Immunometabolic circuits in host defense

Mihai G. Netea

IY31CH17-OGarra ARI 15 February 2013 4:46

Draining Lung lymph node

M. tuberculosis

2 IL-12(p40)2/IL-12p70 1c DC CCL19/CCL21 DC Alveolar 8–12 days 1a macrophage Uptake IL-12p70 PGE2 E Antimicrobial peptides, ferocytosis IL-1α/β, TNF-α, Naive T cell Apoptosis IL-12p40, IL-6, Lipoxins chemokines (LXA4) 3

Necrosis Uninfected 4 1b macrophage Th1 cell

Antimicrobial peptides CXCL10/CCL19/CCL21 (e.g., cathelicidins), Necrosis 14–17 days chemokines, IL-1β Neutrophil

γ Th1 cells

IL-12p70 IFN-

iNOS 5 IL-12p40 TNF-α Macrophage DC

M. tuberculosis dissemination? O’Garra et al, Ann Rev Imm 2013

Figure 4 The cellular immune response to M. tuberculosis. Following aerosol infection with M. tuberculosis, resident lung alveolar macrophages by Radboud Universiteit Nijmegen on 09/06/13. For personal use only. (1a), neutrophils (1b)andlungDCs(1c) can become infected, leading to the production and secretion of antimicrobial peptides, cytokines, and chemokines. The balance of lipid mediators, such as prostaglandin E2 (proapoptotic) or lipoxin (LX) A4 (pronecrotic),

Annu. Rev. Immunol. 2013.31:475-527. Downloaded from www.annualreviews.org within infected macrophages plays a major role in determining downstream pathways leading to the induction of either apoptosis or necrosis. Infected apoptotic cells can be taken up by resident lung DCs or efferocytosed by uninfected lung macrophages (1c). M. tuberculosis–infected DCs migrate to the local lung-draining lymph nodes by 8–12 days post infection. DCs migrate to the lymph nodes under the influence of IL-12(p40)2 and IL-12p70 and that of the chemokines CCL19 and CCL21 (2), to drive naive T cell differentiation toward a Th1 phenotype (3). Protective antigen-specific Th1 cells migrate back to the lungs in a chemokine-dependent manner 14–17 days after the point of initial infection/exposure (4) and produce IFN-γ, leading to macrophage activation, cytokine production, the induction of microbicidal factors including iNOS (5), and bacterial control.

nodes and occurs earlier in resistant C57BL/6 dissemination of M. tuberculosis may aid in mice than in susceptible C3H mice, result- the initiation of an appropriate and timely ing in an earlier immune response in the adaptive immune response, although this re- C57BL/6 mice (134). This finding suggests sponse may be under strict host genetic control that instead of only spreading infection, early (134).

488 O’Garra et al. Immune cell TLR stimulation glucose glucose autophagy

glucose HIF-1α mTOR Akt

glycolysis

PDH Oxidative 2ATP pyruvate Lipid synthesis phosphorylation O2 ACC present AcetylCoA O2 not Krebs cycle present NAD+ HIF-1α Lactic acid Electron transport Histone acetyl NADH transferase 36 ATP Active cells Naïve cells RESEARCH ARTICLE

These phenomena were not observed in our experiments either at early CFU in lung and spleen when compared with mice that received ETH time points (4 hours) or after prolonged MET exposure (24 and 48 hours), alone (Fig. 2, D and E, and fig. S8, C and D). The efficacy of MET was indicating that the drug’smechanismofactionisnotcausingcelldeath further evaluated in the chronic model of Mtb infection. In this experi- of host cells (fig. S7, A to D). ment, mice treated with MET or INH + MET starting from day 42 after infection showed decreased bacillary load in the lungs compared with MET enhances the efficacy of conventional anti-TB drugs untreated or INH-treated mice, respectively (Fig. 2F). Together, these We next evaluated the in vivo efficacy of MET treatment in a mouse experiments suggest that MET represents a potential adjunct drug that model of acute and chronic TB (24). Mtb-infected mice were admin- can enhance Mtb clearance by conventional treatment. istered MET alone or MET in combination with either isoniazid (INH) or ethionamide (ETH), starting on day 7 or 42 after infection. In five dif- MET reduces TB-induced tissue pathology and enhances ferent acute model experiments, mice treated with MET (500 mg/kg) immune response alone had reduced bacillary load in both lung and spleen (Fig. 2, A To significantly improve current TB therapy, anti-Mtb drugs should to E, and fig. S8, A and C). This dose is equivalent to a MET dose of also promote resolution of tissue pathology in addition to accelerating 2430 mg/day for a 60-kg human (http://www.naturalhealthresearch. bacillary clearance (4). Lungs and spleens from Mtb-infected mice treated org/extrapolation-of-animal-dose-to-human/) (25) and lower than with MET were smaller than those of untreated mice at 35 days after the maximum daily dose for MET therapy in diabetic patients infection (Fig. 3, A and B). As expected, mice treated with the conven- (3000 mg/day; http://www.medicines.org.uk/emc/medicine/20952/spc; tional anti-TB drug INH or ETH exhibited a clear reduction in organ http://www.australianprescriber.com/magazine/37/1/2/5). Furthermore, size and tissue lesions, and combination therapy of MET with INH or MET administration enhanced the efficacy of the conventional first- ETH further reduced tissue pathology. Histopathological evaluation line anti-TB drug INH, as demonstrated by decreased bacillary load in of the infected lungs of untreated control mice revealed diffuse coales- the lungs of mice cotreated with INH + MET when compared with mice cent lung lesions with large numbers of infiltrating macrophages and that received INH alone (Fig. 2C and fig. S8, A and B). Indeed, in some lymphocytes and scatteredintracellularacid-fastbacilli(AFB)(Fig.3C). experiments, we were unable to detect any CFU in the lungs of mice In contrast, MET treatment was associated with reduced numbers of that received combined INH + MET therapy (table S3). We next eval- AFB and increased lymphocyte infiltration of the infected tissues uated the efficacy of MET adjunctive therapy when used in combi- (Fig. 3C), which has previously been linked with improved Mtb control nation with the second-line anti-TB drug ETH. In an acute model, in mice (26–28). Mice treated with INH alone had few residual lesions on November 24, 2014 Mtb-infectedMetformin mice cotreated and withhost-directed ETH + MET exhibited therapy decreased in tuberculosisin the lungs with areas of only increased cellularity. No granulomas were observed in the lungs of mice treated with INH + MET, with some areas of the lungs appearing completely normal. Mor- phometric analysis of the histological sec- tions indicated that the percentage of total area of lung tissue involved in TB

