Resea ch pape
The apeu ic e ficacy o pulmona y li e ube culosis accines agains
es ablished as hma by sub e ing local immune en i onmen
Raquel Ta anc
on
a,b
, Elena Ma a
a,b
, San iago U anga
a,b
, Ana Bel
en G
omez
a,b
,
Dessisla a Ma ino a
a,b
, Isabel O al
a,b
, Ca los Ma ín
a,b,c
, Nacho Aguil
o
a,b,
*
a
G upo de Gen
e ica de Micobac e ias, Dp o. Mic obiología, Medicina P e en i a y Salud P
ublica, Uni e sidad de Za agoza, ISS A ag
on, C/ Domingo Mi al s/n, Za a-
goza 50009, Spain
b
CIBER En e medades Respi a o ias, Ins i u o de Salud Ca los III, Mad id 28029, Spain
c
Se icio de Mic obiología, Hospi al Uni e si a io Miguel Se e , ISS A ag
on, Paseo Isabel la Ca 
olica 1-3, Za agoza 50009, Spain
ARTICLE INFO
A icle His o y:
Recei ed 14 May 2020
Re ised 4 Decembe 2020
Accep ed 10 Decembe 2020
A ailable online 18 Janua y 2021
ABSTRACT
Backg ound: Subs an ial ecen ad ances in he comp ehension o he molecula and cellula mechanisms
behind as hma ha e e idenced he impo ance o he lung immune en i onmen o disease ou come, mak-
ing modula ion o local immune esponses an a ac i e he apeu ic a ge agains his pa hology. Li e a en-
ua ed mycobac e ia, such as he ube culosis accine BCG, ha e been classically linked wi h a ype 1
esponse, and p oposed as possible modula o s o he ype 2 esponse usually associa ed wi h as hma.
Me hods: In his s udy we used di e en acu e and ch onic mu ine models o as hma o in es iga e he he a-
peu ic e ficacy o in anasal deli e y o he li e ube culosis accines BCG and MTBVAC by egula ing he lung
immune en i onmen associa ed wi h ai way hype esponsi eness (AHR).
Findings: In anasal adminis a ion o BCG, o he no el ube culosis accine candida e MTBVAC, ab oga ed
AHR-associa ed hallma ks, including eosinophilia and lung emodeling. This co ela ed wi h he e-pola iza-
ion o alle gen-induced M2 mac ophages owa ds an M1 pheno ype, as well as wi h he induc ion o a
s ong alle gen-specific Th1 esponse. Impo an ly, accine ea men was e ec i e in a scena io o es ab-
lished ch onic as hma whe e a s ong eosinophil infil a ion was al eady p esen p io o immuniza ion. We
finally compa ed he nebuliza ion e ficiency o clinical o mula ions o MTBVAC and BCG using a s anda d
comme cial nebulize o po en ial ae osol applica ion.
In e p e a ion: Ou esul s demons a e ha pulmona y li e ube culosis accines e ficien ly e e es ab-
lished as hma in mice. These da a suppo he u he explo a ion o his app oach as po en ial he apy
agains as hma.
Funding: Spanish Minis y o Science [g an numbe s: BIO2014-5258P, RTI2018-097625-B-I00], Ins i u o de
Salud Ca los III, Gobie no de A ag
on/Fondo Social Eu opeo, Uni e si y o Za agoza [g an numbe : JIUZ-
2018-BIO-01].
© 2020 The Au ho (s). Published by Else ie B.V. 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/)
Keywo ds:
Eosinophilia
Es ablished as hma
Li e ube culosis accines
Pulmona y mac ophages
Th2/Th1 lymphocy es
Respi a o y immuniza ion
1. In oduc ion
As hma has eached pandemic le els, wi h mo e han 300 million
indi iduals a ec ed all a ound he wo ld. As hma is especially p e a-
len in de eloped coun ies compa ed o low- and middle-income
scena ios. One o he mos accep ed explana ions o hese di e en-
ces comes om he named “Hygiene hypo hesis”, which sugges s
ha as hma de elopmen is a ou ed by he lowe exposu e o chil-
d en o de e mined en i onmen al ac o s [1]. In his ega d, expo-
su e o ce ain mic oo ganisms and mi es (as hose p esen in a ms)
du ing ea ly li e s ages migh con ibu e o educa ing he immune
sys em, leading o he acquisi ion o highe ole ance o alle gens
[2,3].
As hma is a he e ogeneous disease cha ac e ized by ch onic
ai way inflamma ion and emodeling. E en hough as hma can
be associa ed wi h di e en ypes o inflamma o y esponse, ype
2inflamma ion is p esen in mo e han 80% o as hma cases in chil-
d en. T helpe (Th) lymphocy es wi h a Th2 p ofile a e p esen in
mos o he pa ien s, p oducing cy okines as IL-4, IL-5 o IL-13, which
a e esponsible o some o he cha ac e is ic clinical symp oma ol-
ogy. IL-5 plays a cen al ole in he su i al and ec ui men o eosi-
nophils, one o he main playe s in as hma, and whose p esence in
spu um ep esen s one o he mos accep ed bioma ke s o he
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (N. Aguil
o).
h ps://doi.o g/10.1016/j.ebiom.2020.103186
2352-3964/© 2020 The Au ho (s). Published by Else ie B.V. 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/)
EBioMedicine 64 (2021) 103186
Con en s lis s a ailable a ScienceDi ec
EBioMedicine
jou nal homepage: www.else ie .com/loca e/ebiom
diagnosis o he disease. In addi ion, IL-4 and IL-13 igge ai way
emodeling by inducing p oli e a ion o ai way epi helial cells as well
as exace ba ed mucus p oduc ion [4].
In addi ion o adap i e esponse, o e he las yea s di e en
s udies ha e e idenced he c ucial impo ance o lung inna e popula-
ions o as hma igge ing. Indeed, alle gen p esen a ion h ough
MHC-II molecules om an igen-p esen ing cells (APC) esul s essen-
ial o he induc ion o alle gen-specific T cells [5]. As hma has been
linked wi h a pa hological mac ophage pola iza ion owa ds an M2
pheno ype, as exace ba ed le els o M2 mac ophages ha e been
ound in as hma animal models [6,7] as well as in samples om
pa ien s [8]. M2 mac ophages adop egula o y skills and igge an
immune modula o y en i onmen ha impai s Th1 esponse and
a ou s expansion o Th2 cells [9]. M2 is a simplified e minology
ha encloses di e en subse s o mac ophages wi h egula o y skills.
