The troika host-pathogen-extrinsic factors in tuberculosis: Modulating inflammation and clinical outcomes

Janua y 2018 | Volume 8 | A icle 19481 Re iew published: 09 Janua y 2018 doi: 10.3389/ immu.2017.01948 F on ie s in Immunology | www. on ie sin.o g Edi ed by: Oli ie Ga aud, Ins i u Na ional de la T ans usion Sanguine, F ance Re iewed by: Ma c Veldhoen, Ins i u o de Medicina Molecula (IMM), Po ugal Laleh Majlessi, Ins i u Pas eu , F ance *Co espondence: Iñaki Comas icomas@ib .csic.es; Ma ga ida Sa ai a [email p o ec ed] †These au ho s ha e con ibu ed equally o his wo k. Special y sec ion: This a icle was submi ed o In lamma ion, a sec ion o he jou nal F on ie s in Immunology Recei ed: 25Oc obe 2017 Accep ed: 18Decembe 2017 Published: 09Janua y2018 Ci a ion: Bas osHN, Osó ioNS, GagneuxS, ComasI and Sa ai aM (2018) The T oika Hos –Pa hogen–Ex insic Fac o s in Tube culosis: Modula ing In lamma ion and Clinical Ou comes. F on . Immunol. 8:1948. doi: 10.3389/ immu.2017.01948 The T oika Hos –Pa hogen–ex insic Fac o s in Tube culosis: Modula ing in lamma ion and Clinical Ou comes Helde No ais Bas os1,2,3†, Nuno S. Osó io2,3†, Sebas ien Gagneux4,5, Iñaki Comas6,7*† and Ma ga ida Sa ai a8,9*† 1 Depa men o Pneumology, Cen o Hospi ala do São João, Po o, Po ugal, 2 Li e and Heal h Sciences Resea ch Ins i u e (ICVS), School o Medicine, Uni e si y o Minho, B aga, Po ugal, 3 ICVS/3B’s—PT Go e nmen Associa e Labo a o y, B aga, Po ugal, 4 Depa men o Medical Pa asi ology and In ec ion Biology, Swiss T opical and Public Heal h Ins i u e, Basel, Swi ze land, 5 Uni e si y o Basel, Basel, Swi ze land, 6 Ins i u e o Biomedicine o Valencia (IBV-CSIC), Valencia, Spain, 7CIBER o Epidemiology and Public Heal h (CIBERESP), Mad id, Spain, 8 i3S—Ins i u o de In es igação e Ino ação em Saúde, Uni e si y o Po o, Po o, Po ugal, 9 Ins i u o de Biologia Molecula e Celula (IBMC), Uni e si y o Po o, Po o, Po ugal The al eady eno mous bu den caused by ube culosis (TB) will be u he agg a a ed by he associa ion o his disease wi h mode n epidemics, as human immunode iciency i us and diabe es. Fu he mo e, he inc easingly aging popula ion and he wide use o supp essi e immune he apies hold he po en ial o enhance he incidence o TB. New p e en i e and he apeu ic s a egies based on ecen ad ances on ou unde - s anding o TB a e hus needed. In pa icula , unde s anding he in ica e ne wo k o e en s modula ing in lamma ion in TB will help o build mo e e ec i e accines and hos -di ec ed he apies o s op TB. This e iew in eg a es he impac o hos , pa hogen, and ex insic ac o s on in lamma ion and he almos scien i ically unexplo ed complexi y eme ging om he in e ac ions be ween hese h ee ac o s. We highligh he exci ing da a showing a con ibu ion o his oika o he clinical ou come o TB and he need o inco po a ing i when de eloping no el s a egies o ewi e he immune esponse in TB. Keywo ds: ube culosis, geno ypic di e si y, immune pheno ypes, se e i y o disease, in lamma ion, mic oen i onmen s iNTRODUCTiON Acco ding o cu en es ima es, ube culosis (TB) accoun ed o app oxima ely 1.7 million dea hs in 2016 and a ec ed one-qua e o he wo ld’s popula ion in i s la en o m (1). TB is a he e ogeneous disease, cha ac e ized by a con inuous spec um o in ec ion, o which molecula and clinical bioma ke s o p og ession a e jus s a ing o be un eiled. Se e al s ages o la en TB in ec ion (LTBI) exis and include subclinical o ms o TB wi h an inc eased likelihood o p o- g essing o ac i e disease (2, 3). The clinical mani es a ion o he ac i e disease is highly a iable, wi h mild o ex ensi e pulmona y in ol emen , ex apulmona y, o dissemina ed o ms o TB. Abb e ia ions: CAMP, ca helicidin an imic obial pep ide; COPD, ch onic obs uc i e pulmona y disease; DCs, dend i ic cells; GWAS, genome-wide associa ion s udies; HDTs, hos -di ec ed he apies; HIV, human immunode iciency i us; IFN, in e - e on; IL, in e leukin; iNOS, inducible NO syn hase; IRGM, immuni y- ela ed GTPase M; LTA4H, leuko iene A4 hyd olase; LTBI, la en ube culosis in ec ion; MARCO, mac ophage ecep o wi h collagenous s uc u e; MBL2, mannose-binding lec in 2; MDR, mul id ug esis an ; MSMD, Mendelian suscep ibili y o mycobac e ial disease; MTBC, Mycobac e ium ube culosis complex; NO, ni ic oxide; PGE2, p os aglandin E2; SNP, single-nucleo ide polymo phisms; TB, ube culosis; Th, T helpe ; TLR, oll-like ecep o ; TNF, umo nec osis ac o ; VDR, i amin D ecep o ; XDR, ex ensi ely d ug esis an . FigURe 1 | Impac o he in lamma ion le el in he disease ou come in indi iduals in ec ed by Mycobac e ium ube culosis complex bac e ia. The spec um o ube culosis (TB) disease is s ongly linked wi h he hos immune s a us. The in lamma ion le el esul s om he in e ac ion o hos , pa hogen, and ex insic ac o s. Ve y low and high in lamma ion le els o en associa ed wi h se e e ac i e TB, while balanced immune esponses associa ed wi h mild ac i e TB, la en TB, and possibly TB clea ance. E idence suppo s ha hos -di ec ed he apies (see Table1) ha e he po en ial o success ully modula e in lamma ion and amelio a e disease ou come, by ensu ing a p o ec i e immune esponse. 