pathology at 35 days after infection was stm.sciencemag.org reduced in MET-treated mice compared with untreated control animals (3.8% ver- sus 9.9%, respectively; Fig. 3D) and that combination therapy with MET and INH further reduced areas of lung tissue dam- age compared to INH-alone treatment (0.13% versus 0.70%, respectively; Fig. Downloaded from 3D). We next evaluated MET effects on Thelpercell1(TH1) immune responses in lung tissues because TH1 immunity participates in controlling Mtb infec- tion (29). MET-treated Mtb-infected mice showed a trend of larger CD4+ and RESEARCH ARTICLE CD8+ T cell numbers in the lung com- Fig. 2. Efficacy of MET monotherapy and combinationTUBERCULOSIS therapy in a mouse model of TB. (A) Mtb-infected pared to untreated mice (fig. S9, A to C). mice were treated with MET [250 mg/kg (M250) orMetformin 500 mg/kg as (M500)] adjunct starting antituberculosis 7 days after infection. therapy Bacillary In Mtb-infected mice, MET treatment was loads (CFU) were enumerated in the lungs on daysAmit Singhal, 1, 7, 121,* Liu and Jie,1† Pavanish 35 after Kumar, infection.1† Gan Suay UNT, Hong,2 untreatedMelvin Khee-Shing infected Leow,3,4 associated with an increased percent- Bhairav Paleja,1 Liana Tsenova,5,6 Natalia Kurepina,5 Jinmiao Chen,1 Francesca Zolezzi,1 mice. (B)SpleenbacillaryloadinMtb-infected mice treated5 with MET as1,7 described in2 (A). (C)EfficacyofINH5,8 Barry Kreiswirth, Michael Poidinger, Cynthia Chee, Gilla Kaplan, age and number of mycobacteria-specific Yee Tang Wang,2 Gennaro De Libero1,9* monotherapy (5 mg/kg, I5) and INH + M500 therapy in the lungs of Mtb-infected mice. (D)EfficacyofETH interferon-g (IFN-g)–secreting CD8+ Tcells monotherapy (15 mg/kg, E15) and ETH + M500 therapyThe global burden in the of tuberculosis lungs (TB) of morbidityMtb-infected and mortality mice. remains immense. (E)Spleenbacillary A potential new approach to TB therapy is to augment protective host immune responses. We report that the antidiabetic drug metformin (MET)compared with untreated control animals load in Mtb-infected mice treated as describedreduces in (D). the intracellular (F) Mtb growth-infected of Mycobacterium mice tuberculosis were treated(Mtb) in an AMPK daily (adenosine with monophosphate MET – activated protein kinase)–dependent manner. MET controls the growth of drug-resistant Mtb strains, increases pro-(Fig. 3, E and F). The number, but not the (250 mg/kg, M250), INH (10 mg/kg, I10), and I10duction + M250 of mitochondrial in drinking reactive oxygen water species, and starting facilitates phagosome-lysosome 42 days after fusion. infection. In Mtb-infected mice, + use of MET ameliorated lung pathology, reduced chronic inflammation, and enhanced thespecificimmuneresponsepercentage, of IFN-g–secreting CD4 Tcells Bacillary loads (CFU) were enumerated in theand lungs the efficacy on of conventionaldays 1, TB 21, drugs. 42, Moreover, and in two100 separate after human infection. cohorts, MET treatment In these was associated experiments, each group per time point consistedwith improved of four control of toMtb nineinfection mice. and decreased Data disease are severity. expressed Collectively, these as data means indicate that ± MET isalso a showed an increasing trend in the lungs promising candidate host-adjunctive therapy for improving the effective treatment of TB. SEM. P values are provided in table S12, two-tailed Student’s t test. *P < 0.05; **P < 0.01; ***P < 0.001. of Mtb-infected mice upon MET therapy

INTRODUCTION (ROS) and reactive nitrogen species (RNS), as well as the capacity Mycobacterium tuberculosis (Mtb)istheetiologicalagentoftuberculosis to destroy intracellular pathogens using the phagosomal machinery (TB), a disease that continues to be a leading killer worldwide, with or autophagy pathway. Autophagy is required for the effective control an estimated global mortalitywww. of 1.43ScienceTranslationalMedicine million people every year (1). of intracellular pathogens.org (11, 12 19)andisregulatedbymammaliantarget November 2014 Vol 6 Issue 263 263ra159 3 Although current treatment of drug-susceptible TB is effective, the of rapamycin (mTOR) complex 1 (13), a serine/threonine kinase, and therapy regimen is lengthy (6 to 9 months), and problems with drug adenosine monophosphate–activated protein kinase (AMPK) (13, 14). on November 24, 2014 toxicity and increasing multidrug resistance (MDR) urge the discovery/ Accordingly, perturbations in the autophagy network and AMPK development of new drugs and therapeutic strategies (2, 3). The global signaling have previously been associated with Mtb virulence (15, 16). need for new effective therapies has led to a resurgence in efforts to We therefore screened the capacity of various U.S. Food and Drug Ad- identify additional anti-Mtb agents, several of which are now being eval- ministration (FDA)–approved mTOR-independent autophagy activators uated in clinical trials (4). However, conventional pathogen-targeted (17)andAMPKmodulatorsfortheirability to control intracellular myco- strategies suffer from the serious disadvantage of fostering microbial bacterial growth. We discovered that the AMPK-activating antidiabetic resistance. To circumvent this problem, a new paradigm in drug discov- drug metformin (MET; 1,1-dimethylbiguanide) inhibits the intracellular

ery that involves therapeutic modulation of host cell responses to improve growth of Mtb,restrictsdiseaseimmunopathology,andenhancesthe stm.sciencemag.org pathogen eradication (5–7)hasemerged.These“host-targeted” adjunct efficacy of conventional anti-TB drugs. therapeutic strategies are less likely to engender microbial resistance than direct targeting of the pathogen with conventional drugs (6). We therefore tested whether existing approved drugs with defined effects on RESULTS host cell functions could be repurposed for effective and durable anti-TB therapy. The validation of this approach would have the advantage of re- MET restricts mycobacterial growth by inducing quiring shorter clinical trials than those usually needed for new drugs (4). mitochondrial ROS production Downloaded from An effective host immune system is a crucial factor for both the A total of 13 autophagy and AMPK-activating drugs were tested for control of Mtb growth and its containment as latent TB infection (LTBI). the ability to control the intracellular growth of M. bovis bacillus The success of Mtb in infecting the host cells and maintaining long-term Calmette-Guérin (BCG) in the human monocytic cell line THP-1 using persistent infection is associated with the ability of bacilli to evade host a colony-forming unit (CFU) assay. Eight of these compounds sig- innate as well as adaptive immune responses (8–10). The host cell innate nificantly inhibited the intracellular growth of BCG (fig. S1, A and B, antimicrobial arsenal includes production of reactive oxygen species and table S1). None of these eight compounds displayed direct anti- mycobacterial effects in an in vitro minimum inhibitory concentration (MIC) assay (fig. S1, C to E), suggesting that these compounds restricted 1Singapore Immunology Network, Agency for Science, Technology and Research (A*STAR), Singapore 138648, Singapore. 2Tuberculosis Control Unit, Tan Tock Seng Hospital, BCG growth via effects on host cell–BCG interactions with minimal di- Singapore 308089, Singapore. 3Department of Endocrinology, Tan Tock Seng Hospital, rect effect, if any. In these experiments, rapamycin, a known autoph- Singapore 308433, Singapore. 4Singapore Institute for Clinical Sciences, A*STAR, Singapore agy inducer, was used as a positive control. In subsequent studies, we 117609, Singapore. 5Public Health Research Institute at New Jersey Medical School, Rutgers University, Newark, NJ 07103, USA. 6New York City College of Technology, Brooklyn, NY focused on evaluating the therapeuticpotentialoftheAMPK-activating 11201, USA. 7Department of Biological Sciences, National University of Singapore, agent MET because this drug is already being used extensively in the Singapore 117543, Singapore. 8Bill & Melinda Gates Foundation, Seattle, WA 98109, USA. clinic for the treatment of type 2 diabetes mellitus (DM), and the effects 9 University Hospital Basel, University of Basel, Basel 4031, Switzerland. of the drug on cellular functions including AMPK signaling are known *Corresponding author. E-mail: [email protected] (A.S.); gennaro_ [email protected] (G.D.L.) (18, 19). In our experiments, MET treatment inhibited the growth of in- †These authors contributed equally to this work. tracellular BCG (fig. S2A) as well as the H37Rv strain of Mtb (Fig. 1A) in

