Stronger induction of trained immunity by mucosal BCG or MTBVAC vaccination compared to standard intradermal vaccination

A icle S onge induc ion o ained immuni y by mucosal BCG o MTBVAC accina ion compa ed o s anda d in ade mal accina ion G aphical Abs ac Highligh s dNonhuman p ima es ecapi ula e ained immuni y upon li e a enua ed TB accina ion dIn a enous BCG induces changes in H3K27 ace yla ion and enhances cy okine p oduc ion dMucosal BCG imp o es induc ion o ained immuni y o monocy es o e in ade mal BCG dThe M. ube culosis-de i ed candida e accine MTBVAC appea s equally po en as BCG Au ho s Michel P.M. Vie boom, Ka in Dijkman, Claudia C. Somb oek, ..., Nacho Aguilo, Ca los Ma in, F ank A.W. Ve eck Co espondence ie boom@bp c.nl In B ie Vie boom e al. demons a e he induc ion o ained immuni y in blood and bone ma ow monocy es a e accina ion wi h li e a enua ed TB accines in nonhuman p ima es. Mucosal espi a o y deli e y o BCG o MTBVAC induces ained immuni y mo e e icien ly compa ed o s anda d in ade mal accina ion. Vie boom e al., 2021, Cell Repo s Medicine 2, 100185 Janua y 19, 2021 ª2020 The Au ho (s). h ps://doi.o g/10.1016/j.xc m.2020.100185 ll A icle S onge induc ion o ained immuni y by mucosal BCG o MTBVAC accina ion compa ed o s anda d in ade mal accina ion Michel P.M. Vie boom, 1,8, *Ka in Dijkman, 1 Claudia C. Somb oek, 1 Sam O. Ho man, 1 Cha elle Boo , 1 Richa d A.W. Ve enne, 1 K is a G. Haans a, 1 Maa en an de Sande, 2 Liesbe h an Ems , 3 Jo ge Domı ´nguez-And e ´s, 3 Simone J.C.F.M. Moo lag, 3 Clemens H.M. Kocken, 1 Jelle Thole, 4 Es eban Rod ı ´guez, 5 Eugenia Puen es, 5 Joos H.A. Ma ens, 2 Reinou an C e el, 3 Mihai G. Ne ea, 3 Nacho Aguilo, 6,7 Ca los Ma in, 6,7 and F ank A.W. Ve eck 1 1 Biomedical P ima e Resea ch Cen e, Rijswijk, he Ne he lands 2 Depa men o Molecula Biology, Facul y o Science, Nijmegen Cen e o Molecula Li e Sciences, Radboud Uni e si y, Nijmegen, he Ne he lands 3 Radboud Uni e si y Medical Cen e, Nijmegen, he Ne he lands 4 TuBe culosis Vaccine Ini ia i e, Lelys ad, he Ne he lands 5 Bio ab i, Pon e ed a, Spain 6 Depa men o Mic obiology, Facul y o Medicine, IIS A ago ´n, Uni e si y o Za agoza, Za agoza, Spain 7 CIBERES, Ins i u o de Salud Ca los III, Mad id, Spain 8 Lead con ac *Co espondence: ie boom@bp c.nl h ps://doi.o g/10.1016/j.xc m.2020.100185 SUMMARY BCG accina ion can s eng hen p o ec ion agains pa hogens h ough he induc ion o epigene ic and me a- bolic ep og amming o inna e immune cells, a p ocess called ained immuni y. We and o he s ecen ly demons a ed ha mucosal o in a enous BCG be e p o ec s hesus macaques om Mycobac e ium ube culosis in ec ion and TB disease han s anda d in ade mal accina ion, co ela ing wi h local adap i e immune signa u es. In line wi h p io mouse da a, he e, we show in hesus macaques ha in a enous BCG enhances inna e cy okine p oduc ion associa ed wi h changes in H3K27 ace yla ion ypical o ained immu- ni y. Al e na i e deli e y o BCG does no al e he cy okine p oduc ion o un ac iona ed b onchial la age cells. Howe e , mucosal bu no in ade mal accina ion, ei he wi h BCG o he M. ube culosis-de i ed candida e MTBVAC, enhances inna e cy okine p oduc ion by blood- and bone ma ow-de i ed monocy es associa ed wi h me abolic ewi ing, ypical o ained immuni y. These esul s p o ide suppo o s a egies o imp o ing TB accina ion and, mo e b oadly, modula ing inna e immuni y ia mucosal su aces. INTRODUCTION Canonical accinology aims a he induc ion o long-li ed adap- i e memo y in an igen-speci ic lymphocy e popula ions and T and/o B cells. Inna e immuni y in accinology is p ima ily acknowledged in he con ex o accine o mula ion and he quin essen ial inna e immune ac i a ion o p o essional an igen-p esen ing cells (APCs) ha will p ime and skew he adap i e esponse o lymphocy es. 1 Fo yea s, inna e immune esponses and myeloid monocy e/mac ophage unc ion we e conside ed cons an and in a iable o e ime, bu i is now widely ecognized ha also inna e immune cells can adap and display memo y-like pheno ypes. This inna e immune ‘‘memo y’’ is ypi- cally e e ed o as ained immuni y. 2 Induc ion o ained immuni y is de ined by a unc ional adap- a ion o inna e immuni y a e a p ima y insul esul ing in a mo e e ec i e esponse upon a ‘‘seconda y’’ encoun e wi h an un e- la ed pa hogen. 2 I is cha ac e ized by inc eased cy okine and/o chemokine p oduc ion, 3 me abolic ewi ing, 4 and epigene ic e- p og amming o inna e immune cells. 5 T ained immuni y is no associa ed wi h he cha ac e is ic immune ecep o gene ea - angemen s ha a e associa ed wi h he an igen speci ici y o adap i e memo y, bu a he is d i en by he epigene ic con ol o speci ic immune gene clus e s esul ing in he up egula ion o pa icula p o-in lamma o y cy okines. 6 An epigene ic ma k in pa icula associa ed wi h ained immuni y in monocy es, and used in he p esen s udy, is he ace yla ion o his one 3 a he posi ion o lysine 27 (H3K27ac). 7 Recen wo k in mice has shown ha Bacillus Calme e-Gue ´ in (BCG)-media ed aining o he myeloid monocy e/mac ophage lineage is sus ained by he epigene ic ep og amming o hema opoie ic s em cells (HSCs), he myeloid p ecu so s in bone ma ow (BM). 8 Mycobac e ium bo is-de i ed BCG, cu en ly he only a ail- able accine o igh ube culosis (TB), 9 is he p o o ypical bio- logical agen o which ained immuni y has been demons a ed. T ained immuni y has been linked o he bene icial he e ologous o - a ge e ec o in an BCG accina ion, which om epidemi- ological analyses appea s associa ed wi h educed childhood Cell Repo s Medicine 2, 100185, Janua y 19, 2021 ª2020 The Au ho (s). 1 This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). ll OPEN ACCESS mo ali y om causes o he han TB. 10–13 Eme ging e idence suppo s he hypo hesis ha no only adap i e memo y bu also he memo y-like plas ici y in inna e immuni y posi i ely con- ibu es o immune p o ec ion agains and ea ly clea ance o M. ube culosis (M b) in ec ion. 14,15 Al hough no es ablished as ea u es o inna e aining, a compa a i e analysis in macaque species sugges s ha ma ke s o inna e immuni y and myeloid monocy e unc ion a e associa ed wi h di e en ial TB disease suscep ibili y. 16 Despi e he widesp ead use o li e a enua ed BCG as p ophy- laxis agains TB, i is only pa ially e icacious and by and la ge ails o p o ec adolescen s and adul s om he in ec ious pulmona y mani es a ion o M b in ec ion. 17 Thus, TB con inues o be a majo h ea o global human heal h, and a be e accina ion s a egy is u gen ly needed o elie e he bu den o TB. 18,19 While in ade mal injec ion is he s anda d ou e o BCG accina ion, his o ical da a in hesus macaques ha e shown ha mucosal o in a enous adminis a ion p o ided supe io sig- nals o p o ec ion agains expe imen al in ec ion. 20 We ha e ecen ly demons a ed ha pulmona y mucosal BCG accina ion educed disease in a coho o hesus macaques in which in a- de mal BCG ailed o show any p o ec i e e ec . 