2 Bas os e al. In e ac ions and In lamma ion in TB F on ie s in Immunology | www. on ie sin.o g Janua y 2018 | Volume 8 | A icle 1948 Many known TB p ecipi a ing ac o s, which ei he inc ease he suscep ibili y o TB o he isk o ansi ion om LTBI o ac i e TB, a e connec ed wi h immune imbalances (4, 5). Howe e , he molecula mechanisms go e ning he ansi ions along he TB spec um emain unknown. The immune condi ion o in ec ed hos s is shaped by gene - ics, ex insic ac o s al e ing he local mic oen i onmen , and he he e ogenei y o he in ec ing bac e ia. This oika de e mines he h eshold o he immune esponse gene a ed du ing in ec ion and possibly he disease ou come (Figu e1). Modula ing hese h esholds and unco e ing he links be ween hos , pa hogen, and mic oen i onmen s should allow o he disco e y o solid co ela es o p o ec ion, molecula ma ke s o disease p ognosis, and he de elopmen o sa e and e ec i e hos -di ec ed he apies (HDTs) o TB. This e iew co e s ou cu en unde s anding o he impac o hese elemen s on TB, hei in e ac ions, and how hey may be u he explo ed as a pla o m o de eloping new p oduc s and s a egies agains TB. HOST geNeTiCS: ROLe iN iMMUNOMODULATiON The associa ion o single-gene inbo n e o s ela ed o in e e on (IFN)-γ immuni y wi h se e e childhood TB p o ides he clea - es gene ic basis o TB suscep ibili y (6, 7). These condi ions, globally named Mendelian suscep ibili y o mycobac e ial disease (MSMD) encompass a se ies o ge mline mu a ions in se en au osomal (IFNGR1, IFNGR2, IL12B, IL12RB1, STAT1, IRF8, and ISG15) and wo X-linked (NEMO and CYBB) genes (6, 7). These de ec s a e unc ionally and physiologically ela ed, as hey all esul in an impai men o he CD4 Tcell-media ed immuni y and ha e p o ided decisi e e idence on he c i ical p o ec i e ole o he in e leukin (IL)-12/IL-23/IFN-γ loop in TB (6, 7). This ole was u he con i med in he con ex o seconda y immunode iciencies, such as in human immunode iciency i us (HIV) in ec ion, as discussed below. Gene ic associa ion s udies wi h adul pa ien s showed mo e limi ed success han MSMD, and s ikingly no consis en asso- cia ion o a ian s o genes om he IL-12/IFN-γ axis wi h TB suscep ibili y in adul hood was ound. Ins ead, candida e-based s udies ound a numbe o gene ic a ian s associa ed wi h TB suscep ibili y in humans (7–9). Howe e , esul s emain incon- sis en and ha e no been alida ed in di e en popula ions no in genome-wide associa ion s udies (GWAS) (10–12). This mos likely e lec s he associa ion o adul pulmona y TB wi h com- plex gene ic ai s, whe e he ole o gene ic–ex insic ac o s and gene–gene in e ac ions (epis asis) domina e o e single polymo - phisms on hei own (13, 14). The iden i ica ion o gene ic isk ac o s is also likely masked by he expe imen al design, whe e impo an con ibu o s, such as ex insic ac o s, o pa hogen a iabili y a e la gely neglec ed. Fu he mo e, he ull spec um o TB has been mos ly igno ed in he g oup de ini ion, wi h all pheno ypes being analyzed oge he in wo main s udy g oups: ac i e TB e sus heal hy con ols/LTBI. Finally, combining and in eg a ing gene ic associa ion s udies wi h he in es iga ion o he human epigenome will ce ainly lead o c i ical insigh s in o he gene ic basis o in ec ion and clinical TB. FigURe 2 | Ex insic ac o s associa ed wi h ac i e ube culosis (TB). This igu e depic s ex insic ac o s associa ed wi h p o ec ion o ac i e TB (g een) o suscep ibili y/inc eased isk ( ed) o ac i e TB de elopmen . 3 Bas os e al. In e ac ions and In lamma ion in TB F on ie s in Immunology | www. on ie sin.o g Janua y 2018 | Volume 8 | A icle 1948 THe MODULATiON OF THe iMMUNe ReSPONSe BY eXTRiNSiC FACTORS The lack o clea associa ion o human geno ypes wi h TB suscep- ibili y and he ac ha p og ession om LTBI o pulmona y TB in adul s usually e lec s an impai men o hos esis ance due o non-gene ic ac o s, highligh ing he ele ance o ex insic ac- o s in shaping he hos immuni y wi h an impac on TB ou come (Figu e2). Coin ec ions A majo d i e o he cu en TB epidemics has been he HIV syn- demic, which d ama ically dec eases he hos p o ec i e esponses o TB in a CD4 Tcell coun -dependen manne (15), leading o an accele a ion o bo h diseases. Because he isk o de eloping TB is la gely inc eased in HIV-in ec ed indi iduals e en be o e CD4 Tcell coun s dec ease (16), o he ac o s mus also play a ole. Indeed, a de imen al al e na i e ac i a ion o mac ophages, accompanied by less ni ic oxide (NO) syn hase exp ession and poo ly o med g anulomas was desc ibed in HIV–TB, which in u n down egula ed he Mycobac e ium ube culosis DosR egu- lon (17), a se o genes known o be induced du ing anae obic do mancy (18). The e o e, changes in he hos immuni y esul ing om HIV coin ec ion emodel he bac e ial physiology, u he ewi ing he hos issue mic oen i onmen . Consequen ly, he pa hogenesis o TB is di e en in HIV-coin ec ed indi iduals, esul ing in he lack o comple e ca i a ion and in a highe inci- dence o dissemina ed disease (19, 20). O he less s udied coin- ec ions wi h impac on TB include helmin hs, in luenza, and Helicobac e pylo i. In TB pa ien s coin ec ed wi h helmin hs, a mo e ad anced o m o disease was epo ed (21), possibly ela ed wi h dec eased T helpe (Th) 1 and Th17cell esponses and inc eased sec e ion o IL-10 (22, 23). In he case o in luenza, he inc eased suscep ibili y o M. ube culosis in ec ion is likely media ed by ype I IFN signaling (24). Con as ingly, in ec ion wi h he ubiqui ous bac e ia H. pylo i may help o a oid p og es- sion o ac i e TB in la en indi iduals, due o enhanced IFN-γ and o he Th1-like cy okine esponses gene a ed in esponse o H. pylo i and ha es ain M. ube culosis (25). Ano he example o coin ec ion c oss alk in TB comes om he mouse model, in which p io Helicobac e hepa icus coloniza ion impai ed he immune con ol o M. ube culosis (26). I is in e es ing ha o