Thus, h ee di e en ypes o M2 mac ophages ha e been defined,
M2a, M2b and M2c, each wi h i s own peculia i ies. In he pa icula
case o M2a mac ophages, hei p esence has been linked wi h an
induc ion o Th2 adap i e esponse [10]. Wi h ega d o alle gic
as hma, M2a mac ophages can con ibu e o igge ing he alle gen-
specific T cell esponse by a leas wo di e en ways, including p e-
sen a ion o alle gen-de i ed pep ides o T lymphocy es h ough
MHC-II molecules, and by sec e ion o cy okines such as IL-4, which
d i e T cell esponse owa ds a Th2 p ofile [11]. The apies a ge ing
M2 mac ophages ha e been shown o alle ia e alle gic esponsi e-
ness [12]. Thus, exace ba ed M2 mac ophage ac i a ion ep esen s a
highly a ac i e oppo uni y o design no el immunomodula o y
ea men s a ge ing his misbalance in lung mac ophage popula-
ions [13].
The cu en ube culosis accine BCG is he mos adminis e ed
accine in his o y. As li e-a enua ed Mycobac e ium bo is, BCG ac-
cine is classically conside ed a Th1 esponse-p omo ing s imulus and
as such, he benefi s o in ade mal BCG accina ion o as hma ha e
been widely discussed [14]. Di e en obse a ional s udies sugges
ha accina ion wi h BCG could p o ide p o ec ion in he de elop-
men o ce ain alle gies, including ash ma [15]. Rema kably, an
in e en ional s udy conduc ed in Sou h Ko ea e idenced ha he
g oup o as hma pa ien s ea ed wi h BCG p esen ed an imp o e-
men o lung unc ion in he nex days ollowing accina ion, in com-
pa ison wi h he placebo g oup [16].
A a p eclinical le el, whole-cell BCG, ei he li e o inac i a ed, as
well as di e en mycobac e ial componen s, ha e been ex ensi ely
p o en o be e ficien agains as hma in di e en animal models [17-
19]. Howe e , mos o hese esul s ha e been ob ained wi h BCG
deli e ed p io o o concu en ly wi h alle gen sensi iza ion. As a
esul , he he apeu ic capaci y o BCG o e e es ablished as hma
has no been elucida ed. In addi ion, p e ious s udies ha e ocused
mainly on he Th1/Th2 esponse balance wi hou conside ing o he
componen s o he immune sys em, such as pulmona y mac ophages,
which seem o play majo ole du ing as hma de elopmen .
In he p esen s udy, we mimicked he na u al ou e o ube culo-
sis in ec ion by in anasal deli e y o li e ube culosis accines, o
e alua e he he apeu ic e ficacy o his app oach in di e en models
o ai way hype esponsi eness (AHR). We hypo hesized ha di ec
in e play be ween accines and he lung compa men migh modu-
la e he immune en i onmen associa ed wi h as hma. Ou esul s
e ealed ha BCG was able o e-educa e M2 mac ophages induced
by alle gen adminis a ion owa ds an M1 pheno ype, as well as o
con e alle gen-specific Th2 lymphocy es o Th1. In addi ion, we
assessed o he fi s ime he e ficacy o a li e-a enua ed Mycobac e-
ium ube culosis accine, called MTBVAC, agains as hma. MTBVAC is
cu en ly unde clinical e alua ion, and o da e i emains he fi s
and only li e accine based on a enua ed M. ube culosis ha has
eached he clinic [20-22]. Impo an ly, ou da a showed s ong he -
apeu ic e ficacy o bo h BCG and MTBVAC in alle gen-challenged
mice in a scena io o es ablished disease, demons a ing he po en ial
o li e a enua ed ube culosis accines as he apy o as hma.
2. Ma e ials and me hods
2.1. E hics
Expe imen al wo k was conduc ed in ag eemen wi h he Spanish
Policy o Animal P o ec ion RD53/2013 and he Eu opean Union
Di ec i e 2010/63 o he p o ec ion o animals used o expe imen al
and o he scien ific pu poses. Expe imen al p ocedu es we e
Resea ch in con ex
E idence be o e his s udy
The cu en ube culosis (TB) accine, BCG, is he mos adminis-
e ed accine in his o y. Since BCG, a li e-a enua ed accine, is
classically conside ed a Th1 esponse-p omo e , he benefi s o
in ade mal BCG accina ion o as hma ha e been widely
assessed, al hough epidemiological e idences a e con o e sial,
wi h no clea benefi s demons a ed. Con e sely, di e en
s udies sugges a ela ionship be ween highe ube culosis
in ec ion a es and lowe p e alence o as hma. Indeed, we
ecen ly demons a ed his co ela ion in TB-in ec ed mice sub-
jec ed o an expe imen al model o alle gen-induced as hma.
Since ube culosis in ec ion is mos ly acqui ed by he espi a-
o y ou e, whe eas BCG is gi en in ade mally, e idences om
human and animal s udies sugges ha he beneficial e ec s o
li e mycobac e ia agains as hma migh es on ha bac e ia
physically each he lungs o induce an as hma an agonis ic
esponse a a local le el.
Added alue o his s udy
In he p esen s udy, in an a emp o mimic he na u al ou e o
ube culosis in ec ion, we in es iga ed he he apeu ic e ficacy
o in anasal li e-a enua ed mycobac e ial ube culosis ac-
cines in di e en acu e and ch onic mu ine models o ai way
hype esponsi eness (AHR). Ou esul s showed ha in anasal
accine adminis a ion e e ed AHR-associa ed esponse in all
scena ios assayed, bo h in sho - e m and long- e m acu e
models, and also in an es ablished as hma scena io induced by
ch onic alle gen exposu e, whe e a s ong ai way eosinophilia
was al eady p esen p io o accine adminis a ion. Ou esul s
indica e ha accine ea men sub e ed as hma-associa ed
ype 2 esponse, including epola iza ion o alle gen-specific
Th2 lymphocy es in o Th1. In addi ion, we desc ibed o he
fi s ime he use o he li e-a enua ed mycobac e ial TB ac-
cine MTBVAC agains as hma. MTBVAC is a accine candida e
cu en ly unde clinical e alua ion, and he e o e al e na i e
applica ions o his accine a e ele an om a ansla ional
poin o iew.
