Full Te ms & Condi ions o access and use can be ound a
h ps://www. and online.com/ac ion/jou nalIn o ma ion?jou nalCode=k i 20
Vi ulence
ISSN: 2150-5594 (P in ) 2150-5608 (Online) Jou nal homepage: h ps://www. and online.com/loi/k i 20
Le hali y o B ucella mic o i in a mu ine model o
in ec ion depends on he wbkE gene in ol ed in O-
polysaccha ide syn hesis
Sa ia Ouah ani-Be ache, Ma ía P. Jiménez De Bagüés, Jo ge De La Ga za,
Luca F eddi, Juan P. Bueso, Sébas ien Lyonnais, Sascha Al Dahouk, Daniela
De Biase, S ephan Köhle & Alessand a Occhialini
To ci e his a icle: Sa ia Ouah ani-Be ache, Ma ía P. Jiménez De Bagüés, Jo ge De La Ga za,
Luca F eddi, Juan P. Bueso, Sébas ien Lyonnais, Sascha Al Dahouk, Daniela De Biase, S ephan
Köhle & Alessand a Occhialini (2019) Le hali y o B ucella�mic o i in a mu ine model o in ec ion
depends on he wbkE gene in ol ed in O-polysaccha ide syn hesis, Vi ulence, 10:1, 868-878, DOI:
10.1080/21505594.2019.1682762
To link o his a icle: h ps://doi.o g/10.1080/21505594.2019.1682762
© 2019 The Au ho (s). Published by In o ma
UK Limi ed, ading as Taylo & F ancis
G oup.
View supplemen a y ma e ial
Published online: 02 No 2019. Submi you a icle o his jou nal
A icle iews: 832 View ela ed a icles
View C ossma k da a Ci ing a icles: 3 View ci ing a icles
RESEARCH PAPER
Le hali y o B ucella mic o i in a mu ine model o in ec ion depends on he wbkE
gene in ol ed in O-polysaccha ide syn hesis
Sa ia Ouah ani-Be ache
a
, Ma ía P. Jiménez De Bagüés
b
, Jo ge De La Ga za
a
, Luca F eddi
a
#
, Juan P. Bueso
c
,
Sébas ien Lyonnais
d
, Sascha Al Dahouk
e
, Daniela De Biase
, S ephan Köhle
a
*,
and Alessand a Occhialini
a
*
a
IRIM, CNRS, Uni e si y Mon pellie , INSERM, Mon pellie , F ance;
b
Unidad de Tecnología en P oducción y Sanidad Animal, Cen o de
In es igación y Tecnología Ag oalimen a ia, Ins i u o Ag oalimen a io de A agón, Uni e sidad de Za agoza, Za agoza, Spain;
c
Labo a o io
Ag oalimen a io, Gobie no de A agón, Za agoza, Spain;
d
CEMIPAI, CNRS, Uni e si y Mon pellie , Mon pellie , F ance;
e
Depa men o
Biological Sa e y, Ge man Fede al Ins i u e o Risk Assessmen , Be lin, Ge many;
Depa men o Medico-Su gical Sciences and
Bio echnologies, Sapienza Uni e si y o Rome, Labo a o y a ilia ed o he Is i u o Pas eu I alia –Fondazione Cenci Bologne i, La ina, I aly
ABSTRACT
B ucella mic o i was isola ed a decade ago om wildli e and soil in Eu ope. Compa ed o he
classical B ucella species, i exhibi s a ypical i ulence p ope ies such as inc eased g ow h in
human and mu ine mac ophages and le hali y in expe imen ally in ec ed mice. A spon aneous
ough (R) mu an s ain, de i ed om he smoo h e e ence s ain CCM4915
T
, showed inc eased
mac ophage coloniza ion and was non-le hal in mu ine in ec ions. Whole-genome sequencing
and cons uc ion o an isogenic mu an o B. mic o i and B ucella suis 1330 e ealed ha he
R-pheno ype was due o a dele ion in a single gene, namely wbkE (BMI_I539), encoding a pu a i e
glycosyl ans e ase in ol ed in lipopolysaccha ide (LPS) O-polysaccha ide biosyn hesis.
Complemen a ion o he R-s ains wi h he wbkE gene es o ed he smoo h pheno ype and he
abili y o B. mic o i o kill in ec ed mice. LPS wi h an in ac O-polysaccha ide is he e o e essen ial
o le hal B. mic o i in ec ions in he mu ine model, demons a ing i s impo ance in pa hogenesis.
ARTICLE HISTORY
Recei ed 3 Sep embe 2019
Re ised 11 Oc obe 2019
Accep ed 13 Oc obe 2019
KEYWORDS
B ucella; i ulence;
lipopolysaccha ide (LPS);
O-polysaccha ide;
glycosyl ans e ase; ough
pheno ype; a omic o ce
mic oscopy
In oduc ion
B ucellae a e G am-nega i e acul a i e in acellula coc-
cobacilli causing b ucellosis, a majo bac e ial zoonosis
wi h 500,000 human cases globally epo ed e e y yea
[1]. In he las decade, new species o B ucella,suchas
B ucella mic o i, B ucella inopina a and isola es om
Aus alian oden s and amphibians, ha e been desc ibed
[2]. These s ains a e me abolically mo e ac i e, acid-
esis an and as -g owing when compa ed o he well-
known classical, human-pa hogenic B ucella species,
which include B ucella abo us, B ucella meli ensis,
B ucella suis and B ucella canis [2–7]. Thei isola ion
om hi he o unknown wildli e hos s and he en i on-
men aised he ques ion whe he B ucella may be ans-
mi ed om hese ese oi s o li es ock and humans
li ing in o icially b ucellosis- ee a eas o he wo ld.
