Genome Exp ession Analysis o Nonp oli e a ing In acellula
Salmonella en e ica Se o a Typhimu ium Un a els an Acid pH-
Dependen PhoP-PhoQ Response Essen ial o Do mancy
C is ina Núñez-He nández,
a
* Albe o Tie ez,
a
Ál a o D. O ega,
a
M. G aciela Puccia elli,
a,b
Ma a Godoy,
c
Blanca Eisman,
a
Josep Casadesús,
d
F ancisco Ga cía-del Po illo
a
Depa amen o de Bio ecnología Mic obiana, Cen o Nacional de Bio ecnología-Consejo Supe io de In es igaciones Cien í icas (CNB-CSIC), Mad id, Spain
a
;
Depa amen o de Biología Molecula , Uni e sidad Au ónoma de Mad id, Cen o de Biología Molecula “Se e o Ochoa”-Consejo Supe io de In es igaciones Cien í icas
(CBMSO-CSIC), Mad id, Spain
b
; Se icio de Genómica, Cen o Nacional de Bio ecnología-Consejo Supe io de In es igaciones Cien í icas (CNB-CSIC), Mad id, Spain
c
;
Depa amen o de Gené ica, Facul ad de Biología, Uni e sidad de Se illa, Se ille, Spain
d
Genome-wide exp ession analyses ha e p o ided clues on how Salmonella p oli e a es inside cul u ed mac ophages and epi he-
lial cells. Howe e , in i o s udies show ha Salmonella does no eplica e massi ely wi hin hos cells, lea ing he unde lying
mechanisms o such g ow h con ol la gely unde ined. In i o in ec ion models based on ib oblas s o dend i ic cells e eal
limi ed p oli e a ion o he pa hogen, bu i is p esen ly unknown whe he hese phenomena e lec e en s occu ing in i o.
Fib oblas s a e dis inc i e, since hey ep esen a nonphagocy ic cell ype in which S. en e ica se o a Typhimu ium ac i ely
a enua es in acellula g ow h. He e, we show in he mouse model ha S. Typhimu ium es ains in acellula g ow h wi hin
nonphagocy ic cells posi ioned in he in es inal lamina p op ia. This esponse equi es a unc ional PhoP-PhoQ sys em and is
ep oduced in p ima y ib oblas s isola ed om he mouse in es ine. The ib oblas in ec ion model was exploi ed o gene a e
ansc ip ome da a, which e ealed ha ⬃2% (98 genes) o he S. Typhimu ium genome is di e en ially exp essed in nong ow-
ing in acellula bac e ia. Changes include me abolic ep og amming o mic oae ophilic condi ions, induc ion o i ulence plas-
mid genes, up egula ion o he pa hogenici y islands SPI-1 and SPI-2, and shu down o lagella p oduc ion and chemo axis.
Compa ison o ela i e p o ein le els o se e al PhoP-PhoQ- egula ed unc ions (PagN, PagP, and Vi K) in nong owing in a-
cellula bac e ia and ex acellula bac e ia exposed o di e se PhoP-PhoQ-inducing signals deno ed a egula ion esponding o
acidic pH. These da a demons a e ha S. Typhimu ium es ains in acellula g ow h in i o and suppo a model in which
do man in acellula bac e ia could sense acuola acidi ica ion o s imula e he PhoP-PhoQ sys em o p e en ing in acellu-
la o e g ow h.
Salmonella en e ica se o a s a e ood-bo ne bac e ial pa hogens
ha cause gas oen e i is and sys emic disease ( yphoid e e )
in humans and li es ock (1–3). Salmonellae in ade a a ie y o
euka yo ic cell ypes and ha e been ex ensi ely s udied in animal
models (2,4,5) and in i o in models in ol ing cul u ed mam-
malian cell lines (6). Mas e elemen s o Salmonella pa hogenici y
include wo ype III sec e ion sys ems encoded in he Salmonella
pa hogenici y islands 1 and 2 (SPI-1 and SPI-2), which sec e e
p o eins p omo ing in asion and su i al/p oli e a ion inside he
hos cell. O he widely s udied Salmonella unc ions linked o i -
ulence include egula o y p o eins, such as he wo-componen
sys em PhoP-PhoQ, which o ches a es adap a ion o he pa ho-
gen o he in acellula en i onmen o he in ec ed cell (7,8).
Despi e he bulk o in o ma ion collec ed on pa hogen unc-
ions equi ed o i ulence, i emains la gely unknown how his
pa hogen egula es i ulence unc ions in dis inc hos cell ypes.
Hos cells a ge ed by his pa hogen in i o include epi helial cells,
mac ophages, neu ophils, and dend i ic cells (9,10). Howe e ,
Salmonella p e e en ially esides wi hin mac ophages in bo h
acu e and ch onic in ec ions (11–14). A ea u e ha dis inguishes
he beha io o in acellula bac e ia in i o is hei limi ed capac-
i y o p oli e a e inside hos cells, eaching p ogenies o only 3 o 4
indi iduals pe in ec ed cell (12–15). The mos widely accep ed
model indica es ha S. Typhimu ium colonizes he animal by
inc easing he numbe o in ec ion oci a he han inc easing he
numbe o in acellula bac e ia pe cell. Repe i i e cycles o lim-
i ed p oli e a ion ounds inside mac ophages, exi om he in-
ec ed cells and in ec ion o neighbo cells may accoun o he
inc ease o in ec ion oci (16). O in e es , ecen s udies in cul-
u ed mac ophages and epi helial cells epo ed a ma ked he e o-
genei y in he popula ion o in acellula Salmonella, wi h some
bac e ia unde going ac i e eplica ion while o he s emain in a
nong owing s a e o long pe iods o ime (17,18). This he e oge-
nei y, inhe en o many na u al p ocesses (19), is also known in
o he pa hogens exposed o an imic obials (20) and in he exp es-
sion by Salmonella o ce ain i ulence de e minan s such as he
pa hogenici y island 1, SPI-1 (20,21).
The isualiza ion o low numbe s o in acellula Salmonella
Recei ed 4 Oc obe 2012 Accep ed 15 Oc obe 2012
Published ahead o p in 22 Oc obe 2012
Add ess co espondence o F ancisco Ga cía-del Po illo, [email p o ec ed].
* P esen add ess: C is ina Núñez-He nández, BD Diagnos ics, Mad id, Spain;
Blanca Eisman, Me ck Sha p and Dohme, Mad id, Spain.
Edi o : A. J. Bäumle
A.T. and Á.D.O. con ibu ed equally o his wo k.
Supplemen al ma e ial o his a icle may be ound a h p://dx.doi.o g/10.1128
/IAI.01080-12.
Copy igh © 2013, Ame ican Socie y o Mic obiology. All Righ s Rese ed.
doi:10.1128/IAI.01080-12
154 iai.asm.o g In ec ion and Immuni y p. 154–165 Janua y 2013 Volume 81 Numbe 1
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om
pe in ec ed cell in ch onically in ec ed mice (11) sugges s ha ,
simila o wha has been desc ibed o o he in acellula pa ho-
gens, such as Mycobac e ium ube culosis (22), s a egies limi ing
in acellula g ow h (d i en by he hos and/o by he pa hogen)
may ope a e in ch onic and asymp oma ic in ec ions (15,23–25).