Implica ions o all he a ailable e idences
F om egula o y poin o iew, ae osol (bu no in anasal)
deli e y migh be he only accep ed ou e o adminis a ion o
li e ube culosis accines in o he lung compa men . In his
s udy, we ha e e alua ed he e ficacy o nebuliza ion in i o o
wo clinical o mula ions o MTBVAC and BCG in a s anda d
comme cial nebulize , demons a ing he easibili y o eaching
he apeu ic bac e ial doses using s anda d nebuliza ion de ices.
Al oge he , ou da a suppo he u he explo a ion o pulmo-
na y applica ion o li e-a enua ed mycobac e ia as po en ial
he apy agains as hma.
2R. Ta anc
on e al. / EBioMedicine 64 (2021) 103186
app o ed by he E hics Commi ee o Animal Expe imen s o Uni e -
si y o Za agoza. (p o ocol PI22/15).
2.2. Bac e ia
BCG Danish SSI (Pfize ), BCG Pas eu (s ain 1173P2, Ins i u Pas-
eu Pa is, F ance), MTBVAC(20) (Uni e si y o Za agoza) and
MTBVACDe p [23] (Uni e si y o Za agoza) s ains we e g own a
37 °C in Middleb ook 7H9 b o h (Di co) supplemen ed wi h ADC 10%
(Di co) and 0.05% ( / ) Tween-80 (Sigma), Sau on’s medium o on
solid Middleb ook 7H10 (Di co) aga medium supplemen ed wi h
ADC 10%. BCG Pas eu and MTBVAC we e ans o med wi h he epli-
ca i e pJKD6 plasmid encoding g een fluo escen p o ein (GFP) (a
kind gi om Luciana Lei e, Bu an an Ins i u e, B azil). MTBVAC
hea -killed was ob ained by boiling an MTBVAC cul u e a 100 °C o
15 min. MTBVAC p oduced unde GMP condi ions was p o ided by
Bio ab i (Spain). OncoTICEÒwas pu chased om MSD. Bac e ial sus-
pensions o accina ion we e p epa ed in PBS om glyce ol s ocks
p e iously quan ified by pla ing se ial dilu ions.
2.3. Animal s udies
All mice we e kep unde con olled condi ions and obse ed o
any sign o disease.
Fo induc ion o OVA-specific AHR, 8 o 10 weeks old C57BL/6
(Jan ie Biolabs) emale mice we e sensi ized by wo in ape i oneal
injec ions o 50 mg chicken egg o albumin (lyophilized powde ,
98% (Sigma)) wi h 2 mg aluminum hyd oxide (Sigma, S . Louis, MO)
one week apa . One week a e second sensi iza ion, mice we e
gi en a single in anasal adminis a ion o BCG, MTBVAC o MTBVAC-
de i ed accine a he dose indica ed in he figu e legends, esus-
pended in 40ml o PBS. In he acu e model, ou weeks a e accine
adminis a ion, animals we e in anasally challenged wi h 100 mg
OVA in s e ile PBS o 3 consecu i e days and one day a e he las
challenge, mice we e humanely sac ificed. In some o he expe i-
men s, OVA challenge was delayed un il 4 mon hs a e BCG accina-
ion. In he ch onic OVA model, h ee weeks a e immuniza ion,
mice we e in anasally challenged w h 10 mg OVA wice pe week
du ing eigh weeks. In his case, accines we e adminis e ed a week
9 o he p ocedu e, in he middle o he challenge phase. Fo HDM-
induced ch onic AHR, mice we e in anasally challenged wice a
week o h ee consecu i e weeks wi h 10mg HDM. Vaccines we e
deli e ed a week 4 o he expe imen (i.e., one week a e p ima y
HDM challenge), and one mon h la e , mice we e in anasally chal-
lenged wi h 10 mg HDM o h ee consecu i e days. One day
a e las HDM adminis a ion animals we e humanely sac ificed.
Dexame hasone ea men was gi en acco ding o [24]. B iefly, OVA-
sensi ized mice we e ea ed in ape i oneally wi h 5 mg/kg dexa-
me hasone (DEX) (Dexame hasone-Wa e Soluble, Sigma-Ald ich)
he day p io o ini ia ing he challenge phase, and hen h ee addi-
ional DEX inocula ions we e gi en 1 hou be o e each OVA in ana-
sal adminis a ion.
Fo b onchoal eola la age (BAL) collec ion, achea was cannu-
la ed and BAL pe o med wi h 0.8 ml o ice-cold PBS. Supe na an
was sepa a ed om cells by cen i uga ion o 5 min a 4500 xg.
Lungs we e emo ed asep ically. Fo ob aining cellula suspen-
sions, hey we e added o HEPES bu e (HEPES 10 mM; NaCl 0,15 M;
KCl 5 mM; MgCl2 1 mM; CaCl2 1,8 mM pH 7,4) con aining collage-
nase D 100 mg/ml (Roche) and DNAseI 400 IU (AppliChem), incu-
ba ed a 37 °C o 30 min, and homogenised using Gen leMACS
(Mil enyi Bio ech) dissocia o wi h he lung specific p og am acco d-
ing o manu ac u e ins uc ions. A e wa ds, esidual ed blood cells
we e lysed using Red Blood Cells Lysing Bu e (Sigma) and he
homogenized fil e ed o elimina e issue deb is. Fo bac e ial bu den
de e mina ion, lungs we e homogenized wi h he Gen leMACS, using
he RNA p o ocol, and hen pla ed on o aga medium 7H10
supplemen ed wi h ADC. In he case o his ological analysis, lungs
we e fixed wi h 4% o maldehyde solu ion o 24 h p io o he s ain-
ing p ocedu e.
