazinamide (PZA) or rifampicin (RIF). We chose heat-killed MTB instead of drug-resistant MTB for the following reasons:

(i) We cannot work with multidrug-resistant MTB in our BSL3 facility, and the generation of resistant mutants for each antibiotic would not have been reasonable in terms of time to resubmit a revised version of our manuscript.

(ii) Most of the genes upon that were differentially expressed after MTB infection are also modulated upon stimulation with heat-killed MTB ((Pacis et al., 2015) and Figure 6—figure supplement 1).

(iii) The activation of the lysosomal pathway by BDQ does not require infection with live MTB, as most of the lysosomal genes are also differentially expressed in naïve cells exposed to BDQ (Figure 3A).

Following treatment, only RIF and PZA significantly modulate gene expression in macrophages. 556 and 752 genes were differentially expressed in cells stimulated with heat-killed bacteria and exposed to PZA and RIF respectively (Figure 5A). We classified these genes by performing gene-set enrichment analysis and confirmed that the genes in the lysosomal pathway were not induced upon RIF or PZA treatment. The expression of only 2 genes belonging to this pathway was upregulated by RIF, and only one by PZA, compared to 46 whose expression was modulated by BDQ (FDR<5%, absLogFC>0.1, Figure 5B-C and Figure 5—figure supplement 2).

5) The authors also perform treatment of macrophages with BDQ in the absence of infection, and claim that similar pathways are activated, inferring that these effects are independent of the presence of bacteria. Unfortunately, these comparisons were evaluated at the level of GO pathway annotation, and I believe strongly that if BDQ is directly activating gene expression, perhaps via PTFB, that the comparison should be performed at the gene level, i.e. as done in Figure 2A.

In the revised manuscript, we present new data comparing the gene expression level in all four conditions (+/- BDQ and +/- MTB), and as suggested by reviewer 2, we show a heat-map with the differentially expressed genes associated with the lysosome pathway (Figure 3A). The expression of 70% of the lysosomal pathway genes, differentially expressed in MTB-infected macrophage, were also upregulated in naïve cells upon BDQ treatment.

We also performed additional experiments to demonstrate that the effects of BDQ on host gene expression were independent of infection with live MTB. Briefly, cells were untreated or stimulated with LPS (TLR4 agonist), Pam3CSK4 (TLR1/2 agonist), heat-killed bacteria, drug-susceptible MTB or BDQr-MTB, and treated with BDQ. After 18 h, RNA was collected, and RT-qPCR performed on a panel of lysosomal genes. We also analyzed the intensity of the LysoTracker staining using flow cytometry (Figure 3—figure supplement 1). The results clearly show that the effect of BDQ on lysosome biogenesis/activation is independent of MTB infection and of Toll-like receptor stimulation.

6) I find the conclusion drawn from the apparent synergy btw PZA and BDQ is misleading for two reasons. First of all, given the residual sensitivity of the resistant mutant to high levels of BDQ, it seems imperative that this synergy be tested in culture. Second, given the model that BDQ activates host antimicrobial activities, I expect that BDQ would also synergize with other antibiotics besides PZA.

We tested the synergy between BDQ and PZA in culture liquid medium using increasing concentrations of PZA (from 0 to 400 µg/mL) with or without BDQ (1 µg/mL). At this BDQ concentration, the growth of BDQr-MTB is not affected by BDQ over a period of 12 days (Figure 1—figure supplement 2B) and we observed an increase of the DQ-BSA staining in macrophages (Figure 3F). We observed no synergy between BDQ and PZA in culture liquid medium, suggesting a hostdependent mechanism. We next tested whether BDQ would synergize with the other first-line anti-TB drugs in liquid culture or in BDQr-MTB-infected macrophages and found that BDQ did not potentiate the activity of INH, RIF and EMB in either case. While we cannot exclude the possibility that BDQ may have additive effects with other anti-TB drugs, as has been described (Ibrahim et al., 2007), we saw no evidence of synergism with any of the first-line agents (Figure 4—figure supplement 1).

7) Much of the lysosome results in Figure 2 is complicated by the mild sensitivity of the resistant strain to BDQ. The one experiment that is interesting is the no-infection in panel G, but there is only a correlation, there is no evidence provided that this is the host effect of BDQ.

In the revised version, we add a new supplemental figure describing the lysosomal regulation in uninfected cells (Figure 3—figure supplement 1). As we mention in the text, lysosome biogenesis/activation is increased in both naïve and BDQr-MTB-infected macrophages upon BDQ treatment.