Se o a Typhi es ablishes his condi ion in abou 5% o humans
eco e ing om yphoid e e , and non yphoidal se o a s causing
in ec ions in humans can also pe sis asymp oma ically in li e-
s ock and domes ic owl (23). Despi e he ele ance o hese phe-
nomena, i is s ill unknown whe he Salmonella es ic s in acel-
lula g ow h in i o. The exis ing in o ma ion ela es o ana omical
si es whe e bac e ia a e isualized du ing ch onic in ec ions and
pa hogen unc ions in luencing such a s a e (26,27). Mac o-
phages p esen in he mesen e ic lymph nodes (11), hemophago-
cy ic mac ophages (28), and epi helial cells o he gallbladde (29)
a e p oposed o ac as se o a Typhimu ium ese oi s du ing
ch onic in ec ions. Mic oa ay-based nega i e gene ic selec ions
e ealed ha SPI-1, SPI-2, p ophages, imb ial ope ons, and genes
egula ed by he PhoP-PhoQ wo-componen egula o y sys em
con ibu e o long-las ing coloniza ion o he mouse spleen (30).
Func ions encoded in he genomic island CS54 as ShdA and Ra B
a e also equi ed o e icien ecal shedding in mice (26). O he
unc ions linked o pe sis ence include he sec e ion sys em Zi T/
Zi S, highly exp essed in bac e ia shed in ecal pelle s (31), and
SciS, p oposed o a enua e Salmonella g ow h inside mac o-
phages and o down egula e i ulence in i o (32). Toge he , hese
s udies suppo he idea ha in acellula Salmonella is capable o
limi ing in acellula g ow h in i o.
Ou p e ious s udies e ealed ha S. Typhimu ium does no
p oli e a e inside cul u ed ib oblas s (33,34). The pa hogen con-
ibu es o his condi ion, since bac e ial o e g ow h is obse ed
upon inac i a ion o bac e ial egula o s, such as PhoP-PhoQ, he
sigma ac o RpoS, o he plasmid-encoded egula o Sp R (33).
In addi ion o g ow h es ain , non eplica ing in acellula bac-
e ia ensu e iabili y using SPI-2 and he sigma ac o RpoE (33).
He e, we unde ook a s udy in mice o in es iga e whe he S. Ty-
phimu ium has he capaci y o a enua e in acellula g ow h in
i o. We ob ained e idence o a enua ion o S. Typhimu ium
p oli e a ion in nonphagocy ic cells loca ed in he in es inal lam-
ina p op ia which was dependen on a unc ional PhoP-PhoQ
sys em. The es ablishmen o such a in acellula nonp oli e a i e
s a e was also demons a ed in p ima y ib oblas s isola ed om
in es inal issue. On he basis o his in o ma ion, genome p o il-
ing was de ined in non eplica ing in acellula bac e ia using an in
i o model o pe sis ence. These da a allowed us o u he cha -
ac e ize he mode in which he PhoP-PhoQ sys em is ac i a ed in
in acellula do man bac e ia.
MATERIALS AND METHODS
Bac e ial s ains, cul u e media, and g ow h condi ions. The S. en e ica
se o a Typhimu ium s ains used in his s udy a e shown in Table 1. All
s ains de i e om SV5015, a His
⫹
de i a i e o he mouse- i ulen s ain
SL1344 (36). Bac e ia we e g own in Lu ia b o h (LB) a 37°C. When
app op ia e, kanamycin (30 ␮g/ml) o ampicillin (50 ␮g/ml) was added
o he g ow h media. Fo he ansc ip omic analyses (see below), bac e ia
we e g own in LB medium a 37°C wi h ae a ion (180 pm) o exponen ial
phase (op ical densi y a 600 nm [OD
600
]o ⬃0.2) o o s a iona y phase
( inal OD
600
o ⬃3.0) and main ained in he la e condi ion o an addi-
ional 12 h. To in ec BALB/c mice and euka yo ic cells (see below), bac-
e ia we e g own a 37°C in s anding nonae a ed cul u es ob ained upon
inocula ion o 2 ml o LB medium wi h a bac e ial colony and subsequen
o e nigh incuba ion ( inal OD
600
o ⬃1.0). To analyze gene egula ion
media ed by PhoP-PhoQ, bac e ia we e g own in N minimal medium
(39) con aining 38 mM glyce ol as he ca bon sou ce and supplemen ed
wi h 10 mM o 8 ␮M MgCl
2
as desc ibed p e iously (40). Fo SPI-2-
inducing condi ions, he PCN minimal medium adjus ed o a pH o 5.8
was used as desc ibed p e iously (41).
Cons uc ion o ch omosomal epi ope- agged genes. The s ains
ca ying ch omosomal 3⫻FLAG epi ope- agged genes we e cons uc ed
using he me hod desc ibed by Uzzau e al. (42). Plasmids and oligonu-
cleo ides used o his p ocedu e a e lis ed in Table S8 in he supplemen al
ma e ial. Co ec inse ion o he epi ope a he 3=end o he a ge ed gene
was e i ied in all cases by PCR and sequencing.
Fib oblas cells. NRK-49F no mal a kidney ib oblas s (ATCC CRL-
1570) we e used h oughou he s udy. These ib oblas s we e p opaga ed
in Dulbecco’s modi ied Eagle’s medium (DMEM) (In i ogen, Ca lsbad,
CA) con aining 5% ( ol/ ol) e al bo ine se um (FBS) and 4 mM L-glu-
amine. Mouse in es inal ib oblas s we e isola ed om six C57BL/10 e-
male mice o 12 weeks o age by ollowing he me hod o S ong e al.
desc ibed o human in es inal ib oblas s (43). These p ima y ib oblas s
we e p opaga ed in he p esence o an ibio ics o he i s h ee passages
and hen in an ibio ic- ee medium o a oid in e e ence wi h he bac e-
ial in asion and p oli e a ion assays. The in ec ion expe imen s we e
pe o med be ween passages 5 and 9.
In acellula bac e ial p oli e a ion assays in p ima y in es inal i-
b oblas s and NRK-49F a ib oblas s. Mouse p ima y in es inal ib o-
blas s and NRK-49F no mal a kidney ib oblas s we e in ec ed wi h bac-
e ia o 20 min using a mul iplici y o in ec ion (MOI) o 10:1 (bac e ia o
euka yo ic cells) as p e iously desc ibed (44). A e ex ensi e washing,
in ec ed cells we e incuba ed in esh issue cul u e medium con aining
100 ␮g/ml gen amicin o he i s 2 h pos in ec ion and 10 ␮g/ml o he
emainde o he expe imen . In ec ed cells we e lysed a he desi ed
pos in ec ion imes in phospha e-bu e ed saline (PBS), pH 7.4, 1% T i-
on X-100. The numbe o iable in acellula bac e ia was de e mined by
pla ing. To inhibi acuola acidi ica ion, 100 nM ba ilomycin (BAF) was
added o he in ec ed ib oblas s in he esh issue cul u e medium con-
aining gen amicin o a oid any e ec in bac e ial en y. BAF was main-
ained du ing he incuba ion pe iods wi h high (100 ␮g/ml) and low (10
␮g/ml) doses o gen amicin.
Bac e ial in ec ion o BALB/c mice and immunohis ochemis y.