2.4. Flow cy ome y analysis
10
6
lung o BAL cells we e incuba ed o 15 min a 4 °C wi h
Fc ecep o blocking eagen (Mil enyi Bio ech). Then, eosinophil,
neu ophil and mac ophage p esence was de e mined by ex a-
cellula s aining wi h he ollowing an ibodies: CD45-FITC (RRID:
AB_2658216), siglecF-APC (RRID:AB_2653441), Ly-6G-Vioblue
(RRID:AB_2751964), CD11c-PE (RRID:AB_2654707), CD11b-Pe CP/
Cy5.5 (RRID:AB_2751174) om Mil enyi Bio ech. Eosinophils
we e defined as SSC
high
CD45
+
CD11b
+
SiglecF
+
CD11c

; neu ophils
as CD45
+
Ly6G
+
CD11b
+
CD11c

; and Al eola Mac ophages as
CD45
+
SiglecF
+
CD11c
+
CD11b
dim
cells.
Fo in acellula s aining (ICS), a e labeling memb ane p o eins
wi h he abo e-men ioned an ibodies, in addi ion o MHCII-Vioblue
(RRID:AB_2652908) and CD86-PE (RRID:AB_2660746) (Mil enyi),
and CD206-APC (RRID:AB_2739133) (BD Biosciences), cells we e
fixed and pe meabilized wi h he FoxP3 s aining se (Mil enyi Bio-
ech), acco ding o manu ac u e ins uc ions. As in acellula an i-
bodies, we used iNOS-APC (RRID:AB_2727527) and iNOS-PE (RRID:
AB_2727486) (Mil enyi), and A g1-APC (RRID:AB_2734835)(eBio-
sciences). Cells we e acqui ed using a Gallios flow cy ome e (Beck-
man Coul e ) and analyzed wi h Weasel so wa e.
2.5. Cy okine analysis
Quan ifica ion o IL-5, IL-4, IL-13 and IFN-gwas pe o med using
specific comme cial ELISA ki s ollowing manu ac u e ins uc ions
(Mab ech Bio ech). Cy okine de e mina ion in he lungs was done
om o gan explan s. These we e p epa ed by cu ing he lung in o
small pieces and incuba ing hem o e nigh a 37 °C in 0.5 ml o cul-
u e medium.
To analyze OVA specific esponse, medias inal lymph nodes we e
emo ed asep ically and mechanically dis up ed o cell collec ion.
2£10
6
cells we e incuba ed wi h o wi hou OVA a 1 mg/ml o
96 h. Then, supe na an was collec ed o de e mine cy okine concen-
a ion. OVA-specific esponse o each cy okine was calcula ed as
he di e ence be ween cy okine concen a ion ob ained ollowing
OVA s imula ion minus he uns imula ed con ol. Fo ICS, cells we e
incuba ed wi h 1 mg/ml OVA o 1 mg/ml o an i CD3/CD28 (RRID:
AB_394590; RRID:AB_394763, espec i ely) (BD Biosciences) o
24 h, and 10 mg/ml B e eldin A (Sigma) was added du ing he las six
hou s. Fo su ace s aining, cells we e labelled wi h an i-CD4-FITC
(RRID:AB_394582) (BD Biosciences) and an i-CD3-Pe CPVio700
(RRID:AB_2752207) (Mil enyi Bio ec) in cul u e medium wi h 10%
FCS. Then, cells we e fixed and pe meabilized wi h he Cy ofix/Cy o-
pe m Fixa ion/Pe meabiliza ion Ki (BD Biosciences) ollowing manu-
ac u e ins uc ions, and s ained wi h an i-IFNg-APC (RRID:
AB_2784369) and an i-IL5-PE (RRID:AB_2660015) (Mil enyi Bio ech).
2.6. qRT-PCR
Fo RNA ex ac ion, lungs we e imme sed in o TRIzol eagen
(In i ogen) jus upon ha es ing, and ozen immedia ely on d y ice.
Once hawed, lungs we e homogenised wi h he Gen leMACS, using
he RNA 0.2 p o ocol. 200 ml o chlo o o m we e added pe ml o TRI-
zol and a e igo ous o exing, ubes we e cen i uged a 18,000 x g
du ing one hou a 4 °C. Aqueous uppe phase con aining euka yo ic
RNA was eco e ed, added o 700 ml o isop opanol and cen i uged
a 18,000 x g du ing 10 min a 4 °C. The esul ing pelle was washed
wi h 70% E OH and s o ed a 20 °C. Residual DNA was elimina ed by
DNAse ea men , RNA was pu ified wi h an ex ac ion based on phe-
nol-acid-chlo o o m and p ecipi a ed ON a 20 °C wi h isop opanol
R. Ta anc
on e al. / EBioMedicine 64 (2021) 103186 3
and sodium ace a e. cDNA lib a ies we e cons uc ed o gene
exp ession analysis by RT-qPCR. P ime pai s used in he p esen
s udy we e he ollowing:
Gene 50sequence 30sequence
Ac in 50-ACCAGTTCGCCATGGATGAC 50-TGCCGGAGCCGTTGTC
18S 50-TTCGTATTGCGCCGCTAGA 50-CTTTCGCTCTGGTCCGTCTT
Ga a3 50-GACCCGAAACCGGAAGATGT 50-GCGCGTCATGCACCTTTT
Il12a 50-ACGCAGCACTTCAGAATCACA 50-CACCAGCATGCCCTTGTCTA
Il12b 50-TGGAGCACTCCCCATTCCT 50-TGCGCTGGATTCGAACAA
I ng 50-TTGGCTTTGCAGCTCTTCCT 50-TGACTGTGCCGTGGCAGTA
Il5 50TTGACAAGCAATGAGACGATGAG 50-TCCAATGCATAGCTGGTGATTT
Il4 50-GGAGATGGATGTGCCAAACG 50-CGAGCTCACTCTCTGTGGTGTT
Il13 50-TTGAGGAGCTGAGCAACATCAC 50-CCATGCTGCCGTTGCA
S a 1 50-CTCTGGAATGATGGGTGCATT 50-TTGAGCAGAGCGCGTTCTC
S a 4 50-CATTTGCAACCCAAGGAGATG 50-TGGCAGCCCTCGTTTCC
S a 6 50-AACTGCAACGGCTCTATGTTGA 50-AGCCAGTCAGCCAGGAGATG
Tn 50-CAGCCGATGGGTTGTACCTT 50-GGCAGCCTTGTCCCTTGA
T Be 50-ACCTGTTGTGGTCCAAGTTCAA 50-GCCGTCCTTGCTTAGTGATGA
I nb1 50-CCCTATGGAGATGACGGAGAAG 50-GAGCATCTCTTGGATGGCAAA
Ccl11 50-GACCAGGTTGGGCAAAGAGA 50-GGCATCCTGGACCCACTTCT