Reviewer #3:

[…]

1) The proposed mechanism of enhancement of bactericidal activity is through TFEB, but the only evidence presented is correlative (TFEB immunofluorescence showing TFEB in the nucleus). This must be directly tested using loss-of-function approaches.

To study the role of TFEB in the enhancement of bactericidal activity upon BDQ treatment in more depth, we inactivated TFEB expression in human monocyte-derived macrophages using siRNA-mediated gene silencing (Troegeler et al., 2014). After treating the cells with BDQ, we performed RTqPCR on a panel of lysosomal genes and analyzed the intensity of the LysoTracker staining using flow cytometry. In TFEB-silenced cells, BDQ does not increase LysoTracker staining (Figure 9G), nor the expression of trpm2, scarb1, atp6v0a1 or mcoln1 (Figure 9H). We also infected macrophages with S. aureus and MTB and showed that TFEB activation is required for better control of both pathogens upon BDQ treatment (Figure 9I-J). In addition, we present a chromatin immunoprecipitation assay using an anti-TFEB antibody (please see the comment below).

2) Similarly, the implication from the RNAseq, LysoTracker, and TFEB IF is that TFEB upregulates the genes responsible for lysosomal biogenesis and autophagy, thus increasing bacterial clearance. The inducible recruitment of TFEB to relevant promoters must be directly tested.

We evaluated the recruitment of TFEB to the promoters of some lysosomal genes. Briefly, naïve- and heat-killed-stimulated cells treated with BDQ were lysed, and chromatin was harvested and fragmented using sonication. The chromatin was then subjected to immunoprecipitation using an antiTFEB antibody. The isolated DNA was then analyzed by RT-qPCR with primers specific for gla and mcoln1 promoters. As expected, BDQ induces the recruitment of TFEB to these promoters (Figure 8—figure supplement 1).

3) The mechanism connecting BDQ to calcium release and TFEB activation is not shown. Neither is the requirement for calcineurin.

Unfortunately, we were unable to identify the mechanism connecting BDQ to TFEB activation.

Greenwood et al., recently showed that BDQ accumulated primarily in host cell lipid droplets (Greenwood et al., 2019). It is tempting to speculate that these droplets are actually lysosomes. Due to its high hydrophobicity, BDQ might accumulate in lysosomal membranes and thereby change transmembrane ion permeability. In agreement with this hypothesis, it has been shown that BDQ can accumulate at the lipid membrane of liposomes and act as a H⁺/K⁺ ionophore (Hards et al., 2018). The ensuing lysosomal stress could then facilitate the dissociation of mTOR from the lysosomal membrane (Plescher et al., 2015) and the activation of TFEB. Interestingly, BDQ increases the expression of mucolipin 1 gene (mcoln1, Figure 3A and Figure 3—figure supplement 2A). Recently, Medina et al. showed that lysosomal Ca²⁺ is released through MCOLN1 and activates calcineurin, which binds and dephosphorylates TFEB (Medina et al., 2015). These two hypotheses are now mentioned in the Discussion section.

4) Although the synergistic effect of BDQ with PZA seems to approach 1 log (~10% survival) killing of M. tuberculosis, the effect size of BDQ alone (Figure 3F) is quite small. This raises questions of how significant these observations are to TB. This is important, since the whole setup is to find new therapeutic agents to control MDR TB.

In the former Figure 3F, we used a low concentration of BDQ (1 µg/mL) in order to discriminate between additive and synergic effect. At higher concentration of BDQ (5 µg/mL, the concentration detected in the plasma of TB patients treated with BDQ (Andries et al., 2005)), we observed a 50% decrease in the number viable bacteria (now in Figure 6F).

5) The authors use BAPTA (extracellular) to eliminate calcium and test the effect of loss of TFEB activation on bacterial CFU. This is done with S. aureus (not TB). Also, if they really are using BAPTA, they would not be inhibiting intracellular calcium release but import from extracellular stores, changing the interpretation of the experiment.

The calcium chelation experiments were performed with BAPTA-AM, not BAPTA. We would like to apologize for this error and have corrected it in the revised manuscript. Unlike BAPTA, which is a membrane-impermeable calcium chelator that binds extracellular calcium ions, BAPTA-AM is a cell permeable analog of BAPTA that binds calcium only after the acetoxymethyl group is removed by cytoplasmic esterases (Tymianski et al., 1994). It is highly selective for Ca²⁺ over Mg²⁺, and It is commonly used to evaluate the role of intracellular Ca²⁺ in cell signaling.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Reviewer #1:

[…]

My only query relates to new data (Figure 7) showing that human macrophages induce nitrite, which the authors claim is derived from host NO. As far as I am aware, there has been considerable difficulty demonstrating induction of NO by human macrophages in culture. In some cases, the NO2- observed has actually been produced by the bacteria themselves (see Bussel, Zhang and Nathan, 2013). The authors may want to take this into account.