B. mic o i was isola ed om common ole, ed ox, wild
boa , and soil in Cen al Eu ope and, mo e ecen ly, om
a domes ic ma sh og a m [8,9]. Phylogene ically, his
species is close o hose pa hogenic o human and
li es ock han o he g oup o newly desc ibed a ypical
species/s ains [2,10]. Howe e , in he absence o clinical
epo s, he pa hogenic po en ial o B. mic o i emains o be
e i ied. We we e he i s o desc ibe ha , unlike he
classical B ucella species, B. mic o i is le hal in mice when
injec ed in ape i oneally (i.p.) a a s anda d dose [11]. The
le halpheno ypeinmicedependson he ypeIVsec e ion
sys em Vi B [12] and is also a gene al unambiguous c i e -
ion o es ablish i a speci ic B ucella gene plays a ole in
i ulence o B. mic o i, as wild- ype bac e ia kill he mu ine
hos a he in ec ion dose o 10
5
CFU (colony- o ming
uni s); in con as , in classical species i ulence has been
co ela ed o he capaci y o a s ain o es ablish o main-
ain a ious deg ees o ch onic in ec ion o he spleen and/
o he li e , necessi a ing epea ed bac e ial enume a ion in
hese o gans o ollow up he cou se o in ec ion. On he
o he hand, a sub-le hal doses (≤10
4
CFU), B. mic o i is
apidly clea ed om in ec ed mice, ne e gi es ise o
ch onic in ec ion and con e s p o ec ion [11]. Le hali y in
mice was la e also demons a ed o B. inopina a BO1 and
CONTACT S ephan Köhle [email p o ec ed]
*These au ho s con ibu ed equally o his wo k.
#
Cu en a ilia ion: Uni é des Zoonoses Bac é iennes, ANSES, Maisons-Al o , F ance.
supplemen al da a o his a icle can be accessed he e.
VIRULENCE
2019, VOL. 10, NO. 1, 868–878
h ps://doi.o g/10.1080/21505594.2019.1682762
© 2019 The Au ho (s). Published by In o ma UK Limi ed, ading as Taylo & F ancis G oup.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed
use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.
B ucella s ain 83–210 [13].Weando he sassumed ha
he abili y o hese B ucella species o kill he mu ine hos
may be due o di e ences in su ace an igens, in pa icula
he s uc u e o lipopolysaccha ide (LPS) wi h a possibly
highe endo oxic po en ial [13,14]. Because o i s low endo-
oxici y, he LPS o classical B ucella species is conside ed
as non-canonical in compa ison wi h ha o Esche ichia
coli and o he pa hogenic bac e ia, enabling B ucella o
es ablish ch onic in ec ions and e ade TLR4 de ec ion
[15–17]. The LPS is a majo componen o he ou e
memb ane and consis s o h ee key elemen s: (1) he
lipid A, which p o ides he hyd ophobic LPS ancho in
heou e memb ane,(2)aninne andou e co ecomposed
o b anched-chain oligosaccha ides, and (3) an
O-polysaccha ide (O-PS), linked o he ou e co e and
p o uding in o he ex acellula en i onmen . In
B ucella, he O-PS is cha ac e ized by a homopolyme ic
linea chain o N- o myl-pe osamine esidues linked ia α-
1,2 and/o α-1,3 glycosidic bonds [18]. Depending on he
ela i e abundance and dis ibu ion o hese bonds, he
O-PS p o ides he A, M, and common (C) epi opes widely
used o se o yping [19,20]. Depending on he p esence o
absence o he O-PS, he colony pheno ype is ei he smoo h
(S) o ough (R). All B ucella species ha in ec humans
and li es ock a e na u ally S, excep o B ucella o is and
B. canis [21]. Fo accina ion o li es ock agains b ucello-
sis, S- and R-s ains ha e been used [22].
No ably, a speci ic in e ac ion be ween in ac LPS
and he lipid a s in phagocy ic cells is esponsible
o he selec i e en y o B ucella S-s ains in o he
hos cells and a icking along he endocy ic pa hway
[23–26]. In con as , R s ains do no en e he cell
h ough he lipid a s and a e apidly elimina ed [25].
In his s udy, we cha ac e ized a spon aneous
R-mu an (BmR
SM
) o he B. mic o i e e ence s ain
CCM4915
T
. I s comple e genome sequence helped o
iden i y a mu a ion inac i a ing he wbkE gene, known
o be in ol ed in he syn hesis o O-PS [27]. To co e-
la e his mu a ion wi h he R pheno ype and i ulence,
we cons uc ed a knock-ou mu an (BmR
ΔwbkE
)by
allelic exchange. The a e o R
SM
,R
ΔwbkE
and hei
complemen ed s ains in cellula and mu ine in ec ion
models was s udied and compa ed o ha o he zoo-
no ic s ain B. suis 1330.