Ym1 50-GTCTGGCCCCTGGACATG 50AGAGGGAAATGTCTCTGGTGACA
Il1b 50-AGTTGACGGACCCCAAAAGA 50-GGACAGCCCAGGTCAAAGG
Re nla 50-CAGCTGATGGTCCCAGTGAA 50TTCCTTGACCTTATTCTCCACGAT
Nos2 50-GGATCTTCCCAGGCAACCA 50-TCCACAACTCGCTCCAAGATT
A g1 50-GCTCCAAGCCAAAGTCCTTAGA 50-CCTCGAGGCTGTCCTTTTGA
2.7. Nebuliza ion s udies
MTBVACÒand OncoTICEÒGMP ials we e esuspended wi h 1 ml/
ial o eluen indica ed by manu ac u e s, concen a ions no malized
a 10
7
CFU/ml, and 2 ml o each accine p epa a ion placed in he es-
e oi o he clinical nebulize U100 (OMRON). Nebulize was con-
nec ed wi h a plas ic ube o a gas washing flask wi h 5 ml o s e ile
wa e , coupled wi h a acuum pump o eco e he nebulized ac-
ion. Bac e ia we e nebulized o 5 min and bo h nebulized and ese -
oi ac ions we e pla ed on solid aga 7H10 medium supplemen ed
wi h ADC. Nebuliza ion e ficacy index o each accine s ain was cal-
cula ed as he pe cen age o bac e ia in he nebulized ac ion com-
pa ed o ha in he ese oi .
2.8. S a is ics
Mice we e andomly dis ibu ed in g oups o 6 animals pe cage
p io o s a expe imen al p ocedu es. Resul s we e no blinded o
analysis. No s a is ical me hod was used o calcula e sample size in
animal expe imen s. G aphP ism so wa e was used o s a is ical
analysis. S a is ical es s used o each expe imen a e indica ed in
he figu e legends. All s a is ical es s used we e wo- ailed. Ou lie
alues we e de e mined applying he G ubb’s es o all da a se s,
and disca ded om he final s a is ical analysis. Di e ences we e
conside ed significan a p<0.05.
2.9. Role o unding sou ce
The unde s had no ole in s udy design, da a collec ion, in e p e-
a ion and analysis, decision o publish o p epa a ion o he manu-
sc ip .
3. Resul s
3.1. In anasal BCG and MTBVAC p e en AHR in alle gen-sensi ized
mice
Ini ially, we es ed accine e ficacy in an acu e model o AHR
d i en by o albumin (OVA) adminis a ion. Appea ance o as hma
symp oms in humans comes p eceded by an asymp oma ic phase o
alle gen sensi iza ion, which is usual o occu a an ea ly li e s age
du ing childhood [25]. Thus, o make ou conclusions mo e
ep esen a i e o he eal si ua ion, we deli e ed accines o e p e i-
ously sensi ized animals (Fig. 1a). Wi h he objec i e o each he p i-
ma y o gan (lungs) a ec ed du ing as hma episodes, we inocula ed
he accines by he in anasal ou e. Ou long expe ience wi h his
ou e o adminis a ion in mouse models indica e ha a subs an ial
p opo ion o bac e ia eaches he lungs unde he expe imen al se -
ings used. The dose o BCG ini ially adminis e ed was 10
6
CFUs. As
p ima y ma ke o as hma ic esponsi eness we measu ed eosinophil
infil a ion in lungs and espi a o y ai ways he day a e las chal-
lenge (Fig. 1b, 1c, S1), ele an om a clinical poin o iew. Ou da a
e ealed a s ong abili y o in anasal BCG o ab oga e eosinophilia
bo h in BAL and lungs (Fig. 1c-e). In addi ion, we ound no changes in
o he myeloid popula ions, as neu ophils and al eola mac ophages,
sugges ing ha BCG did no igge an uncon olled cellula infil a-
ion in o he espi a o y ai ways and indeed, he numbe o o al leu-
kocy es (CD45+ cells) we e significan ly lowe in he BCG OVA g oup
compa ed o OVA (Fig. 1e). Ano he ypical as hma hallma k is he
emodeling o he ai way epi helium, which is cha ac e ized by a
hickening o he al eola walls and he p oli e a ion o goble cells,
esponsible o mucus p oduc ion and sec e ion. These ea u es we e
obse ed in he lungs om OVA g oup a e s aining wi h Pe iodic
Acid Schi (PAS) echnique, whe e di e en laye s o epi helial cells
could be dis inguished in he al eola walls, wi h an impo an p o-
po ion o PAS-posi i e cells (goble cells). In addi ion, p esence o
mucosubs ances was ound in he ai ways. Rema kably, BCG ea -
men p e en ed his pheno ype. Goble cells we e ha dly isible and
no mucus sec e ion was obse ed in al eola spaces, whe eas ai way
epi helium was es ic ed o one laye (Fig. 1 ).
We also assessed in anasal BCG in a long- e m AHR model, in
which OVA challenge was pe o med ou mon hs a e BCG immuni-
za ion. BCG also educed eosinophilia in his model (Fig. 1g). In e es -
ingly, bac e ial coun ing in lungs a he ime o sac ifice indica ed
ha BCG pe sis ence was d ama ically educed compa ed o he one-
mon h model (Fig. S2a), indica ing ha in anasal BCG could modu-
la e as hma-associa ed immune en i onmen e en a e almos com-
ple e bac e ial clea ance, which migh sugges a con ibu ion o
memo y esponse o he accine-p o ec i e e ec .
