Vi ulence P o iles o Bac e emic Ex ended-Spec um b-
Lac amase-P oducing
Esche ichia coli
: Associa ion wi h
Epidemiological and Clinical Fea u es
Jesu
´s Rod ı
´guez-Ban
˜o
1,2
*, Jesu
´s Mingo ance
3
, Na alia Fe na
´ndez-Rome o
3
, La a Se ano
1
, Lo ena Lo
´pez-
Ce e o
1
, Al a o Pascual
1,4
, he ESBL-REIPI g oup
"
1Unidad Clı
´nica de En e medades In ecciosas y Mic obiologı
´a, Hospi al Uni e si a io Vi gen Maca ena, Se illa, Spain, 2Se icio de Mic obiologı
´a, Hospi al Uni e si a io La
Paz - IdiPAZ, Mad id, Spain, 3Depa amen o de Medicina, Uni e sidad de Se illa, Se illa, Spain, 4Depa amen o de Mic obiologı
´a, Uni e sidad de Se illa, Se illa, Spain
Abs ac
The e is sca ce da a abou he impo ance o phylog oups and i ulence ac o s (VF) in bloods eam in ec ions (BSI) caused
by ex ended-spec um b-lac amase-p oducing Esche ichia coli (ESBLEC). A p ospec i e mul icen e Spanish coho including
191 cases o BSI due o ESBLEC was s udied. Phylog oups and 25 VF genes we e in es iga ed by PCR. ESBLEC we e classi ied
in o clus e s acco ding o hei i ulence p o iles. The associa ion o phylog opus, VF, and clus e s wi h epidemiological
ea u es we e s udied using mul i a ia e analysis. O e all, 57.6%, 26.7%, and 15.7% o isola es belonged o A/B1, D and B2
phylog oups, espec i ely. By mul i a ia e analysis (adjus ed OR [95% CI]), i ulence clus e C2 was independen ly
associa ed wi h u ina y ac sou ce (5.05 [0.96–25.48]); clus e C4 wi h sou ces o he han u ina y o bilia y ac (2.89 [1.05–
7.93]), and clus e C5 wi h BSI in non-p edisposed pa ien s (2.80 [0.99–7.93]). Isola es p oducing CTX-M-9 g oup ESBLs and
om phylog oup D p edomina ed among clus e C2 and C5, while CTX-M-1 g oup o ESBL and phylog oup B2
p edominan es among C4 isola es. These esul s sugges ha hos ac o s and p e ious an imic obial use we e mo e
impo an han phylog oup o speci ic VF in he occu ence o BSI due o ESBLEC. Howe e , some associa ions be ween
i ulence clus e s and some speci ic epidemiological ea u es we e ound.
Ci a ion: Rod ı
´guez-Ban
˜o J, Mingo ance J, Fe na
´ndez-Rome o N, Se ano L, Lo
´pez-Ce e o L, e al. (2012) Vi ulence P o iles o Bac e emic Ex ended-Spec um b-
Lac amase-P oducing Esche ichia coli: Associa ion wi h Epidemiological and Clinical Fea u es. PLoS ONE 7(9): e44238. doi:10.1371/jou nal.pone.0044238
Edi o : Axel Cloeckae , Ins i u Na ional de la Reche che Ag onomique, F ance
Recei ed June 11, 2012; Accep ed July 30, 2012; Published Sep embe 7, 2012
Copy igh : ß2012 Rod ı
´guez-Ban
˜o e al. 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, 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 au ho and sou ce a e c edi ed.
Funding: This s udy was unded by he Minis e io de Ciencia e Inno acio
´n, Ins i u o de Salud Ca los III - co- inanced by Eu opean De elopmen Regional Fund ‘‘A
way o achie e Eu ope’’ ERDF, Spanish Ne wo k o Resea ch in In ec ious Diseases (REIPI RD06/0008), Fondo de In es igacio
´n Sani a ia (g an s 070190, 10/02021,
10/01955, and 10/00795), and Jun a de Andalucı
´a (g an s 0048/2008, and CTS-5259). The unde s had no ole in he s udy design, da a collec ion and analysis,
decision o publish, o p epa a ion o he manusc ip .
Compe ing In e es s: The au ho s ha e ead he jou nal’s policy and ha e he ollowing con lic s: J. Rod ı
´guez-Ban
˜o has been a consul an o Wye h, Me ck,
and P ize , has se ed as speake o Wye h, Me ck, P ize , As a-Zeneca and GlaxoSmi hKline, and has ecei ed esea ch suppo om Me ck and Wye h.J.
Mingo ance has ecei ed esea ch suppo om Roche and P ize . A. Pascual has been a consul an o Me ck and P ize , has se ed as speake o Wye h, As a-
Zeneca, Me ck, and P ize and has ecei ed esea ch suppo om Me ck and P ize and Wye h. All o he au ho s had no con lic o in e es . This does no al e he
au ho s’ adhe ence o all he PLoS ONE policies on sha ing da a and ma e ials.
* E-mail: [email p o ec ed]
"Membe ship o he ESBL-REIPI g oup is p o ided in he Acknowledgmen s.
In oduc ion
Mos ex ain es inal in ec ions due o Esche ichia coli a e caused
by isola es de i ed om he so-called i ulen phylogene ic g oups
(PG) B2 and D, which exhibi mo e i ulence ac o s (VF) han
o he PGs such as A and B1, hence conside ed ‘‘low i ulence’’ o
‘‘commensal’’ PGs [1,2]. Speci ically, in s udies on bloods eam
in ec ions (BSI) caused by E. coli, .70% o he isola es belonged o
PGs B2 (which was p edominan ) and D [3–8]. Isola es om PG
B2 ha e been associa ed wi h BSI wi h a u ina y ac sou ce
[4,5,7] and in non-p edisposed pa ien s [7,9], while PG A has been
ound wi h inc eased equency in nosocomial BSI, comp omised
hos s [4,6,10] and in BSI caused by an ibio ic- esis an isola es
[5,6,8].