hese h ee in ec ions, only in luenza is also a lung disease, hus indica ing ha dis an e en s shape he lung mic oen i onmen . The Hos Mic obiome Di e en mouse models o in ec ion deple ed o commensal gu mic obio a a e an ibio ic ea men showed an inc eased isk o coloniza ion by espi a o y pa hogens, such as S ep ococcus pneumoniae (27), S aphylococcus au eus (28), and Klebsiella pneumoniae (29). Al e a ions in he gu mic obio a also al e he suscep ibili y o TB (30), and an inc ease in he lung 4 Bas os e al. In e ac ions and In lamma ion in TB F on ie s in Immunology | www. on ie sin.o g Janua y 2018 | Volume 8 | A icle 1948 bac e ial bu den ea ly pos M. ube culosis in ec ion was epo ed in ge m- ee mice (31). Limi ed a ailable da a sugges ha he gu mic obio a o mice in ec ed wi h M. ube culosis is d as ically educed a e ini ial in ec ion (32). Gu di e si y in M. ube culosis-in ec ed mice is eco e ed abou he ime ha he adap i e immune sys em is onse , al hough signi ican di - e ences in axa composi ion emain in p e- and pos in ec ion samples (32). In addi ion, expe imen emo al o he gu mic o- bio a wi h an ibio ic ea men leads o highe suscep ibili y o M. ube culosis in ec ion in mice (30). How he in es inal mic o- bio a dis ally a ec s pulmona y immuni y, and e en ually he cou se o in ec ion and disease, and how he gu –lung axis may impac TB awai u he esea ch. We a e also s ill a om unde s anding he composi ion and he impac o changes in he o al–nasal ca i y and lung mic obio a on TB. M. ube culosis es ablishes in ec ion in he lowe espi a- o y ac , and as such has o ini ially e ade mic obio a-ac i a ed mac ophages o he uppe espi a o y ac (33). I emains o be seen i he p esence o absence o ce ain mic obial species in he uppe espi a o y ai way gene a es a mo e pe missi e en i onmen o he es ablishmen o TB in ec ion. I his is he case, es o ing key playe s o he mic obio a may help o igh in asion o e en imp o e immuni y, he apeu ic possibili ies no ye explo ed in TB in ec ion. Non-Communicable Como bidi ies Among non-communicable como bidi ies, he p esence o diabe es emains he majo isk ac o o TB. Owing o he dimension o he diabe es epidemics, i s o eseen impac on he global numbe s o TB cases is ala ming. Diabe ic pa ien s ha e a h ee imes highe isk o de eloping TB han heal hy indi iduals (34). Diabe es also wo sens disease se e i y (35), is a isk ac o o dea h in TB pa ien s (36, 37), and is associa ed wi h inc eased ailu e o s anda d TB ea men (35, 38, 39). P e ious s udies sugges ha he in e ac ion o M. ube culosis wi h mac ophages and dend i ic cells (DCs) is impai ed in he con ex o diabe es, leading o an ini ial hypo-in lamma o y s a e (40–43). Once he in ec ion is es ablished, he e is e idence o inc eased in lamma ion in TB/diabe es pa ien s, as an augmen ed le el o p o-in lamma o y cy okines in he pe iphe al blood is measu ed, likely due o hype ac i e Tcell esponses (41–43). I is possible ha many o he al e a ions seen and hei impac on TB a e ac ually in e connec ed o changes in he composi ion o he human gu mic obio a imposed by diabe es (44). Subjec s wi h ch onic obs uc i e pulmona y disease (COPD) p esen g ea e isks o de eloping ac i e TB (45, 46) and TB dea h as compa ed wi h TB pa ien s wi hou his como bidi y (36, 45). Al e ed immune esponses likely unde lie he mechanisms linking COPD and TB. COPD is cha ac e ized by a dis up ion o inna e de ense mechanisms in he ai ways, including dec eased mucocilia y clea ance and impai ed mac ophage phagocy osis (47, 48). Fu he mo e, an accumula ion o lung egula o y Tcells and he inc ease o ci cula ing IL-10 and TGF-β (49) we e desc ibed in COPD pa ien s. Thus, COPD limi s he e ec o unc ion o Tcells in esponse o pa hogens, which may explain he inc eased suscep ibili y o lowe espi a o y ac bac e ial in ec ions, including M. ube culosis. en i onmen al Fac o s Smoking exposu e is an independen isk ac o o M. ube - culosis in ec ion, p og ession o ac i e disease, and o poo ea men ou comes (50–52). The unde lying immunological mechanisms a e jus s a ing o be un eiled and include educed p oduc ion o umo nec osis ac o (TNF), IL-1β, and IFN-γ by in i o in ec ed al eola mac ophages (53), dec eased numbe o DCs (54, 55), and comp omised ec ui men o IFN-γ-p oducing CD4 Tcells o he lung, hus weakening he o ma ion o g anu- loma (56). Mo e ecen ly, al eola mac ophages om smoke s we e ound o exhibi lysosomal accumula ions o obacco smoke pa icula es, which impai ed hei mig a ion owa d M. ube culosis-in ec ed cells (57). Malnu i ion has also been associa ed wi h an inc eased isk o ac i e TB, al hough i emains unclea whe he he nu i ional s a us is a cause o a consequence o he disease. A s ong link be ween TB, malnu i ion, and immune dys egula ion is in place (58). This associa ion is e en wo se in he amewo k o HIV in ec ion (59). Nu i ional s a us no only a ec s he unc ion o se e al immune cells, including Tcells, bu addi ionally impac s he pha macodynamics o he d ugs (60). Less clea is he ole o anemia, a common symp om, and p ognosis ma ke o TB (36), pa icula ly linked o i on dep i a ion, in acili a ing o exace ba ing TB disease. An eme ging ole is being ecognized o hepcidin, a p o ein ha egula es he homeos asis and cell ype dis ibu ion o i on in he body and ha plays a ole on inna e immune esponses o mycobac e ial in ec ion (61, 62). In ag eemen , in epidemiological s udies hepcidin le els ha e been posi i ely co ela ed wi h inc eased isk o mo ali y in TB–HIV coin ec