Ma e ial and me hods
Bac e ial s ains, cul u e condi ions and pheno ypic
cha ac e iza ion
E. coli and B ucella s ains (Table 1)we eg ownunde
ae obic condi ions a 37°C in Lu ia Be ani (LB,
In i ogen) and T yp ic Soy (TS, Di co) media, espec i ely.
When necessa y, media we e supplemen ed wi h kanamy-
cin o ampicillin a 50 µg/ml, o wi h chlo amphenicol a
25 µg/ml. All expe imen s wi h iable B ucella we e pe -
o med in a BSL-3 acili y. The smoo h (S) and ough (R)
pheno ypes o B ucella we e assessed by c ys al iole s ain-
ing [28] and by agglu ina ion es s using an i-R polyclonal
an ise um and an i-A and an i-M monospeci ic se a
(ANSES, F ance). Bac e ial mo phology was obse ed by
a omic o ce mic oscopy (AFM).
DNA analysis and mu an s ains cons uc ion
Genomic DNA o B. mic o i was isola ed using he
Qiagen Mini Ki . Whole-genome sho gun sequencing
o he spon aneous ough s ain (BmR
SM
) was pe -
o med using Illumina pai ed-end sequencing wi h
a lib a y inse size o 300 bp and an a e age a ge
co e age o 484 x (GATC). BmR
ΔwbkE
o B. mic o i was
ob ained by eplacing an in e nal po ion o gene
BMI_I539 wi h a Kan
R
casse e. A ecombinan wbkE-
Kan
R
-con aining plasmid de i ed om pGEM®-T was
c ea ed as p e iously desc ibed [5,30]. B ie ly, wo
DNA agmen s (A, 526 bp and B, 595 bp) each ca y-
ing an EcoRI es ic ion si e in a 46-bp homology
egion a he 3ʹ- and 5ʹ-end, espec i ely, we e ampli-
ied by PCR. The agmen s we e used in a second
PCR, o yield agmen AB (1075 bp) con aining he
EcoRI si e in he middle and missing 608 ou o 1110 bp
o he a ge gene BMI_I539 (i.e. om posi ions 71 o
678; Table 1). Following cloning o agmen AB in
pGEM-T, he esul ing plasmid (pGEM-T-AB;
Table 1) was diges ed wi h EcoRI and liga ed wi h he
Kan
R
casse e (1282 bp) excised om plasmid pUC4K.
The esul ing plasmid (pGEM-T-ABKan, 5357 bp;
Table 1) was elec opo a ed in o B. mic o i. The
Kan
R
/Amp
S
clones a ising om double-c osso e we e
selec ed and e i ied by PCR (p ime s in Table 1). Same
cons uc ions and p o ocols we e used o ob ain he
ΔwbkE mu an o B. suis (BsR
ΔwbkE
;Table 1).
Using elec opo a ion, he h ee mu an s ains
(BmR
SM
,BmR
ΔwbkE
,BsR
ΔwbkE
) we e complemen ed
wi h pBBR1MCS-wbkE ec o , con aining he wild-
ype B. mic o i wbkE gene and i s up- and downs eam
egions (Table 1).
A omic o ce mic oscopy
Bac e ia we e g own o s a iona y phase in TS, washed
in PBS, ixed o 1 h in 2.5% glu a aldehyde and s o ed
in PBS a 5 × 10
9
bac e ia/ml. Fluo oDish™cell cul u e
dishes (Wo ld P ecision Ins umen s, UK) we e coa ed
o e nigh a 4°C wi h 0.1% poly-L-lysine, washed wi h
PBS, ai d ied and s o ed a 4°C. Bac e ia we e dilu ed
VIRULENCE 869
20- old in PBS and added o he unc ionalized dish.
Images we e eco ded wi h qp-BioAC CB2 can ile e s
using he quan i a i e imaging mode a ailable on he
NanoWiza d IV AFM (JPK Ins umen s –B uke ). The
applied o ce was kep a 0.3 nN, and a cons an
app oach/ e ac speed o 80 µm/s (z ange o 800 nm).
Mac ophage in ec ion wi h B ucella s ains
Mu ine J774A.1 mac ophage-like cells we e in ec ed
wi h B. mic o i and B. suis s ains a a mul iplici y o
in ec ion (MOI) o 20 as desc ibed p e iously [11]. All
expe imen s we e pe o med in iplica e.
In ec ion o Balb/c mice wi h B. mic o i s ains
App o ed animal expe imen a ion guidelines we e ol-
lowed in he mouse expe imen s and he wo king p o-
ocol was app o ed by he CITA e hical animal
expe imen commi ee. A p ocedu e o he assessmen
o pain, dis ess and discom o in expe imen al ani-
mals, adap ed om [31] was ollowed, assigning a sco e
o each animal ega ding se e al a iables (weigh loss,
appea ance, spon aneous beha io , esponses o ex e -
nal s imuli and clinical signs). I he sco e ose o 15–20
poin s p io o spon aneous dea h, animals we e eu ha-
nized and conside ed as ha ing succumbed o in ec ion.
Bac e ia we e inocula ed i.p [11]. To es he le hali y o
he bac e ial s ains, g oups o six 9-weeks-old Balb/c
emale mice (Jan ie Labs) each we e in ec ed i.p. wi h
10
5
CFU o he wild- ype, BmR
ΔwbkE
and complemen-
ed BmR
ΔwbkE
s ains o wi h 10
8
and 10
9
CFU o
BmR
SM
and BmR
ΔwbkE
mu an s ains. Mice su i al
was moni o ed o e a pe iod o 25 days pos -in ec ion
(d.p.i.).