Wild- ype and phoP mu an bac e ia g own o e nigh in LB medium a
TABLE 1 S. en e ica se o a Typhimu ium s ains used in his s udy
a
S ain Rele an geno ype
Sou ce o
e e ence
SL1344 hisG64, psL, mouse i ulen isola e 36
SV4056 SL1344 phoP7953::Tn10 33
SV4386 SL1344 ssaC (spiA)::KIXX 33
MD1682 SL1344 glpK::3⫻FLAG-Kn 38
SV5015 Wild ype, SL1344 His
⫹
37
MD1120 phoP7953::Tn10 37
MD1178 pagC::3⫻FLAG-Kn This s udy
MD1182 pagC::3⫻FLAG-Kn phoP7953::Tn10 This s udy
MD1942 mg C::3⫻FLAG-Kn This s udy
MD1951 mg C::3⫻FLAG-Kn phoP7953::Tn10 This s udy
MD2926 glpK::3⫻FLAG-Kn This s udy
MD2927 glpK::3⫻FLAG-Kn phoP7953::Tn10 This s udy
MD3702 i K::3⫻FLAG-Kn This s udy
MD3729 i K::3⫻FLAG-Kn phoP7953::Tn10 This s udy
MD3703 pagN::3⫻FLAG-Kn This s udy
MD3726 pagN::3⫻FLAG-Kn phoP7953::Tn10 This s udy
MD3727 pagP::3⫻FLAG-Kn This s udy
MD3723 pagP::3⫻FLAG-Kn phoP7953::Tn10 This s udy
a
Unless o he wise indica ed, all s ains a e isogenic o he wild- ype s ain SV5015
(SL1344 His
⫹
).
T ansc ip ome o Nong owing In acellula Salmonella
Janua y 2013 Volume 81 Numbe 1 iai.asm.o g 155
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om
37°C in nonshaking condi ions we e collec ed by cen i uga ion (5,000 ⫻
g, 10 min, 4°C), washed wice in cold s e ile PBS, pH 7.4, and suspended a
a densi y o ⬃8⫻10
10
CFU/ml. Se ial dilu ions we e used o in ec o ally
(25 ␮l) g oups o 8-week-old emale BALB/c mice as desc ibed p e iously
(45). The p o ocols used in hese s udies we e app o ed by he Comi é
É ico de Expe imen ación o he Consejo Supe io de In es igaciones
Cien í icas (CSIC). The in ec ious dose was in he ange o 10
8
o 10
9
iable bac e ia pe mouse, as con i med by pla ing and coun ing o CFU.
A dis inc pos in ec ion imes, 6 and 24 h, a lapa o omy was pe o med o
localize he smallin es ine. Ileumwas ex ac edasep ically and ixed in4%
pa a o maldehyde (PFA) du ing1ha oom empe a u e. A e h ee
washes wi h PBS, pH 7.4, bu e , he ileum was incuba ed a 4°C o e nigh
in a 20% suc ose-PBS, pH 7.4, solu ion. Tissue blocks o ca. 1 cm
3
we e
embedded in Jung issue eezing medium (Leica) and apidly deep- o-
zen in cold ace one (⫺50 o ⫺60°C). Fi een-␮m- hick sec ions we e
ob ained by c yo omy and moun ed in glass slides p e ea ed wi h Supe -
F os -Plus (Menzel-Glase ). Blocks and slices we e kep a ⫺80°C un il
u he p ocessing. Fo immunohis ochemis y, glass slides con aining
he issue sec ion we e hawed o e nigh a oom empe a u e o inc ease
issue adhesi eness. Sec ions we e u he incuba ed in cold ace one
(⫺20°C) du ing 10 min, he ace one was e apo a ed a oom empe a u e
o 15 min, and inally he sec ions we e ehyd a ed in PBS, pH 7.4, o 15
min. P io o immunos aining, sec ions we e incuba ed o 1 h a oom
empe a u e in blocking solu ion (10% e al bo ine se um [FBS], 0.2%
saponin). P ima y and seconda y an ibodies we e dilu ed as app op ia e
in 2% FBS, 0.2% saponin. Sec ions we e incuba ed wi h he p ima y an-
ibodies o 48 h a 4°C, ollowed by h ee washes wi h PBS, pH 7.4, and
u he incuba ion wi h seconda y luo och ome-conjuga ed an ibodies
o 1 h. Nuclei we e s ained wi h he cyanine luo och ome To-P o3 (In-
i ogen) a a 1:200 dilu ion du ing 20 min a oom empe a u e. Sec ions
we e blo ed o emo e excess PBS bu e . A d op o inclusion medium
con aining poly inyl alcohol and DABCO (Fluka) was added, and he
sec ions we e co e ed wi h 24- by 60-mm glass co e slips (Menzel-Gla-
se ). Samples we e isualized in a Zeiss Axio e 200 luo escence mic o-
scope equipped wi h a con ocal Radiance 2100 uni (Bio-Rad). Lase -
Sha p 2000 so wa e was used o cap u e he image, and Lase Pix and
Adobe Pho oshop we e used o image p ocessing.
La ge-scale in ec ion o ib oblas s o ob ain RNA and p o ein om
in acellula bac e ia. NRK-49F no mal a ib oblas s we e seeded in
BioDish-XL 500-cm
2
pla es ( e e ence 351040; BD Biosciences) a a den-
si y o 2 ⫻10
7
cells pe dish and in ec ed a an MOI o 10:1 (bac e ia:
ib oblas ) and 40 o 60% con luence. A e 20 min, he in ec ed cells we e
washed i e imes wi h p ewa med Hank’s balance sal solu ion (HBSS).
These cells we e incuba ed un il 1 h pos in ec ion in esh cul u e medium
con aining 100 ␮g/ml o gen amicin. The cul u e medium was hen e-
placed wi h new esh medium con aining 10 ␮g/ml gen amicin un il he
desi able pos in ec ion ime (1, 2, 6, 8, o 24 h). In ec ed ib oblas s we e
p ocessed as desc ibed o he Salmonella-mac ophage in ec ion model
(46), wi h sligh modi ica ions. B ie ly, in ec ed ib oblas s we e washed
i e imes wi h cold PBS, pH 7.4, and lysed (a 30 ml pe pla e) in a
solu ion con aining 0.4% SDS, 1% acidic phenol, and 19% e hanol in
wa e . A e 30 min o incuba ion a 4°C, in acellula bac e ia we e col-
lec ed by cen i uga ion (27,500 ⫻g, 4°C, 30 min) and washed h ee imes
wi h 1 ml o a 1% acidic phenol, 19% e hanol solu ion. Fo each in ec ion
ime poin a which RNA o p o ein was ex ac ed, ou BioDish-XL 500-
cm
2
pla es usually we e pooled. Fo mic oa ay hyb idiza ions, RNA o
nong owing in acellula bac e ia was ex ac ed and pooled om a min-
imum o 20 independen expe imen s wi h ou BioDish-XL 500-cm
2
pla es each. Fo p o ein ex ac ion, in acellula bac e ia we e washed
wice wi h cold PBS, pH 7.4, eco e ed by cen i uga ion (15,000 ⫻g, 4°C,
10 min), and p ocessed as desc ibed p e iously (37,38). RNA and
p o ein ex ac ion in ex acellula bac e ia was pe o med as desc ibed
p e iously (37,38,46) by ollowing he same ea men as ha o
in acellula bac e ia wi h a solu ion con aining 1% acidic phenol,
19% e hanol, and 0.4% SDS.
Genome exp ession analyses and RT-qPCR. To al RNA pu i ied
om in acellula and ex acellula bac e ia we e p ocessed as p e iously
desc ibed o gene a e he co esponding cDNAs (47). The Salgenomics
mic oa ay used o hese s udies has been desc ibed p e iously and con-
ains 70-me an isense oligonucleo ides speci ic o 4,369 open eading
ames (ORFs), 21 RNAs, 86 RNAs, and 47 sRNAs iden i ied in he
genome o S. Typhimu ium s ain SL1344 (47). The hyb idiza ion condi-
ions, da a acquisi ion, no maliza ion, and s a is ical analyses ha e been
desc ibed elsewhe e (37,47). Valida ion assays we e pe o med by quan-
i a i e e e se ansc ip ion-PCR (RT-qPCR) as desc ibed p e iously
(37), using ompA as an in e nal con ol.