ion usually associa ed wi h mo e dea h- h ea ening mani es a ion o he disease, as ex apulmona y and milia y TB (63). Ano he nu ien , i amin D, has been he ocus o enewed a en ion by esea che s. His o ically, bo h i amin D and exposu e o sunligh , which endogenously p omo es he con e sion in he skin o 7-dehyd ocholes e ol in o p e- i amin D3, we e used in he ea men o TB (64). Insu iciency o his molecule has been linked o highe isk o ac i e TB (65, 66) and inc eased p opensi y o ex apulmona y in ol emen (67). The immunomodula o y ole o i amin D is well es ablished leading o se e al changes in immune esponses, including he induc ion o ca helicidin an imic obial pep ide, be a-de ensin, and he p omo ion o au hophagy and/o bac e ial killing (68, 69). Howe e , i is impo an o men ion ha despi e se e al epo s on he p o ec i e ole o au ophagy du ing M. ube cu- losis in ec ion (70), a ecen s udy based on a gene ic app oach a ge ing mul iple au ophagy- ela ed genes concluded ha he cellula au ophagic capaci y did no co ela e wi h he ou come o M. ube culosis in ec ion (71). The e o e, i is possible ha he impac o au ophagy du ing M. ube culosis in ec ion esul s om he use o in i o models, hus calling o u he in i o s udies when co ela ing p o ec i e media o s wi h induc ion o au ophagy. Vi amin D le els ha e been shown o impac adap- i e immune esponses by in luencing Th cell unc ion and by p omo ing T egs (72). I hus seems ha i amin D p omo es he mac ophage e ec o unc ion, while a he same ime keeping he immune esponse a check h ough i s ac ion on Tcells. Se e al clinical ials ha e shown ha supplemen a ion o d ug egimens FigURe 3 | Le els o gene ic di e si y ac oss he Mycobac e ium ube culosis complex (MTBC) and i s epidemiological and clinical impac . The e a e di e en le els o di e si y ac oss he MTBC. Wi hin a hos o ansmission chain, M. ube culosis isola es ypically di e in less han 25 single-nucleo ide polymo phisms (SNPs). Di e si y inc eases when compa ing isola es wi hin he same lineage (a ound 25–1,000 SNPs) o wi hin di e en lineages o he MTBC (a ound 1,000–2,000 SNPs). This di e si y impac s hos /pa hogen in e ac ions, pa icula ly he in ensi y and quali y o he immune esponse and he clinical ou come, a he le els o d ug acquisi ion, adap a ion o di e en popula ions, ansmissibili y, o disease mani es a ion. 5 Bas os e al. In e ac ions and In lamma ion in TB F on ie s in Immunology | www. on ie sin.o g Janua y 2018 | Volume 8 | A icle 1948 wi h i amin D does no imp o e TB ou comes in he gene al popula ion (73), bu i has been shown o accele a e spu um con e sion in pa ien s wi h he geno ype o he Taql i amin D ecep o polymo phism (74). I emains o be seen i i amin D can play a ole in p e en ing in ec ion o p og ession o ac i e TB. I is possible ha h ough he modula ion o i amin D, a ia ions in sunligh may unde lie some o he di e ences in TB incidence a e ac oss he globe (65). This no only includes na u al seasonal a ia ions bu also a i icial a ia ion due o human-associa ed ac i i ies, such as pollu ion. M. ube culosis Di eRSiTY AND iMMUNOMODULATiON: PARADigM CHANgiNg e iDeNCe How a iable is M. ube culosis? Tube culosis is caused by a g oup o phylogene ically closely ela ed bac e ia, collec i ely known as he M. ube culosis com- plex (MTBC), now known o encompass se en main phylogene ic lineages o human-adap ed bac e ia (75). Wi hin his complex, M. ube culosis and Mycobac e ium a icanum (in Wes A ica) a e esponsible o he la ge majo i y o human cases o TB (76). The gene ic di e si y wi hin he MTBC is highe han o iginally expec ed and can be obse ed a di e en e olu iona y and geog aphical scales (Figu e3). Mos o he di e si y obse ed is likely due o gene ic d i , i.e., s ochas ic a ia ion o di e si y due o limi ed popula ion sizes, o o neu al a ia ion wi h no impac on he i ness o he bac e ia (77). Di e si ica ion o he ini ial in ec ing bac e ia in subpopula ions wi hin a single pa ien has also been epo ed (78), pa o i is likely due o an ibio ic selec ion p essu es (79, 80). Al hough s ill unclea , addi ional di e si y maybe selec ed independen ly o an ibio ic p essu e (80). I is possible ha some o he di e si y wi hin he hos e lec s he he e ogeneous immune esponses associa ed wi h di e en lung lesions (81). The complexi y o he lung mic os uc u e is now acknowledged, wi h he immune esponse being spa ially sepa a ed e en in single g anulomas (82). wha e olu iona y Fo ces Shape MTBC Di e si y? In he long- e m e olu ion o he pa hogen, he main d i ing o ces shaping MTBC di e si y we e a balance be ween gene ic d i , du ing he global expansion o he bac e ia and hos - o- hos ansmission, and posi i e and pu i ying selec ion (77, 83). A long pe iod o pa allel e olu ion o he pa hogen wi h di e en popula ions o mode n humans ollowing he ou -o -A ica mig a ions (84) esul ed in di e en bac e ial lineages o p e ail in di e en geog aphical egions o he wo ld. Echoes o his ype o in e ac ion a e s ill obse ed in cosmopoli an se ings, 6 Bas os e al. In e ac ions and In lamma ion in TB F on ie s in Immunology | www. on ie sin.o g Janua y 2018 | Volume 8 | A icle 1948 whe e he phylogeny o he in ec ing MTBC isola es co ela es wi h he pa ien ’s e hnic o igin (85, 86). A ecen s udy showed ha e en wi hin one speci ic MTBC lineage, i.e., Lineage 4 (o Eu o-Ame ican), a phylogeog aphical s uc u e can be ound, wi h some sublineages being geog aphically es ic ed and o he s occu ing globally (87). Whe he he immune sys em is also