To s udy he cou se o in ec ion in mice, Balb/c we e
inocula ed i.p. wi h a dose o 10
4
CFU o B. mic o i
s ains. Fi e mice pe s ain we e sac i iced a 3, 14 and
21 d.p.i. Following mice eu hanasia by CO
2
asphyxia-
ion, spleens we e asep ically collec ed, weighed, homo-
genized, se ially dilu ed and pla ed on o TS aga o
iable coun s o B ucella. The signi icance o di e ences
be ween s ains was analyzed by he S uden - es .
P alues ≤0.05 we e conside ed signi ican .
Sequence accession numbe
The genomic DNA sequence o BmR
SM
has been depos-
i ed in he SRA da abase (NCBI) unde he accession
numbe PRJNA545613.
Table 1. Bac e ial s ains, plasmids, and p ime s used in his s udy.
Bac e ial s ains Ac onyms Desc ip ion Re e ence
E. coli DH5αE. coli supE44 ΔlacU169(φ80lacZΔM15)hsdR17 ecA1endA1 gy A96 hi-1 elA1λpi In i ogen
B. mic o i CCM4915
T
BmS
WT
Wild- ype e e ence s ain, smoo h pheno ype [8]
B. mic o i R
SM
BmR
SM
Spon aneous mu an o BmS
WT
, ough pheno ype This wo k
B. mic o i R
SM
+pBBR-wbkE
BmR
SM
p
wbkE
Complemen ed s ain o BmR
SM
ca ying he wbkE gene in plasmid pBBR1MCS This wo k
B. mic o i ΔwbkE BmR
ΔwbkE
Dele ion mu an o BmS
WT
in which he wbkE gene is eplaced by a kanamycin casse e This wo k
B. mic o i ΔwbkE
+pBBR-wbkE
BmR
ΔwbkE
p
wbkE
Complemen ed s ain o BmR
ΔwbkE
ca ying he wbkE gene in plasmid pBBR1MCS This wo k
B. suis 1330 BsS
WT
Wild- ype e e ence s ain, smoo h pheno ype ATCC
23444
B. suis ΔwbkE BsR
ΔwbkE
Dele ion mu an o BsS
WT
in which he wbkE gene is eplaced by a kanamycin casse e This wo k
B. suis ΔwbkE
+pBBR-wbkE
BsR
ΔwbkE
p
wbkE
Complemen ed s ain o BsR
ΔwbkE
ca ying he wbkE gene in plasmid pBBR1MCS This wo k
Plasmids
pGEM®-T T/A cloning ec o wi h ampicillin esis ance ma ke P omega
pUC4K Plasmid ec o ca ying a kanamycin esis ance casse e (KanR) GE
Heal hca e
pBBR1MCS E. coli/B ucella shu le ec o wi h chlo amphenicol esis ance ma ke [29]
pGEM-T-AB pGEM-T ca ying he AB PCR- agmen wi h sequences up- and downs eam o B ucella wbkE This wo k
pGEMT-AB-Kan pGEM-T-AB ca ying KanR in EcoRI si e o he AB agmen This wo k
pBBR1MCS-wbkE pBBR1MCS ca ying he wbkE PCR- agmen including he na i e gene wi h 398 bp up- and 600 bp
downs eam egions cloned in o XhoI-SacI-si es
This wo k
P ime s F agmen Sequence (5ʹ-3ʹ)
1
Size (base
pai s)
A-BMI_I539-Fo A GCAGTGGATCGTGGTGTATG 526 bp
A-BMI_I539 EcoRI-Re TGAGGTTTCATAGGCCCATCGAATTCCATGAATGGTTCGCTCAATG
B-BMI_I539-EcoRI-Fo B CATTGAGCGAACCATTCATGGAATTCGATGGGCCTATGAAACCTCA 595 bp
B-BMI_I539-Re ACATTAATCGCCCGACACTC
BMI_I539-XhoI-Fo wbkE GCGCCTCGAGAGTTGCCATCATGAGCTTGT 1928 bp
BMI_I539-SacI-Re GCGCGAGCTCTATCGGAAACAGTCGTGGTC
1
Res ic ion si es a e unde lined, and non-homologous egions a e indica ed by bold ype. Size o he used AB PCR- agmen ob ained using bo h p ime s
A-BMI_I539-Fo and B-BMI_I539-Re , was 1075 bp. All PCRs we e pe o med wi h P x DNA polyme ase (In i ogen) using B. mic o i genomic DNA as ma ix.
870 S. OUAHRANI-BETTACHE ET AL.
Resul s
A spon aneous ough mu an o B. mic o i shows
inc eased coloniza ion o mac ophages and is
a i ulen in mice
Following he i s pla ing o B. mic o i CCM4915
T
on TS
aga and s aining wi h c ys al iole , a ough colony was
obse ed. The colony was picked, subcul u ed h ee imes
and s ained again wi h c ys al iole : he ough pheno ype
emained s able o all colonies on pla e. This s ain was
named B. mic o i ough spon aneous mu an and abb e-
ia ed BmR
SM
.