21 In a subse- quen s udy, exploi ing epea ed limi ing dose (RLD) a he han single dose M b challenge, we ha e shown ha mucosal bu no in ade mal BCG p o ides signals o p e en ion o in ec ion. 22 Mos ecen ly, Da ah and colleagues 23 ha e co obo a ed he su- pe io p o ec i e capaci y o BCG when gi en in a enously. Bo h s udies did no iden i y an immune bioma ke in he blood ha co ela ed wi h he p o ec i e e icacy o ei he mucosal o in a- enous BCG accina ion. Locally, howe e , an igen-speci ic in e leukin-17A-posi i e (IL-17A + ) poly unc ional T helpe (Th) cells and IL-10 sec e ion in he ai ways did co ela e wi h p o ec- ion, 22 and he a icle by Da ah e al. 23 sugges s ha issue- esi- den lymphocy es may be pi o al in con e ing p o ec ion. While om hese non-human p ima e (NHP) s udies uncon en ional adminis a ion o BCG appea s o be a mos in e es ing lead o imp o ing ou cu en TB p ophylaxis, i emains o be es ab- lished, howe e , i and how al e na i e ou ing o BCG a ec s inna e and ained immuni y. Va ious candida e TB accines, anging om subuni o mula- ions o li e a enua ed and ecombinan mycobac e ia, a e a a ious s ages o clinical es ing in he cu en accine-de elop- men pipeline (www. b i.eu/wha -we-do/pipeline-o - accines/). O he li e a enua ed candida es, MTBVAC is unique in ha i is de i ed om a clinical isola e o M b (M 103) a he han om M. bo is (which is he pa en al s ain o BCG and ecombi- nan BCG candida es). 24,25 MTBVAC is a enua ed by dele ion o he phoP and he adD26 gene. The eby, he exp ession o i u- lence ac o s is c i ically dis up ed, while an in insically b oade an igen epe oi e in compa ison o BCG is p ese ed. 26 Aguilo e al. 27 demons a ed ha he exp ession o ESAT6 and CFP10 by MTBVAC is key o i s p o ec i e capaci y in a mouse models o TB. MTBVAC has shown p eclinical e icacy be e han BCG in animal models. 24,27 MTBVAC accina ion p o ec s hesus macaques agains ae osol challenge wi h M b and induces im- mune signa u es analogous o hose obse ed in clinical s udies. 28 I has es ed sa e and immunogenic in adul and in an human popula ions so a . 9,29 Simila o BCG, MTBVAC is able o gene a e ained immuni y h ough he induc ion o glycolysis and glu aminolysis, and he accumula ion o his one me hyla ion ma ks a he p omo e s o p oin lamma o y genes. Recen ly, Ta anco ´n e al. 30 showed ha MTBVAC-induced he e ologous p o ec ion agains a le hal challenge wi h S ep ococcus pneu- moniae in an expe imen al mu ine model o pneumonia. In he p esen s udy, we sough o in es iga e he impac o li e a enua ed mycobac e ial accina ion ia he espi a o y mucosa on ained inna e immuni y in NHPs. To his end, we ollowed a 2 32 ac o ial design s a egy o an immuno- genici y analysis in adul hesus macaques, accina ing ani- mals by in ade mal injec ion o by endob onchial ins illa ion ei he wi h BCG o wi h MTBVAC. We con i med he dis inc- i e adap i e esponse p o ile in he ai ways upon pulmona y mucosal deli e y o BCG and MTBVAC, and analyzed he p oduc ion o inna e and adap i e cy okines a e in i o s imula ion o cells ob ained om lung, blood, and BM. We pilo ed he induc ion o ained immuni y a e in a enous BCG injec ion, demons a ing he inc eased acquisi ion o H3K27ac ma ks and esul ing in p ominen inc eased cy o- kine p oduc ion a e he e ologous s imula ion in monocy es isola ed om blood and BM. We subsequen ly demons a ed ha espi a o y mucosal adminis a ion o li e a enua ed mycobac e ial accines mo e e icien ly induced ained im- muni y in blood and BM monocy es, compa ed o in ade mal immuniza ion, whe eby MTBVAC was equally po en as BCG. Ou indings unde pin he inna e immune s imula o y po en ial o he candida e TB accine, MTBVAC. Fo li e a enua ed mycobac e ial accines in gene al, hey sugges he enhance- men o inna e immune aining ia espi a o y mucosal ac- cine adminis a ion. RESULTS In he p esen s udy, we se ou o in es iga e whe he mucosal adminis a ion o li e a enua ed TB accines in he- sus macaques could modula e inna e immune esponses and induce ained immuni y. To his end, we selec ed heal hy, pu pose-b ed adul hesus macaques (Macaca mula a), bo h males and emales, and s a i ied hem in o compa able g oups o 6 animals each (Table S1). By andom assignmen in a 2 32 ac o ial design s a egy, hese g oups ecei ed ei he a s anda d human dose o BCG o he equi alen dose o he li e a enua ed candida e accine, MTBVAC, by in ade mal injec ion (BCG.id; MTBVAC.id) o by endob on- chial ins illa ion in o he lowe igh lung lobe (BCG.muc; MTBVAC.muc). Fo he assessmen o adap i e and/o inna e immuni y, ele an samples we e collec ed be o e and a e accina ion om he ai ways by b onchoal eola la age (BAL), om he blood by enipunc u e, and om he BM by needle aspi a ion. Be o e analyzing he in ade mal and mucosal accina ion g oups, and since i had been epo ed as an e ec i e esea ch s a egy o measu ing ained immu- ni y in mice, 8 a sepa a e g oup o 3 hesus macaques we e accina ed by in a enous BCG injec ion (BCG.i ) o pilo - ained immuni y p o iling o monocy es (2 weeks a e acci- na ion). A schema ic diag am o accina ion and sampling o e ime is displayed in Figu e 1. 