www.ScienceTranslationalMedicine.org 19 November 2014 Vol 6 Issue 263 263ra159 1 Metformin and host-directed therapy in tuberculosis RESEARCH ARTICLE

flammation (40). MET treatment also promotes the expansion of Mtb- specific IFN-g–secreting T cells in the lungs of infected mice. The expansion of lung IFN-g–secreting CD8+ Tcellsinuninfectedmicein- dicates that MET has an impact on the lung immune cells indepen- dently of current infections. These latter effects are consistent with MET-induced expansion of CD8+ memory T cells (41), another known consequence of AMPK activation (42). Both CD4+ and cytotoxic T cells are key to controlling primary Mtb infection in human subjects (29, 43–45). Notably, our data also showed increased numbers of Mtb-specific T cells in MET-treated diabetic patients with LTBI. Fig. 5. Effect of MET in TBRESEARCH patients ARTICLE with DM as comorbidity. (A)Percent- ages of TB patients with DM as comorbidity showing detectable pulmonary MET therapy combined with standard TB treatment regimens was TUBERCULOSIS cavities determined by CXRMetformin at the time as of adjunct first visit antituberculosis to TB clinic. MET-receiving therapy associated with beneficial consequences on clinical outcomes in active

(M) and alternative drug–receiving1 (Non-M)1† TB + DM1† patients.2 *P =0.041;OR,3,4 Amit Singhal, * Liu Jie, Pavanish Kumar, Gan Suay Hong, Melvin Khee-Shing Leow, TB disease, as observed in the retrospective study presented here. Fur- Bhairav Paleja,1 Liana Tsenova,5,6 Natalia Kurepina,5 Jinmiao Chen,1 Francesca Zolezzi,1 0.6; 95% CI, 0.36 to 0.97. (B)MortalityamongMandNon-MTB+DMpatients5 1,7 2 5,8 Barry Kreiswirth, Michael Poidinger, Cynthia Chee, Gilla Kaplan, thermore, MET treatment exerted favorable effects on LTBI in diabetic during anti-TB therapy. *PYee=0.039;OR,0.29;CI,0.14to0.95. Tang Wang,2 Gennaro De Libero1,9* The global burden of tuberculosis (TB) morbidity and mortality remains immense. A potential new approach to TB patients. Several issues remain to be addressed before introduction of therapy is to augment protective host immune responses. We report that the antidiabetic drug metformin (MET) reduces the intracellular growth of Mycobacterium tuberculosis (Mtb) in an AMPK (adenosine monophosphate– MET in TB therapy. First, although our retrospective clinical data are activated protein kinase)–dependent manner. MET controls the growth of drug-resistant Mtb strains, increases pro- duction of mitochondrial reactive oxygen species, and facilitates phagosome-lysosome fusion. In Mtb-infected mice, use of MET ameliorated lung pathology, reduced chronic inflammation, and enhanced thespecificimmuneresponse very encouraging, they might be subjected to confounders, which are and the efficacy of conventional TB drugs. Moreover, in two separate human cohorts, MET treatment was associated with improved control of Mtb infection and decreased disease severity. Collectively, these data indicate that MET is a difficult to precisely identify. Second, the exact dose of MET to be used promising candidate host-adjunctive therapy for improving the effective treatment of TB. in the adjunctive therapy remains to be evaluated in TB patients. Third, INTRODUCTION (ROS) and reactive nitrogen species (RNS), as well as the capacity Mycobacterium tuberculosis (Mtb)istheetiologicalagentoftuberculosis to destroy intracellular pathogens using the phagosomalit has machinery to be investigated whether integration of MET with current ther- (TB), a disease that continues to be a leading killer worldwide, with or autophagy pathway. Autophagy is required for the effective control an estimated global mortality of 1.43 million people every year (1). of intracellular pathogens (11, 12)andisregulatedbymammaliantargetapeutic treatments may also improve prophylactic control of TB and Although current treatment of drug-susceptible TB is effective, the of rapamycin (mTOR) complex 1 (13), a serine/threonine kinase, and therapy regimen is lengthy (6 to 9 months), and problems with drug adenosine monophosphate–activated protein kinase (AMPK) (13, 14). on November 24, 2014 toxicity and increasing multidrug resistance (MDR) urge the discovery/ Accordingly, perturbations in the autophagyreduce network and AMPK duration of treatment. All theseissuesrequireaccurateprospec- development of new drugs and therapeutic strategies (2, 3). The global signaling have previously been associated with Mtb virulence (15, 16). need for new effective therapies has led to a resurgence in efforts to We therefore screened the capacity of various U.S.tive Food clinicaland Drug Ad- trials to properly test the efficacy of MET as adjunct TB ther- identify additional anti-Mtb agents, several of which are now being eval- ministration (FDA)–approved mTOR-independent autophagy activators uated in clinical trials (4). However, conventional pathogen-targeted (17)andAMPKmodulatorsfortheirability to controlapy intracellular in different myco- environmental conditions. strategies suffer from the serious disadvantage of fostering microbial bacterial growth. We discovered that the AMPK-activating antidiabetic resistance. To circumvent this problem, a new paradigm in drug discov- drug metformin (MET; 1,1-dimethylbiguanide) inhibits the intracellular

ery that involves therapeutic modulation of host cell responses to improve growth of Mtb,restrictsdiseaseimmunopathology,andenhancesthe stm.sciencemag.org pathogen eradication (5–7)hasemerged.These“host-targeted” adjunct efficacy of conventional anti-TB drugs. on November 24, 2014 Fig. 6. Reduced incidencetherapeutic of strategies LTBI are less in likely toMET-treated engender microbial resistance DM patients. (A) than direct targeting of the pathogen with conventional drugs (6). We Frequency of LTBI among DMtherefore patients tested whether existing receiving approved drugs with MET defined (M) effects on orRESULTS alternative drugs host cell functions could be repurposed for effective and durable anti-TB MATERIALS AND METHODS (Non-M), as determined bytherapy. T-SPOT The validation IFN- of thisg approtest.ach would The have the frequency advantage of re- MET of restricts T-SPOT mycobacterial posi- growth by inducing quiring shorter clinical trials than those usually needed for new drugs (4). mitochondrial ROS production Downloaded from tivity was reduced in the MET-treatedAn effective host immune group. system is a crucial *P = factor 0.041; for both the OR,A total 0.44; of 13 autophagy CI, 0.20 and AMPK-activating to drugs were tested for control of Mtb growth and its containment as latent TB infection (LTBI). the ability to control the intracellular growthStudy of M. bovis bacillus design 0.95, multiple logistic regressionThe success of Mtb test.in infecting (B the) host Scatterplot cells and maintaining long-term showingCalmette-Guérin the (BCG) number in the human monocytic cell line THP-1 using persistent infection is associated with the ability of bacilli to evade host a colony-forming unit (CFU) assay. Eight of these compounds sig- innate as well as adaptive immune responses (8–10). The host cell innate nificantly inhibited the intracellular growth ofThe BCG (fig. objective S1, A and B, of this study was to identify FDA-approved drugs that can of ELISpot forming units (SFU)antimicrobial in arsenal individual includes production DM of reactive patient oxygen species inand response table S1). None of these to eight the compounds displayed direct anti- mycobacterial effects in an in vitro minimum inhibitorybe repurposed concentration to modulate host immune system and capable of con- Mtb ESAT-6 and CFP-10 antigens in the blood as measured(MIC) by assay T-SPOT (fig. S1, C to E), IFN- suggestigng that these compounds restricted 1Singapore Immunology Network, Agency for Science, Technology and Research (A*STAR), Singapore 138648, Singapore. 2Tuberculosis Control Unit, Tan Tock Seng Hospital, BCG growth via effects on host cell–BCG interactions with minimal di- test. Vertical and horizontalSingapore do 308089,tted Singapore. lines3Department represent of Endocrinology, Tan Tock the Seng Hospital, cutoffrect effect, of if≥ any.5SFUper In these experiments, rapamycin,trolling a known autoph- the growth of intracellular Mtb.First,wescreened13autophagy Singapore 308433, Singapore. 4Singapore Institute for Clinical Sciences, A*STAR, Singapore agy inducer, was used as a positive control. In subsequent studies, we 5 117609, Singapore. 5Public Health Research Institute at New Jersey Medical School, Rutgers 2.5 × 10 cells. MET-treatedUniversity, (M) Newark, patients, NJ 07103, USA. 6Newn York=48;patientstreatedbyNon-MET City College of Technology, Brooklyn, NY focused on evaluating the therapeuticpotentialoftheAMPK-activatingand AMPK-activating drugs for their effect on the intracellular growth 11201, USA. 7Department of Biological Sciences, National University of Singapore, agent MET because this drug is already being used extensively in the 8