MTBVAC is a no el li e ube culosis accine, a enua ed om a
clinical isola e o Mycobac e ium ube culosis, he pa hogen causa i e
o TB in humans, whe eas BCG is a de i a i e accine om Mycobac-
e ium bo is [20]. MTBVAC has demons a ed highe e ficacy and
immunogenici y han BCG agains ube culosis in di e en p eclinical
models [26], and is cu en ly unde clinical e alua ion in newbo n
[22] (ClinicalT ials.go Iden ifie NCT03536117) and adul popula-
ions [21] (NCT02933281) in TB-endemic coun ies in A ica, showing
accep able sa e y and immunogenici y p ofile o da e. In he p esen
s udy, we e alua ed in anasal MTBVAC o he fi s ime in a model
o as hma. Using he OVA-d i en acu e AHR model desc ibed abo e,
we obse ed a s ong capaci y o in anasal MTBVAC o ab oga e
eosinophil infil a ion in o ai ways, compa able o ha obse ed
wi h BCG (Fig. 1h).
To add ess whe he accine pe sis ence could a ec p o ec ion
agains as hma, we compa ed MTBVAC wi h wo o he MTBVAC-
de i ed accines: MTBVACDe p s ain, which con ains an addi ional
dele ion o he gene e p esul ing in a hype a enua ed pheno ype
wi h lowe pe sis ence as p e iously desc ibed [23] and demon-
s a ed in his s udy (Fig. S2b); and MTBVAC killed by hea (MTBVAC
HK). MTBVAC HK did no educe eosinophils ollowing OVA chal-
lenge, whe eas MTBVACDe p showed significan educ ion as com-
pa ed o he OVA g oup bu lowe han ha obse ed wi h MTBVAC
(Fig. 1i). These esul s sugges ed an influence o bac e ial pe sis ence
on p o ec ion. To s udy his in mo e de ail, we conduc ed a dose-
esponse expe imen wi h in anasal BCG and MTBVAC. We obse ed
obus co ela ion o he dose wi h a s onge eosinophilia educ ion,
bo h in BAL (Fig. 1J) and lungs (Fig. 1k). In he case o he lungs, he
highes dose g oups (10
7
MTBVAC and 10
6
BCG) showed a le el o
4R. Ta anc
on e al. / EBioMedicine 64 (2021) 103186
Fig. 1. In anasal BCG and MTBVAC p e en alle gic ai way esponsi eness in alle gen-sensi ized mice. (a) Eosinophils we e de e mined by flow cy ome y in an OVA-d i en acu e
AHR model, wi h mice ea ed in anasally wi h 10
6
BCG CFUs one week a e second sensi iza ion. (b) A ep esen a i e hema oxilin-eosin lung image o OVA-challenged mice. P es-
ence o eosinophils is highligh ed wi h a ows. (c) Eosinophils we e quan ified by flow cy ome y, co esponding o a popula ion SSC
high
SiglecF+CD11b
+
CD11c

. A ep esen a i e
diag am is shown in he figu e. (d) Pe cen age o eosinophils in BAL and lungs wi h espec o CD45+ cells. (e) To al numbe o leukocy es, eosinophils, neu ophils and al eola mac-
ophages (AMs) in BAL de e mined by flow cy ome y. ( ) Rep esen a i e images o PAS-s ained fixed lungs om OVA-challenged mice un ea ed (1,2) o BCG- ea ed (3,4). (g) Eosi-
nophils in BAL using long- e m AHR model, in which OVA challenge was pe o med 4 mon hs a e BCG immuniza ion. (h) Eosinophils in BAL ollowing in anasal ea men wi h
10
6
MTBVAC o BCG CFU. (i) To al numbe o eosinophils in BAL ollowing in anasal ea men wi h 10
6
MTBVAC, o he de i a i e s ains MTBVACDe p and MTBVAC Hea -Killed
(HK). (j, k) Eosinophils in BAL (j) and lungs (k) ollowing in anasal ea men wi h inc easing doses o MTBVAC o BCG. (l) BAL eosinophils compa ison ollowing BCG, MTBVAC and
R. Ta anc
on e al. / EBioMedicine 64 (2021) 103186 5

eosinophils simila o he nega i e con ol, and he e o e we used
hese doses o he subsequen expe imen s. Finally, we assessed
eosinophilia ollowing in anasal MTBVAC o BCG in compa ison
wi h sys emic adminis a ion o dexame hasone (DEX), which is a
s anda d d ug used in as hma ic pa ien s. O esul s showed compa-
able eosinophil educ ion induced by in anasal MTBVAC and BCG
and sys emic DEX adminis a ion (Fig. 1l).
3.2. In anasal accina ion sub e s as hma-associa ed lung immune
en i onmen
We hypo hesized ha he beneficial e ec s agains AHR obse ed
upon in anasal accina ion could be ela ed wi h he abili y o he
accine o locally in e ac wi h esiden phagocy ic cells in he lungs.
To assess in i o in ec ed cell popula ions ollowing in anasal immu-
niza ion, we deli e ed a BCG s ain ha exp essed he g een fluo es-
cen p o ein (GFP). Ou da a demons a ed ha le el o BCG in ec ion
was e ficien enough o disce n in ec ed cells by flow cy ome y
(Fig. 2a). Using a panel o an ibodies o de ec myeloid cell ma ke s,
we iden ified ha he majo i y o BCG-in ec ed cells co esponded o
al eola mac ophages (AMs). No ewo hy, an impo an p opo ion
o in ec ed AMs exp essed high le els o iNOS and CD86, wo well-
known ma ke s o classical mac ophage ac i a ion (o M1 pheno-
ype) (Fig. 2a). We nex compa ed mac ophage pola iza ion o o al
lung mac ophages in OVA-challenged mice ea ed o no wi h BCG.