Some VFs ha e been assigned a pa hogenic ole in ex ain es-
inal in ec ions based on compa isons wi h ec al isola es,
associa ion wi h in ec ions in non-p edisposed pa ien s, and animal
models [11]. In s udies dealing wi h BSI, se e al VF has been
ound o be associa ed wi h speci ic epidemiologic ea u es, bu i is
papGII ha has been mo e consis en ly associa ed wi h u ina y
ac sepsis as opposed o o he sou ces, and in pa ien s wi hou
p edisposing ac o s [4,9,12].
Ex ended-spec um be a-lac amase-p oducing Esche ichia coli
(ESBLEC) a e inc easing wo ldwide as a cause o communi y
and nosocomial BSI, equen ly a ec ing pa ien s wi h p edispos-
ing condi ions [12,13]. The e a e sca ce da a abou he
dis ibu ion o PGs and VF in BSI due o ESBLEC [14] and
hei associa ion wi h an imic obial esis ance. Also, o he bes o
ou knowledge, he impac o speci ic VF in he epidemiology o
BSI due o ESBLEC has no been s udied. Finally, he e is some
con o e sy abou he eal i ulence o ESBLEC, including isola es
p oducing CTX-M-15 belonging o he wo ldwide sp ead clone
ST131 [16–18].
The objec i es o his s udy we e o in es iga e he PGs and
p e alence o VF genes in a well cha ac e ized collec ion o
ESBLEC causing BSI, and whe he some PGs and VF (indi id-
PLOS ONE | www.plosone.o g 1 Sep embe 2012 | Volume 7 | Issue 9 | e44238
ually o in clus e s) we e associa ed wi h he epidemiology,
pa ien s’ ea u es and sou ce o BSI.
Me hods
S udy Design and Pa ien s
Da a and isola es om a p ospec i e coho including 191 cases
o BSI due o ESBLEC om 13 Spanish hospi als we e used o
his analysis. The epidemiology, clinical ea u es, ou comes, ypes
o ESBL and suscep ibili y da a o his coho we e p e iously
epo ed [13,14]. B ie ly, all monomic obial BSI in pa ien s wi h
sign o symp oms o sys emic in ec ion caused by ESBLEC
diagnosed in he pa icipa ing hospi als be ween Oc obe 2004
and Janua y 2006 we e included. The cases we e de ec ed by daily
e iew o mic obiological esul s o blood cul u es a each cen e .
Da a collec ed included demog aphics, acquisi ion classi ied as
communi y, heal hca e-associa ed o nosocomial [13], ch onic
unde lying diseases, se e i y o unde lying condi ion acco ding o
Cha lson index [19], in asi e p ocedu es, exposu e o an ibio ics
in he p eceding 2 mon hs, and sou ce o BSI acco ding o clinical
and mic obiological c i e ia.
Fo his analysis, pa ien s wi h any o he ollowing we e
conside ed o ha e sys emic p edisposing ea u es o BSI: diabe es
melli us, li e ci hosis, ch onic enal insu iciency, inmunosup-
p esi e he apy, and neu openia. Pa ien s wi h a p ocedu e-
associa ed BSI (including ascula ca he e , u ina y ca he e ,
endoscopic p ocedu es and su ge y), o any u ina y o bilia y ac
BSI in he p esence o obs uc i e diseases o hese ac s we e
conside ed o ha e local p edisposing ac o s o BSI. The s udy
was app o ed by he E hics Commi ee o Hospi al Uni e si a io
Vi gen Maca ena which wai ed he need o ob ained consen
because all da a we e analysed anonymously and he obse a ional
na u e o he s udy.
Mic obiological S udies
Me hods o bac e ial iden i ica ion, suscep ibili y s udies and
ESBL con i ma ion and cha ac e iza ion we e p e iously epo ed
[13,14]. B ie ly, ESBL p oduc ion and suscep ibili y by mic o-
dilu ion o ce u oxime, ce o axime, ce azidime, ce epime, amox-
icillin-cla ulanic acid, pipe acillin- azobac am, cip o loxacin, gen-
amycin, ob amycin, amikacin, e apenem, imipenem, me ope-
nem, ime hop im-sul ame hoxazol, os omycin, and igecycline
we e s udied acco ding o CLSI ecommenda ions [20]; a
esis ance sco e (numbe o an imic obials o which he isola e
was esis an ) was calcula ed o each isola e. b-lac amase
cha ac e iza ion was ca ied ou by isoelec ic ocusing, PCR o
he bla genes, and sequencing. ST131 clone was s udied by O25b
yping [17] and analysis o allele 3 o pabB [21]; he phylogene ic
g oup was de e mined by mul iplex PCR [22].
Twen y- i e genes codi ying o pu a i e VF we e s udied,
including adhesins (papC, papGI, papGII, papGIII, imH, s aD/E,
a aB/C, iha); oxins (cn 1, cd B, sa , hlyA); ela ed o i on acquisi ion
(iucD, i oN, iu A, i eA, and yuA); p o ec ins (kps MTII, aT, c aC, and
ompT); and miscellaneous (ibeA, maIX, s g, and usp). The p esence/
absence o VF genes was s udied by PCR using p e iously
desc ibed p ime s [2,23–31]. To al DNA was pu i ied om each
s ain wi h he Ul aClean Mic obial DNA pu i ica ion ki (MO
BIO Labo a o ies Inc., Ca lsbad, CA). DNAs we e dis ibu ed in
96-well mas e pla es and PCRs we e done in 50 mL mix u es
con aining 5 ml (20 ng) empla e DNA, 0.2 mM o each p ime ,
0.2 mM mix dNTPs and 1U DNA polyme ase (Bio ools S. L.,
Spain) in 1X bu e wi h MgCl
2
. PCR condi ions we e as ollows:
5 min a 95uC, ollowed by 30 cycles o 30 s a 95uC, 30 s a
annealing empe a u e o each p ime pai , 1 min a 72uC, and a
inal 5 min incuba ion a 72uC. The PCR p oduc s we e analyzed
by elec opho esis in 96-well aga ose gels (VG-FAST, Fishe
Bioblock Scien i ic) s ained wi h GelRed
TM
(Bio ium Inc.). A
i ulence sco e (numbe o VF genes) was calcula ed o each
isola e. The simila i y o he isola es acco ding o hei VF
geno ypes was s udied by cons uc ing a dendog am using he
bina y pa e s (0, 1) o VF o each isola e; clus e s o isola es we e
iden i ied using he Dice simila i y coe icien . A e e iewing he
da a ob ained, a 70% simila i y h eshold was used a e e iewing
he da a ob ained (a 60% h eshold was no disc imina i e enough,
since only one clus e included 72.2% isola es; and a 80%
h eshold ound only 4 clus e s wi h .5 cases including 32.4% o
he isola es).