2Cell Repo s Medicine 2, 100185, Janua y 19, 2021 A icle ll OPEN ACCESS BCG and MTBVAC a e immunogenic and bo h display an adap i e immune p o ile speci ic o he ou e o accina ion To con i m he e iciency o accina ion, we pe o med a speci ic in e e on-g(IFN-g) enzyme-linked immunoso ben spo (ELISpo ) assay a e in i o ecall s imula ion o pe iphe al blood mononuclea cells (PBMCs) wi h he p o ein-pu i ied de i a i e (PPD) o M b (o cul u e medium as a nega i e con ol). Fo bo h BCG and MTBVAC and by ei he ou e o deli e y, we ound an inc ease in an igen-speci ic IFN-g elease a e p ima y acci- na ion. In line wi h p e ious obse a ions, 21,22 he accine- induced IFN-g esponse was highe a e in ade mal o e mucosal BCG deli e y, and a compa able esponse pa e n was obse ed wi h MTBVAC (Figu es 2A and 2B). When s imu- la ing wi h a ecombinan usion p o ein o ESAT6 and CFP10, wo an igens encoded in he egion o dele ion (RD)1 ha is absen om M. bo is BCG bu p esen in MTBVAC, posi i e IFN-g elease signals we e ob ained only om MTBVAC- acci- na ed animals, as expec ed (Figu e 2C). To u he con i m he adap i e esponse p o ile ha we egis- e ed p e iously a e pulmona y mucosal BCG as a co ela e o he p e en ion o M b in ec ion, we assessed he cy okine p o- duc ion p o ile o CD4 + T lymphocy es om he ai ways by low cy ome y 22 a e ex i o s imula ion wi h PPD. As expec ed, we de ec ed he mos p ominen le els o poly unc ional Th17 cells (p oducing IL-17A, IFN-g, umo nec osis ac o a [TNF-a], and IL-2) in BALs ha we e collec ed 8 weeks a e p ima y accina ion wi h BCG ia he pulmona y mucosal a he han he in ade mal ou e (Figu e 2D). M b-de i ed MTBVAC was ound o induce a simila esponse pa e n and o be a leas as po en as BCG. Rega dless o he accine, signi ican numbe s o poly unc ional Th17 cells we e also eco e ed om he lowe le lung lobe (i.e., opposi e o he si e a ge ed by en- dob onchial immuniza ion), sugges ing ha he poly unc ional Th17 esponse is no con ained o he si e o accina ion bu dis- semina es locally. Es ablishing ained immuni y in monocy es a e in a enous immuniza ion wi h BCG To e alua e he induc ion o ained immuni y a e s anda d in- ade mal e sus mucosal accina ion wi h li e a enua ed TB accines, we es ablished he easibili y o he de ec ion o ained immuni y in he nonhuman p ima e model. Fo his, we immunized h ee animals wi h BCG in a enously, a p o ocol ha was success ully explo ed in mice by Kau mann e al. 8 To minimize he in e e ence o a de eloping adap i e immune esponse while s ill being able o de ec obus inna e immune e- sponses, we ook samples om di e en compa men s a week 2 a e accina ion. Nex o he analysis o BAL cells om he lung, we pe o med a mo e selec i e analysis o myeloid CD14 + monocy es pu i ied om esh PBMCs (PBMC.mo) and BM (BM.mo), he p o o ypic cells in which he ained immuni y pheno ype has been desc ibed. CD14 + monocy e popula ions we e ob ained by posi i e selec ion (using magne ic bead-cell so ing) and ypically eached a pu i y o ±90% (Figu e S1). Bo h un ac iona ed BAL cells and en iched monocy es om blood and BM we e subsequen ly analyzed o signals o ained immuni y a e in i o es imula ion wi h M b-de i ed whole-cell lysa e (WCL) o he he e ologous Esche ichia coli de i ed Toll- like ecep o (TLR)-4 agonis lipopolysaccha ide (LPS). A e 24 h, supe na an s we e ha es ed, ozen, and analyzed la e o cy okine p oduc ion, indica i e o ained inna e immuni y, including TNF-a, IL-6, and IL-1b. A e in a enous BCG accina ion, we ound ma ginal sup- p ession o cy okine sec e ion, i any e ec a all, a e compa ing M b o LPS s imula ion o BAL (Figu es 3A and 3B). S imula ion o PBMC.mo and BM.mo wi h M b e ealed ma kedly inc eased le els o TNF-a2 weeks a e BCG.i (Figu e 3A), bu no appa en modula ion o IL-6 and IL-1bsec e ion le els. Howe e , he e ol- ogous s imula ion wi h LPS demons a ed p ominen inc eased p oduc ion o all h ee cy okines by bo h monocy es om he blood and BM (wi h he excep ion o a single animal wi h high baseline le els in BM.mo; Figu e 3B, PBMC.mo and BM.mo). Me abolic ewi ing, speci ically a swi ch om oxida i e phos- pho yla ion o glycolysis unde ae obic condi ions, 4 has been desc ibed as an ini ia ing e en unde lying ained immuni y. The e o e, we measu ed lac a e, a me aboli e o glycolysis, le els in he supe na an s o s imula ed blood- and BM-de i ed monocy es as a su oga e ma ke o me abolic ewi ing, and hus ained immuni y. Lac a e p oduc ion was obse ed o 2 o 3 LPS-s imula ed BM.mo, bu unde ec able o PBMC.mo (Figu e 3C). Ul ima ely, we de e mined whe he BCG.i accina ion e- sul ed in epigene ic ep og amming, ano he hallma k o ained inna e immuni y, in blood-de i ed monocy es (PBMC.mo). The e o e, we examined he dynamics o his one modi ica ion H3K27ac (an epigene ic ma k associa ed wi h ac i e ch oma in in he con ex o ained immuni y 7 ) a e BCG.i by ch oma in immune p ecipi a ion, ollowed by high- h oughpu sequencing (ch oma in immunop ecipi a ion sequencing [ChIP-seq]) o assess he changes associa ed wi h BCG.i accina ion. H3K27ac changes we e de ec ed a many loci (Figu es 3D–3F). In o al, we de ec ed 792 egions inc eased and 646 egions dec eased in H3K27ac (Figu e 3D). While he egions showing inc eased ace yla ion we e associa ed wi h genes ela ed o he cell cycle, signaling, and ac i a ion o he immune esponse, Figu e 1. Schema ic ep esen a ion o he imeline o accina ion in he di e en acci- na ion s a egies The BCG.i g oup (n = 3) was analyzed a weeks 1 and 2. The g oups ha we e immunized wi h BCG.id (n = 6), BCG.muc (n = 6), MTBVAC.id (n = 6), and MTBVAC.muc (n = 6) we e analyzed a weeks 1 and 8. Only om mucosally accina ed animals BAL was collec ed om he igh lung as well as he le lung. See also Table S1. Cell Repo s Medicine 2, 100185, Janua y 19, 2021 3 A icle ll OPEN ACCESS egions dec eased in ace yla ion we e associa ed wi h genes in ol ed in cellula di e en ia ion (Figu e 3E). Examples o bo h an ac i e egion, IFN egula o y ac o 3 (IRF3), and sup- p essed egion, syndecan 2 (SDC2) a e gi en (Figu e 3F). These da a a e in line wi h accina ion-induced changes in he ace y- lome and he cellula p og am ela ed o he induc ion o ained immuni y. P o-in lamma o y cy okine p oduc ion a e s imula ion o BAL cells wi h M b o LPS We hen sough o compa e he induc ion o ained immuni y a - e s anda d in ade mal e sus mucosal accina ion wi h he li e a enua ed TB accines BCG and MTBVAC. We analyzed cy o- kine/chemokine p oduc ion in BAL cells a e ex i o s imula ion wi h homologous M b WCL o he e ologous LPS. Using a ailo ed mul iplex ki o cy okine and chemokine measu emen , we assessed he p oduc ion o molecules ha ha e p e iously been associa ed wi h lymphocy e ac i a ion and unc ion (IFN-g, IL-2, IL-17A), in lamma ion, leukocy e di e en ia ion, mig a ion, and/o wi h ained immuni y (TNF-a, IL-6, IL-1b, g anulocy e-mac ophage-colony-s imula ing ac o [GM-CSF], mac ophage in lamma o y p o ein 1a[MIP-1a; CCL3], and IFN- g-induced p o ein 10 [IP-10; CXCL10]). 