a d i ing o ce o M. ube culosis di e si y emains unclea . On one hand, ex ao - dina y insigh s can be gained, as shown by he iden i ica ion o single nucleo ide polymo phisms (SNPs) in he wo-componen egula ion sys em PhoP/PhoR impac ing Lineage 6 and Mycobac e ium bo is s ains (88). On he o he hand, while mos pa hogens ha e e ol ed o e ade hos immuni y by an igenic a ia ion, M. ube culosis seems o apply a di e en s a egy, hus ques ioning he ole o Tcells in d i ing M. ube culosis e olu- ion. Genome sequencing o M. ube culosis showed ha he known human Tcell epi opes a e e olu iona ily hype conse ed, wi h he la ge majo i y o indi idual epi opes analyzed showing no amino acid change a all (89). I is in e es ing o no e ha epi opes appea o be signi ican ly mo e conse ed han he mean o he genome, and his esul is obus o he numbe o s ains analyzed (89, 90). Fu he mo e, hese hype conse ed epi opes all on genes ha a e no biased owa d pa icula unc ional ca ego ies, bu in which he majo common heme is ha hey encode pep ides ha a e ecognized by he immune sys em (90). This hypo hesis s ill awai s unc ional and expe imen al con i - ma ion, bu based on insilico analyses i is emp ing o specula e ha , despi e he gene al conse a ion o he M. ube culosis genome as compa ed wi h o he bac e ia, Tcell esponses may d i e selec i e o ces owa d he hype conse a ion o epi opes. The ele ance o hos Tcell esponses o he na u al his o y o M. ube culosis in ec ion comes om indings in HIV+ pa ien s, showing ha less equen ca i a ion, in a CD4 Tcell-dependen equency, associa es wi h lowe ansmissibili y (as in e ed by looking a he numbe o household con ac s in ec ed by an HIV+ pa ien ) (15). The e o e, i is concei able ha he Tcell esponse ha locally ies o elimina e he bacilli, which esul s in ca i a ion and hus in ansmission. Ensu ing conse a ion o T cell esponses could hus be a mechanism explo ed by M. ube culosis o ensu e ansmission. None heless, ou lie epi opes o he gene al ule o hype conse a ion ha e been desc ibed bo h in p e iously known and newly iden i ied an i- genic egions (89, 91). Mo eo e , amino acid subs i u ions in hese a iable epi opes we e shown o impac he hos esponse wi h some pa ien s esponding only o he wild- ype epi ope a i- an and o he s only o he mu a ed o ms (89). This is impo an , as he al e a ion o a single amino acid in an epi ope can impac i s a ini y o a speci ic HLA molecule, hus in luencing he Tcell synapse and modula ing he le el and ype o immune esponse elici ed (92, 93). These a iable epi opes could be po en ially exploi ed as accine componen s o inc ease he p o ec ion p o ided by he exis ing TB accine. impac o Bac e ia geno ypes on immune Responses and Clinical Ou comes Mycobac e ium ube culosis complex di e si y impac s he hos immune esponse, as ce ain clinical isola es a e mo e po en han o he s in inducing he sec e ion o immune media o s by in ec ed monocy es (94–99) and in expe imen al in ec ions (100–104). The impac o bac e ial di e si y is no only e lec ed a he le el o soluble immune media o s bu also in he abili y o i ulen mycobac e ia o inhibi apop osis, while igge ing nec osis o hos mac ophages o p omo e an inna e delay in he ini ia ion o adap i e immuni y (105). Clea ly, unbalancing he immune esponse is a s a egy used by M. ube culosis o inc ease i s i ulence, and his s a egy migh be modula ed di e en ly by di e se s ains. The ealiza ion o he MTBC di e si y in all i s ex en has also led o mul iple s udies explo ing he impac o his a ia ion on he clinical ou come o TB (Figu e3). Fo example, Lineage 2 s ains (which includes he Beijing amily) ha e been epea edly associ- a ed wi h ea men ailu e and elapse (106–108). One in i o s udy u he demons a ed ha Lineage 2 may acqui e d ug esis ance mo e apidly han Lineage 4 (109), in line wi h epo s om clinical se ings whe e mul id ug- esis an (MDR) isola es we e mo e likely o belong o he Beijing lineage han o Lineage 4 (110, 111). High ansmissibili y o Beijing amily Lineage 2 s ains was shown in San F ancisco (112), whe eas Lineage 3 was epo ed o be less ansmissible han Lineages 1, 2, and 4 in TB pa ien s om Mon eal, Canada (113), and s iking di e ences in ansmissibili y among sublineages o Lineage 5 in Benin and Nige ia we e epo ed (114). Lineages 2 and 4 we e shown o ha e a highe a e o p og ession o ac i e TB as compa ed wi h Lineage 6 in he Gambia (115). In addi ion, mo e debili a - ing symp oms (such as weigh loss) ha e been associa ed wi h Lineage 4 s ains in Tanzania (116). The same lineage was linked mainly o pulmona y TB (117), while Lineages 2 (117–119) and 3 (120) we e epo ed o associa e wi h ex apulmona y disease. S ikingly, wi hin-hos bac e ial di e si y also seems o con ibu e o disease mani es a ion. Indeed, he p esence o dis inc bac e ial subpopula ions is associa ed wi h poo clinical ou comes (121), esul ing in di e en ial esolu ion o g anulomas (122), ha may e en con as he o e all end o disease p og ession (123–125). BACTeRiA AND HOST geNOTYPe iNTeRACTiONS The e is eme ging e idence showing an impac o hos –pa hogen geno ype in e ac ions in hos immune esponses, TB ansmission, and disease p esen a ion (76). This is he case o he associa ion obse ed be ween he p esence o he T597C allele o he oll-like ecep o (TLR) 2 gene and suscep ibili y o dissemina ed disease, in a Vie namese popula ion upon in ec ion by MTBC Lineage 2 (117). Th ee s udies conduc ed in Ghana u he suppo geno ype–geno ype in e ac ions. Fi s , he a ian G57E o he