An ibodies and immuno luo escence mic oscopy. The ollowing
p ima y an ibodies we e used o Wes e n assays and immuno luo es-
cence mic oscopy s udies: abbi polyclonal an i-Salmonella lagellin
(FliC/FljB) (48); abbi polyclonal KH1331 an i-TlpA (gi om Reini
Hu me, Ka olinska Ins i u e , S ockholm, Sweden); mouse monoclonal
an i-FLAG epi ope (clone M2; Sigma); abbi polyclonal an i-S. Typhi-
mu ium lipopolysaccha ide (LPS), g oup B, ac o s 1:4:5:12 (Di co Lab-
o a o ies); mouse monoclonal an ibac e ial RNA polyme ase sigma S
subuni , RpoS (clone 1RS1; San a C uz Bio echnology); abbi polyclonal
an i-OmpA (gi o H. Schwa z, Tübingen, Ge many); abbi polyclonal
an i-calnexin (S essgen); a monoclonal an i-CD18 (clone M18/2; De-
elopmen al S udies Hyb idoma Bank [DSHB], IA); a monoclonal an i-
CD45 (clone 30-F11; BD Pha Mingen); and mouse monoclonal an i-al-
pha ac in o smoo h muscle (␣-SMA) conjuga ed o Cy3 (clone 1A4;
Sigma). Fo immuno luo escence mic oscopy, he ollowing seconda y
an ibodies we e used a a 1:500 dilu ion: goa polyclonal an i- abbi IgG
conjuga ed o Alexa 488 (Molecula P obes), goa polyclonal an i- a IgG
conjuga ed o Alexa 594 (Molecula P obes), and goa polyclonal an i-
mouse IgG conjuga ed o Alexa 594 (Molecula P obes). Goa polyclonal
an i-mouse IgG conjuga ed o ho se adish pe oxidase (HRP; Bio-Rad)
was used as seconda y an ibody a a 1:5,000 dilu ion o Wes e n assays.
Polyclonal abbi HRP-conjuga ed an i-G oEL (Sigma) was also used.
In ec ed NRK-49F ib oblas s and mouse in es inal p ima y ib oblas s
we e ixed and p ocessed o immuno luo escence mic oscopy as p e i-
ously desc ibed (44). Cells we e examined in a Leica luo escence in e ed
mic oscope (DMI6000B).
S a is ical analysis. Da awe e analyzed wi hG aphPad P ism5.0 so -
wa e (G aphPad Inc., San Diego, CA) using S uden ’s es . Di e ences in
alues wi h P⬍0.05 we e conside ed signi ican .
Accession numbe s. The cha ac e is ics and con igu a ion o he Sal-
genomics mic oa ay we e deposi ed in he MIAME da abase (h p:
//www.ebi.ac.uk/miamexp ess) unde accession numbe A-MEXP-846.
Gene exp ession da a we e deposi ed in he A ay Exp ess da abase (h p:
//www.ebi.ac.uk/a ayexp ess) unde accession numbe s E-MEXP-1774
(in acellula phoP ansc ip ome), E-MEXP-1775 (in acellula wild-
ype ansc ip ome), and E-MEXP-1776 (ex acellula wild ype, s a ion-
a y phase).
RESULTS
S. Typhimu ium a enua es g ow h in nonphagocy ic cells lo-
ca ed in he lamina p op ia o in es inal illi. S. Typhimu ium
uses he PhoP-PhoQ sys em o a enua e g ow h inside cul u ed
ib oblas s (33,34). To de e mine whe he he pa hogen also ig-
ge san a enua ion esponse in i o,BALB/c micewe e challenged
o ally wi h wild- ype and phoP isogenic bac e ia. Hos cell ypes
ha bo ing bac e ia we e sea ched by mic oscopy in in es inal is-
sue sec ions. Due o he lack o cell ma ke s sui able o dis in-
guishing by immunohis ochemis y and in an exclusi e manne
he ib oblas popula ions p esen in he in es inal issue (49), we
examined he dis ibu ion o CD18 and CD45, wo panleukocy e
cell ma ke s. CD18 is he ␤2 chain o he leukocy e-speci ic in eg-
ins LFA-1, Mac-1, gp150, and gp95 (50). CD45, also known as
leukocy e common an igen, is a ansmemb ane glycop o ein
Núñez-He nández e al.
156 iai.asm.o g In ec ion and Immuni y
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om
p esen exclusi ely in nuclea ed cells o hema opoie ic o igin (51).
Mo e han 200 sec ions o he in es inal issue we e examined a
he mic oscope o each pos in ec ion ime and bac e ial s ain
(i.e., wild- ype and phoP mu an s ains). In he Peye ’s pa ches,
wild- ype and phoP mu an bac e ia we e isualized inside CD18
⫹
CD45
⫹
cells loca ed unde nea h he in es inal epi helium (Fig.
1A). A simila inspec ion in he in es inal illi esul ed in he i-
sualiza ion o CD18
⫹
CD45
⫹
and CD18
⫺
CD45
⫺
cells posi ioned
in he lamina p op ia (Fig. 1B). Su p isingly, he nonphagocy ic
CD18
⫺
CD45
⫺
-nega i e cells con ained a highe in acellula
bac e ial load only in he phoP mu an s ain-in ec ed mice (Fig.
1B). Such a di e ence was consis en ly obse ed in all issue sec-
ions in which in ec ed cells we e p esen in he lamina p op ia o
he illi. Thus, o a o al o 14 CD18
⫺
CD45
⫺
cells isualized in he
FIG 1 S. Typhimu ium a enua es g ow h inside nonphagocy ic cells posi ioned in he lamina p op ia o in es inal illi. (A) Tissue sec ions o he in es inal ileum
co esponding o Peye ’s pa ch a eas we e labeled wi h an ibodies ecognizing S. Typhimu ium lipopolysaccha ide (LPS) and he panphagocy ic ma ke CD18
o CD45. To-p o3 was used o s ain nuclei. Samples we e collec ed a 6 o 24 h pos challenge o BALB/c mice wi h he SV5015 (wild- ype) and MD1120 (phoP
mu an ) s ains. A eas in boxes in uppe panels a e magni ied in he lowe panels. (B) Tissue sec ions showing bac e ium-con aining cells in he lamina p op ia
o in es inal illi. Samples we e collec ed a 6 o 24 h pos in ec ion as desc ibed o panel A and we e labeled wi h an ibodies agains S. Typhimu ium LPS, CD18,
o CD45. No e he p esence o nonphagocy ic s omal cells con aining la ge numbe s o in acellula phoP mu an bac e ia. A eas in boxes in uppe panels a e
magni ied in he lowe panels. (C) Mo phology o p ima y in es inal ib oblas s isola ed om in es inal issue. These p ima y ib oblas s we e in ec ed wi h he
SV5015 (wild- ype) o MD1120 (phoP mu an ) s ain. In pa allel, NRK-49F ib oblas s we e also in ec ed wi h he same s ains. Bac e ia we e de ec ed wi h
an i-S. Typhimu ium LPS an ibodies, and nuclei we e s ained wi h 4=,6=-diamidino-2-phenylindole (DAPI). No e he simila bac e ial pheno ypes in bo h ypes
o ib oblas s. L, in es inal lumen.