I has been epo ed ha ough mu an s ains o
B. suis and B. meli ensis exhibi educed in acellula
su i al in in ec ed mac ophages, hough en y is
imp o ed [28]. BmR
SM
indeed en e ed mu ine
J774A.1 mac ophage cells app oxima ely 100- old be -
e han he wild- ype s ain and B. suis 1330, which
was used as s anda d e e ence (Figu e 1). In con as
o B. suis and B. meli ensis ough s ains [28], BmR
SM
eplica ed 20-30- old, a leas up o 24 hou s pos -
in ec ion (Figu e 1).
To cha ac e ize he beha io o BmR
SM
in i o,Balb/c
mice we e injec ed i.p. wi h he suble hal dose o 10
4
bac-
e ia, as p e iously published by he au ho s o he
B. mic o i wild- ype s ain [11]. The numbe o bac e ia
eco e ed om spleen and li e 3 days a e inocula ion,
co esponding o he peak o in ec ion o B. mic o i [11],
was educed by 3.5 and 2.2 logs (P < 0.001), espec i ely,
compa ed o hose p e iously ob ained wi h he wild- ype
s ain (Table 2) and also con i med in his wo k (see la e
sec ion on “acu e mu ine in ec ion”). The e o e, he ough
mu an colonized hese o gans signi ican ly less han he
wild- ype, esul ing in he lack o a ansien acu e phase o
in ec ion.
BmR
SM
s ain is cha ac e ized by a mu a ion o he
glycosyl ans e ase-encoding gene wbkE
To iden i y he mu a ion(s) esponsible o he ough
pheno ype in BmR
SM
, i s genome was sequenced and
compa ed wi h ha o B. mic o i CCM4915
T
, accessible
in he NCBI da abase. Ou o 48 a ian s, 28 we e
assigned o SNV (Single Nucleo ide Va ian s) and 20
o InDel (Inse ion/Dele ion) a ian s. Sc u iny o he
a ian s esul ed in e aining o 3 SNV and 4 InDels,
ul illing all he ollowing c i e ia: (1) loca ed wi hin
open eading ames o in he immedia e ups eam
icini y, (2) causing amino acids subs i u ions, ame-
shi - o s op-mu a ions in he co esponding genes, (3)
no loca ed in pseudogenes, and (4) ep esen ing he
mos p e alen a ian acco ding o he numbe o
sequencing eads (≥90%) wi h espec o he e e ence
sequence (Supplemen a y Table S1). Only ou mu a-
ions we e in agenic and a ec ed he ollowing genes:
BMI_I525 (2 SNVs), BMI_I539 (1 InDel) and
BMI_I1103 (1 SNV), encoding a ansposase (ISBm1),
a glycosyl ans e ase (wbkE) and a queuine RNA-
ibosyl ans e ase ( g ), espec i ely. The mu a ion
ound in wbkE was ega ded as he mos plausible
cause o he ough pheno ype, because he homolo-
gous gene BMEI1393 o B. meli ensis, loca ed in
Time pos in ec ion (hou s)
1,5 7,0 24,0 48,0
B ucella/well (log10 CFU)
2
3
4
5
6
7
8
Figu e 1. In acellula eplica ion o smoo h B. mic o i CCM4915
T
( iangle down), he spon aneous ough mu an o B. mic o i
(BmR
SM
; iangle up), and smoo h B. suis 1330 (ci cle), in mu ine J774A.1 mac ophage-like cells. The numbe o colony o ming uni s
(CFU) was de e mined by pla ing se ial dilu ions on TS aga pla es a e 2 o 3 days o incuba ion a 37°C o B. mic o i and B. suis,
espec i ely. The expe imen s we e pe o med h ee imes in iplica e each. Da a a e p esen ed as mean alues ± SD o one
expe imen (in iplica e).
VIRULENCE 871

a highly conse ed clus e o he majo (wbk) gene ic
egion o LPS syn hesis, pa icipa es in O-PS biosyn h-
esis [2,27]. P o ein sequences o BMI_I539 and
BMEI1393 a e iden ical o 368 ou o 369 amino
acids. In BmR
SM
, dele ion o a T a posi ion 452 o
wbkE causes a ameshi and he gene a ion o
a p ema u e s op codon a posi ion 622. The esul ing
p o ein sequence is he e o e expec ed o be unca ed
a posi ion 207.
Dele ion o he B. mic o i wbkE gene con i ms i s
ole in smoo h (S-)LPS biosyn hesis and esul s in
enhanced mac ophage up ake
To con i m ha he absence o a unc ional wbkE is
esponsible o he ough pheno ype and he educed
i ulence o BmR
SM
, we cons uc ed a mu an
(BmR
ΔwbkE
) by eplacing a 608-bp in e nal agmen
o wbkE wi h a Kan
R
casse e in he pa en al s ain.