mannose-binding Lec in (P o ein C) 2 (MBL2) gene was associ- a ed wi h TB caused by M. a icanum, bu no by M. ube culosis sensu s ic o (126). Second, he a ian 261TT o he immuni y- ela ed GTPase M (IRGM) gene was p o ec i e agains TB caused by Lineage 4, bu no o disease caused by o he MTBC lineages (127). Thi d, M. a icanum was signi ican ly mo e common in TB pa ien s belonging o he Ewe e hnic g oup, an associa ion mainly d i en by Lineage 5 (128, 129). In a Sou h A ican popula ion o mixed ances y, an associa ion be ween di e en HLA class I TABLe 1 | E idence suppo ing HDTs o TB. HDT mechanism examples o po en ial HDTs agen s e idence on hos e ec Re e ence Reducing excessi e issue damaging in lamma ion Ibup o en (NSAIDs)aInhibi s p os aglandin p oduc ion by inhibi ing cyclooxygenase. Reduces lung pa hology and Mycobac e ium ube culosis bu den in mouse models Vilaplana e al. (135) Zileu on (leuko iene syn hesis inhibi o s)aInhibi s lipoxygenase ac i i y, blocking leuko iene p oduc ion, and inc easing PGE2 le els. P e en s ype I IFN-d i en acu e mo ali y o M. ube culosis-in ec ed mice Maye -Ba be e al. (136) To aci inib ( y osine kinases inhibi o s)aJAK blocke wi h an i-in lamma o y p ope ies (JAK/STAT pa hway is downs eam he ac i a ion o mos cy okine ecep o s), sho ens he ime equi ed o lung s e ili y in a ch onic TB mouse model Maiga e al. (137) Adalimumab (an i-TNFα)bLi e- h ea ening pulmona y TB a ibu able o he eco e y o TNF-dependen in lamma ion caused by wi hd awal o adalimumab. Lung in lamma ion wo sened despi e clea ance o iable M. ube culosis om spu um and lung issue by an imic obial he apy. Clinical imp o emen did no occu un il adalimumab ea men was esumed Wallis e al. (138) P ednisolone (glucoco icoids)dModula e ex eme immunopa hological eac ions and imp o ed mo ali y o TB pe ica di is and meningi is. Possible bene i in pulmona y TB. Adjunc i e ea men wi h co icos e oids may imp o e he clinical ou come and may accele a e spu um smea con e sion om HIV coin ec ed pa ien s E ans (139); C i chley e al. (140); Bilaçe oğlu e al. (141); Mayanja-Kizza e al. (142) Modula ing inna e and adap i e immune esponses Sim as a in (s a ins)aInhibi s he 3-hyd oxy-3-me hylglu a yl coenzyme educ ase, educing he choles e ol le els wi hin phagosomal memb anes, which p omo es phagosomal ma u a ion and au ophagy. Reduces bac e ial bu den in human PBMCs and MDMs. Imp o es his opa hologic indings, wi h educed lung M. ube culosis bu dens in expe imen al mu ine in ec ion Pa iha e al. (143) Ca bamazepine (an icon ulsan s)aSodium-channel blocke , capable o enhancing au ophagic killing o in acellula M. ube culosis in mac ophages h ough cellula myoinosi ol deple ion. In mice in ec ed wi h a highly i ulen MDR s ain, ca bamazepine ea men educed bac e ial bu den, imp o ed lung pa hology, and s imula ed adap i e immuni y Schieble e al. (144) Me o min (biguanides, an idiabe ic d ugs)cIn e up s he mi ochond ial espi a o y chain, inc eases p oduc ion o mi ochond ial eac i e oxygen species, and acili a es phagosome–lysosome usion, leading o enhanced killing o in acellula M. ube culosis. In he mouse model, Tcell esponses and he e icacy o con en ional TB d ugs a e imp o ed, wi h esul an educed lung pa hology. In wo sepa a e human coho s, me o min associa es wi h dec eased TB se e i y and imp o ed clinical ou come in ac i e TB and is associa ed wi h enhanced M. ube culosis-speci ic Tcell immune esponse in LTBI Singhal e al. (145) Vi amin D3dInduces he gene exp ession o be a-de ensin 2 and human ca helicidin LL-37 ha a e able o supp ess he g ow h o M. ube culosis and modula e an imic obial esponses. Adjunc he apy wi h i amin D3 enhanced in acellula mycobac e ial killing in mac ophages, inc eased spu um cul u e con e sion, and educed clinical symp oms in TB pa ien s Mily e al. (146); Rahman e al. (147) Immune checkpoin inhibi ion Ni olumab and pemb olizumab (an i-PD-1)aPD-L1 gene exp ession is ele a ed in pa ien s wi h ac i e TB disease. Human gene exp ession o PD-1 and PD-L1 in whole-blood dec ease du ing success ul TB ea men . In ec ions wi h li e M. ube culosis up egula ed PD-L1 exp ession on monocy es. In i o PD-1 blocking escued M. ube culosis-speci ic IFN-γ-p oducing Tcells om unde going apop osis. PD-1 blockade po en ia es he speci ic deg anula ion o CD8+ Tcells Singh e al. (148); Ju ado e al. (149); Hassan e al. (150) Immune ac i a ion, cy okine he apy Recombinan human IFN-γbIFN-γ adminis a ion in a pa ien wi h MSMD caused by IL-12Rβ1 de iciency p o ided a no iceable clinical e ec , wi h no addi ional ad e se e ec s Alanga i e al. (151) (Con inued) 7 Bas os e al. In e ac ions and In lamma ion in TB F on ie s in Immunology | www. on ie sin.o g Janua y 2018 | Volume 8 | A icle 1948 HDT mechanism examples o po en ial HDTs agen s e idence on hos e ec Re e ence Cell-based he apy Au ologous BM-MSCsdBM-MSCs ha e immunomodula o y p ope ies ha can educe damaging in lamma ion, induce issue egene a ion, and es o e p oduc i e immune esponses. Single-dose au ologous BM-MSC is a sa e adjunc he apy o pa ien s wi h MDR o XDR-TB in combina ion wi h s anda d d ug egimens and econs i u ed an i-M. ube culosis Tcell esponses in a phase 1 ial Sk ahin e al. (152) An imic obial-po en ia ing e ec Ve apamil (calcium-channel blocke s)aBlocks e lux pump, esul ing in highe in acellula an imycobac e ial d ug le els and enhanced d ug ac i i y. Accele a es bo h he bac e icidal and he s e ilizing ac i i ies o he egimen in a mouse model. Adjunc