T ansc ip ome o Nong owing In acellula Salmonella
Janua y 2013 Volume 81 Numbe 1 iai.asm.o g 157
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om
illi o mice challenged wi h wild- ype bac e ia, none o hem
exhibi ed massi e amoun s o bac e ia. In con as , all he 17
CD18
⫺
CD45
⫺
cells obse ed in he illi o mice challenged wi h
he phoP mu an ha bo ed la ge quan i ies o in acellula bac e-
ia. Addi ional assays p o ed ha he s omal cells in which he
phoP mu an o e g ows we e nega i e o smoo h muscle ac in
(SMA) (see Fig. S1 in he supplemen al ma e ial). SMA is a ma ke
p esen in smoo h muscle cells, myo ib oblas s, and in e s i ial
cells o Cajal (ICC) bu is absen om leukocy es and in e s i ial
s omal ib oblas s (49,52). To assess he possibili y ha he
CD45
⫺
CD18
⫺
cells in which S. Typhimu ium es ic s g ow h
co espond o in e s i ial ib oblas s, we isola ed p ima y ib o-
blas s om he in es inal lamina p op ia o he ileum. When p op-
aga ed in i o, in es inal p ima y ib oblas s exhibi ed uni o m
mo phology (Fig. 1C) and we e in aded a simila a es by wild-
ype and phoP mu an s ain bac e ia (da a no shown). In con-
as o wild- ype bac e ia, he phoP mu an s ain p oli e a ed
ex ensi ely wi hin hese in es inal p ima y ib oblas s in a ashion
eminiscen o ha exhibi ed by he same mu an in CD18
⫺
CD45
⫺
nonphagocy ic cells o he lamina p op ia (Fig. 1B). A
simila beha io was also obse ed o he phoP mu an upon
in asion o cul u ed NRK-49F a ib oblas s (Fig. 1C) in which
his in acellula g ow h-a enua ing esponse was unco e ed
(33). To ou knowledge, hese da a p o ided he i s in i o e i-
dence ha S. Typhimu ium can es ain in acellula g ow h in
he hos .
Genome-wide exp ession p o iling o nong owing in acel-
lula Salmonella loca ed inside ib oblas s. The ma ked pa allel-
ism ound in he beha io o in acellula bac e ia in i o and in
cul u ed ib oblas s (Fig. 1B and C) led us o exploi he in i o
model o dissec he g ow h-a enua ing esponse a he ansc ip-
ome le el. To al RNA was ex ac ed om nong owing wild- ype
bac e ia collec ed a 24 h pos in ec ion o NRK-49F ib oblas s.
RNA ex ac ion om in acellula bac e ia loca ed inside ib o-
blas s equi ed op imiza ion o he p o ocol p e iously desc ibed
o he Salmonella-mac ophage in ec ion model (46). Gi en he
low numbe o nong owing wild- ype bac e ia esiding inside i-
b oblas s (a e age o ⬃2 o 3 bac e ia pe in ec ed cell), we in-
ec ed ca. 10
9
ib oblas s o ob ain he minimal amoun o RNA
equi ed o hyb idiza ion pu poses. T ansc ip omic da a we e
ob ained using he Salgenomics 70-me oligonucleo ide mic oa -
ay pla o m (37,47), which co e s he genome o S. Typhimu-
ium s ain SL1344. Only ela i e exp ession changes g ea e han
4- old we e conside ed signi ican . To al RNA was also pu i ied
om ex acellula wild- ype and phoP mu an s ain bac e ia in an
ac i e (exponen ial) phase o g ow h and o wild- ype bac e ia in
s a iona y phase. The la e condi ion was included o di e en-
ia e genes genuinely exp essed in nong owing condi ions inside
he ib oblas . The ini ial assays e ealed ha he ansc ip omes
o wild- ype and phoP mu an bac e ia we e a he simila in ex-
acellula condi ions (LB medium), wi h only h ee genes, phoP,
STM0939 (ybjD), and STM0940 (ybjX), exhibi ing exp ession di -
e ences g ea e han 4- old (see Table S1 in he supplemen al
ma e ial). The exp ession p o ile o ex acellula wild- ype bac e-
ia g owing exponen ially in LB was used as a compa a o o
nong owing wild- ype bac e ia in in acellula (inside ib oblas s)
and ex acellula (s a iona y-phase) en i onmen s (Fig. 2A; also
see Table S2). The exp ession p o ile o he phoP mu an in expo-
nen ial phase in LB medium was also compa ed o ha o he same
mu an in he in acellula (o e g owing) condi ion (Fig. 2A; also
see Table S2). A o al o 98 genes (ca. 2% o he genome) showed
di e en ial egula ion in nong owing wild- ype bac e ia loca ed
inside he ib oblas (Fig. 2A; also see Table S3). O hese, 51 genes
we e up egula ed (see Table S4) and 48 down egula ed (Fig. 2A;
also see Table S5). Some imb ia- ela ed genes, such as s bB,s bC,
lp A, and imF, we e exp essed a highe le els in nong owing in-
acellula wild- ype bac e ia han in he o e g owing phoP mu-
an (Fig. 2B; also see Table S2). Simila ly, mos o he pSLT i u-
lence plasmid genes we e exp essed a highe le els in nong owing
in acellula wild- ype bac e ia (Fig. 2B; also see Table S2). Up-
egula ion o i ulence plasmid unc ions was con i med a he
p o ein le el o PSLT048 (TlpA), a p o ein egula ed by PhoP-
PhoQ ha was de ec ed only in nong owing in acellula wild-
ype bac e ia (Fig. 2B). Chemo axis and lagella genes we e also
s ongly down egula ed in in acellula bac e ia (see Table S5),
which ag eed wi h he absence o lagellin no ed by Wes e n assays
in wild- ype and phoP mu an bac e ia loca ed inside he ib o-
blas (Fig. 2C). This esponse seems o make sense o bac e ia
pe sis ing wi hin a igh ly apposed memb ane-bound acuole
and, he e o e, no equi ing mo ili y. This obse a ion con as s
wi h he up egula ion o lagellin occu ing a la e in ec ion imes
in Salmonella p oli e a ing wi hin epi helial cells (18,53). Non-
g owing in acellula bac e ia also up egula ed me abolic unc-
ions esponding o low-oxygen condi ions. Examples a e dmsB
and STM1499, which encode subuni s o he anae obic dime hyl
sul oxide educ ase, and glpB, encoding a subuni o he anae obic
glyce ol-3-phospha e dehyd ogenase (see Table S2). Genes en-
coding unc ions ela ed o he u iliza ion o p opanediol (pduT)
o e hanolamine (eu G and eu S) and ce ain hea shock p o eins
(ibpB) we e also exclusi ely up egula ed in nonp oli e a ing wild-
ype bac e ia (see Table S2). Con e sely, gn T, a gene encoding a
high-a ini y glucona e pe mease and p e iously epo ed o be
induced by S. Typhimu ium inside mac ophages (46), was
s ongly down egula ed by nong owing bac e ia inside ib oblas s
(see Table S2). O e all, hese da a indica ed ha he nonp oli e a-
i e li es yle o S. Typhimu ium in ol es a ansc ip ional p o ile
dis inc om hose epo ed o mac ophages and epi helial cells.
Some ea u es o his unique li es yle include me abolic ep o-
g amming o mic oae ophilic condi ions and gene exp ession
changes ha can be en a i ely in e p e ed as ene ge ic es ain .
Cha ac e iza ion o he Salmonella PhoP-PhoQ egulon in
nong owing in acellula bac e ia. As expec ed, some known
PhoPQ- egula edgenes, suchas pagC andmg C, we eup egula ed
by in acellula wild- ype bac e ia in he nong owing do man
s a e, while such up egula ion was no obse ed in o e g owing
phoP mu an bac e ia (see Table S2 in he supplemen al ma e ial).