Colony s aining wi h c ys al iole and agglu ina ion
wi h an i-R an ise um [32] con i med ha BmR
ΔwbkE
was ough as BmR
SM
(Table 3). In pa allel, we pe -
o med a su ace analysis o smoo h and ough bac e -
ial s ains by a omic o ce mic oscopy (AFM). AFM has
been es ablished as a powe ul imaging echnique and
allows cha ac e iza ion o he su ace mo phology o
mic obial cells a he nanoscale [33]. We used a o ce-
cu e-based imaging mode whe e he AFM ip is
pushed owa d an a ea o he cell su ace and e ac ed
om i , gene a ing a o ce s. sepa a ion dis ance cu e
encoding in o ma ion abou heigh , adhesion o elas i-
ci y o each image pixel. Su ace opog aphy o
B. mic o i wild- ype and BmR
ΔwbkE
e ealed
a uni o m, smoo h s uc u e o B. mic o i wild- ype,
in con as o a jagged, i egula s uc u e o he
BmR
ΔwbkE
mu an , wi h signi ican ly inc eased ough-
ness (Figu e 2(a,b); Table 3). Mapping o he adhesion
o ces be ween he ip and he bac e ial su ace also
e ealed he p esence o la ge pa ches o inc eased
adhesion a he su ace o BmR
ΔwbkE
when compa ed
wi h he su ace o wild- ype bac e ia, sugges ing
impo an di e ences in he molecula s uc u e o he
BmR
ΔwbkE
mu an su ace (Figu e 2(a,c); Table 3).
Bo h ough mu an s BmR
ΔwbkE
and BmR
SM
we e
hen complemen ed wi h an in ac copy o wbkE cloned
in ec o pBBR1MCS, which es o ed he smoo h phe-
no ype, as e idenced by lack o c ys al iole s aining
and by he agglu ina ion wi h an i-M an ise um only
[8](Table 3). In addi ion, AFM con i med a smoo h
su ace s uc u e o complemen ed BmR
ΔwbkE
, wi h
oughness and adhesion o ces back o wild- ype le els
(Figu e 2,Table 3).
The BmR
ΔwbkE
mu an en e ed J774A.1 cells o an
ex en simila o ha o he BmR
SM
mu an (100 imes
mo e e icien han he pa en al s ain) and eplica ed 60-
old o e 30 hou s (Figu e 3(a)). In con as , BmR
ΔwbkE
and BmR
SM
s ains complemen ed wi h an in ac copy o
wbkE in ec ed mac ophages like he pa en al s ain and
Table 2. Balb/c mice li e and spleen coloniza ion by B. mic o i
S and R
SM
s ains 3 days pos -inocula ion.
S ains
Bac e ia/spleen (log10
CFU)
Bac e ia/li e (log10
CFU)
B. mic o i CCM4915
T
Smoo h
6.52 ± 0.16
a
5.43 ± 0.21
a
B. mic o i R
SM
3.09 ± 0.59 3.27 ± 0.31
Resul s ep esen means ± SD. Di e ences be ween bo h s ains a e sig-
ni ican in bo h o gans (P < 0.001).
a
P e iously published da a [11]
Table 3. Pheno ypes o S and R s ains o B. mic o i and B. suis.
A omic Fo ce Mic oscopy
S ains C ys al iole s aining
a
Se um agglu ina ion
b
Roughness, nm ± SD Adhesion, pN ± SD
B. mic o i CCM4915
T
Smoo h
- M 2.2 ± 0.67
124.6 ± 26
B. mic o i R
SM
+RND
c
ND
B. mic o i R
SM
+
pBBR-wbkE
- M ND ND
B. mic o i ΔwbkE + R 7.8 ± 4.35 446 ± 120.4
B. mic o i
ΔwbkE + pBBR-
wbkE
- M 2.6 ± 1.57 166.1 ± 34.9
B. suis 1330
Smoo h
- A ND ND
B. suis 1330
ΔwbkE
+ R ND ND
B. suis 1330
ΔwbkE + pBBR-
wbkE
- A ND ND
a
+, up ake; -, no up ake o c ys al iole by he colonies
b
wi h monospeci ic A, M, o R polyclonal an ise a
c
no de e mined
872 S. OUAHRANI-BETTACHE ET AL.
eplica ed 400- and 250- old, espec i ely (Figu e 3(a)).
An isogenic wbkE mu an o B. suis 1330 (BsR
ΔwbkE
)was
also cons uc ed in o de o compa e i s beha io o ha
o o he ough mu an s desc ibed in he pas [25,28]. As
expec ed, he BsR
ΔwbkE
mu an e ained c ys al iole and
agglu ina ed wi h an i-R an ise um (Table 3). I colo-
nized he mac ophages 12 imes mo e e icien ly han
he wild- ype, bu in con as o R-s ains o B. mic o i,
a 3- old educ ion in in acellula su i al was obse ed
wi hin a pe iod o 30 h, esul ing in 100 imes lowe
Figu e 2. A omic Fo ce Mic oscopy (AFM) images o B. mic o i wild- ype (Bm WT), ΔwbkE mu an (Bm RΔwbkE) and he comple-
men ed ΔwbkE mu an (Bm RΔwbkE compl). (a) Each column shows om op o bo om he e ical de lec ion image (heigh ) o he
whole bac e ia and 0.3 × 0.3 µm
2
a eas o he cell su ace, ep esen ing oughness and adhesion eco ded on he shown bac e ia
(blue squa e). Quan i a i e oughness (b) and adhesion (c) measu emen s o Bm WT, Bm RΔwbkE and complemen ed Bm RΔwbkE:
0.5 × 0.5 µm
2
images we e eco ded and used o measu emen s o 0.25 × 0.25 µm
2
a eas o quan i y a i hme ic oughness R
a
and
adhesion (Peak- o-Valley). n = 9 bac e ia/s ain. S a is ical di e ences we e analyzed by - es and yielded P alues < 0.001 when
compa ing Bm WT o Bm R
ΔwbkE
compl wi h Bm R
ΔwbkE
. Image analysis was done wi h Gwyddion [34].