S a is ical Analysis
Pe cen ages we e compa ed using he chi squa ed es o he
Fishe exac es , as app op ia e, and con inuous a iables using
he Mann-Whi ney U es . Mul i a ia e analysis we e pe o med
by logis ic eg ession; a iables wi h a uni a ia e p alue ,0.1
we e in oduced in he models, and selec ed using a s epwise
backwa d p ocess; 0.1 was se as he limi o emo al o e ms. All
es s we e pe o med using SPSS 18.0.
Resul s
Among he 191 ESBLEC bac e emic isola es, 55 (28.8%) belong
o PG A, 55 (28.8%) o PG B1, 51 (26.7%) belong o PG D, and 30
(15.7%) o B2. The median (IQR) i ulence sco e we e 10 (9–12)
o B2, 8 (7–9) o D, 5 (4–6) o B1, and 4 (2–6) o A. Among he
B2 isola es, 21 (70% o B2, 10.9% o he whole se ies) we e O25b
and pabB3 posi i e and we e conside ed as belonging o ST131.
Fo easie unde s anding, PGs A and B1 we e analyzed oge he
because bo h showed simila i ulence sco es, equency o VF
(only imH,iucD, and i oN we e signi ican ly mo e equen among
B1 han among A), ESBLs (p edominance o CTX-M-14), and
esis ance pa e ns.
The equency o VF genes acco ding o PG is shown in able 1.
In summa y, papC, a aB/C, iha, imH, sa , yuA, iu A, kps MTII,
ompT, ibeA, usp, and maIX we e mo e equen among in B2 and D
han in A/B1; addi ionally, papGIII, hlyA, cn 1, and cd B, al hough
in equen , we e mo e p e alen in B2 han in A/B1; and papGII
and i eA we e mo e equen in D han in A/B1. Finally, s aD/E,
a aB/C, sa , yuA, kps MTII, ompT, usp, and maIX we e mo e
equen in B2 han in D, while only papGII was mo e equen in
D han in B2.
The dis ibu ion o he ESBL g oups p oduced and esis ance
p o iles o an imic obials acco ding o PG a e also shown in
able 1. ESBLs om he CTX-M-9 g oup (mainly CTX-M-14)
we e he mos equen among D and A/B1 isola es, while hose
om he CTX-M-1 g oup (mainly CTX-M-15) we e he mos
equen among B2 isola es. As ega ds an imic obial esis ance,
isola es om he B2 PG showed a highe esis ance sco e han A/
B1 and we e mo e equen ly esis an o amoxicillin-cla ulanic
acid, ob amycin, and amikacin, while hose om PG D we e
mo e equen ly esis an o co- imoxazole and less o ce azidime.
Among he B2 isola es, 21 (70%) we e ST131. The ESBLs
p oduced by ST131 and non-ST131 B2 isola es di e ed; hus,
CTX-M-15 was p oduced by 15/21 o ST131 isola es (71.4%) and
by 0/9 o non-ST131 B2 isola es (p = 0.0007), while he numbe s
o CTX-M-14 we e 2/21 o ST131 and 5/9 o non-ST131 B2
isola es (9.5% s. 55.5%, p = 0.01). In compa ison wi h non-
ST131 B2 isola es, ST131 ha bou ed mo e equen ly a aA/B
(42.9% s. 0, p = 0.02), iha (66.7% s. 22.2%, p = 0.04), and sa
Vi ulence o Bac e emic ESBL-E. coli
PLOS ONE | www.plosone.o g 2 Sep embe 2012 | Volume 7 | Issue 9 | e44238
Table 1. Vi ulence ac o genes, ESBL g oups and an imic obial esis ance o 191 ESBL-p oducing E. coli isola es causing BSI
acco ding o phylog oups.