Pe cen age o iNOS- and CD86-posi i e mac ophages was subs an-
ially highe in he accina ed g oup, whe eas no di e ence was
ound in he case o MHC-II, wi h mos o he mac ophages posi i e
o his ma ke in bo h g oups (Fig. 2b). Impo an ly, exp ession o
CD206, a classical ma ke associa ed wi h M2 pola iza ion, was
g ea e in mac ophages om OVA g oup compa ed o hose o he
accina ed g oup, which exp essed simila le els o CD206 as naï e
cells (Fig. 2c). To complemen hese da a, we pe o med ano he
expe imen in which we isola ed lung RNA om un ea ed o BCG-
ea ed OVA-challenged mice. Exp ession analysis o genes linked
wi h mac ophage pola iza ion clea ly demons a ed ha BCG immu-
niza ion led o an up egula ion o he M1 p ofile-associa ed genes
Nos2 and Il1b, whe eas i inhibi ed exp ession o he M2 ac i a ion
ma ke s Ym1,A g1 and Re lna (Fig. 2d). In e es ingly, in a dose-
esponse expe imen wi h MTBVAC and BCG, iNOS exp ession was
ound highes in he g oups ea ed wi h he mos p o ec i e accine
doses (10
7
MTBVAC and 10
6
BCG) (Fig. 2e), sugges ing a co ela ion
be ween M1 pheno ype and p o ec ion agains as hma. Al oge he ,
ou da a demons a ed he en ichmen o M2-like mac ophages du -
ing as hma ic esponsi eness, and sugges ed ha he he apeu ic
e ficacy o li e a enua ed mycobac e ia could es on he e e sion
o his esponse by inducing classical ac i a ion o lung mac ophages.
We nex s udied pulmona y Th1 and Th2 esponses, whose misbal-
ance owa ds he la e one has been classically connec ed wi h he
immune pa hogenesis o as hma. Gene exp ession analysis o lung
RNA e ealed ha BCG accina ion led o induc ion o Th1-associa ed
genes as I ng,Il12a, o he ansc ip ion ac o Tbe . Con e sely, genes
codi ying o ypical Th2 cy okines and chemokines, as IL-5, IL-4, IL-13
o CCL-1, we e down modula ed by BCG (Fig. 3a) compa ed o
un ea ed mice. In consonance wi h hese esul s, IL-5 and IL-4 cy o-
kines we e ound ele a ed in BAL and lungs espec i ely in he OVA
con ol g oup, whe eas IFNgwas inc eased when ea ed wi h BCG
(Fig. 3b, c). In addi ion o o al cy okine le els, we also s udied alle -
gen-specific T cells a e ex i o OVA s imula ion o ha es ed lympho-
cy es om medias inal lymph nodes. Da a showed a highe IL-4, IL-5
and IL-13, and lowe IFNgOVA-specific p oduc ion in he OVA g oup
compa ed o BCG- accina ed mice (Fig. 3d-g). In anasal MTBVAC
showed a simila abili y as BCG o shi CD4+ cells owa ds Th1 p ofile
(Fig. S3). We also ound a co ela ion be ween accine-induced Th1
p ofile and bac e ial dose, indica ing he impo ance o bac e ial pe -
sis ence o he shi o Th2 esponse owa d Th1 (Fig S4).
Impo an ly, using in acellula s aining and flow cy ome y we
di ec ly isualized cy okine-p oducing CD4+ Tcells exp essing he
memo y ma ke CD44. Bo h ollowing s imula ion wi h CD3/CD28 o
wi h OVA, we obse ed an opposi e T cell esponse pola iza ion
be ween OVA con ols and BCG- ea ed mice. BCG ea men ab o-
ga ed OVA-specific IL-5-p oducing cells, whe eas i igge ed alle -
gen-specific Th1 IFNg-p oducing CD4+ lymphocy es (Fig. 3h, i). This
sugges s ha T cell-d i en IFNgp oduc ion in BCG- ea ed mice is
no only gene a ed by expanded lymphocy es ha ecognize myco-
bac e ial an igens, bu also by alle gen-specific T cells, ha migh be
e-shaped om a Th2 owa ds a Th1 p ofile due o accine-associa ed
inflamma o y en i onmen .
3.3. BCG and MTBVAC inhibi eosinophilia and lung emodeling in a
scena io o es ablished as hma
Wi h he aim o elucida e he he apeu ic po en ial o li e a enu-
a ed accines agains es ablished as hma, we conduc ed expe imen s
using a ch onic AHR model wi h mul iple OVA challenges, whe e ac-
cines we e deli e ed hal way h ough he challenge phase (Fig. 4a).
BAL eosinophils we e de e mined in an addi ional g oup sac ificed
he day p io o accina ion, a week 8, confi ming eosinophil infil a-
ion a he ime o accine adminis a ion (Fig. 4b). Da a a he end o
he p ocedu e, a week 13, e ealed ha eosinophils we e educed in
he ea ed g oup, indica ing he capaci y o BCG o o e come es ab-
lished alle gen-induced eosinophilia (Fig. 4c).
In ano he expe imen , we also obse ed ha MTBVAC was able
o e e eosinophilia unde simila expe imen al se ings as BCG
(Fig. 4d). In his expe imen we addi ionally assessed lung emodel-
ing by PAS s aining and his ological e alua ion. Ou da a e ealed an
impo an diso ganiza ion o he al eola epi helium su ace in he
OVA con ol g oup, as well as a subs an ial p esence o goble cells.
This pheno ype was no obse ed ollowing MTBVAC ea men
(Fig. 4e). We also analyzed by flow cy ome y exp ession o he M2
and M1 pola iza ion ma ke s A g-1 and iNOS, espec i ely. A signifi-
can p opo ion o A g-1-posi i e mac ophages was ound in he OVA
g oup bu no in he MTBVAC- ea ed mice. Con e sely, iNOS-posi-
i e mac ophages we e de ec ed in he MTBVAC g oup, sugges ing
he abili y o he accine o e-pola ize as hma-associa ed M2 mac o-
phages owa ds M1 p ofile (Fig. 4 , g).