i e use o e apamil dec eases he MIC o bedaquiline in he wild- ype s ain M. ube culosis H37R and also in d ug-suscep ible and d ug- esis an clinical isola es. Po en ia es he ac i i y o bedaquiline agains M. ube culosis in an in i o mouse model. Pe mi s lowe doses o bedaquiline and he eby educe i s dose- ela ed oxici ies Gup a e al. (153) BM-MSCs, bone ma ow-de i ed mesenchymal s omal cells; MDMs, monocy e-de i ed mac ophages; MDR, mul id ug esis an ; MIC, minimum inhibi o y concen a ion; PBMCs, pe iphe al blood mononuclea cells; NSAIDs, non- s e oidal an i-in lamma o y d ugs; XDR, ex ensi ely d ug esis an ; TB, ube culosis; LTBI, la en TB in ec ion; HDT, hos -di ec ed he apy; IFN, in e e on; MSMD, Mendelian suscep ibili y o mycobac e ial disease; IL, in e leukin; TNF, umo nec osis ac o ; PGE2, p os aglandin E2; HIV, human immunode iciency i us. Following a e he ypes o s udies. aP eclinical. bCase epo s. cObse a ional s udies. dRandomized ials. TABLe 1 | Con inued 8 Bas os e al. In e ac ions and In lamma ion in TB F on ie s in Immunology | www. on ie sin.o g Janua y 2018 | Volume 8 | A icle 1948 ypes and disease caused by di e en MTBC s ain amilies was epo ed (130). Polymo phisms in he mac ophage ecep o wi h collagenous s uc u e (MARCO) gene, a ecep o in ol ed in M. ube culosis phagocy osis (131), p e e en ially associa ed wi h Lineage 2 o e Lineage 1 o 4, implying ha he hos MARCO geno ypes may in e ac wi h M. ube culosis o he Lineage 2 geno ype o inc ease suscep ibili y o TB (132). The i s epo o a M. ube culosis lineage-based GWAS was ecen ly published (133). In his s udy, an SNP on ch omosome 1p13, nea he CD53 gene, was speci ically associa ed wi h non-Beijing lineage-in ec ed old age onse cases (133). Al oge he , hese s udies demons a e ha in e ac ions be ween bac e ial and human gene ic loci exis and join ly in luence clinical pheno ypes. This e idence calls o he need o in eg a ing he pa hogen geno ype in human gene ic associa ion s udies, as well as in he s udy o TB immuni y. RewiRiNg THe iMMUNe ReSPONSe iN TB: THe NeeD TO iNCORPORATe Di eRSiTY iN HOST, PATHOgeN, AND eN iRONMeNT Hos -di ec ed he apies a e gaining momen um in he ield o TB ea men (Table1). HDTs aim a modula ing hos in lamma ion as a way o imp o e he e icacy o cu en ea men s, while sho ening he du a ion o hese ea men s, lowe ing oxici y and dec easing a es o esis ance acquisi ion (134). A pa icula bene i o HDTs would be hei applica ion o ea MDR- and ex ensi ely d ug esis an -TB, whe e an ibio ics ha e limi ed e ec i eness and immunopa hological in lamma ion, issue damage and high a ali y a e obse ed. Al hough HDTs ake ad an age o immunomodula o y mechanisms, hey a ely accoun o he in e ac ion be ween he di e si y o he bac e ia, he hos , and he en i onmen . Howe e , in some clinical si es, indi ec da a sugges ha success ul HDTs ha e o be ailo ed owa d hese a iables. Fo example, he e is a la ge amoun o da a showing Lineage 2 s ains as being low-cy okine induce s, bu po en induce s o ype I IFN (97, 98, 104, 154), which is a leas in pa media ed by he di - e en ial ac i a ion o TLR2 and TLR4 ecep o s (98). In his way, Lineage 2 s ains may sub e an i ube culous hos de enses by inhibi ing he enzyme inducible NO syn hase, as well as IL-1β, IL-18, and IL-12p40, while inducing he immunosupp essi e media o s IL-10 and IL-1 ecep o an agonis (4, 155). In addi- ion, he e a e da a om TB pa ien s in Vie nam showing an associa ion be ween indi iduals wi h a T597C allele in he TLR2 gene and suscep ibili y o in ec ion by he Lineage 2/Beijing geno ype (117); and an associa ion o indi iduals wi h an SNP in he leuko iene A4 hyd olase (LTA4H) p omo e wi h an excess o in lamma ion and TB se e i y (156). How all he combina ions o human/pa hogen geno ypes in e ac in Vie nam emains o be elucida ed, bu i will su ely impac he ou come o HDTs in TB, as well as eaching impo an lessons. Fu he mo e, HDTs will need o deal wi h ou pa ial unde s anding o wha cons i u es p o ec i e immuni y o TB. Fo example, he ole o ype I IFNs is now being econside ed. A ecen s udy showed ha in ce ain scena ios, namely, in he absence o IFN-γ, he induc ion o ype 9 Bas os e al. In e ac ions and In lamma ion in TB F on ie s in Immunology | www. on ie sin.o g Janua y 2018 | Volume 8 | A icle 1948 I IFN is ac ually p o ec i e o he hos , as i allows he con ol o he swi ch o M1 (e ec o ) o M2 (de imen al) mac ophages (157). The ac ha ype I IFN plays dynamic oles du ing in ec- ion (157, 158) may ha e implica ions in he use o eicosanoid modula o s o enhance p os aglandin E2 le els and dec ease he un a o able ype I IFN esponse (136). So, HDTs designed o modula e excessi e ype I IFN esponse, like he one associa ed wi h Lineage 2 s ains, will need o conside hos -speci ic cha - ac e is ics, such as a high o low abili y o di e en ia ing IFN-γ esponses, as well as he ime o in e en ion. CONCLUSiON The hos immune s a us is igh ly linked o he spec um o TB in ec ion and disease. Howe e , he molecula de e minan s b idging in lamma o y h esholds and TB ou comes emain elusi e. Likely, his is due o he ac ha wha inally dic a es disease ou comes and ansmission is no a single ac o , bu he in e ac ing (an agonis ically o syne gis ically) ac ion o mul iple ac o s. Di e si y in hos , pa hogen, and ex insic ac o s needs o be s udied in conce a he han indi idually, so ha he ull ex en o he biological in e play unde lying he immune esponse can be cap u ed and modeled. In he long e m, hese