All isola es
(n = 191)
Phylog oup B2
(n = 30)
Phylog oup D
(n = 51)
Phylog oups A/B1
(n = 110)
Vi ulence sco e, median (IQR) 6 (4–8) 9 (8–13)
a
7 (6–9) 6 (2–7)
VF
papC 45 (24) 8 (27) 23 (45)
d
14 (13)
papGI 00 0 0
papGII 31 (16) 3 (10) 20 (39)
c,d
8 (7)
papGIII 7 (4) 4 (13)
a
1(2) 2(2)
s aD/E 4(2) 1(3) 1(2) 2(2)
a aB/C 13 (7) 9 (30)
a,b
4(8)
d
0
iha 30 (16) 16 (53)
a,b
10 (20)
d
4 (4)
imH 160 (84) 28 (93)
a
48 (94)
d
84 (76)
hlyA 4 (2) 3 (10)
a
1(2) 0
cn 1 4 (2) 3 (10)
a
1(2) 0
cd B 4 (2) 3 (10)
a
1(2) 0
sa 39 (20) 19 (63)
a,b
17 (33)
d
2 (3)
yuA 98 (51) 28 (93)
a,b
37 (73)
d
33 (30)
iu A 157 (82) 28 (93)
a
46 (90)
d
83 (76)
iucD 140 (73) 23 (77) 41 (80) 76 (69)
i oN 118 (62) 17 (57) 29 (57) 72 (66)
kps MTII 44 (23) 20 (67)
a,b
19 (37)
d
5 (5)
aT 141 (74) 28 (83) 37 (72) 79 (72)
c aC 66 (35) 7 (23) 16 (31) 43 (39)
ompT 102 (53) 27 (90)
a,b
32 (63)
d
43 (39)
ibeA 15 (8) 7 (23)
a
7 (14)
d
1 (1)
usp 33 (17) 26 (87)
a,b
7 (14)
d
3 (3)
maIX 75 (39) 28 (93)
a,b
34 (67)
d
13 (12)
s g 3 (2) 2 (7) 0 1 (1)
i eA 29 (15) 3 (10) 19 (37)
c,d
7 (6)
ESBL g oup
CTX-M-9 g oup
g
122 (64) 10 (33) 41 (80)
c,d
71 (65)
CTX-M-1 g oup
h
42 (22) 17 (57)
a,b
8 (16) 17 (16)
SHV g oup
i
33 (17) 4 (13) 5 (10) 24 (22)
TEM g oup 1 (1) 0 0 1 (1)
Resis ance o
Ce o axime 184 (96) 28 (93) 50 (98) 106 (96)
Ce azidime 70 (37) 14 (47)
b
12 (24) 44 (40)
e
Ce epime 124 (65) 19 (63) 35 (69) 70 (64)
Amoxicillin/cla ulanic acid 73 (38) 19 (63)
a,b
20 (39) 34 (31)
Pipe acillin/ azobac am 16 (8) 5 (17) 3 (6) 8 (7)
Cip o loxacin 129 (68) 18 (60) 36 (71) 75 (68)
Gen amycin 39 (20) 4 (13) 14 (28) 21 (19)
Tob amicin 34 (18) 13 (43)
a,b
10 (20) 11 (10)
Amikacin 3 (2) 2 (7)
a
1(2) 0
Co- imoxazole 115 (60) 19 (63) 38 (75)
d
58 (53)
Resis ance sco e, median (IQR) 5 (4–7) 6 (4–7)
a
5 (4–6) 5 (4–5)
a
Highe in B2 s A/B1 (p,0.05).
b
Highe in B2 s D (p,0.05).
c
Highe in D s B2 (p,0.05).
d
Highe in D s A/B1 (p,0.05).
e
Highe in A/B1 s D (p,0.05).
All o he compa isons, p.0.05.
Vi ulence o Bac e emic ESBL-E. coli
PLOS ONE | www.plosone.o g 3 Sep embe 2012 | Volume 7 | Issue 9 | e44238
(81% s. 22.2%, p = 0.004), and less equen ly papGII (0 s. 33.3,
p = 0.02) and i eA (0 s. 33.3%, p = 0.02).
The ea u es o he pa ien s acco ding o PG a e shown in
able 2. Isola es om PG B2 and D did no seem o be ela ed o
lowe equency o p edisposing ea u es o in asi e in ec ions
han isola es om PG A/B1. The only signi ican di e ence was
cance , which was less equen among pa ien s wi h B2 isola es
han among hose wi h A/B1. Also, he e we e no signi ican
di e ences in he epidemiological ea u es o sou ces o BSI. E en
when B2 and D isola es we e g ouped, he only signi ican
di e ence wi h A/B1 isola es was ha he o me mo e equen ly
occu ed in nu sing home esiden s (9/81 [11.1%] s 3/110
[2.7%], p = 0.03).
The associa ion o all 25 speci ic VF genes s udied wi h
p edisposing ac o s o BSI, ype o acquisi ion, p e ious
an ibio ic use, o sou ce o BSI was s udied. O e all, no associa ion
was ound (da a no shown) wi h 2 excep ions: papGII was mo e
equen in pa ien s wi hou any p edisposing ac o (local o
sys emic) han in pa ien s wi h hem (25% s 12%, p = 0.02), while
he opposi e occu ed wi h sa (25% s 40%, p = 0.03). We also
pe o med s a i ied analysis acco ding o sou ce. Among pa ien s
wi h a u ina y ac sou ce o BSI, hose wi hou any local o
sys emic p edisposing ea u e had isola es wi h a highe p e alence
o papC and papGII han hose wi h any p edisposing ac o (46% s
10%; p = 0.01, and 36% s 7%; p = 0.001, espec i ely). No
signi ican associa ions we e ound be ween VF and o he sou ces
o BSI.
The p o iles o VF genes we e ex emely di e se: he 191 s ains
showed 159 di e en p o iles, o hese 134 we e unique, 21 p o iles
appea ed wice, 2 appea ed h ee imes, one p o ile was epea ed
ou imes and ano he one appea ed i e imes. Such di e si y
p omp ed us o classi y hem in clus e s; 29 clus e s we e ound
using a 70% simila i y h eshold; 6 clus e s a bi a ily named C1–
C6 g ouped 128 isola es (67%). PGs, ESBLs, an imic obial
esis ance, and epidemiological da a acco ding o clus e a e
shown in Table 3; dis ibu ion o VF among he clus e s a e shown
in Figu e 1. In summa y, isola es om C1 caused in ec ions in
younge pa ien s; hose in C2 we e associa ed wi h highe
equency o u ina y ac sou ce; C4 isola es showed highe
equency o CTX-M-1 g oup o ESBLs, esis ance o amoxicillin/
cla ulanic acid and ob amycin, and bac e emia om sou ces
o he han u ina y o bilia y ac s; hose in C5 had he lowe
equency o local p edisposing ac o s; and C6 isola es showed less
equen esis ance o cip o loxacin. Six een o he 22 isola es
om C4 (73%) belonged o ST131; also, 76% isola es om ST131
belonged o C4.