Singapore 117543, Singapore. Bill & Melinda Gates Foundation, Seattle, WA 98109, USA. clinic for the treatment of type 2 diabetes mellitus (DM), and the effects stm.sciencemag.org drugs (Non-M), n =14. 9 University Hospital Basel, University of Basel, Basel 4031, Switzerland. of the drug on cellular functions including AMPKof signalingM. are bovis known BCG, a surrogate for Mtb, in vitro in human cells. These ex- *Corresponding author. E-mail: [email protected] (A.S.); gennaro_ [email protected] (G.D.L.) (18, 19). In our experiments, MET treatment inhibited the growth of in- †These authors contributed equally to this work. tracellular BCG (fig. S2A) as well as the H37Rv strainperiments of Mtb (Fig. 1A) in led to the identificationofMETasthemostappropriatecan- potentially (i) augment the efficacy of antibiotic-basedwww.ScienceTranslationalMedicine treatment,.org 19 November 2014 Vol 6didate. Issue 263 263ra159 Next,1 detailed mechanistic studies were performed to elucidate (ii) shorten treatment regimens for drug-susceptible and/or drug-resistant the mechanism of MET in controlling Mtb growth in vitro and in mouse Mtb infections, (iii) reduce the immunopathology associated with TB, and models of Mtb infection. MET was then evaluated as adjunctive treat- (iv) promote development of immunological memory that protects ment together with standard TB therapeutic drugs in mouse Mtb infec- against relapse. A number of in vitro and animal studies have attempted tion models. Mice were infected and randomized in different groups to address these issues by repurposing existing drugs, but mixed out- 1daybeforethestartoftreatment.Finally,torevealthepossibleeffect Downloaded from comes were reported (4, 35–37). Some of these drugs, such as phospho- of MET in TB patients, retrospective data were mined from two inde- diesterase 4 inhibitors, alter tissue inflammation associated with Mtb pendent cohorts (TB and diabetic cohorts). infection, and the anti-inflammatoryeffectswereconsideredbeneficial (26, 38). Despite the studies in the field, it remains to be clarified how Drugs protective immunity and pathological immunity in Mtb infection are The following drugs were used: MET(Sigma,no.D150959),INH(Sigma, differentiated, and whether these two aspects of immune response can no. I3377), ETH (Sigma, no. E6005), streptomycin sulfate (Sigma, no. be separated (34). Our data suggest that they can be independently S9137), Bay K8644 (Sigma, no. B112), R-(+)-Bay K8644 (Sigma, no. B132), modulated with beneficial effects during Mtb infection because MET S-(−)-Bay K8644 (Sigma, no. B133), FPL 64176 (Sigma, no. F131), 2′,5′- enhances Mtb-specific host immunity, reduces inflammation, promotes dideoxy-adenosine (Merck, no. 288104), loperamide hydrochloride disease resolution, and improves TB treatment outcome. (Sigma, no. L4762), nimodipine (Sigma, no. N149), nitrendipine (Sigma, At the cellular level, MET differentially affects immune response and no. N144), KT5720 (Merck, no. 420323), rapamycin (Sigma, no. R8781), inflammation through different mechanisms. The protective effect is verapamil hydrochloride (Sigma, no. V4629), and amiodarone hydro- mediated by increased host cell production of mROS and increased acid- chloride (Sigma, no. A8423). ification of mycobacterial phagosome. Indeed, mROS produced upon mitochondrial recruitment to phagosomes is instrumental in the killing Cell culture of intracellular bacteria by macrophages (39). The anti-inflammatory THP-1 cells (human monocytic cell line) were maintained in RPMI effect is mediated by activation of AMPK, a negative regulator of in- 1640 supplemented with 10% fetal bovine serum (FBS), 1% sodium

www.ScienceTranslationalMedicine.org 19 November 2014 Vol 6 Issue 263 263ra159 6 F O CUS Metformin and host-directed therapy in tuberculosis

Pulmonary TB Extrapulmonary TB

Heart Brain Lung Spine Kidney Reduction of Lymph node Bone marrow inflammation

Reduction of 2+ Ca channel Mtb burden Insulin receptor

Macrophage CD8+ T cell Mtb

Prostaglandin E2 Verapamil Metformin

Activation HDAC inhibitors ￿ chain mTOR Metformin MHC class I ￿￿TCR AMPK

Zileuton COX-2 inhibitors HDAC Nivolumab T cell activation 5-LO and cytolytic activity on November 19, 2014 PDL-1 PD-1 Nucleus RESEARCH ARTICLE

Fig. 1. Healing TB: Many targets, one goal. Host-directed therapiesTUBERCULOSIS in TB target infl ammatory immune responses, which may be associated with increased Mtb proliferation and enhanced bacterial mutation rates. MetforminOn the left is an Mtb- asinfected adjunct macrophage antituberculosis that is activating a cytotoxic therapy T lym- phocyte on the right. Pulmonary TB is the most common form of Mtb infection; lesions show di erent stages of infl ammation, and infection with multiple Mtb strains is possible, which complicates further the challengeAmit of Singhal, adequately1* Liu timing Jie,1† Pavanish host-directed Kumar, therapies.1† Gan Suay For Hong, extrapulmonary2 Melvin Khee-Shing TB lesions, Leow, 3,4 tissue- and organ-specifi c di erences can cause alternative editing Bhairavof Mtb– Paleja,directed1 Liana immune Tsenova, responses.5,6 Natalia In Kurepina,boxes are5 Jinmiao host-directed Chen,1 Francescadrugs that Zolezzi, act 1