Based on hese obse a ions, we easoned ha compa ison o he
exp ession p o iles o nong owing wild- ype bac e ia (in acellu-
la and ex acellula , s a iona y phase) o ha o o e g owing in-
acellula phoP mu an bac e ia could e eal ea u es o he non-
p oli e a i e in acellula li es yle (Fig. 3A). A o al o 160 genes
we e ound o be di e en ially exp essed in nong owing in acel-
lula bac e ia compa ed o ex acellula s a iona y-phase bac e ia,
he e o e hey we e conside ed o espond genuinely o in acel-
lula cues (Fig. 3A; also see Table S6). On he o he hand, non-
g owing wild- ype bac e ia di e ed in he exp ession o 270 genes
compa ed o o e g owing in acellula phoP mu an bac e ia (Fig.
3A; also see Table S7). O hese 270 genes, 60 esponded in a
PhoP-PhoQ-dependen manne o bo h ai s, he in acellula
en i onmen and a nong owing condi ion (Fig.3A). In e es ingly,
Núñez-He nández e al.
158 iai.asm.o g In ec ion and Immuni y
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om

some o hese genes we e no p e iously asc ibed o he PhoP-
PhoQ egulon in ex acellula g owing condi ions (Fig. 3A). To
alida e hese obse a ions, we quan i ied he ela i e ansc ip
le els o mg C and pagC in in acellula and ex acellula bac e ia.
Ano he gene, ushA, no p e iously assigned o he PhoP-PhoQ
egulon and displaying al e ed exp ession exclusi ely in nong ow-
ing in acellula bac e ia (Fig. 3A; also see Table S2), was also
included in he analysis. ushA encodes a pu a i e UDP-suga hy-
d olase/5=-nucleo idase ha is p esen in S. en e ica and Esche-
ichia coli (54). RT-qPCR assays con i med ha he exp ession o
mg C,pagC, and ushA in nonp oli e a ing in acellula bac e ia
wasPhoP-PhoQ dependen (Fig. 3B). In e es ingly, he ushA allele
ha bo ed by he s ain used in ou s udy (SV5015; a His
⫹
de i a e
o SL1344) encodes a p o ein no con aining he S139Y missense
mu a ion epo ed o ab oga e he ac i i y o his enzyme in o he
S. Typhimu ium s ains, such as LT2 (54) (see Fig. S2). In addi ion
o ushA, we also alida ed glpK, a gene encoding a pu a i e glyc-
e ol-kinase ha displayed nega i e egula ion by he PhoP-PhoQ
sys em in in acellula bac e ia (see Tables S2 and S7). Using a
glpK::3⫻FLAG- agged s ain om a p e ious p o eomic s udy
(38), we cons uc ed a de i a e phoP glpK::3⫻FLAG isogenic
s ain o de e mine ela i e le els o GlpK in in acellula and
ex acellula bac e ia. In conco dance wi h he ansc ip omic
da a, GlpK le els we e ound o be highe in he phoP mu an han
in wild- ype bac e ia (Fig. 3B). The ansc ip omes ob ained wi h
RNA ex ac ed om wild- ype and phoP mu an s ains he e o e
p o ide a aluable sou ce o iden i y new genes hi he o no as-
signed o he PhoP-PhoQ egulon.
Kine ics o induc ion o he PhoP-PhoQ sys em in in acel-
lula nong owing bac e ia. Using ecombinan S. Typhimu ium
s ains ha bo ing a phoP::GFP ansc ip ional usion, Ma in-
O ozco e al. showed in cul u ed mac ophages ha phoP exp es-
sion is up egula ed in in acellula bac e ia om 30 min pos en y
(55). A his ime, GFP-de i ed luo escence was de ec ed in
FIG 2 Genome-wide exp ession analyses e eal unique signa u es in nong owing in acellula wild- ype S. Typhimu ium. (A) Numbe o genes displaying
exp ession changes highe han 4- old (log
2
⫽M alue o ⱕ⫺2o ⱖ2) in in acellula nong owing bac e ia (wild- ype s ain SV5015) o in acellula
p oli e a ing bac e ia (phoP mu an s ain MD1120) a 24 h pos in ec ion o NRK-49F ib oblas s. A hi d sample, co esponding o ex acellula bac e ia g own
o e nigh o s a iona y phase in LB medium in shaking condi ions, was included o compa ison. The h ee samples a e e e enced o he exp ession pa e n
displayed by ac i ely g owing ex acellula bac e ia g own in LB medium o an OD o 0.2 (exponen ial phase) (see he ex o de ails). The numbe o genes ha
displayed up- and down egula ion compa ed o bac e ia g own o exponen ial phase a e also indica ed. (B) Hea map showing he up egula ion in in acellula
bac e ia o genes mapping in he i ulence plasmid pSLT. None ha , ela i e o ex acellula bac e ia, pSLT plasmid genes a e up egula ed o a g ea e ex en in
nong owing in acellula wild- ype bac e ia han in in acellula phoP mu an bac e ia (also see Table S2 in he supplemen al ma e ial). A ep esen a i e case
con i ming his di e ence is shown o he TlpA (PSLT048) plasmid p o ein. (C) Hea map showing he exp ession changes o gene clus e s encoding dis inc
imb iae o lagella p o eins. No e he down egula ion in he exp ession o lagella and chemo axis genes in in acellula wild- ype and mu an bac e ia (bo h
wild- ype and phoP mu an s ains) (see Table S2). A Wes e n assay demons a ing he ma ked d op in lagellin (FliC/FljB) ela i e le els in in acellula bac e ia
is shown.
T ansc ip ome o Nong owing In acellula Salmonella
Janua y 2013 Volume 81 Numbe 1 iai.asm.o g 159
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om
⬃50% o he in ec ed mac ophages. These kine ics we e mo e
apid han hose obse ed in ex acellula bac e ia exposed o
PhoP-PhoQ-inducing signals a a low magnesium concen a ion,
which equi ed abou 1h(55). Based on his, we in es iga ed
whe he ac i a ion o he PhoP-PhoQ sys em inside ib oblas s
sha es simila i ies wi h mac ophages. To ha aim, we enginee ed
s ains ca ying mg C::3⫻FLAG and pagC::3⫻FLAG alleles agged
in hei 3=ends and in hei espec i e ch omosomal loca ions.
Unde hese condi ions, he egula o y scheme emains unal e ed.
P o ein ex ac s we e p epa ed om ex acellula bac e ia g own
in inducing (8 ␮MMg
2⫹
) and ep essing (10 mM Mg
2⫹
) condi-
ions o he PhoP-PhoQ sys em. Rela i e p o ein le els we e
compa ed o hose de ec ed in in acellula bac e ia collec ed a
ea ly and la e pos in ec ion imes (1 and 24 h). Bo h p o eins,
FIG 3 Cha ac e iza ion o he S. Typhimu ium PhoP-PhoQ egulon in do man nong owing in acellula bac e ia. (A) Venn diag am showing he o e lapping
among ansc ip omes ob ained om in acellula nong owing bac e ia (wild- ype s ain SV5015), in acellula p oli e a ing bac e ia (phoP mu an s ain
MD1120), and ex acellula wild- ype bac e ia g own o s a iona y phase in LB medium. Each o hese ansc ip omes e e s o he exp ession o ex acellula
bac e ia g own o exponen ial phase (see Table S2 in he supplemen al ma e ial). Numbe s o genes di e ing in exp ession among ansc ip omes by mo e han
4- old a e indica ed. Highligh ed in boxes a e some o he genes displaying di e en ial exp ession among ansc ip omes (in a-WT e sus ex a-WT and
in a-WT e sus in a-phoP), which espond o bo h he in acellula en i onmen and he unc ional s a us o he PhoP-PhoQ sys em. Genes p e iously
epo ed o be egula ed by PhoP-PhoQ a e indica ed in blue (see Tables S6 and S7 o de ails). (B) Valida ion da a ob ained by RT-qPCR o wo PhoP-PhoQ-
egula ed genes, mg C and pagC. Two o he genes hi he o no assigned o he PhoP-PhoQ egulon, ushA and glpK, we e also alida ed a he ansc ip and
p o ein le els, espec i ely. mg C and pagC da a e e o di e en pos in ec ion imes (1, 4, 8, and 24 h) upon in asion o NRK-49F ib oblas s and a e ela i e o
exp ession le els de ec ed in ex acellula bac e ia g owing o exponen ial phase. ushA da a a e ela i e o he exp ession le els egis e ed a 24 h pos in ec ion in
nong owing in acellula bac e ia. In he RT-qPCR assays, exp ession alues we e no malized o hose ob ained o he ompA gene used as an in e nal con ol.