3,11,31 In line wi h he low cy ome ic de ec ion o poly unc ional Th17 cells, we ound signi ican ly ele a ed le els o IL-17A, IFN-g, IL-2, and TNF-a sec e ion om M b WCL-s imula ed BAL cells—in pa icula a e mucosal accina ion wi h BCG o MTBVAC (Figu e 4A). A signi ican inc ease in he sec e ion o IL-2 and TNF-a(bu no IL-17A and IFN-g) was also measu ed a e in ade mal accina- ion. As he esponse le els o BCG and MTBVAC in his ega d we e indis inguishable, we ha e depic ed esul s and pe o med s a is ical analyses on he 12 animals o bo h BCG and MTBVAC ABC D Figu e 2. Mucosal accina ion es ablishes a unique local adap i e signa u e de ined by poly unc ional Th17 cells (A–C) Indi idual IFN-gELISpo esponses a e in i o ecall s imula ion o PBMCs agains PPD (A) be o e accina ion o (B) 8 weeks a e accina ion o wi h (C) ESAT6-CFP10 8 weeks a e accina ion. Ho izon al lines in ba s indica e g oup medians; n = 6 animals/g oup. ANOVA adjus ed o mul iple compa isons; Dunn’s mul iple compa ison es . (D) The pe cen ages o poly unc ional IL-17A + PPD-speci ic CD4 + T cells (also p oducing IFN-g, TNF-a, and IL-2) we e de e mined by low cy ome ic analysis o BAL cells be o e (PRE) and 8 weeks a e in ade mal (.id) o mucosal (.muc) accina ion. Cells we e ypically collec ed om he lowe igh lung lobe (R); a ge ed o accina ion) and o mucosally accina ed animals also om he le lung (L) o es ablish dissemina ion o he immune esponse. The da a a e p esen ed as medium con ol co ec ed (cc); n = 6 animals/g oup; Wilcoxon ma ched-pai s signed ank es . ULD, uppe limi o de ec ion; LLD, lowe limi o de ec ion. See also Figu e S7. 4Cell Repo s Medicine 2, 100185, Janua y 19, 2021 A icle ll OPEN ACCESS AB D C E F (legend on nex page) Cell Repo s Medicine 2, 100185, Janua y 19, 2021 5 A icle ll OPEN ACCESS a ms oge he , pe ou e o adminis a ion. The elease o p o-in- lamma o y IL-6, a e in i o s imula ion wi h M b WCL, compa ing p e- and pos - accina ion ime poin s, was a he he e ogeneous. In con as , IL-1b, GM-CSF, and MIP-1a (CCL3) sec e ion was, by end o wi h s a is ical signi icance, dec eased in mucosally accina ed animals (Figu e 4A). CXCL- 10 (IP-10), howe e , was ma kedly and signi ican ly inc eased upon M b WCL s imula ion, ega dless o he na u e o he ac- cine o he immuniza ion ou e. This cy okine/chemokine sec e ion p o ile o BAL cells a e s imula ion wi h a p epa a ion as c ude as he WCL o M b, con- aining bo h p o ein an igens and (non-p o ein) inna e ecep o ligands, is he ne esul o adap i e and inna e ac i a ion o all subse s ha a e p esen in he un ac iona ed BAL cell popula- ion. In addi ion, i mus be no ed ha he ela i e abundance o immune cell subse s in BAL is a ec ed by pulmona y mucosal bu no by in ade mal BCG accina ion owa d a ela i e in- c ease in T lymphocy es o e al eola mac ophage (Figu e S2). To a leas ci cum en he adap i e componen o his esponse p o ile and owa d he assessmen o ained immuni y by he e - ologous inna e s imula ion, we also s imula ed BAL cells wi h LPS o measu e he ensuing cy okine p oduc ion capaci y. Upon LPS, and in con as o M b WCL s imula ion, IL-17A, IFN-g, and TNF-aconcen a ions a e accina ion we e no di e en om wha was ound a baseline (Figu e 4B). Mos likely as a esul o he bys ande ac i a ion o T lymphocy es speci - ically ec ui ed in o he ai ways a e mucosal accina ion, IL-2 elease upon LPS s imula ion was inc eased wi h BAL cells om mucosally accina ed animals (Figu e 4B), bu o a lesse ex en compa ed o M b WCL s imula ion (Figu e 4A). Also compa ed o M b WCL s imula ion, he accina ion-modula ed GM-CSF sec e ion o BAL cells was los wi h LPS, and ha o CXCL-10 was much lowe and he e ogeneous (Figu e 4B). LPS-s imula ed IL-6 le els we e compa ably he e ogeneous as wi h M b WCL s imula ion. IL-1band MIP-1a, as wi h M b WCL s imula ion, showed diminished sec e ion upon he LPS s imula- ion o BAL cells a e accina ion, ega dless o he ou e o ac- cine deli e y (Figu e 4B). This cy okine/chemokine sec e ion analysis o un ac iona ed BAL cells upon he e ologous LPS (o homologous M b WCL) s imula ion did no e eal a signal o ained immuni y in he ai - ways a e (mucosal) li e a enua ed mycobac e ial accina ion. Induc ion o ained immuni y in monocy es a e mucosal accina ion wi h BCG and MTBVAC Since BAL cells did no p o ide a signal indica i e o ained im- muni y, we sough o compa e he induc ion o ained immuni y h ough ex i o he e ologous LPS s imula ion o blood- and BM-de i ed monocy es, as was shown o BCG.i be ween in- ade mal and mucosal immunized animals, wi h li e a enua ed BCG o MTBVAC. To demons a e ha he cy okine p oduc ion is due o ex i o he e ologous LPS s imula ion and exclude he possibili y o con inued inna e ac i a ion by he (pe sis ing) ac- cine, uns imula ed medium con ols we e included ha showed minimal o no p oduc ion o cy okines (Figu e S3). While T cell-de i ed cy okines IL-17A, IFN-g, and IL-2 we e no de ec ed upon monocy e s imula ion (no shown), he enhance- men o LPS-s imula ed sec e ion o TNF-a, IL-6, and IL-1bwas e iden in some bu no all animals ha we e accina ed by in a- de mal injec ion (Figu e 5A). Speci ically, 4 o 11 animals (2 om each o he BCG- and he MTBVAC- accina ed g oups) showed ele a ed TNF-ap oduc ion wi h an e ec size (independen o baseline p oduc ion) o >2- old inc ease a e BCG.id/ MTBVAC.id (Figu e 5B; due o a echnical e o , we missed he baseline alues o a single animal o he MTBVAC.id g oup). Simila ly, 4 o 11 showed he opposi e e ec o dec eased TNF-ale els a e BCG.id/MTBVAC.id, while 3 o 11 showed no conside able modula ion o LPS-s imula ed TNF-aa e in a- de mal accina ion (Figu e 5B). O no e, M b-de i ed MTBVAC appea ed equally as po en as BCG in he induc ion o ained immuni y in PBMC.mo (Figu e S4A). No ably, he sec e ion le els o p o-in lamma o y IL-6 and IL-1bshowed a pa e n simila o ha o TNF-a. The a io o he p oduc ion o he 3 signa u e cy- okines (TNF-a, IL-6, and IL-1b) did no de ia e o e a pe iod o 8 weeks in a coho o un accina ed animals (Figu e S4C). In con as o in ade mal BCG/MTBVAC accina