5 1,7 2 5,8 stm.sciencemag.org at various sites in the immune-activation process. PGE2 prevents uncontrolledBarry Kreiswirth, infl ammationMichael Poidinger,and contributesCynthia to the Chee, balanceGilla of Kaplan, IL-1 and type I IFNs, 2 1,9 keeping Mtb in control. Verapamil a ects mycobacterial e ux pumps,Yee a ects Tang bacterial Wang, Gennaro drug tolerance, De Libero and* enhances anti-TB drug e cacy. Metformin Mtb 1 interacts with mTOR (mammalian target of rapamycin) and activates TheAMPK global both burden in macrophages of tuberculosis (TB) and morbidity in T cells, and mortalityleading remains to control immense. of A potential growth new ( approach). to TB PGE2 supplementation—either directly or by 5-lipoxygenase (5-LO) therapyinterference is to augment (via the protective drug hostzileuton immune or responses.COX-2 inhibitors)—results We report that the antidiabetic in a reduction drug metformin (MET) in type I IFNs, a shift that reduces Mtb virulence and TB progression. Histonereduces the modifi intracellular cation, growth either of inMycobacterium macrophages tuberculosis or T cells,(Mtb leads) in an to AMPK changes (adenosine in gene monophosphate – expression, antigen presentation, and epigenetic modifi cation in exhaustedactivated T protein cells. T kinase) cell activation–dependent manner.can be METaugmented controls the by growth antibodies of drug-resistant to PD-1Mtb or (drug-strains, increases pro- duction of mitochondrialMtb reactive oxygen species, and facilitates phagosome-lysosome fusion. In Mtb-infected mice, mediated) AMPK stimulation, leading to increased numbers of Th1-producinguse of MET ameliorated-reactive lung T pathology, cells. TCR, reduced T cell chronic receptor; inflammation, PDL-1,and ligand enhanced for PD-1. thespecificimmuneresponse and the efficacy of conventional TB drugs. Moreover, in two separate human cohorts, MET treatment was associated

with improved control of Mtb infection and decreased disease severity. Collectively, these data indicate thatDownloaded from MET is a e TB bactericidal drug pyrazinamide of cAMP with low-dosepromising candidatecyclophosphamide host-adjunctive therapyactivated for improving protein the effective kinase treatment (AMPK), of TB. a cellular contributes to host-cell immune responses may also help to remove regulatory T cells, a energy-sensor that regulates cell growth and in standard antimycobacterial chemother- strategy used in INTRODUCTIONcancer vaccination and for survival (10), and initially(ROS) was and developed reactive nitrogen for species (RNS), as well as the capacity apy by down-regulation of proin amma- the clinical managementMycobacterium of severe tuberculosis in(Mtb uenza)istheetiologicalagentoftuberculosis the treatment of hyperglycemiato destroy intracellular and type pathogens 2 using the phagosomal machinery (TB), a disease that continues to be a leading killer worldwide, with or autophagy pathway. Autophagy is required for the effective control tory cytokines tumor necrosis factor– and infection. Histonean estimateddeacetylase global (HDAC) mortality of in- 1.43 milliondiabetes people (T2D). every year More (1). recently,of intracellular epidemiologi- pathogens (11, 12)andisregulatedbymammaliantarget IL-6 and by induction of autophagy in Mtb- hibitors might modulateAlthough current Mtb- treatmentinfected of cells drug-susceptible cal studies TB is effective, pinpointed the of a rapamycin decrease (mTOR) in cancer complex 1 (13), a serine/threonine kinase, and infected host cells, an integral component of by increasing majortherapy histocompatibility regimen is lengthy (6 tocom- 9 months),incidence and problems in withdiabetes drug patientsadenosine monophosphatewho were tak-–activated protein kinase (AMPK) (13, 14). on November 24, 2014 toxicity and increasing multidrug resistance (MDR) urge the discovery/ Accordingly, perturbations in the autophagy network and AMPK the drug’s e ectiveness (7). e second mes- plex (MHC) expressiondevelopment and of newpresentation drugs and therapeutic of ing strategies metformin,which (2, 3). The global ledsignaling to deep have mechanistic previously been associated with Mtb virulence (15, 16). senger cyclic adenosine monophosphate alternate peptideneed repertoires for new effective and therapiese ector has T led toanalyses a resurgence of inmetformin efforts to Weaction. therefore On screened the basis the capacity of of various U.S. Food and Drug Ad- identify additional anti-Mtb agents, several of which are now being eval- ministration (FDA)–approved mTOR-independent autophagy activators (cAMP) suppresses innate immune re- cells by altering epigeneticuated in clinical programs trials (4). However, associ- conventionaldata from pathogen-targeted the epidemiological(17)andAMPKmodulatorsfortheirabil studies, met- ity to control intracellular myco- sponses via down-regulation of proin am- ated with T cell anergystrategies ( suffer9). from the serious disadvantageformin of fostering is now microbial being testedbacterial in growth. more We than discovered 50 that the AMPK-activating antidiabetic

SCIENCE TRANSLATIONAL MEDICINE TRANSLATIONAL SCIENCE matory cytokines and phagocytic functions. TB, meet T2D.resistance. Produced To circumvent by the this French problem, a newcancer paradigm clinical in drug discov-trials anddrug in metformin individuals (MET; 1,1-dimethylbiguanide) who inhibits the intracellular ery that involves therapeutic modulation of host cell responses to improve growth of Mtb,restrictsdiseaseimmunopathology,andenhancesthe stm.sciencemag.org Sildena l (phosphodiesterase inhibitor type lilac Galega o cinalis,pathogen eradicationthe biguanide (5–7)hasemerged.These met- are “athost-targeted increased” adjunct risk forefficacy developing of conventional cancer— anti-TB drugs. 5) and cilostazol (phosphodiesterase inhibi- formin is one oftherapeutic the most strategies widely are less used likely re- to engenderfor instance, microbial resistance cancer survivors, individuals tor type 3) decrease pulmonary damage and purposed drugs, thanwith direct up targetingto 120 ofmillion the pathogen pre- with conventionalwith potentially drugs (6 ).precancerous We (bronchial) le- therefore tested whether existing approved drugs with defined effects on RESULTS facilitate Mtb clearance (8), most likely via scriptions dispensedhost cell worldwide. functions could Metformin be repurposed for effectivesions, andor durablemen in anti-TB active surveillance programs