(C) Rela i e le els o he Mg C-3⫻FLAG- and PagC-3⫻FLAG- agged p o eins de ec ed in in acellula nong owing wild- ype bac e ia and in he o e g owing
phoP mu an a he indica ed pos in ec ion imes. As a con ol o canonical PhoP-PhoQ egula ion, hese wo p o eins we e also moni o ed in ex acellula
bac e ia g own in inducing (8 ␮MMg
2⫹
) o ep essing (10 mM Mg
2⫹
) condi ions. (D) Le els o he al e na i e sigma ac o RpoS de ec ed in in acellula
bac e ia a di e en pos in ec ion imes upon en y in o NRK-49F ib oblas s. DnaK, OmpA (bac e ial p o eins), and calnexin (euka yo ic p o ein) we e used as
loading con ols.
Núñez-He nández e al.
160 iai.asm.o g In ec ion and Immuni y
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om
PagC and Mg C, we e de ec ed in he nong owing in acellula
wild ype om 2 h pos in ec ion, wi h a p og essi e inc ease in
p o ein le els o e ime (Fig. 3C). As expec ed, PagC and Mg C
le els we e negligible a all pos in ec ion imes es ed in he o e -
g owing phoP mu an collec ed om ib oblas s (Fig. 3C). Taken
oge he , ou da a show ha in ib oblas s he PhoP-PhoQ sys em
is ac i a ed in in acellula S. Typhimu ium in he absence o any
no iceable inc ease in bac e ial g ow h, mos p obably be ween 1
and 2 h pos in ec ion.
We nex ocused on ein o cing he idea ha in acellula S.
Typhimu ium adap s o a nong owing s a e inside he ib oblas s.
We hypo hesized ha he al e na i e sigma ac o RpoS, equi ed
o adap a ion o bac e ia o nonp oli e a ing (s a iona y phase)
condi ions (56) and egula ed pos ansc ip ionally by PhoP-
PhoQ in Salmonella (57), could be p oduced in la ge amoun s by
nong owing wild- ype bac e ia. Ou ea ly s udies also e ealed
ha S. Typhimu ium poS mu an s o e g ow inside ib oblas s
(33). Wes e n assays showed ha nong owing in acellula wild-
ype bac e ia con ained la ge amoun s o RpoS han he o e -
g owing phoP mu an bac e ia (Fig. 3D). This di e ence was mo e
e iden a 24 h pos in ec ion, when wild- ype bac e ia may equi e
mo e RpoS o ace s esses linked o long-las ing esidence in he
in ec ed cell. These da a suppo he exis ence o a posi i e egu-
la ion o he PhoP-PhoQ sys em o e RpoS in bac e ia pe sis ing
inside he ib oblas .
Induc ion o PhoP-PhoQ in nonp oli e a ing in acellula S.
Typhimu ium loca ed inside ib oblas s esponds o acuola
acidi ica ion. The S. Typhimu ium egula o y sys em PhoP-
PhoQ has been shown o espond in i o o di e se signals, in-
cluding Mg
2⫹
limi a ion (58,59), an imic obial pep ides (60,61),
and acidic pH (62). The ex en o which hese signals ac i a e
PhoP-PhoQ in bac e ia loca ed in he phagosome is s ill a ma e
o deba e (8,63). Gi en ha PhoP-PhoQ induc ion could be easily
moni o ed a he p o ein le el in nong owing in acellula bac e-
ia (Fig. 3), we sough o de e mine in ib oblas he signals sensed
by his sys em by compa ing dis inc inducing condi ions in ex a-
and in acellula bac e ia. Fo his pu pose, we gene a ed addi-
ional epi ope- agged s ains in h ee genes posi i ely egula ed by
PhoP-PhoQ, namely, i K,pagN, and pagP. The ela i e le els o
3⫻FLAG- agged Vi K, PagN, and PagP p o eins we e quan i ied
in in acellula bac e ia a 24 h pos in ec ion o NRK-49F and
compa ed o ex acellula bac e ia g own in 8 ␮Mo 10mM
Mg
2⫹
concen a ion. We also included in he analysis bac e ia
g own in PCN minimal medium con aining 1 mM Mg
2⫹
and
adjus ed o wo dis inc pH alues, 5.8 o 7.4. The induc ion pa -
e n obse ed in nong owing in acellula bac e ia indica ed ha
PagN was s ongly induced inside he ib oblas as well as Vi K,
al hough in he la e case he induc ion was less p onounced (Fig.
4). In e es ingly, he ela i e le el o PagP did no inc ease in in-
acellula bac e ia, al hough i s exp ession was clea ly PhoP-
PhoQ dependen (Fig. 4). The exp ession pa e n o Vi K, PagN,
and PagP in in acellula bac e ia was ep oduced in ex acellula
bac e ia g own in acidi ied (pH 5.8) PCN medium bu no in a low
(8 ␮M) Mg
2⫹
concen a ion. Thus, Vi K and PagN we e induced
in bo h condi ions while PagP esponded exclusi ely o Mg
2⫹
lim-
i a ion (Fig. 4). Taken oge he , hese obse a ions sugges ha
induc ion o PhoP-PhoQ by nong owing do man in acellula S.
Typhimu ium esiding in acuoles o ib oblas s occu s upon
sensing o acidic pH.
An acidi ied acuole di ec s su i al o nong owing in acel-
lula S. Typhimu ium inside he ib oblas . Since ou da a sup-
po ed he idea o PhoP-PhoQ being induced upon bac e ial sens-
ing o in a acuola acidic pH (Fig. 4), we easoned ha loss o
such acidi ica ion could ha e consequences in PhoP-PhoQ unc-
ion and i ness o in acellula bac e ia. Indeed, ea ly s udies e-
po ed a key ole o acid pH o PhoP-PhoQ induc ion in S. Ty-
phimu ium loca ed inside mac ophages (35,55). To es whe he
a simila phenomenon occu s in ib oblas s, we moni o ed Mg C
and PagC ela i e le els in S. Typhimu ium esiding wi hin NRK-
49F ib oblas s ha we e ea ed wi h he in a acuola acidi ica-
ion inhibi o ba ilomycin A1 (64) a e bac e ial en y. Con ol
expe imen s showed ha no cy o oxici y occu ed in he ib o-
blas s a he concen a ion o he d ug used (100 nM). Dissipa ion
o in a acuola acidi ica ion esul ed in lowe p oduc ion o
FIG 4 Regula ion exe ed by he PhoP-PhoQ sys em in nong owing do man in acellula bac e ia ma ches he egula o y pa e n obse ed in ex acellula
bac e ia incuba ed in acidi ied g ow h medium. Wes e n assays showing he ela i e le els o h ee dis inc 3⫻FLAG- agged p o eins egula ed by he PhoP-
PhoQ sys em (Vi K, PagN, and PagP) in ex acellula and in acellula bac e ia. Induc ion in in acellula bac e ia was moni o ed by analysis o p o ein le els in
ex acellula bac e ia used o in ec he NRK-49F ib oblas s (inoculum, nonshaking g ow h condi ions) and in acellula bac e ia collec ed a 24 h pos in ec ion.