To u he in es iga e he associa ion o C2 wi h u ina y ac
sou ce, mul i a ia e analysis we e pe o med. We in oduced he
ollowing a iables: age, gende , acquisi ion, local p edisposing
ac o , sys emic p edisposing ac o , PG, clus e , papGII, VF sco e,
ESBLs, and an imic obial esis ance sco e. C2 was independen ly
associa ed wi h u ina y ac sou ce a e con olling o age, local
and sys emic p edisposing ac o s, while PGs, papGII, VF sco e o
an imic obial esis ance we e no ( able 4). We did he same o
in es iga e he associa ion o C4 wi h sou ces o he han u ina y o
bilia y ac s. C4 was independen ly associa ed, while again PGs,
speci ic VF, VF sco e, and an imic obial esis ance sco e we e no
( able 4).
Finally, we analyzed he associa ion be ween di e en mic obi-
ological ea u es and absence o p edisposing sys emic and local
ea u es o BSI. In he uni a ia e analysis, clus e 5, papC, papGII,
sa , emale gende , lowe age, communi y, sou ce, and no eceip
o p e ious an imic obial use showed a p alue ,0.1 and we e
in oduced in he mul i a ia e analysis. The a iables selec ed as
independen p edic o s o BSI in non-p edisposed pa ien s we e
lowe age, communi y-acqui ed BSI, no eceip o p e ious
an imic obials, and clus e 5 ( able 5).
Discussion
Ou s udy showed ha he phylogene ic backg ound o
i ulence p o iles o ESBLEC causing BSI in Spain we e di e en
o wha would be expec ed o bac e emic E. coli. The ac ha
isola es om he so-called ‘‘low i ulen ’’ PGs A and B1
p edomina ed as caused o BSI is in con as wi h p e ious s udies
including mainly non-ESBL-p oducing isola es, in which B2 ad D
we e p edominan [3–8]. As a consequence, he p e alence o all
VF s udied was much lowe among ESBLEC isola es han among
p e ious collec ions o blood isola es o E. coli excep o iu A, i oN,
aT, and c aC [2–4,6,7,12]. The e a e, o ou knowledge, sca ce
p e ious da a on collec ions o blood ESBLEC isola es. In a s udy
om The Ne he lands including 41 ESBLEC blood isola es, only
22% o isola es belong o A o B1 PGs [15]. Simila o ou esul s,
PG A was p edominan in he subg oup o ESBL-p oduce s om
a F ench s udy; howe e , only 19 ESBLEC we e included [8].
Th ee ac s may explain ou esul s. Fi s , mos cases occu ed
in pa ien s wi h local o sys emic p edisposing ac o s o BSI;
hence, less i ulence ac o s would be equi ed o cause in asi e
in ec ion in such pa ien s. Second, p e ious an ibio ic ea men
was common, which would ha e selec ed o ESBLEC because o
hei mul id ug- esis an na u e ega dless hei i ulence p o ile.
Al hough an imic obial esis ance has been equen ly shown o be
mo e equen among isola es om he A and B1 PGs han among
B2 isola es [11], B2 isola es we e mo e equen ly esis an o
se e al an imic obials (pa icula ly amoxicillin-cla ulanic acid and
ob amycin) han isola es om o he PGs. This e lec s he
esis ance p o ile o isola es o ST131 p oducing CTX-M-15
[32,33], which comp ised mos B2 isola es in ou se ies. And hi d,
in a ecen s udy pe o med in F ance, non-ST131 B2 E. coli
isola es we e ound o a ely p oduce CTX-M enzymes [34]; his,
oge he wi h he ac ha ST131 was no p edominan in ou
se ies, would p o ide an addi ional explana ion o he low a e o
B2 isola es.
E en hough ESBLEC om he B2 and D PGs showed, as
expec ed, a much highe con en in VF, we did no ind B2 and D
isola es o ha e caused in ec ions in clea ly less p edisposed
pa ien s han A/B1 isola es, wi h he excep ion o cance (less
equen among B2). A ecen s udy on non-ESBL-p oducing E.
7 isola es p oduced .1 ESBL.
g
Mainy CTX-M-14.
h
Mainly CTX-M-15.
i
Mainly SHV-12.
Da a a e p esen ed as numbe o isola es (pe cen age) excep whe e speci ied.
doi:10.1371/jou nal.pone.0044238. 001
Table 1. Con .
Vi ulence o Bac e emic ESBL-E. coli
PLOS ONE | www.plosone.o g 4 Sep embe 2012 | Volume 7 | Issue 9 | e44238
coli ound ha B2 isola es we e p edominan as cause o
bac e emia and spon aneous pe i oni is in pa ien s wi h li e
ci hosis [35]; o no e, li e ci hosis was mo e equen among
pa ien wi h B2 isola es han hose wi h D o A/B1 isola es in ou
se ies, bu he di e ences did no each s a is ical signi icance.
Also, we ound ha PGs o speci ic VFs we e no independen ly
associa ed o sou ce o BSI. In p e ious s udies o E. coli
bac e emic isola es, hose om PG B2 had been associa ed wi h
u ina y ac sou ce o BSI [4,5,7]. As ega ds speci ic VF, se e al
s udies ha e in es iga ed hei associa ion wi h BSI sou ces; he
s udies we e di e en in popula ions, de ini ions, and VF s udied,
making i di icul o d aw clea conclusions [4,9,12]. Howe e ,
papGII has been mo e consis en ly associa ed wi h u ina y ac
sou ce in hese s udies. We did no ind such associa ion, al hough
papGII was mo e equen in c ude analysis among non-
p edisposed pa ien s wi h u ina y ac BSI.
O e all, hese esul s sugges ha hos ac o s and p e ious
an imic obial use we e mo e impo an han phylog oup back-
g ound, i ulence sco e o speci ic VF in he occu ence o BSI
due o ESBLEC. Howe e , by in es iga ing he exis ence o
clus e s o isola es acco ding o hei VF con en , we ound some
associa ions be ween i ulence backg ound and some speci ic
epidemiological ea u es. Thus, clus e C2 (mainly PG D, CTX-
M-14 p oduce s) was independen ly associa ed wi h u ina y ac
Table 2. Compa ison o p edisposing ea u es acco ding o phylog oup among 191 pa ien s wi h bac e emia due o ESBL-
p oducing E. coli.