CREDIT: H. MCDONALD/ H. CREDIT: improved bacterial phagocytosis. Reduction activates adenosinetherapy. monophosphate The validation of this (AMP)– approach wouldfor have low-risk the advantage disease of re- in METprostate restricts cancer. mycobacterial growth by inducing quiring shorter clinical trials than those usually needed for new drugs (4). mitochondrial ROS production Downloaded from An effective host immune system is a crucial factor for both the A total of 13 autophagy and AMPK-activating drugs were tested for www.controlScienceTranslationalMedicine of Mtb growth and its containment.org as latent 19 November TB infection 2014 (LTBI). Volthe 6 Issue ability 263 to control263fs47 the intracellular 2 growth of M. bovis bacillus The success of Mtb in infecting the host cells and maintaining long-term Calmette-Guérin (BCG) in the human monocytic cell line THP-1 using persistent infection is associated with the ability of bacilli to evade host a colony-forming unit (CFU) assay. Eight of these compounds sig- innate as well as adaptive immune responses (8–10). The host cell innate nificantly inhibited the intracellular growth of BCG (fig. S1, A and B, antimicrobial arsenal includes production of reactive oxygen species and table S1). None of these eight compounds displayed direct anti- mycobacterial effects in an in vitro minimum inhibitory concentration (MIC) assay (fig. S1, C to E), suggesting that these compounds restricted 1Singapore Immunology Network, Agency for Science, Technology and Research (A*STAR), Singapore 138648, Singapore. 2Tuberculosis Control Unit, Tan Tock Seng Hospital, BCG growth via effects on host cell–BCG interactions with minimal di- Singapore 308089, Singapore. 3Department of Endocrinology, Tan Tock Seng Hospital, rect effect, if any. In these experiments, rapamycin, a known autoph- Singapore 308433, Singapore. 4Singapore Institute for Clinical Sciences, A*STAR, Singapore agy inducer, was used as a positive control. In subsequent studies, we 117609, Singapore. 5Public Health Research Institute at New Jersey Medical School, Rutgers University, Newark, NJ 07103, USA. 6New York City College of Technology, Brooklyn, NY focused on evaluating the therapeuticpotentialoftheAMPK-activating 11201, USA. 7Department of Biological Sciences, National University of Singapore, agent MET because this drug is already being used extensively in the Singapore 117543, Singapore. 8Bill & Melinda Gates Foundation, Seattle, WA 98109, USA. clinic for the treatment of type 2 diabetes mellitus (DM), and the effects 9 University Hospital Basel, University of Basel, Basel 4031, Switzerland. of the drug on cellular functions including AMPK signaling are known *Corresponding author. E-mail: [email protected] (A.S.); gennaro_ [email protected] (G.D.L.) (18, 19). In our experiments, MET treatment inhibited the growth of in- †These authors contributed equally to this work. tracellular BCG (fig. S2A) as well as the H37Rv strain of Mtb (Fig. 1A) in

www.ScienceTranslationalMedicine.org 19 November 2014 Vol 6 Issue 263 263ra159 1 RESEARCH

◥ ulation by microbial pathogens, increased REVIEW SUMMARY production of inflammatory mediators, and enhanced capacity to eliminate infection. Mechanistic studies have demonstrated that INNATE IMMUNITY trained immunity is based on epigenetic re- programming, which is broadly defined as sustained changes in transcription programs Trained immunity: A program of and cell physiology that do not involve per- manent genetic changes, such as mutations and recombination. Histone innate immune memory in ◥ ON OUR WEBSITE modifications with chro- matin reconfiguration have health and disease Read the full article at http://dx.doi. proven to be a central proc- org/10.1126/ Mihai G. Netea,* Leo A. B. Joosten, Eicke Latz, Kingston H. G. Mills, ess for trained immunity, science.aaf1098 but other mechanisms— Gioacchino Natoli, Hendrik G. Stunnenberg, Luke A. J. O’Neill, Ramnik J. Xavier ...... such as DNA methylation or modulation of microRNA and/or long non- BACKGROUND: Host immune responses dependently of T and B lymphocytes. These coding RNA expression—are also expected to are classically divided into innate immune observations led to the hypothesis that innate be involved. This leads to transcriptional pro- responses, which react rapidly and nonspe- immunity can display adaptive characteris- grams that rewire the intracellular immune cifically upon encountering a pathogen, and tics after challenge with pathogens or their signaling of innate immune cells but also in- adaptive immune responses, which are slower products. This de facto immunological mem- duce a shift of cellular metabolism from oxida- to develop but are specific and build up im- ory has been termed “trained immunity” or tive phosphorylation toward aerobic glycolysis, munological memory. The dogma that only “innate immune memory.” thus increasing the innate immune cells’ ca- adaptive immunity can build immunological pacity to respond to stimulation. Trained im- on April 21, 2016 memory has recently been challenged by studies ADVANCES: In recent years, emerging evi- munity programs have evolved as adaptive showing that innate immune responses in dence has shown that after infection or vac- states that enhance fitness of the host (e.g., plants and invertebrates (organisms lacking cination, prototypical innate immune cells protective effects after infection or vaccina- adaptive immune responses) can mount re- (such as monocytes, macrophages, or natural tion, or induction of mucosal tolerance toward sistance to reinfection. Furthermore,Trained in certain immunitykiller cells) display versus long-term changes tolerance in their colonizing microorganisms). Proof-of-principle mammalian models of vaccination, protection functional programs. These changes lead to in- experimental studies support the hypothesis from reinfection has been shown to occur in- creased responsiveness upon secondary stim- that trained immunity is one of the main immunological processes that mediate the nonspecific protective effects against infec- tions induced by vaccines, such as bacillus Calmette-Guérin or measles vaccination. How-

ever, when inappropriately activated, trained http://science.sciencemag.org/ immunity programs can become maladaptive, as in postsepsis immune paralysis or auto- inflammatory diseases.

OUTLOOK: The discovery of trained immunity has revealed an important and previously un-

recognized property of human immune re- Downloaded from sponses. This advanceopensthedoorfor future research to explore trained immunity’s effect on disease, for both diseases with im- paired host defense, such as postsepsis immune paralysis or cancers, and autoinflammatory dis- eases, in which there is inappropriate activation of inflammation. These findings have consider- able potential for aiding in the design of new therapeutic strategies, such as new generations of vaccines that combine classical immunologi- cal memory and trained immunity, the activa- tion of trained immunity for the treatment of postsepsis immune paralysis or other immune deficiency states, and modulation of exaggerated Netea et al, Science,Innate immune2016 activation by infections or vaccinations leads to histone modifications and inflammation in autoinflammatory diseases. functional reprogramming of cells (such as monocytes, macrophages, or NK cells) termed ▪ “trained immunity” or “innate immune memory.” Trained immunity evolved to lead to adaptive states that protect the host during microbial colonization or after infections. However, in certain The list of author affiliations is available in the full article online. *Corresponding author. Email: [email protected] situations, trained immunity may result in maladaptive states such as postsepsis immune paralysis Cite this article as M. G. Netea et al., Science 352,aaf1098 or hyperinflammation. miRNA, microRNA. (2016). DOI: 10.1126/science.aaf1098

SCIENCE sciencemag.org 22 APRIL 2016 • VOL 352 ISSUE 6284 427 Trained immunity in humans

Small shot, big impact

Some vaccines seem to provide us with a host of extra New Scientist benefts. Michael Brooks sees the cornerstone of Aug. 2013 modern medicine in a new light