ion, he mucosal deli e y o hese li e a enua ed accines esul ed in a obus and signi ican enhancemen o TNF-a, IL-6, and IL-1b p oduc ion a e LPS s imula ion in 10 o 12 animals (Figu es 5A and 5B; 5 animals om he BCG and he MTBVAC accina ed g oup each). The median e ec size was 15.1, 6.2, and 5.9 o he old inc ease in TNF-a, IL-6, and IL-1bp oduc ion, espec i ely (Figu e 5B). Fo GM-CSF and MIP-1a, 2 cy okines ha ha e been associa ed wi h ained immuni y, a simila pa e n o enhanced sec e ion a e p io mucosal a he han in ade mal accina ion was ob ained (Figu es 5A and 5B). This enhance- men , howe e , was less p ominen han o TNF-a, IL-6, and IL-1b, wi h a median old-inc ease o GM-CSF and MIP-1aa e BCG.muc/MTBVAC.muc o 2.1 and 2.9, espec i ely (Figu e 5B). CXCL-10 was no de ec ed om hese monocy e s imula ion analyses (Figu e 5A). Compa ing BCG o MTBVAC wi hin each ou e o deli e y (id e sus muc) s a egy demons a ed ha bo h accines we e equally po en in media ing ained immuni y (Figu e S5A). We explo ed he mycobac e icidal ac i i y wi h Figu e 3. T ained immuni y a e in a enous BCG accina ion (A and B) F eshly isola ed BAL cells ( om he igh lung) and monocy es om blood (PBMC.mo) and bone ma ow (BM.mo) we e s imula ed o 24 h wi h (A) M b whole-cell lysa e (25 mg/mL) o (B) LPS (0.1 mg/mL) be o e and 2 weeks a e in a enous accina ion wi h BCG. The da a a e p esen ed as medium con ol co ec ed. (C) Lac a e p oduc ion as an indica o o me abolic ewi ing was measu ed in 24-h supe na an om LPS-s imula ed monocy es pu i ied om blood (PBMC.mo) o BM (BM.mo). See also Figu e S1. (D) Hea map o H3K27ac eads ( ed) o e BCG-speci ic peaks. The in ensi y o e he cen e o he peak ±12 kb is depic ed. (E) The op GO pa hways associa ed wi h he nea es genes o dynamic H3K27ac, wi h adjus ed p alues. (F) H3K27ac dynamics a in e e on egula o y ac o 3 (IRF3) and syndecan 2 (SDC2) locus o PBMC.mo p e- and pos - accina ion. The comple e lis o genes and alues can be ound a GEO: GSE159046. 6Cell Repo s Medicine 2, 100185, Janua y 19, 2021 A icle ll OPEN ACCESS A B (legend on nex page) Cell Repo s Medicine 2, 100185, Janua y 19, 2021 7 A icle ll OPEN ACCESS PBMC.mo isola ed om esidual ozen PBMCs, bu wi hin he limi s o ou e o s, we did no de ec mycobac e icidal ac i i y by ained monocy es (Figu e S6). Nex o PBMC.mo, we add essed he LPS-s imula ed cy o- kine/chemokine sec e ion po en ial o BM-de i ed CD14 + monocy es (BM.mo) be o e and a e BCG o MTBVAC Figu e 4. Cy okine p oduc ion by BAL cells F eshly isola ed BAL cells we e s imula ed (A) wi h M b whole cell lysa e o (B) LPS, be o e (PRE) and 8 weeks a e in ade mal (.id) o mucosal (.muc) accina ion (wi h ei he BCG, in ci cles, o MTBVAC, in iangles). Cells we e ypically collec ed om he lowe igh lung lobe (R), he lobe a ge ed by accina ion, and o mucosally accina ed animals also om he le lung (L), o es ablish dissemina ion o he immune esponse. The da a a e p esen ed as medium con ol co ec ed (mcc); n = 12 animals/g oup; Wilcoxon ma ched-pai s signed ank es . p alues a e indica ed a he op o each g aph. See also Figu e S2. A B Figu e 5. T ained immuni y a e mucosal accina ion in pe iphe al blood monocy es Cy okine p oduc ion was measu ed a e 24 h o s imula ion wi h LPS (0.1 mg/mL). (A) Pai ed display o cy okine le els be o e and 8 weeks a e accina ion. The da a a e p esen ed as medium con ol co ec ed (mcc; BCG in ci cles; MTBVAC in iangles). S a is ical signi icance was calcula ed by Wilcoxon’s non-pa ame ic pai ed analysis es (n = 12 animals/g oup; PRE e sus 8 weeks pos - acci- na ion). p alues a e indica ed a he op o each g aph. (B) Median old inc ease in cy okine p oduc ion is indica ed in i alics; signi ican p < 0.05 is indica ed in bold; n = 12 animals/g oup; Wilcoxon signed ank es . See also Figu es S1,S3,S4,S5, and S6. 8Cell Repo s Medicine 2, 100185, Janua y 19, 2021 A icle ll OPEN ACCESS STAR+METHODS KEY RESOURCES TABLE REAGENT o RESOURCE SOURCE IDENTIFIER An ibodies an i-CD3 – AF700 (clone SP34-2) BD Biosciences Ca #: 557917; RRID: AB_396938 an i-CD4 – Pe CP.Cy5.5 (clone L200) BD Biosciences Ca #: 552838; RRID: AB_394488 an i-CD8a– APC-H7 (clone SK1) BD Biosciences Ca #: 641400; RRID: AB_164536 an i-CD14 – BV421 (clone M5E2) BD Biosciences Ca #: 301830; RRID: AB_10959324 an i-CD20 – BV421 (clone 2H7) Biolegend Ca #: 302330; RRID: AB_10965543 an i-CD28 – ECD (clone CD28.2) IOTes Ca #: 6607111; RRID: AB_1575955 an i-CD45RA – PE-CF594 (clone 5H9) BD Biosciences Ca #: 565419; RRID: AB_2739229 an i-CD95 – BV605 (clone DX2) Biolegend Ca #: 305628; RRID: AB_2563825 an i-IL-2 – AF488 (clone MQ1-17H12) Biolegend Ca #: 500314; RRID: AB_493368 an i-IL-17A – PE-Cy7 (clone ebio64DEC17) Biolegend Ca #: 25-7179-42; RRID: AB_11063994 an i-TNF-a– BV650 (clone Mab11) BD Biosciences Ca #: 502938; RRID: AB_2562741 an i-IFN-g– BV711 (clone 4S.B3) BD Biosciences Ca #: 502540; RRID: AB_2563506 an i-CD3ε– FITC (clone SP34) BD Biosciences Ca #: 556611; RRID: AB_396484 an i-CD14 – V450 (T€ uk4) Mil eny Bio ec Ca #: 130-113-152; RRID: AB_10831023 an i-CD20 – BV605 (clone 2H7) Biolegend Ca #: 302334; RRID: AB_2563398 an i-CD45 – BV786 (clone D058-1283) BD Biosciences Ca #: 563861; RRID: AB_2738454 an i-CD3 – BV421 (clone SP34-2) BD Biosciences Ca #: 562877; RRID: AB_2737860 an i-CD14 – BV786 (clone M5E2) BD Biosciences Ca #: 301840; RRID: AB_2563425 an i-CD16 – PE-CY7 (clone 3G8) BD Biosciences Ca #: 560716; RRID: AB_1727433 an i-CD20 – BV605 (clone 2H7) Biolegend Ca #: 302334; RRID: AB_2563398 an i-CD45 – AF700 (clone D058-1283) BD Biosciences Ca #: 561288; RRID: AB_10613813 an i-CD66 – PE (clone TET2) Mil eny Bio ec Ca #: 130-093-133; RRID: AB_871699 an i-CD206 – Pe CP (clone 15-2) BioLegend Ca #: 321122; RRID: AB_10899411 B bu e BioLegend Ca #: 563794; RRID: AB_2869750 VIVID – BV421 The mo ishe Ca #: L34955 Li e/Dea h – eFluo 506 eBioscience Ca #: 65-0866-14 an i-H3K27ac Diagenode Ca #: pab-196-050; RRID: AB_2637079 Bac e ial and i us s ains BCG (s ain So ia; 5 310 5 CFU) In e Vax L d Ca #: no applicable MTBVAC (M. ub s ain; 5 310 5 CFU) Bio ab i Ca #: no applicable M b E dmann BEI Resou ces Ca #: NR-50781 Chemicals, pep ides, and ecombinan p o eins Pu i ied P o ein De i a e (PPD; M. ub) AJ Vaccines Ca #: 2391 Pu i ied P o ein De i a e (PPD; M.bo ) Li e Technologies NV Ca #: 760060 Pu i ied P o ein De i a e (PPD; M.a ) Li e Technologies NV Ca #: 760065 Whole Cell Lysa e (WCL; M b HN878) BEI Resou ces Ca #: NR-14824 Lipopolysaccha ide (0111:B4 s ain; Ul apu e) In i ogen Ca #: l l-3pelps PMA Sigma-Ald ich Ca #: P8139 Ionomycine Sigma-Ald ich Ca #: I0634 Roswell Pa k Memo ial Ins i u e (RPMI) Li e Technologies NV Ca #: 52400041 Lymphop epTM Axis-Shield Ca #: AXI-1114547 Fe al Cal e Se um (FCS) Li e Technologies NV Ca #: 10270106 (Con inued on nex page) Cell Repo s Medicine 2, 100185, Janua y 19, 2021 e1 A icle ll OPEN ACCESS RESOURCE AVAILABILITY Lead con ac Fu he in o ma ion and eques s o esou ces and eagen s should be di ec ed o and will be ul illed by he Lead Con ac Michel PM Vie boom ([email p o ec ed]). Ma e ials a ailabili y This s udy did no gene a e new unique eagen s. Da a and code a ailabili y Raw da a iles o he ChIPseq analysis on BCG ained monocy es bego e and a e accina ion ha e been deposi ed in he NCBI Gene Exp ession Omnibus unde accession numbe GEO: GSE159046. EXPERIMENTAL MODEL AND SUBJECT DETAILS Animals and e hics All animal expe imen s we e in acco dance wi h Du ch Law on animal expe imen a ion, which is in acco dance wi h he EU Di ec i e 2010/63/EU on he p o ec ion o animals used o scien i ic pu poses. The s udy was app o ed by he ins i u ional animal wel a e body (in Du ch: Ins an ie oo Die welzijn, I D; accession numbe CCD009D) o BPRC. A coho o 8 non accina ed con ol animals was used o Figu e S4C. The s udy in which hese animals we e used was app o ed by he ins i u ional animal Con inued REAGENT o RESOURCE SOURCE IDENTIFIER Pe chlo ic acid Sigma-Ald ich Ca #: 244252-1L Fo maldehyde (16%) The mo Scien i ic Ca #: 28906 P o ease Inhibi o Cock ail Sigma-Ald ich Ca #: P8465 Dynabeads p o ein A In i ogen Ca #: 10002D Dynabeads p o ein G In i ogen Ca #: 10004D BSA Sigma-Ald ich Ca #: A7030 Tagmen DNA bu e Nex e a Ca #: 20034197 AMPu eXP beads Beckman Coul e Ca #: A63880 KAPA HiFi Ho s a Ready Mix KAPA Biosys ems Ca #: KK2601 Nex e a DNA Lib a y P ep Ki Illumina Ca #: FC-121-1031 KAPA Lib a y P epa a ion Ki KAPA Biosys ems Ca #: KK8400 TMB (ELISPOT subs a e) Mab ech Ca #: 3651-10 C i ical comme cial assays CD14 mic obeads Mil eny Bio ec Ca #: 130-091-097 Cus om NHP Legendplex (11-plex) Biolegend Ca #: 92919 Monkey IFN-gELISPOT U-CyTech Ca #: CT610-10 Lac a e-ELISA P omokine Ca #: PK-CA577-K627 Deposi ed da a ChIP sequencing da a be o e and a e BCGi accina ion This pape GEO: GSE159046 Expe imen al models: o ganism/s ain Rhesus monkeys (Indian); male and emale; adul (> 4 yea s); pu pose b ed BPRC b eeding colony no applicable So wa e and algo i hms Eli.Analyze (ELISPOT; 6.1) A.EL.VIS GmbH no applicable FACSDi a So wa e 8.0.1 (BD LSRII) BD Biosciences SCR_001456 Flowjo so wa e 10 T ees a SCR_000410 LEGENDplexTM Da a Analysis So wa e (V8.0) Biolegend/Vigene Tech no applicable G aphPad P ism 8.4.2 G aphPad So wa e h ps://www.g aphpad.com e2 Cell Repo s Medicine 2, 100185, Janua y 19, 2021 A icle ll OPEN ACCESS wel a e body (Accession numbe CCD009F). The BPRC is acc edi ed by he Ame ican Associa ion o Acc edi a ion o Labo a o y Animal Ca e (AAALAC) and has an app o ed Assu ance (#A5509-1) o he ca e and use o animals on ile a he Na ional Ins i u es o Heal h (NIH). The hesus monkeys we e housed in socially compa ible pai s a he Biomedical P ima e Resea ch Cen e (BPRC; animal biosa e y le el ABSL-3). The animals we e o e ed a daily die consis ing o monkey ood pelle s (Hope Fa ms, Woe den, he Ne he lands), ui and ege ables o he season, and b ead. D inking wa e was a ailable ad libi um ia au oma ic wa e sys ems. The s udy was ca ied ou in 27 heal hy pu pose-b ed, pedig eed, male and emale hesus monkeys (Macaca mula a) o Indian o igin (5.99 – 13,50 kg; see Table S1). All animals we e sc eened o be nega i e o p e-exis ing immuni y agains mycobac e ial an- igens as de e mined by in e e on gamma (IFN-g) ELISpo assay a e in i o ecall s imula ion wi h Pu i ied P o ein De i a e (PPD) o M. ube culosis,M.a ium and M.bo is. All animal handling and biosampling was pe o med unde ke amine seda ion (10 mg kg1, by in amuscula injec ion). Fo b on- choal eola la age (BAL) and in ab onchial accina ion inocula ion wi h BCG o MTBVAC ke amine (5 mg kg1) was supplemen ed wi h in amuscula mede omidine (0.04 mg kg1) and an analgesic sp ayed in o he la ynx. Vaccines and accina ion BCG s anda d dose & MTBVAC s anda d dose. Animals immunized wi h Bacillus Calme e–Gue in (BCG) s ain So ia (In e Vax L d.) ecei ed a single s anda d adul human dose o 5 310 5 CFU. Animals immunized wi h MTBVAC ecei ed an equi alen human dose o 5 310 5 CFU. Vaccines we e deli e ed in a olume o 0.1 mL in ade mally and 10 mL in a enously o in ab onchially. Mucosal accina ion in he lung was pe o med by in ab onchial deli e y o he same dose in 10mL o s e ile saline solu ion o a segmen al b onchus in he lowe igh lung lobe using a b onchoscope. Vaccina ion was execu ed o all animals in a single session in andom o de wi hin 2-3 hou s om accine p epa a ion METHOD DETAILS Biological sample collec ion and p ocessing Cells om he pulmona y mucosa we e eco e ed by BAL a week 1 o all g oups om he lowe igh lung lobe; a week 2 a e in a enous BCG and week 8 a e in ade mal BCG om lowe igh lung lobe and week 8 a e mucosal accina ion om bo h he lowe igh and le lung lobe. Th ee olumes o 20 mL o p ewa med 0.9% saline solu ion we e consecu i ely ins illed and eco e ed. BAL luid was ha es ed by cen i uga ion o BAL samples o 10 min a 400 g a e 100 mm il a ion. Supe na an was subsequen ly decan ed and s o ed a 80C pending u he analysis. The BAL cell pelle was aken up in Roswell Pa k Memo ial Ins i u e (RPMI) medium supplemen ed wi h 10% FCS, glu aMAX, and penicillin/s ep omycin ( om now on e e ed o as R10) and used eshly in downs eam assays. Hepa inized blood o immune moni o ing, was collec ed by enipunc u e. Bone ma ow was ob ained h ough needle ex ac ion o he capu hume us. PBMCs and BM cells we e ob ained by densi y g adien cen i uga ion wi h Lymphop ep lymphocy e sepa a ion medium (Axis-Shield) and esuspended in R10 o u he analysis. Mycobac e ium up ake and killing assay To assess mycobac e ium up ake, PBMC.mo (30.000 cells/well) we e incuba