models will lead he disco e y o solid co ela es o p o ec ion, bioma ke s o p ognosis, he apeu ic a ge s, and mo e accu a e epidemiol- ogy models. The g ea challenge is now o in eg a e his oika o in e ac ions in he de elopmen o HDTs, a a pe sonalized le el, o de elop s e ilizing he apies. Mos likely, we will be soon alking abou pe sonalized medicine o ea TB, by ewi ing he immune esponse h ough hos –pa hogen–en i onmen di ec ed he apies. Only hen will we be able o ansla e he many yea s o esea ch de o ed o he s udy o he p o ec i e immune esponse in o eal clinical applica ions, i.e., be e accines, he apeu ics and no el bioma ke s o p ognosis. AUTHOR CONTRiBUTiONS All au ho s concei ed and w o e he pape . SG, IC, and MS coo dina ed and comple ed he inal e sion. FUNDiNg HN-B acknowledges he eceip o esea ch schola ships om Bolsa D. Manuel de Mello and he Po uguese Socie y o Pneumology. NO acknowledges FCT IF/00474/2014. SG is unded by he Eu opean Resea ch Council (g an numbe 309540-EVODRTB); he Swiss Na ional Science Founda ion (g an numbe 310030_166687) Sys emsX.ch. IC lab is inanced by Minis e io de Economía y Compe i i idad (Spanish Go e nmen ) esea ch g an SAF2013-43521-R, SAF2016- 77346-R, and PROMETEO/2016/122 om Gene ali a Valenciana and he Eu opean Resea ch Council (ERC) (638553-TB-ACCELERATE). The MS lab is inanced by FEDER—Fundo Eu opeu de Desen ol imen o Regional unds h ough he COMPETE 2020—Ope acional P og amme o Compe i i eness and In e na ionalisa ion (POCI), Po ugal 2020, and by Po uguese unds h ough FCT in he amewo k o he p ojec “Ins i u e o Resea ch and Inno a ion in Heal h Sciences” (POCI-01-0145-FEDER-007274). MS is an FCT P incipal In es iga o . Some g aphical elemen s in he igu es we e designed by kjpa ge e /F eepik. The unding agencies had no ole in he design o he manusc ip . ReFeReNCeS 1. WHO. Global Tube culosis Repo 2017. (2017). 2. Esmail H, Lai RP, Lesosky M, Wilkinson KA, G aham CM, Coussens AK, e  al. Cha ac e iza ion o p og essi e HIV-associa ed ube culosis using 2-deoxy-2-[18F] luo o-d-glucose posi on emission and compu ed omog- aphy. Na Med (2016) 22:1090–3. doi:10.1038/nm.4161 3. Zak DE, Penn-Nicholson A, Sc iba TJ, Thompson E, Suliman S, Amon LM, e  al. A blood RNA signa u e o ube culosis disease isk: a p ospec i e coho s udy. Lance (2016) 6736:1–11. doi:10.1016/S0140-6736(15)01316-1 4. O’Ga a A, Red o d PS, McNab FW, Bloom CI, Wilkinson RJ, Be y MPR. The immune esponse in ube culosis. Annu Re Immunol (2013) 31:475–527. doi:10.1146/annu e -immunol-032712-095939 5. Ba y CE, Bosho H, Da ois V, Dick T, Eh S, Flynn J, e al. The spec um o la en ube culosis: e hinking he goals o p ophylaxis. Na Re Mic obiol (2009) 7:845–55. doi:10.1038/n mic o2236 6. Bus aman e J, Boisson-Dupuis S, Abel L, Casano a JL. Mendelian suscep i- bili y o mycobac e ial disease: gene ic, immunological, and clinical ea u es o inbo n e o s o IFN-γ immuni y. Semin Immunol (2014) 26:454–70. doi:10.1016/j.smim.2014.09.008 7. Abel L, El-Baghdadi J, Bous iha AA, Casano a J-L, Schu E. Human gene ics o ube culosis: a long and winding oad. Philos T ans R Soc Lond B Biol Sci (2014) 369:20130428. doi:10.1098/ s b.2013.0428 8. Qu H-Q, Fishe -Hoch SP, McCo mick JB. Knowledge gaining by human gene ic s udies on ube culosis suscep ibili y. J Hum Gene (2011) 56:177–82. doi:10.1038/jhg.2010.164 9. Azad AK, Sadee W, Schlesinge LS. Inna e immune gene polymo phisms in ube culosis. In ec Immun (2012) 80:3343–59. doi:10.1128/IAI.00443-12 10. Cu is J, Luo Y, Zenne HL, Cuche -Lou enço D, Wu C, Lo K, e  al. Suscep ibili y o ube culosis is associa ed wi h a ian s in he ASAP1 gene encoding a egula o o dend i ic cell mig a ion. Na Gene (2015) 47:523–7. doi:10.1038/ng.3248 11. G an AV, Sab i A, Abid A, Abde ahmani Rho i I, Benki ane M, Souhi H, e  al. A genome-wide associa ion s udy o pulmona y ube culosis in Mo occo. Hum Gene (2016) 135:299–307. doi:10.1007/s00439-016-1633-2 12. Meye CG, Thye T. Hos gene ic s udies in adul pulmona y ube culosis. Semin Immunol (2014) 26:445–53. doi:10.1016/j.smim.2014.09.005 13. de Wi E, an de Me we L, an Helden PD, Hoal EG. Gene-gene in e ac ion be ween ube culosis candida e genes in a Sou h A ican popula ion. Mamm Genome (2011) 22:100–10. doi:10.1007/s00335-010-9280-8 14. Daya M, Van De Me we L, Van Helden PD, Mölle M, Hoal EG. In es iga ing he ole o gene-gene in e ac ions in TB suscep ibili y. PLoS One (2015) 10:e0123970. doi:10.1371/jou nal.pone.0123970 15. Kwan C, E ns JD. HIV and ube culosis: a deadly human syndemic. Clin Mic obiol Re (2011) 24:351–76. doi:10.1128/CMR.00042-10 16. Sonnenbe g P, Glynn JR, Fielding K, Mu ay J, God ey-Fausse P, Shea e S. How soon a e in ec ion wi h HIV does he isk o ube culosis s a o inc ease? A e ospec i e coho s udy in Sou h A ican gold mine s. J In ec Dis (2005) 191:150–8. doi:10.1086/426827 17. Wal e ND, De Jong BC, Ga cia BJ, Dolgano GM, Wo od ia W, Byanyima P, e al. Adap a ion o Mycobac e ium ube culosis o impai ed hos immuni y in HIV-in ec ed pa ien s. J In ec Dis (2016) 214:1205–11. doi:10.1093/in dis/ jiw364 18. Leis ikow RL, Mo on RA, Ba ek IL, F impong I, Wagne K, Voskuil MI. The Mycobac e ium ube culosis DosR egulon assis s in me abolic homeo- s asis and enables apid eco e y om non espi ing do mancy. J Bac e iol (2010) 192:1662–70. doi:10.1128/JB.00926-09 19. Died ich CR, Flynn JAL. HIV-1/Mycobac e ium ube culosis coin ec ion immunology: how does HIV-1 exace ba e ube culosis? In ec Immun (2011) 79:1407–17. doi:10.1128/IAI.01126-10