>SB
^
illh
^q
vlro
ibcq
pelriabo7
fc
vlr
^
jbj_bo
lc
qeb
TEL
`ljjfqqbb+
^ob
m^pq
vlro
jfa*qtbkqfbp
fq
^ijlpq
ÎS^``fkbp
e^sb
_bbk
^
c^kq^pqf`
pr``bpp)
H`boq^fkiv
_b^op
^
`fo`ri^o
p`^o+
Al
vlr
_rq
tb
`^k
mol_^_iv
al
jr`e
_bqqbo
_v
objbj_bo
elt
fq
dlq
qebob<
Vlr
nrbrba
rm
fk
q^hfkd
klk*pmb`fcf`
bccb`qp
fkql
^``lrkq+
>k
qeb
p`elli
e^ii)
mboe^mp)
lo
lrqpfab
qeb
kropbÑp
bu^jfk^qflk
lc
qebpb
fpprbp
fp
ilkd
lsboarb+Ï lccf`b)
t^q`efkd
vlro
cofbkap
or__fkd
qebfo
@lkpfabofkd
s^``fkbp
e^sb
_bbk
rpba
pfk`b
^ojp
^p
qebv
t^ihba
^t^v)
obifbsba
^q
e^sfkd
qeb
.5--p
^ka
^ob
qeb
`bkqo^i
mi^kh
lc
lro
prosfsba
qebfo
?@D
g^_+ mr_if`
eb^iqe
pvpqbj)
fq
j^v
pbbj
e^oa
ql
Qeb
?^`fiib
@^ijbqqb*DrŠofk
s^``fk^qflk
_bifbsb
qe^q
pr`e
molclrka
bccb`qp
`lria
e^sb
t^p
dfsbk
ql
molqb`q
vlr
colj
qr_bo`rilpfp+
dlkb
fdkloba
^ii
qefp
qfjb+
Fk
c^`q)
^k
b^oiv
Te^q
tb
^ob
lkiv
grpq
ob^ifpfkd
fp
qe^q)
fk
/-qe
`bkqrov
Ptbafpe
mevpf`f^k
`^iiba
@^oi
`ljjlk
tfqe
pbsbo^i
lqebo
s^``fkbp)
fq
j^v
K†pirka
afa
klqf`b
pljbqefkd
t^p
rm
^cqbo
e^sb
alkb
c^o
jlob
qe^k
qe^q+
qeb
?@D
s^``fkb
t^p
fkqolar`ba
fk
efp
`lrkqov+
Qebob
fp
doltfkd
bsfabk`b
qe^q
s^``fkbp
S^``fk^qba
`efiaobk
e^a
^
jr`e
efdebo
`e^k`b
e^sb
^
tfabo*o^kdfkd
fkcirbk`b
lk
qeb
lc
ob^`efkd
qebfo
cfopq
_foqea^v
Ì
bsbk
qelrde
fjjrkb
pvpqbj
qe^k
tb
qelrdeq+
Fk
>cof`^)
Q?
kloj^iiv
hfiip
liabo
`efiaobk+
clo
fkpq^k`b)
pqrafbp
e^sb
peltk
qe^q
jb^pibp
Fk
qeb
.61-p
^ka
2-p)
qof^ip
fk
qeb
RP
^ka
RH
s^``fkb
`rqp
ab^qep
colj
^ii
lqebo
fkcb`qflkp
prddbpqba
qe^q
?@D*s^``fk^qba
`efiaobk
e^a
^
`lj_fkba
_v
^
qefoa)
j^fkiv
_v
molqb`qfkd
/2
mbo
`bkq
iltbo
ab^qe
o^qb
colj
afpb^pbp
lqebo
^d^fkpq
mkbrjlkf^)
pbmpfp
^ka
af^ooelb^+ qe^k
Q?+
?rq
kl
lkb
qllh
jr`e
klqf`b
rkqfi
Bsbk
fk
qeb
Tbpq)
tebob
fq
fp
c^o
ibpp
`ljjlk
0-
vb^op
^dl)
tebk
^
A^kfpe
^kqeolmlildfpq
clo
`efiaobk
ql
afb
colj
fkcb`qflrp
fiikbppbp)
`^iiba
Mbqbo
>^_v
_bd^k
tlohfkd
fk
qeb
Tbpq
qebob
^ob
pqfii
promofpfkd
_bkbcfqp7
pljb
>cof`^k
pq^qb
lc
Drfkb^*?fpp^r+
Fk
.646
eb
s^``fkbp
pbbj
ql
obar`b
lro
prp`bmqf_fifqv
ql
tfqkbppba
^
pbsbob
jb^pibp
lrq_ob^h
qe^q
b`wbj^
^ka
^pqej^+
Bu^`qiv
te^q
`^rpbp
qebpb
hfiiba
.
fk
1
fkc^kqp
^ccb`qba+
>^_v
^oo^kdba
clo
Îklk*pmb`fcf`
bccb`qpÏ)
^p
qebv
^ob
qbojba)
fp
^
jb^pibp
s^``fk^qflk
ql
_b
fkqolar`ba)
_rq
t^p
jvpqbov+
?rq
pljb
p`fbkqfpqp
^ob
^odrfkd
qe^q)
promofpba
ql
pbb
qe^q
bsbk
^cqbo
qeb
bmfabjf`
abpmfqb
qeb
rk`boq^fkqfbp)
fq
fp
qfjb
ql
pq^oq
^_^qba)
fjjrkfpba
`efiaobk
tbob
jlob
ifhbiv
e^okbppfkd
qebj
jlob
bccb`qfsbiv+
ql
prosfsb
`efiaella+
>^_v
_bd^k
afddfkd)
^ka
Qeb
Tloia
Eb^iqe
Lod^kfw^qflk)
tef`e
fp
afp`lsboba
pqrafbp
colj
bipbtebob
fk
>cof`^)
qeb
j^fk
molsfabo
lc
s^``fkbp
fk
absbilmfkd
^p
tbii
^p
?^kdi^abpe
^ka
E^fqf)
qe^q
^ipl
`lrkqofbp)
e^p
^phba
^
dolrm
lc
s^``fkb
bumboqp
prddbpqba
jb^pibp
s^``fkb
dfsbp
^
tfabo
hfka
ql
dbq
ql
qeb
_lqqlj
lc
fq+
ÎQefp
`lria
e^sb
lc
molqb`qflk+
ÎTb
^ob
`liib`qfkd
jlob
^ka
erdb
fjmif`^qflkp
clo
dil_^i
eb^iqe`^ob)Ï
jlob
a^q^
`lkpfpqbkq
tfqe
klk*pmb`fcf`
p^vp
@eofpqfkb
?bkk)
^
pbkflo
obpb^o`ebo
^q
bccb`qp
_bfkd
sbov
fjmloq^kq)Ï
p^vp
>^_v+

BRIAN LAROSSA BRIAN qeb
Pq^qbkp
Pborj
Fkpqfqrqb
fk
Abkj^oh
^ka
Te^q
`lria
qeb
bumi^k^qflk
_b<
Pbsbo^i
>

38 | NewScientist | 17 August 2013 17 August 2013 | NewScientist | 39 Does this happen in vivo in humans?

M. tuberculosis S. aureus C. albicans 12 9 12 8 * 10 ** 10 7 8 * 6 8 5 6 ** 6 4 4 3 4 2 2 2 1 0 0 0 before BCG 2 weeks 3 months before BCG 2 weeks 3 months before BCG 2 weeks 3 months

Kleinnijenhuis et al, PNAS, 2012 BCG enhances monocyte-derived cytokines

M. tuberculosis S. aureus C. albicans 3.0 2.5 2.5 * 2.5 * * * 2.0 * 2.0 * 2.0 1.5 1.5 1.5 1.0 1.0 1.0 0.5 0.5 0.5 0.0 0.0 0.0 before BCG 2 weeks 3 months before BCG 2 weeks 3 months before BCG 2 weeks 3 months

M. tuberculosis S. aureus C. albicans 2.5 3 1.4 * 1.2 * 2.0 * 2 1.0 1.5 0.8 1.0 0.6 1 0.4 0.5 0.2 0.0 0 0.0 before BCG 2 weeks 3 months before BCG 2 weeks 3 months before BCG 2 weeks 3 months

Kleinnijenhuis et al, PNAS, 2012 BCG effects on epigenecs and transcripon

M. tb S. a C. a (hyph)

500 1000 1050 350 IL1" mRNA 550 500 200 50 50 20 10 30 TNF! mRNA 20 10 10

0 0 0 before vaccination 3 months before vaccination 3 months before vaccination 3 months

Before BCG 3 months post BCG 1000 *

100 *

10 TNF! IL6 TLR4 Kleinnijenhuis et al, PNAS, 2012 a 10 kb c 10 kb e 50 kb _ - 69- control 147 control 142 control 3 3 3 e e e 1_ 1_ 1_ m m _ m - - 142

4 4 4 -glucan 69 B-glucan 147 B-glucan B K K K

3 3 1_ 3 _ 1_ 1 H LTA H IL18 H TLR4 TNF

10 kb 20 kb 50 kb b 32 - d g 133 - control control - control 3 3 3 107 e e e 1_ What arem the pathwaysm - distinguishingm _ _ 1 1 - 4 4 133 -glucan 4 32 - B 107 B-glucan K B-glucan K K 3 3 3 1_ 1_ H H H 1_ CLEC7A MYD88 TrainingIL6 vs Tolerance ?