ed o 1h a 37C wi h M b E dmann a a MOI o 5. Non-phagocy osed mycobac e ium was washed away and monocy es we e incuba ed wi h high dose (30 mg/mL) gen amycin o 10-minu es. PBMC.mo we e subsequen ly lysed wi h 0.05% SDS in wa e , i a ed and pla ed on o 7H10 aga o bac e ial enume - a ion. To asses killing, PBMC.mo we e incuba ed wi h M b E dmann a a MOI o 5 o 1h a 37C and subsequen ly cul u ed wi h low gen amycin (5 mg/mL) o 24 h . Bac e ia we e enume a ed as desc ibed be o e. As a con ol, bac e ia we e incuba ed wi hou PBMC.mo. Da a a e p esen ed as % o con ol (Figu e S5). Monocy e isola ion CD14+ cells (monocy es) we e labeled wi h magne ic CD14 Mic oBeads. The cell suspension was loaded on o a MACSColumn which is placed in he magne ic ield o a MACS Sepa a o . The magne ically labeled CD14+ cells a e e ained in he column. The unlabeled cells un h ough, his cell ac ion is deple ed o CD14+ cells. A e emo al o he column om he magne ic ield, he magne ically e ained CD14+ cells can be elu ed as he posi i ely selec ed cell ac ion. Pu i y was analyzed by lowcy ome y (Figu e S1). IFN-gELISPOT A NHP speci ic IFN-gELISpo assay was used on PBMC, acco ding o he manu ac u e p o ocol (U-CyTech, U ech ), o de e mine he equency o an igen speci ic IFN-gp oducing cells. In b ie , 200.000 eshly isola ed PBMC we e incuba ed in iplica e o 24 h wi h speci ied an igens o con ol s imuli. Subsequen ly, supe na an was collec ed and s o ed (80C), and cells we e ans e ed o speci ic an i-IFN-gcoa ed il e pla es (PVDF, Millipo e) o an addi ional o e nigh (18 h) incuba ion. Cells we e disca ded and memb ane bound IFN-gwas de ec ed using bio inyla ed an i-IFN-gan ibody, s ep a idin-ho se adish pe oxidase conjuga e Cell Repo s Medicine 2, 100185, Janua y 19, 2021 e3 A icle ll OPEN ACCESS and e ame hylbenzidine (TMB) subs a e ( he la e om MAbTech, S ockholm). Spo s we e quan i ied using an au oma ed eade (AELVIS, Hanno e ). Flow cy ome y An igen-speci ic T cell esponses we e de e mined a WK-1 (PRE) and week 8 indica ed ime poin s by low cy ome y locally using BAL cells. Cells we e s imula ed o e nigh in he p esence o Golgiplug anspo inhibi o . The ollowing mAb s aining T cell panel was used: In acellula s aining: CD3 AF700 (SP34-2; BD), TNF-a PE-CY7 (Mab11; BD), IL-2 BV510 (MQ1-17H12; BioLegend), IFN-g APC (4S.B3; BD); IL-17A PE-Cy7 (ebio64DEC17; The mo ishe ). Su ace s aining: CD4 Pe CP-Cy5.5 (L200; BD), CD8a APC-H7 (SK1; BD), CD14 BV421 (M5E2; BD), CD20 BV421 (2H7; BioLegend), CD45 BV786 (D058-1283; BD). Ga ing s a egy o T cells (Figu e S7; Rela ed o Figu e 2D) T cells we e ga ed as Single s/Lymphocy es/Viable/ CD14-CD20-/CD45+ /CD3+ be o e CD4 and CD8 ga ing was applied. A e double exclusion, lymphocy es we e ga ed based on size and g anula i y. Any anomaly indica i e o uns able signal acquisi ion was excluded using he ime pa ame e . E en s om he combined ime-ga es we e plo ed agains he dump channel con aining he iabili y, CD14 and C20 ma ke s and subsequen ly ga ed as iable, CD14- and CD20-. Cells we e u he selec ed o CD45 and CD3 posi i i y be o e CD4 and CD8 ga ing was applied. Boolean ga ing o any cy okine exp ession o IL-2/IFN-g/TNFa/IL-17A. BAL cell composi ion analysis Cells we e s imula ed o e nigh wi h WCL. The ollowing mAb s aining panel was used o analyze BAL cell composi ion: CD45 AF700 (D058-1283; BD); CD66 PE (TET2; Mil enyi); CD206 Pe CP (15-2;Biolegend); CD16 PE-CY7 (3G8; BD); CD3 BV421 (SP34-2;BD); CD20 BV605 (2H7; Biolegend); CD14 BV786 (M5E2; BD). Ga ing s a egy o BAL cell composi ion (Figu e S2B) Single s/ Leucocy es (CD45)/Viable/Time ga e: CD206+/CD14+ (Al eola Mac ophages); CD3 (T cells); CD20 (B cells); CD16+ (NK-cells) and CD66+ (Neu ophils). BAL cell composi ion is exp essed as % om he CD45+/ iable ac ion. Isola ion o CD14+ monocy es PBMC o BM cells we e incuba ed wi h CD14 magne ically labeled Mic obeads (Mil enyi Bio ec; Ne he lands). The cell suspension is loaded on o a MACSColumn which is placed in he magne ic ield o a MACS Sepa a o . The magne ically labeled CD14+ cells a e e ained in he column. The column is washed once. The column was emo ed om he magne ic ield and he magne ically e ained CD14+ cells we e hen elu ed as he posi i ely selec ed cell ac ion. Pu i y o CD14+ ac ion was subsequen ly es ed by low cy ome y. Ga ing s a egy o CD14+ monocy es (Figu e S1B) Single s/CD45+/ Viable/CD14+. CD14+ cells a e exp essed as % om he CD45+/ iable ac ion. Mul iplex cy okine assay Cells om di e en sou ces (PBMC.mo, BM.mo and BAL we e s imula ed wi h WCL and i a ed amoun o LPS (E.coli; Ul apu e). F eshly isola ed cells 25000 cells we e incuba ed ex- i o o 24 h wi h each s imulus a e which he supe na an was collec ed. Medium alone was aken along o con ol. Cy okine p oduc ion was assessed wi h mul i-analy e low assay ki using a cus om nonhuman p ima e Legendplex TM (Biolegend; USA). Assays we e pe o med acco ding o he manu ac u e ’s ins uc ions. B ie ly, he supe na an s o s imula ed cells we e incuba ed wi h beads coa ed wi h cy okine-speci ic an ibodies. Bound cy okines we e isualized using bio in-coupled de ec ion an ibodies and R-phycoe y h in-conjuga ed s ep a idin. Beads we e acqui ed on 3 lase , 14 colo LSR-II low cy ome e (BD Biosciences). A NHP cus om made 11-plex consis ed o TNF-a, IL-6, IL-1bcha ac e izing ained immuni y, GM-CSF, CCL3 (MIP-1a), CXCL10 (IP-10) and cy okines associa ed wi h adap i e immuni y IFN-g, IL-2, IL-17A. Lac a e measu emen s Supe na an collec ed om he 24 h inna e s imula ion was analyzed o he p oduc ion o lac a e as a su oga e ma ke o he me a- bolic ewi ing (Wa bu g-e ec ) unde lying ained immuni y. P o eins in he supe na an , in e e ing wi h he lac a e analysis, we e dena u ed p ecipi a ed wi h pe cholo ic acid (PCA) and emo ed by high speed cen i uga ion. The clea dep o einized supe na an was collec ed and neu alized wi h NaOH. Samples we e subsequen ly dilu ed 15- old and analyzed in a L-Lac a e Assay Ki I (P omokine; P omoCell GmbH; Ge many) ac- co ding o manu ac u e ’s ins uc ion. Ch oma in immunop ecipi a ion Cell pelle s we e dissol ed in PBS and c osslinked in solu ion by using o maldehyde (1% inal olume, shaking 10 min a oom empe a u e). Glycine (0.125 M) was added o quench he eac ion. Cells we e hen