We nex assessed he he apeu ic e ficacy o in anasal BCG and
MTBVAC in a model o es ablished AHR induced by house dus mi es
(HDM), a ele an alle gen in clinical as hma. Alle gen-induced ai -
way esponsi eness was igge ed by successi e in anasal HDM
challenges (Fig. 5a), which elici ed a s ong eosinophil infil a ion in
he BAL (Fig. 5b). BCG and MTBVAC adminis a ion esul ed in abol-
ishmen o BAL eosinophilia, as measu ed in equency (Fig. 5b) and
absolu e numbe o cells (Fig. 5c). PAS-s ained lungs showed ha
HDM exposu e in un ea ed mice s ongly induced p oli e a ion o
he al eola epi helium and appea ance o goble cells, a pheno ype
ha was no obse ed in he accine- ea ed g oups (Fig. 5d).
Finally, we obse ed ha accine adminis a ion diminished p o-
duc ion o Th2 cy okines and subs an ially inc eased IFNgin BAL
(Fig. 5e, g) and lymph node cells ollowing s imula ion wi h HDM
(Fig. 5 , h).
3.4. MTBVAC and BCG ae osoliza ion easibili y wi h a clinical nebulize
The e a e s ong egula o y sa e y conce ns abou in anasal
deli e y o accines in humans [27]. Howe e , ae osol adminis a ion
dexame hasone (DEX) ea men . Da a in he g aphs a e pooled means§SEM om h ee independen expe imen s (d, e, h), wo (i), o one (g, j, k, l). A minimum o 6 mice we e used
pe g oup and expe imen . *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, by wo-way ANOVA (a, e,) o by one-way ANOVA (g, h, i, j, k, l), wi h Bon e oni pos - es .
6R. Ta anc
on e al. / EBioMedicine 64 (2021) 103186
Fig. 2. BCG in anasal in ec s lung esiden mac ophages and induces classical ac i a ion. (a) G oups o mice we e immunized wi h 10
6
CFU o GFP-exp essing BCG. One mon h la e ,
in ec ed cells we e moni o ed and cha ac e ized by flow cy ome y. Exp ession o M1-pola iza ion ma ke s iNOS and CD86 was analyzed. Rep esen a i e diag ams a e shown in he
figu e. (b) Pe cen age o iNOS-, CD86- and MHC- posi i e lung mac ophages in OVA-challenged un ea ed o BCG- ea ed. (c) Su ace exp ession o he M2-ac i a ion ma ke CD206.
Rep esen a i e o e lay his og am showing CD206 su ace exp ession o indica ed expe imen al g oups. G aph shows compa ison o Mean Fluo escence In ensi y (MFI) co espond-
ing o CD206 le el o exp ession. (d) M1 and M2 ac i a ion ma ke s measu ed by qRT-PCR in lungs om OVA-challenged mice, un ea ed o BCG- ea ed. (e) Pe cen age o iNOS-
posi i e cells in lung mac ophages ollowing in anasal ea men wi h inc easing doses o MTBVAC o BCG. Da a in he g aphs a e ep esen a i e mean§SEM om wo independen
expe imen s (b, c) o one (d, e). A minimum o 6 mice was used pe g oup and expe imen . *p<0.05; **p<0.01; ***p<0.001; by unpai ed single (b, c) o mul iple (d) -s uden es ; o
by one-way ANOVA (e) wi h Bon e oni pos - es .
R. Ta anc
on e al. / EBioMedicine 64 (2021) 103186 7
Fig. 3. BCG in anasal eshapes alle gen-specific esponse in o a Th1 p ofile. (a) Th1 and Th2 ac i a ion ma ke s measu ed by qRT-PCR in lungs om OVA-challenged mice,
un ea ed o ea ed wi h BCG one week a e sensi iza ion. (b) IL-5 de e mina ion in BAL. (c) IL-4 and IFNgde e mina ion in lung explan s (d-g) Alle gen-specific IL-4, IL-5, IL-13
and IFNgp oduced by medias inal lymph node cells, ollowing ex i o s imula ion wi h OVA. Da a a e ep esen ed ollowing sub ac ion o he alue ob ained in he absence o
alle gen. (h, i) IL-5- and IFNg- p oducing cells isualized by in acellula s aining and flow cy ome y ollowing ex i o s imula ion wi h an i CD3/CD28 o wi h OVA. Rep esen a i e
diag ams a e shown. Da a in he g aphs a e pooled means§SEM om h ee independen expe imen s (b-e, g) o one (a, i). A minimum o 6 mice was used pe g oup and expe i-
men . *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, by mul iple -s uden es (a), one-way ANOVA wi h Bon e oni pos - es (b, D-g), and wo-way ANOVA wi h Bon e oni pos -
es (c, i).
8R. Ta anc
on e al. / EBioMedicine 64 (2021) 103186
Fig. 4. BCG and MTBVAC in anasal e e es ablished alle gic ai way esponsi eness in an OVA-d i en ch onic model. (a) Scheme o he expe imen based on an OVA-induced
ch onic model. Sensi ized mice we e challenged om week 3 o 13 wi h wo in anasal inocula ions weekly o OVA 10mg. Vaccines we e deli e ed a week 9, in he hal o he chal-
lenge phase. (b) To al numbe o eosinophils in BAL a week 8, he week be o e accine immuniza ion. (c) To al numbe o eosinophils in BAL a week 13, ou weeks a e BCG ea -
men . (d) To al numbe o eosinophils in BAL a week 13, ou weeks a e ea men wi h 10
7
CFU o MTBVAC. (e) Rep esen a i e images o PAS-s ained fixed lungs om OVA-
challenged mice un ea ed o MTBVAC- ea ed. ( , g) In acellula exp ession o iNOS and A g-1 in lung mac ophages a week 13. Rep esen a i e do -plo diag ams a e shown. Da a
in he g aphs a e pooled means§SEM om wo independen expe imen s (b, c), o one (d, g). A minimum o 6 mice was used pe g oup and expe imen . *p<0.05; **p<0.01;
***p<0.001, by -s uden es (b, e), one-way ANOVA wi h Bon e oni pos - es (b, c), and mul iple -s uden es (g).
R. Ta anc
on e al. / EBioMedicine 64 (2021) 103186 9