All isola es
(n = 191)
Phylog oup B2
(n = 30)
Phylg oup D
(n = 51)
Phylog oups A/B1
(n = 110)
Age in yea s, median (IQR) 71 (55–78) 72 (58–82) 71 (58–78) 69 (54–77)
Male gende 107 (56) 20 (66.7) 28 (54.9) 59 (53.6)
Acquisi ion
Communi y 23 (12) 4 (13.3) 3 (5.9) 16 (14.5)
Heal hca e-associa ed 72 (37.6) 12 (40) 24 (47.0) 36 (32.7)
Nosocomial 96 (50.2) 14 (46.7) 24 (47.9) 58 (52.7)
Nu sing home esiden 12 (6.2) 2 (6.7) 7 (13.7) 3 (2.7)
Cha lson index, median (IQR) 2 (1–4) 2.5 (1–4) 2 (1–5) 2 (1–4)
Diabe es melli us 52 (27.2) 9 (30) 12 (23.5) 31 (28.2)
Ch onic pulmona y disease 34 (17.8) 4 (13.3) 9 (17.6) 21 (19.1)
Cance 55 (28.7) 4 (13.3)
a,b
15 (29.4) 36 (32.7)
Li e ci hosis 18 (9.4) 5 (16.7) 4 (7.8) 9 (8.2)
Ch onic enal insu iciency 28 (14.6) 3 (10) 5 (9.8) 20 (18.2)
Inmunosupp esi e he apy 27 (14.1) 4 (13.3) 9 (17.6) 14 (12.7)
Obs uc i e u ina y disease 43 (22.5) 7 (23.3) 7 (13.7)
c
29 (26.4)
Bilia y ac disease 18 (9.4) 2 (6.7) 3 (5.9) 13 (11.8)
Neu openia 10 (5.2) 1 (3.3) 3 (5.9) 6 (5.5)
U ina y ca he e 66 (34.5) 13 (43.3) 17 (33.3) 36 (32.7)
Cen al enous ca he e 53 (27.7) 5 (16.7) 12 (23.5) 36 (32.7)
Mechanical en ila ion 8 (4.1) 1 (3.3) 3 (5.9) 4 (3.6)
P e ious su ge y 44 (20.9) 6 (20) 14 (27.5) 24 (21.8)
P edisposing ac o , local 122 (63.8) 19 (63.3) 30 (58.8) 73 (66.4)
P edisposing ac o , sys emic 100 (52.3) 16 (53.3) 24 (47.1) 60 (54.5)
P edisposing ac o s, sys emic o local 158 (82.7) 24 (80) 41 (80.4) 93 (84.5)
P e ious an ibio ic use, any 107 (56) 18 (60) 28 (54.9) 61 (55.5)
Fluo oquinolones 49 (25.6) 7 (23.3) 12 (23.5) 30 (27.3)
Cephalospo o ins 53 (27.7) 9 (30) 15 (29.4) 29 (26.4)
Amoxicillin/cla ulanic acid 20 (10.4) 3 (10) 3 (5.9) 14 (12.7)
Sou ce
U ina y ac 90 (47.1) 12 (40) 25 (49) 53 (48.2)
Bilia y ac 24 (12.5) 3 (10) 5 (9.8) 16 (14.5)
O he s
d
77 (40.3) 15 (50) 21 (41.2) 41 (37.3)
a
P alue o B2 s A/B1 = 0.03.
b
P alue o B2 s D = 0.09.
c
P alue o D s A/B1 = 0.07.
All o he compa isons, P alue $0.1.
d
O he sou ces we e: unkown, 25 pa ien s; in aabdominal (non-bilia ), 24; espi a o y ac , 10; ca he e - ela ed, 9; miscellaneous, 8.
Da a a e exp essed as numbe o cases (%) excep whe e speci ied.
doi:10.1371/jou nal.pone.0044238. 002
Vi ulence o Bac e emic ESBL-E. coli
PLOS ONE | www.plosone.o g 5 Sep embe 2012 | Volume 7 | Issue 9 | e44238

BSI; C4 (mos ly B2 and ST131, CTX-M-15 p oduce s) wi h non-
u ina y o bilia y ac sou ces; and C5 (mos ly D and CTX-M-14
p oduce s) wi h BSI in non-p edisposed pa ien s. All hese clus e s
had mode a e o high i ulence sco es. The classi ica ion o isola es
in o clus e s acco ding o VF had been p e iously ca ied ou by
Johnson e al. acco ding o clonal g oups [36]; howe e , we
cons uc ed he clus e s by conside ing exclusi ely he VF con en
o he isola es and wi hou aking in o accoun nei he he
phylog oups no any o he clonal ela ionship among isola es
because ou aim was o speci ically in es iga e he in luence o FV
Figu e 1. Dis ibu ion o i ulence ac o s acco ding o clus e s. Pe cen age o isola es: whi e: 0–25%; pale g ey, 26–50%; da k g ey, 51–75%;
black, .75%.
doi:10.1371/jou nal.pone.0044238.g001
Vi ulence o Bac e emic ESBL-E. coli
PLOS ONE | www.plosone.o g 6 Sep embe 2012 | Volume 7 | Issue 9 | e44238
con en by i sel in he epidemiology o he in ec ions. Hypo he ical
implica ions om ou da a a e ha accines de eloped agains
speci ic VFs migh no be e icacious in a oiding in asi e in ec ions
due o ESBLEC in p edisposed pa ien s, and ha educing he
an ibio ic p essu e in such pa ien s migh be a mo e impo an
measu e o y and educe such in ec ions in hese pa ien s.