washed wo imes wi h cold PBS. Cell pelle s we e lysed in a olume o 110ml using lysis bu e (20 mM HEPES pH 7.6, 1% SDS, 1 3P o ease Inhibi o Cock ails). Samples e4 Cell Repo s Medicine 2, 100185, Janua y 19, 2021 A icle ll OPEN ACCESS we e sonica ed using he Bio up e Pico (Diagenode) wi h eigh cycles (30 s on/30 s o ). A e wa d, he samples we e spun a 16,000 3g o 5 min a oom empe a u e and he supe na an was s o ed a 80C. Ch oma in om 100,000 cells was used o ChIPmen a ion. ChIPmen a ion was pe o med as desc ibed by Schmidl e al., 46 wi h se e al modi ica ions. In sho , he ch oma in was incuba ed o e nigh a 4C o a ing in dilu ion bu e (1% T i on X-100, 1.2 mM EDTA, 16.7 mM T is pH 8, 167 mM NaCl), 1 3P o ease Inhibi o Cock ail 47 and 1mg o an ibody [H3K27ac (C15410196, Diagenode)] in a o al olume o 300 ml. The nex day pe ChIP 10ml p o ein A Dynabeads and 10ml p o ein G Dynabeads (bo h In i ogen) we e added. Beads we e washed be o e use wi h dilu ion bu e (+0.15% SDS, +0.1% BSA) and incuba ed wi h he ch oma in and an ibody mix o 60 min a 4C o a ing. A e wa d, he beads we e washed a 4C once wi h ChIP wash bu e 1 (2 mM EDTA, 20 mM T is pH 8, 1%T i on X-100, 0.1% SDS, 150 mM NaCl), wice wi h ChIP washbu e 2 (2 mM EDTA, 20 mM T is pH 8, 1% T i on X-100, 0.1%SDS, 500 mM NaCl), and wice wi h TE (1 mM EDTA, 10 mM T ispH 8). Beads we e esuspended in 24 ml Tagmen DNA bu e (Nex e a) and 1ml Tn5 enzyme (p oduced in-house) and incuba ed o 10 min a 37C wi h 550 pm shaking. A e wa d, he beads we e washed wice wi h WBI (20 mM HEPES, 150 mM NaCl, 0.1% SDS, 0.1% DOC,1% T i on X-100, 1 mM EDTA, 0.5 mM EGTA) and wice wi h WBIV (20 mM HEPES, 1 mM EDTA, 0.5 mM EGTA) wi h 5 min o a ing a oom empe a u e in be ween. Samples we e dec osslinked o 1 h a 55C 1,000 pm shaking ollowed by an o e nigh incuba ion a 65C using elu ion bu e (0.5% SDS, 300 mM NaCl,5 mM EDTA, 10 mM T is pH 8) and p o einase K. Samples we e incuba ed one addi ional hou he nex day wi h elu ion bu e and p o einase K o 1 h a 55C. The samples we e pu i ied using 2 3SPRI AMPu eXP beads. qPCR was used o de e mine he su icien amoun o PCR cycles needed o ampli y he lib a y. The lib a ies we e ampli ied using he KAPA HiFi Ho s a Ready Mix (KAPA Biosys ems) and Nex e a Index Ki 1 (i7) and 2 (i5) p ime s (Illumina). Ampli ied lib a ies we e pu i ied using a 0.65 3SPRI AMPu eXP beads incuba ion ollowed by a 1.8 3SPRI AMPu eXP beads incuba ion. Lib a y concen a ion was measu ed using he KAPA Lib a y Quan i ica ion Ki (KAPA Biosys ems); lib a y size was de e mined using he BioAnalyze High Sensi i i y DNA Ki (Agilen ). Sequencing was pe o med using an Illumina HiSeq 2000, and 50-bp pai ed-end eads we e gene a ed. ChIP-seq analysis Sequenced eads we e quali y and adap o immed wi h im galo e and cu adap (K uege , Felix, T im galo e. ‘‘A w appe ool a ound Cu adap and Fas QC o consis en ly apply quali y and adap e imming o Fas Q iles 516 (2015): 517. ‘‘Pica d Toolki .’’ 2020. B oad Ins i u e, Gi Hub; Ma in, Ma cel. Cu adap emo es adap o sequences om high- h oughpu sequencing eads. EMBne .jou nal). These eads we e aligned agains he UCSC Rhesus e e ence genome (RheMac10) wi h Bu ows-Wheele Aligne (BWA) p og am 48 wi h de aul pa ame e s. Duplica e eads we e emo ed wi h pica d Ma kDuplica es. Peaks we e called wi h MACS2 and de aul se ings 49 and used o de ec he binding si es o ChIP-seq acks. Tags wi hin a gi en egion we e coun ed and adjus ed o ep esen he numbe o ags wi hin a 1 kb egion. Subsequen ly he pe cen age o hese ags as a mea- su e o he o al numbe o sequenced ags o he sample was calcula ed and displayed as hea maps using lu . 50 To de e mine genomic loca ions o binding si es, he peak ile was analyzed using a sc ip ha anno a es binding si es acco ding o all Re Seq genes. Nec opsy and pos -mo em e alua ion Rhesus monkeys we e scheduled by p o ocol o eu hanasia and nec opsy a 2 weeks (3 animals o BCG IV ) o 8 weeks (BCG ID , BCG MUC , MTBVAC ID , MTBVAC MUC ) a e accina ion. A he end o he expe imen blood (< 45 ml) and bone ma ow was collec ed be o e eu hanasia wi h sodium pen oba bi al (> 50 mg/kg). Samples we e s o ed a 80C o analysis a la e ime poin . QUANTIFICATION AND STATISTICAL ANALYSIS Da a handling o measu emen s o cy okine p oduc ion by Mul iplex The Legendplex TM is a bead-based assay o low cy ome y allowing o he measu emen o mul iple cy okines in he same sam- ple. Bead coun pe analy e was ±300/sample. Da a ob ained wi h a bead coun a bi a ily se a < 25 was excluded. S anda d cu e (pg/mL) aken along on each pla e: 10000, 2500, 625, 156, 39.1, 9.8, 2.4 pg/mL. PRE and WK2/WK8 samples om indi idual animals we e measu ed on he same pla e o a oid in e -pla e a ia ion. Samples om animals om di e en ea men g oups we e andomly dis ibu ed o e pla es. I lowe limi o quan i ica ion (LLOQ) was < 2.4 pg/mL, he lowes de ec able concen a ion was se a 2.4 pg/mL. I LLOQ was > 2.4 pg/mL, he lowes de ec able concen a ion was se a LLOQ. This da a p ocedu e ensu ed a conse a i e es ima ion o he old change be ween PRE and WK2/WK8 pos accina ion. S a is ics compa ing ou es o accina ion, cy okine, and lac a e p oduc ion S a is ical pa ame e s including numbe o animals, pe cen age o cells le els o cy okine p oduc ion, old inc ease o cy okine p oduc ion and absolu e inc ease in lac a e p oduc ion a e epo ed in he Figu es and Figu e legends. Da a was analyzed by ANOVA adjus ed o mul iple compa isons, Wilcoxon’s non-pa ame ic pai ed analysis es , Wilcoxon signed ank es o whe e applicable. S a is ical analyses we e conduc ed using G aphPad P ism 7 so wa e ( e sion 7.0d; G aphPad So wa e, La Jolla, CA). Cell Repo s Medicine 2, 100185, Janua y 19, 2021 e5 A icle ll OPEN ACCESS S a is ics epigene ics Peaks we e called wi h MACS2 and de aul se ings 49 and used o de ec he binding si es o ChIP-seq acks. All peaks we e me ged using bed ools and o each sample ags pe peak we e coun ed. Di e en ial peaks we e called by compa ing he wo se s o samples using hese c i e ia: he sum o coun ed ags wi hin one peak using all samples should be > 200 and he old di e ence should be > median plus 2xSD. The es ing o gene se /pa hway en ichmen s was done using Fishe ’s exac es ing. e6 Cell Repo s Medicine 2, 100185, Janua y 19, 2021 A icle ll OPEN ACCESS