VIRULENCE 873
iable coun s when compa ed o he wild- ype
(Figu e 3(b)). This was e y simila o he beha io o he
manB
co e
mu an o B. suis 1330 [25]. The complemen ed
BsR
ΔwbkE
s ain agglu ina ed speci ically wi h an i-A an i-
se um as he wild- ype s ain (Table 3) and showed wild-
ype le els o in acellula in ec ion and eplica ion,
despi e a s onge ansi ional dec ease in he ea ly
phase o in ec ion (Figu e 3(b)).
The B. mic o i wbkE gene is indispensable o acu e
mu ine in ec ion
In ec ion o Balb/c wi h 10
4
CFU o B. mic o i s ains
showed a 3-logs educ ion o BmR
ΔwbkE
in he spleen a
day 3 pos -injec ion, as compa ed o he wbkE-comple-
men ed mu an and wild- ype s ains (Figu e 4(a)), con-
i ming he incapaci y o a ΔwbkE mu an s ain o
B ucella/well (log10 CFU)
Time pos in ec ion (hou s)
1,5 7,0 24,0 30,0
B ucella/well (log10 CFU)
2
3
4
5
6
7
8
Time pos in ec ion (hou s)
1,5 7,0 24,0 30,0
2
3
4
5
6
7
8
ba
Figu e 3. In acellula eplica ion o smoo h and ough s ains o B. mic o i (a) and B. suis (b) in mu ine J774A.1 mac ophage-like
cells. (a) B. mic o i CCM4915
T
wild- ype ( illed iangle down), he spon aneous R-s ain BmR
SM
(open iangle up), he complemen ed
BmR
SM
mu an ( illed iangle up), he cons uc ed R-s ain BmR
ΔwbkE
(open ci cle), and he complemen ed BmR
ΔwbkE
mu an ( illed
ci cle). (b) B. suis 1330 wild- ype ( illed iangle down), he cons uc ed R-s ain BsR
ΔwbkE
(open ci cle), and he complemen ed BsR
ΔwbkE
mu an ( illed ci cle). The complemen ed s ains exp essed na i e wbkE cloned in o he eplica i e plasmid pBBR1MCS. The
expe imen s we e pe o med h ee imes in iplica e each. Da a a e p esen ed as mean alues ± SD o one expe imen (in iplica e).
Time (days)
3 d 14 d 21 d
Bac e ia / spleen (log10 CFU)
0
1
2
3
4
5
6
***
ab
Time (days)
3 d 14 d 21 d
Spleen weigh (g ams)
0,0
0,1
0,2
0,3
0,4
0,5
*** *** ** *
Figu e 4. In ec ion o Balb/c mice wi h B. mic o i s ains: g ow h and su i al o B. mic o i s ains in he spleen (a) and spleen weigh s
o in ec ed animals (b) a e i.p. inocula ion o 10
4
bac e ia. The numbe o iable B. mic o i CCM4915
T
wild- ype (black ba s),
BmR
ΔwbkE
s ain (open ba s), and complemen ed BmR
ΔwbkE
mu an (g ey ba s) was de e mined a days 3, 14, and 21 pos -in ec ion.
The a ow indica es he in ec ion dose o 10
4
bac e ia. Fi e mice we e sac i iced pe bac e ial s ain and ime poin , and alues
ep esen means ± SD. As e isks indica e a iable signi icance o he di e ences be ween he R-s ain and he wild- ype (nex o le
ba ) o R-s ain and he complemen ed mu an (nex o igh ba ), o be ween he R-s ain and bo h he wild- ype and he
complemen ed mu an (abo e middle ba ): * P < 0.05; ** P < 0.005; *** P < 0.001.
874 S. OUAHRANI-BETTACHE ET AL.
es ablish an acu e phase o in ec ion in he hos (see also
Table 2).Theacu ein ec ionphaseobse edwi h hewild-
ype and he wbkE-complemen ed s ains was ollowed by
an inc ease o he spleen weigh un il a leas day 14,
e lec ing an in lamma o y esponse (Figu e 4(b)). In con-
as , mice in ec ed wi h BmR
ΔwbkE
did no gain spleen
weigh . A simila inding was ob ained when he s ain
BmR
SM
was injec ed i.p. a 10
4
CFU, as spleen weigh s
emained unchanged a day 3 and day 14 (0.09 ± 0.007
and 0.10 ± 0.015, espec i ely; P < 0.001), con i ming
a educed immune esponse induc ion wi h he wbkE-
mu an ough s ains.
The wbkE gene is essen ial o he le hal cha ac e
o B. mic o i in ec ions in Balb/c mice
Ou p e ious wo k showed ha he in a-pe i oneal injec-
ion o 10
5
CFU o B. mic o i CCM4915
T
caused he dea h
o 83%o heBalb/cmicewi hin ou dayso in ec ion[11].