S eng hs o ou s udy include i s mul icen e na u e, clinical
da a a e comp ehensi e and we e p ospec i ely collec ed, and
isola es a e well cha ac e ized. Howe e , i has some limi a ions:
we could no compa e he ESBLEC p o iles wi h a con ol g oup
o non-ESBL p oduce s and hus used collec ions om o he
s udies as a e e ence; we s udied he genes codi ying o VF, bu
Table 3. Phylog oups, i ulence sco e, ESBLs, selec ed an imic obial esis ance, and associa ed pa ien s’ ea u es o ESBL-
p oducing E. coli isola es causing BSI acco ding o i ulence p o ile clus e s.
Clus e C1
(n = 55)
Clus e C2
(n = 11)
Clus e C3
(n = 8)
Clus e C4
(n = 22)
Clus e C5
(n = 25)
Clus e C6
(n = 7)
O he isola es
(n = 63)
Phylog oups
B2 0* 0* 25 90.9* 0* 57.1 6.3
D 10.9 54.4 75* 9.1 72* 42.9 15.9
A 23.6 18.2 0* 0* 16 0* 57.1
B1 65.5* 27.3 0* 0* 12 0* 20.6
Vi ulence sco e, median (IQR) 5 (5–6)
*
8 (7–9)* 6 (6–7) 10 (10–12)* 8 (8–9)* 14 (12–15)* 4 (2–5)
ESBL g oups
CTX-M-1 g oup 16.4 18.2 50 68.2* 8 28.6 12.7
CTX-M-9 g oup 70.9 72.7 62.5 27.3
*
84 57.1 61.9
SHV g oup 16.4 9.1 0 4.5 16 14.3 27
An imic obial esis ance
Cip o loxacin 72.7 45.5 50 81.8 64 28.6
*
69.8
Amoxicillin-cla ulanic acid 29.1 45.5 37.5 68.2
*
32 57.1 34.0
Tob amycin 7.3 18.2 12.5 54.5
*
0
*
14.3 22.2
Pa ien s’ ea u es
Age in yea s, median (IQR) 68 (50–76)
*
73 (66–78) 76 (70–83) 72 (57–79) 71 (52–77) 80 (58–84) 70 (58–80)
Male gende 58.2 54.5 50 63.6 48 42.9 57.1
Communi y-onse 50.9 45.5 62.5 50 56 28.6 47.6
Nu sing home esiden 1.8 9.1 12.5 13.6 12 0 4.8
Median Cha lson index (IQR) 2 (2–5) 2 (1–5) 2 (1–4) 2 (2–4) 3 (2–4) 3 (1–3) 2 (1–4)
P edisposing ac o , local 72.7 72.7 75 59.1 44
*
85.7 60.3
P edisposing ac o , sys emic 60 37.3 50 50 40 71.4 54
U ina y ac sou ce 52.7 81.8
*
37.5 40.9 48 28.6 41.3
Bilia y ac sou ce 7.3 0 25 0 12 28.6 20.6
O he sou ces 40 18.2 37.5 59.1
*
40 42.9 38.1
*P alues ,0.05 in compa ison wi h isola es no included in he clus e .
Da a a e p esen ed as pe cen age o isola es in each clus e excep whe e speci ied.
doi:10.1371/jou nal.pone.0044238. 003
Table 4. Mul i a ia e analysis o a iables associa ed wi h
speci ic sou ces o bloods eam in ec ion.
OR (95% CI) P
U ina y ac sou ce
Age (pe yea ) 1.02 (1.00–1.04) 0.009
Local p edisposing ac o 2.10 (1.11–3.98) 0.002
Sys emic p edisposing ac o 0.57 (0.31–1.04) 0.07
Clus e C2 5.05 (0.96–26.48) 0.05
Non u ina y o bilia y ac sou ces
Age (pe yea ) 0.96 (0.94–0.98) 0.001
Local p edisposing ac o 0.27 (0.14–0.54) ,0.001
Sys emic p edisposing ac o 2.75 (1.41–5.36) 0.003
Clus e C4 2.89 (1.05–7.93) 0.03
doi:10.1371/jou nal.pone.0044238. 004
Table 5. Mul i a ia e analysis o a iables associa ed wi h
absence o local o sys emic p edisposing condi ions.
OR (95% CI) P
Absence o sys emic and local p edisposing ac o s
Age (pe yea ) 0.97 (0.94–1.00) 0.04
Communi y-onse BSI 2.62 (1.02–6.76) 0.04
No p e ious an ibio ics 5.69 (2.24–14.45) ,0.001
Clus e 5 2.80 (0.99–7.93) 0.05
doi:10.1371/jou nal.pone.0044238. 005
Vi ulence o Bac e emic ESBL-E. coli
PLOS ONE | www.plosone.o g 7 Sep embe 2012 | Volume 7 | Issue 9 | e44238
his does no necessa ily e lec he exp ession o hese VF du ing
in ec ion; and esul s migh no be applicable o a eas wi h a
di e en epidemiology o ESBLEC.
In conclusion, bac e emic ESBLEC mo e equen ly belonged
o PGs A and B1 and hus had a lowe i ulen con en han
expec ed; nei he PGs o speci ic VF we e consis en ly associa ed
wi h p edisposing ea u es o sou ces o BSI; and some clus e s o
isola es iden i ied acco ding o hei i ulence p o ile we e
iden i ied and associa ed wi h speci ic sou ce o acquisi ion o
BSI in he absence o p edisposing ac o s.