We thank the reviewer for his/her positive comments and suggestions, which have enabled us to improve our manuscript significantly. We agree the role of NO in the inhibition of M. tuberculosis growth is controversial in human. However, in Figure 7D, we quantified NO2- in uninfected cells upon BDQ treatment, which excludes nitric oxide produced by bacteria.

Author details

1. Alexandre Giraud-Gatineau

   1. Unit for Integrated Mycobacterial Pathogenomics, CNRS UMR 3525, Institut Pasteur, Paris, France
   2. Université Paris Diderot, Sorbonne Paris Cité, Cellule Pasteur, Paris, France

   Contribution

   Conceptualization, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

   Competing interests

   No competing interests declared

2. Juan Manuel Coya

   Mycobacterial Genetics Unit, Institut Pasteur, Paris, France

   Contribution

   Formal analysis, Validation, Investigation, Visualization, Methodology

   Contributed equally with

   Alexandra Maure and Anne Biton

   Competing interests

   No competing interests declared

3. Alexandra Maure

   1. Unit for Integrated Mycobacterial Pathogenomics, CNRS UMR 3525, Institut Pasteur, Paris, France
   2. Université Paris Diderot, Sorbonne Paris Cité, Cellule Pasteur, Paris, France

   Contribution

   Formal analysis, Investigation

   Contributed equally with

   Juan Manuel Coya and Anne Biton

   Competing interests

   No competing interests declared

4. Anne Biton

   Bioinformatics and Biostatistics, Department of Computational Biology, USR 3756 CNRS, Institut Pasteur, Paris, France

   Contribution

   Formal analysis, Investigation, Methodology

   Contributed equally with

   Juan Manuel Coya and Alexandra Maure

   Competing interests

   No competing interests declared

5. Michael Thomson

   MRC Centre for Molecular Bacteriology and Infection, Department of Life Sciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom

   Contribution

   Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

6. Elliott M Bernard

   Host-Pathogen Interactions in Tuberculosis Laboratory, The Francis Crick Institute, London, United Kingdom

   Contribution

   Formal analysis, Investigation

   Competing interests

   No competing interests declared

7. Jade Marrec

   Mycobacterial Genetics Unit, Institut Pasteur, Paris, France

   Contribution

   Formal analysis, Investigation

   Competing interests

   No competing interests declared

8. Maximiliano G Gutierrez

   Host-Pathogen Interactions in Tuberculosis Laboratory, The Francis Crick Institute, London, United Kingdom

   Contribution

   Funding acquisition, Investigation

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-3199-0337

9. Gérald Larrouy-Maumus

   MRC Centre for Molecular Bacteriology and Infection, Department of Life Sciences, Faculty of Natural Sciences, Imperial College London, London, United Kingdom

   Contribution

   Funding acquisition, Investigation

   Competing interests

   No competing interests declared

10. Roland Brosch

    Unit for Integrated Mycobacterial Pathogenomics, CNRS UMR 3525, Institut Pasteur, Paris, France

    Contribution

    Supervision, Funding acquisition

    Contributed equally with

    Brigitte Gicquel

    Competing interests

    No competing interests declared

11. Brigitte Gicquel

    1. Mycobacterial Genetics Unit, Institut Pasteur, Paris, France
    2. Department of Tuberculosis Control and Prevention, Shenzhen Nanshan Center for Chronic Disease Control, Shenzhen, China

    Contribution

    Supervision, Funding acquisition

    Contributed equally with

    Roland Brosch

    Competing interests

    No competing interests declared

12. Ludovic Tailleux

    1. Unit for Integrated Mycobacterial Pathogenomics, CNRS UMR 3525, Institut Pasteur, Paris, France
    2. Mycobacterial Genetics Unit, Institut Pasteur, Paris, France

    Contribution

    Conceptualization, Formal analysis, Validation, Investigation, Visualization, Methodology, Supervision, Funding acquisition, Writing - original draft, Project administration, Writing - review and editing

    For correspondence

    ludovic.tailleux@pasteur.fr

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0003-3931-1052

• 3,898

  Page views

• 518

  Downloads

• 44

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.