Toin es iga e hepossiblein ol emen o hewbkE gene in
he le hal ou come o a B. mic o i in ec ion, he suscep -
ibili y o Balb/c mice in ec ed wi h BmR
SM
o BmR
ΔwbkE
was compa ed o ha obse ed ollowing in ec ion wi h he
B. mic o i CCM4915
T
wild- ypeand hecomplemen ed
BmR
ΔwbkE
s ains, o e a moni o ing pe iod o 25 days
(Table 4): 67% and 83% o he mice in ec ed wi h he
s anda d dose o 10
5
CFU o he wild- ype o he comple-
men ed BmR
ΔwbkE
s ain, espec i ely, died be ween days 2
and 6 pos -inocula ion. In s iking con as , all he mice
in ec ed wi h 10
5
CFU o BmR
ΔwbkE
su i ed wi hou any
symp oms. 100% su i al was also obse ed o bo h
R-mu an s ains BmR
SM
and BmR
ΔwbkE
a e he injec ion
o 10
8
CFU. Howe e , when inocula ed wi h 10
9
CFU o
ei he R-s ain, all mice died be ween days 2 and 6 pos -
inocula ion. These esul s a e e en mo e no able i com-
bined wi h ou p elimina y obse a ion (no shown) ha
BmR
SM
in Balb/c mice p o ided p o ec ion agains
B. mic o i wild- ype, B. abo us, B. meli ensis and B. suis
1330.
Discussion
The non-canonical LPS o classical b ucellae lacks endo oxici y
and possesses a pa icula co e s uc u e helping o e ade he
hos ’s immune sys em [35,36]. The phenomenon o dissocia-
ion, esul ing in he con e siono S- oR-pheno ype,hasbeen
i s desc ibed o B ucella in 1933 [37]. Mo e ecen ly,
B. abo us, B. meli ensis and B. suis R-mu an s de oid o
O-PS ha e been s udied in cellula and mu ine models o
in ec ion [25,27,28,38,39]. These s udies consis en ly showed
he ele ance o O-PS o i ulence. The educed i ulence o
R-mu an s has been a ibu ed o (1) high sensi i i y o com-
plemen -media ed lysis in mice [40,41], (2) lipid a -
independen en y in o mac ophages esul ing in enhanced
phagolysosome usion [25], and (3) lack o in acellula eplica-
ion due o mac ophage ac i a ion [28]. Monoclonal an ibodies
speci ic o common O-PS epi opes o A- o M-dominan
classical s ains also ecognize LPS o he M-dominan
B. mic o i e e ence s ain [14], indica ing a conse ed s uc u e
o O-PS. Howe e , an i-R-LPS monoclonal an ibodies do no
eac wi h B. mic o i LPS, sugges ing s uc u al speci ici ies in
he co e-lipid A moie y o i s LPS [14], which may esul in
enhanced endo oxic p ope ies and possibly explain he killing
abili y o B. mic o i in he mu ine model o in ec ion.
Because o he lack o da a a ailable on a ypical species
mu an s a ec ed in O-PS biosyn hesis, we in es iga ed he
i ulence p ope ies o a spon aneous R-mu an o
B. mic o i, which was o ui ously isola ed. Mu ine in ec-
ion expe imen s showed a loss o le hali y o his mu an ,
and subsequen analysis by whole-genome sequencing
allowed o link his abili y o he wbkE gene, encoding
a glycosyl ans e ase loca ed in he majo O-PS biosyn hesis
egion, as p e iously desc ibed o B. meli ensis [27].
Complemen a ion o R-mu an s es o ed wild- ype p ope -
ies including smoo h cha ac e , educed mac ophage en y
and mu ine le hali y, demons a ing ha he BmR
SM
s ain
was a ec ed in a single LPS biosyn hesis gene. In con as ,
spon aneous R-mu an s o B. abo us and B. meli ensis iso-
la ed om mac ophage and mu ine in ec ions we e o en
simul aneously a ec ed in se e al loci [42]. Cell su ace
s uc u e analysis by AFM con i med he smoo h cha ac e
o he wild- ype and complemen ed BmR
ΔwbkE
s ains,
whe eas a dis inc i egula su ace was eco ded o
BmR
ΔwbkE
, possibly due o exposu e o ou e memb ane
p o eins and lipid A/ou e co e disaccha ides in he absence
o he O-chain [27]. This exposu e may also explain he
inc eased adhesion o ces obse ed du ing he in e ac ion
o he AFM ip wi h he R-mu an su ace. Con e sely, in
a ecen epo , AFM analysis o wo B. abo us s ains,
a wild- ype and i s isogenic mu an Δgmd R lacking
O-chain, shows simila deg ees o oughness [43].
S uc u al di e ences in he LPS co e-lipid A moie ies o
Table 4. Le hali y o B. mic o i S and R s ains in Balb/c mice
S ains In ec ion dose (i.p.) % Mo ali y
ab
B. mic o i CCM4915
T
Smoo h 10
5
67
B. mic o i R
SM
10
8
0
10
9
100
B. mic o i ΔwbkE10
5
0
10
8
0
10
9
100
B. mic o i ΔwbkE+ pBBR-wbkE10
5
83
a
Each bac e ial s ain was inocula ed o a g oup o six 9-weeks-old Balb/c
emale mice
b
O e a 25-days pe iod o moni o ing, mu ine dea h occu ed be ween days
2 and 6 pos -inocula ion
VIRULENCE 875