ATF3 CFLAR HSPA5 BCL2L1 BCL2A1 f CAPG TNFRSF1B MAP3K5 RGL1 DDIT3 GSN RAS DUSP6 FAS apoptosis HK3 MEF2C RASAL2 PKM2 NF1 ENO1 HBEGF 2A B PPP2CA BCAT1 oxidative stress JUN IGF2R CBLB C7orf10 ELF1 SPRY2 NPL RAP2B D0 D6 LPS BG D0 D6 LPS BG D0 D6 LPS BG D0 D6 LPS BG D0 D6 LPSFOS BG D0 D6 LPS BG glycolysis insulin PDGF EGF CYP27A1 aminoacid metabolism PLA2G4A ARHGAP10 RAPGEF1 GSK3B ACe3 FDX1 VitD angiogenesis ACp3 ABL2 GNAQ RPS6KA1 LIMS1 NCK1 CTNNB1 HIF1A ARHGEF2 G-protein signalingARF6 integrin signaling LPXN ARHGEF11 CALM3 ACe4 ACp2 axon guidance RHOA AKAP13 NRP1 T cellHLA-DMB PRKACB CLTA ARPC5 HLA-DPA1 RGS1 ACp4 HLA-DMA RGS18 interleukin FOXO3 CD86 CFL1 RhoA regulation GNG5 ACe2 EVL SPI1 ACTG1 Hunt. Dis.signaling IL10RA ACTR3 IL10 IL18 ARL8B ROCK1 inflamation REL ACp1 ITGB2 FPR3 ACe7 TLR1 IFNGR1 PTEN C5AR1 ACe5 CCR1 TLR4 Genes hits Cluster of ACe6 in pathway pathways TNFAIP3 Toll like receptor 18 Immune Defense Color code ACe1 13 Signaling pathway H3K27ac Metabolism H3K4me1 11 H3K4me3 9 Structure ! DNAseI 8 6 5 Figure 5 3 Saeed, Quintin et al, Science, 2014 2 C D heterochromatin Transcribed region H3K27ac H3K4me3 H3K4me1 Repressed promoter Active promoter Regulatory element Scale 2 kb Scale10 kb hg19 hg19 chr17: chr14: 61954763 75 _ 75 _ D0 D0 whole genome

1 1 75 _ 75 _ D6 D6 ACp4 1 1 75 _ 75 _ LPS LPS ACp3 1 1 75 _ BG 75 _ BG N ACp2 1 1_ 120 _ 120 _ _ D0 D0 ACp1 1_ 1 120 _ 120 _ _ D6 D6 _ ACe7 1_ 150 _ 1_ _ LPS 150 _ LPS _ ACe6 1_ 120 _ 1_ u BG 120 _ BG 1_ ACe5 1 75 _ D0 75 _ D D0 1_ ACe4 1_ 75 _ D6 D 75 _ D D6 1_ ACe3 75 _ LPS 1_ _ 75 _ _ 1_ LPS ACe2 75 _ 1_ N BG 75 _ 1_ BG LOC284080 1 ACe1 ITGA3 PRKCH _ 0 20 40 60 80 100 % total regions a 10 kb c 10 kb e 50 kb _ - 69- control 147 control 142 control 3 3 3 e e e 1_ 1_ 1_ m m _ m - - 142

4 4 4 -glucan 69 B-glucan 147 B-glucan B K K K

3 3 1_ 3 _ 1_ 1 H LTA H IL18 H TLR4 TNF

10 kb 20 kb 50 kb b 32 - d g 133 - control control - control 3 3 3 107

e What aree the pathwayse distinguishing 1_ m _ m - m 1_ 1 - 4 4 133 -glucan 4 32 - B 107 B-glucan K B-glucan K K 3 3 3 1_ 1_ H H Training vs ToleranceH ? 1_ CLEC7A MYD88 IL6

ATF3 CFLAR HSPA5 BCL2L1 metabolism BCL2A1 f CAPG TNFRSF1B MAP3K5 RGL1 DDIT3 GSN RAS DUSP6 FAS apoptosis HK3 MEF2C RASAL2 PKM2 NF1 ENO1 HBEGF PPP2CA BCAT1 oxidative stress JUN IGF2R CBLB C7orf10 ELF1 SPRY2 NPL RAP2B FOS glycolysis insulin PDGF EGF CYP27A1 aminoacid metabolism PLA2G4A ARHGAP10 RAPGEF1 GSK3B FDX1 VitD angiogenesis ABL2 GNAQ RPS6KA1 LIMS1 NCK1 CTNNB1 HIF1A ARHGEF2 G-protein signalingARF6 integrin signaling LPXN ARHGEF11 CALM3 axon guidance RHOA AKAP13 NRP1 T cellHLA-DMB signaling PRKACB CLTA ARPC5 HLA-DPA1 RGS1 HLA-DMA RGS18 interleukin FOXO3 CD86 CFL1 RhoA regulation GNG5 EVL SPI1 ACTG1 Hunt. Dis. IL10RA ACTR3 IL10 IL18 ARL8B ROCK1 inflamation REL ITGB2 FPR3 TLR1 IFNGR1 PTEN C5AR1 CCR1 TLR4 Genes hits Cluster of in pathway pathways TNFAIP3 Toll like receptor 18 Immune Defense 13 Signaling pathway Metabolism Saeed, Quintin et al, Science, 2014 11 9 Structure ! 8 6 5 Figure 5 3 2 Glucose metabolism and host defense

• The metabolic pathways through which immune cells metabolize glucose are crucial for immune activation • Cellular metabolism has a signaling function linking immune siganls and long-term epigenetic reprogramming of cell function • Understanding cellular metabolism of immune cells during infection (TB) is crucial to fully describe the cell function • Modulation of cell metabolism may represent a novel therapeutic target in improving host defenese in tuberculosis Thank you !

Our lab Max Plack - Berlin Ekta Lachmandas Macarena Beigier-Bompadre Rob Arts Stefan Kauffman Bas Blok Harvard University Johanneke Kleinnijenhuis Clary Clish James Cheng Ramnik Xavier Jos W.M. van der Meer Leo Joosten University of Minho, Braga Reinout van Crevel Agosnho Carvalho Ricardo Silvestre Universityy of Groningen Fernando Rodrigues Vinod Kumar Cisca Wijmenga Trinity College Luke O’Neill Dept. Molecular Biology - Radboud Sadia Saeed Athens University Joost Martens Evangelos Giamarellos Colin Logie Henk Stunnenberg