These samples a e ma ked as ex a and in a, espec i ely. The posi i e egula ion o hese h ee p o eins by PhoP-PhoQ was es ed in low Mg
2⫹
concen a ions
andacidpHusing he N and PCN media, espec i ely (see Ma e ials and Me hods). Shown a e hele elso Vi K, PagN, and PagP de ec ed in ex acellula bac e ia
g own in inducing (ei he 8 ␮MMg
2⫹
o pH 5.8) o ep essing (ei he 10 mM Mg
2⫹
o pH 7.4) condi ions. No e ha he esponse obse ed in acidi ied PCN
medium ma ches, o a la ge ex en , ha obse ed in in acellula bac e ia. Howe e , he ma ked inc ease o PagP le els obse ed in 8 ␮MMg
2⫹
is no obse ed
in nonp oli e a ing in acellula bac e ia. Loading con ols based on DnaK a e shown o he pagN::3⫻FLAG- agged s ains wi h equi alen esul s ob ained o
he o he se s o s ains shown.
T ansc ip ome o Nong owing In acellula Salmonella
Janua y 2013 Volume 81 Numbe 1 iai.asm.o g 161
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om
Mg C o PagC by nong owing in acellula bac e ia, con i ming
he impo an ole played by acid pH in PhoP-PhoQ induc ion
(Fig. 5A). This e ec was especially e iden in he case o PagC,
which was p oduced by nong owing in acellula bac e ia a
ela i ely highe le els han Mg C in he un ea ed ib oblas s
(Fig. 5A). P e ious s udies in mac ophages showed ha S. Typhi-
mu ium iabili y elies on he main enance o an in a acuola
pH (65). Coun ing o iable in acellula bac e ia a 2 and 24 h
pos in ec ion o ib oblas s also e ealed ha inhibi ion o in a-
acuola acidi ica ion esul s in loss o iabili y o nonp oli e a -
ing in acellula bac e ia (Fig. 5B). Dissipa ion o in a acuola
acidi ica ion also dec eased he g ow h a e o he phoP mu an in
he ib oblas (Fig. 5B). In e es ingly, ba ilomycin A1 did no a -
ec su i al o an SPI-2 mu an (sseC)(Fig. 5B) which is known o
lose iabili y inside no mal un ea ed ib oblas s (33). Taken o-
ge he , hese da a indica e ha PhoP-PhoQ induc ion, SPI-2 ac-
i a ion, and main enance o su i al by nong owing in acellula
S. Typhimu ium a e in e connec ed phenomena equi ing acu-
ola acidi ica ion.
DISCUSSION
This s udy epo s he i s genome-wide exp ession analysis pe -
o med in in acellula S. Typhimu ium while pe sis ing in a non-
g owing s a e wi hin he in ec ed hos cell. The occu ence o neg-
a i e egula ion o Salmonella in acellula p oli e a ion due o he
ac ion o pa hogen unc ions was en isioned in ou ea ly s udies
on ib oblas s, which un a eled he equi emen o he PhoP-
PhoQ sys em o es ic bac e ial g ow h (33,34). O he au ho s
epo ed mu an s exhibi ing inc eased loads o in acellula bac-
e ia in mac ophages (32). Howe e , none o hese s udies in es-
iga ed he physiology o nonp oli e a ing in acellula bac e ia
and he basis o g ow h es ain . In addi ion, he issue(s) and cell
ype(s) in which Salmonella may ac i a e in i o hese in acellula
esponses we e unknown. Mic oscopy analyses shown he e un-
equi ocally demons a e ha nonphagocy ic cells posi ioned in
he lamina p op ia o in es inal illi ha bo bac e ia ha a enua e
in acellula g ow h in a PhoP-PhoQ-dependen manne . Aside
om phagocy ic cells such as neu ophils, T and B lymphocy es,
monocy es, and dend i ic cells, ib oblas s a e he only cells
known o popula e he lamina p op ia o in es inal illi. The no-
ion ha Salmonella is capable o in ec ing ib oblas s in his loca-
ion is suppo ed by he ema kable iden i y in he pheno ypes
exhibi ed by wild- ype and phoP mu an bac e ia in p ima y ib o-
blas s isola ed om he lamina p op ia (Fig. 1). Recen s udies o
he s ep omycin mouse model o S. Typhimu ium dia hea e-
ealed ha he pa hogen a ge s epi helial cells and lamina p op ia
phagocy es (66). Ou assays we e, howe e , pe o med in a y-
phoid in ec ion model, o which only sca ce in o ma ion exis s
on he ea ly e en s occu ing in he in es ine and he in es inal cell
ypes colonized by S. Typhimu ium. In he la e model, pa allel
ou es in ol ing a ic and dissemina ion o he pa hogen
h ough he lympha ic sys em a e being inges ed by dend i ic
cells o CD18
⫹
phagocy es (monocy es o DCs) seem o occu
(10). Howe e , a de ailed mic oscopy analysis a he le el o hos
cell popula ions con aining he pa hogen is no a ailable ye . Al-
hough he lack o highly speci ic ib oblas ma ke s sui able o
immunohis ochemis y makes his ype o s udy di icul , ou
indings p o ide he i s in i o e idence o an S. Typhimu ium
esponse di ec ed o es ain g ow h wi hin he in ec ed cell.
The ansc ip ome p o ile ob ained om cul u ed ib oblas s
sha ed some ea u es wi h genome-wide exp ession da a epo ed
o S. Typhimu ium p oli e a ing inside mac ophages and epi he-
lial cells (46,53,67). Examples included s ess- ela ed unc ions o
he amily o phage-shock p o eins (Psp) ha espond o impai ed
memb ane unc ion (68) and unc ions egula ed posi i ely by
PhoP-PhoQ, such as hose o Mg B, Mg C, PagC, Mig-3, and
PhoN. An in iguing obse a ion was he la e exp ession (24 h
pos in ec ion) o SPI-1 genes by do man nonp oli e a ing in a-
FIG 5 Ac i i y o he PhoP-PhoQ sys em in nonp oli e a ing in acellula S. Typhimu ium esponds o in a acuola acidic pH. (A) E ec o he dissipa ion o
in a acuola acidi ica ion on he induc ion o he PhoP-PhoQ sys em. Shown a e he ela i e le els o he 3⫻FLAG- agged p o eins Mg C and PagC p oduced
by in acellula bac e ia isola ed om NRK-49F ib oblas s ha we e le un ea ed o we e ea ed wi h 100 nM ba ilomycin (BAF), an inhibi o o acuola
acidi ica ion. OmpA (bac e ial p o ein) and calnexin (euka yo ic p o ein) we e used o loading con ols. (B) E ec o loss o acuola acidi ica ion on he
iabili y o in acellula bac e ia. Shown a e he a ios o iable in acellula bac e ia enume a ed a 24 h e sus 2 h. Da a a e he means and s anda d de ia ions
om h ee independen expe imen s. **, P⫽0.001 o 0.01; ***, P⬍0.001; n.s., no signi ican by a S uden es .
Núñez-He nández e al.
162 iai.asm.o g In ec ion and Immuni y
on July 18, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://iai.asm.o g/Downloaded om