Acknowledgmen s
O he pa icipan s om he ESBL-REIPI/GEIH g oup a e: Paloma Gijo´n
(Hospi al Uni e si a io G ego io Ma an˜o´n, Mad id, Spain), Jose´ Ramo´n
He na´ndez (Hospi al Uni e si a io Vi gen Maca ena, Se illa, Spain), Jose
M. Cisne os (Hospi al Uni e si a io Vi gen del Rocı
´o, Se illa, Spain),
Ca men Pen˜a (Hospi al Uni e si a io de Bell i ge, Ba celona, Spain),
Manuel Almela (Hospi al Clinic, Ba celona, Spain), Beni o Almi an e
(Hospi al Uni e si a io Vall d’Heb o´n, Ba celona, Spain), Fabio G ill
(Hospi al Uni e si a io Ramo´n y Cajal, Mad id; p esen add ess, Hospi al
Uni e si a io La Paz, Mad id, Spain), Ja ie Colomina (Hospi al de la
Ribe a, Alzi a, Valencia, Spain), Monse a Gime´nez (Hospi al Ge mans
T ias i Pujol, Badalona, Spain), An onio Oli e (Hospi al Son Espases,
Palma de Mallo ca, Spain), Juan Pablo Ho cajada (Hospi al Uni e si a io
Ma que´s de Valdecilla, San ande ; p esen add ess, Hospi al del Ma ,
Ba celona, Spain), Gemma Na a o (Co po acio Sani a ia Pa c Taulı
´,
Sabadell, Spain), Ana Coloma (Hospi al San a C eu i San Pau, Ba celona,
Spain).
Au ho Con ibu ions
Concei ed and designed he expe imen s: JRB JM AP. Pe o med he
expe imen s: NFR LS LLC. Analyzed he da a: JRB JM LLC AP.
Con ibu ed eagen s/ma e ials/analysis ools: JRB NFR LS LLC. W o e
he pape : JRB JM AP.
Re e ences
1. Pica d B, Se ali J, Gou iou S, Du iez P, B ahimi N, e al. (1999) The link
be ween phylogeny and i ulence in Esche ichia coli ex ain es inal in ec ion.
In ec Immun 67: 546–553.
2. Johnson JR, S ell AL (2000) Ex ended i ulence geno ypes o Esche ichia coli
s ains om pa ien s wi h u osepsis in ela ion o phylogeny and hos
comp omise. J In ec Dis 181: 261–272.
3. Sannes MR, Kuskowski MA, Owens K, Gajewski A, Johnson JR (2004)
Vi ulence ac o p o iles and phylogene ic backg ound o Esche ichia coli isola es
om e e ans wi h bac e emia and unin ec ed con ol subjec s. J In ec Dis 190:
2121–2128.
4. Mo eno E, Planells I, P a s G, Planes AM, Mo eno G, And eu A (2005)
Compa a i e s udy o Esche ichia coli i ulence de e minan s in s ains causing
bac e emia e sus s ains causing pyeloneph i is and o he sou ces o
bac e emia. Diag Mic obiol In ec Dis 53: 93–99.
5. Bukh AS, Schønheyde HC, Emme sen JM, Søgaa d M, Bas holm S, e al.
(2009) Esche ichia coli phylogene ic g oups a e associa ed wi h si e o in ec ion and
le el o an ibio ic esis ance in communi y-acqui ed bac e aemia: a 10 yea
popula ion-based s udy in Denma k. J An imic ob Chemo he 64: 163–168.
6. Cooke NM, Smi h SG, Kellehe M, Roge s TR (2010) Majo di e ences in
equencies o i ulence ac o s and mul id ug esis ance be ween communi y
and nosocomial Esche ichia coli bloods eam isola es. J Clin Mic obiol 48: 1099–
1104.
7. Le o A, Panha d X, Cle mon O, Woe he P-L, B ange C, e al. (2011) Hos
ac o s and po al o en y ou weigh bac e ial de e minan s o p edic he
se e i y o Esche ichia coli bac e emia. J Clin Mic obiol 49: 777–783.
8. Cou pon-Claudinon A, Le o A, Pahna d X, Cle mond O, Do nic Q, e al.
(2011) Bac e aemia caused by hi d-gene a ion cephalospo in- esis an Esche-
ichia coli in F ance: p e alence molecula epidemiology, and clinical ea u es.
Clin Mic obiol In ec 17: 557–565.
9. Johnson JR, Kuskowski MA, O’B yan TO, Maslow JN (2002) Epidemiological
co ela es o i ulence geno ype and phylogene ic backg ound among Esche ichia
coli blood isola es om adul s wi h di e se-sou ce bac e emia. J In ec Dis 185:
1439–1447.
10. Mo eno E, P a s G, Planells I, Planes AM, Pe´ ez T, e al. (2006)
Cha ac e iza ion o Esche ichia coli isola es de i ed om phylogene ic g oups A
and B1 causing ex ain es inal in ec ion. En e m In ecc Mic obiol Clin 24: 483–
489.
11. Johnson JR, Russo TA (2005) Molecula epidemiology o ex ain es nal
pa hogenic (u opa hogenic) Esche ichia coli. In J Med Mic obiol 295: 383–404.
12. Wang MC, Tseng CC, Chen CY, Wu JJ, Huang JJ (2002) The ole o bac e ial
i ulence and hos ac o s in pa ien s wi h Esche ichia coli bac e emia who ha e
acu e cholangi is o uppe u ina y ac in ec ion. Clin In ec Dis 35: 1161–1166.
13. Rod ı
´guez-Ban˜o J, Pico´n E, Gijo´n P, He na´ndez JR, Ruı
´z M, e al. (2010)
Communi y-onse bac e emia due o ex ended-spec um be a-lac amase-p o-
ducing Esche ichia coli: isk ac o s and p ognosis. Clin In ec Dis 50: 40–48.
14. Rod ı
´guez-Ban˜o J, Pico´n E, Gijo´n P, He na´ndez JR, Cisne os JM, e al. (2010)
Risk ac o s and p ognosis o nosocomial bloods eam in ec ions caused by
ex ended-spec um b-lac amase-p oducing Esche ichia coli. J Clin Mic obiol 48:
1726–1731.
15. an de Bij AK, Pei ano G, Goessens WH, an de Vo m ER, an Wes eenen
M, e al. (2011) Clinical and molecula cha ac e is ics o ex ended-spec um-
be a-lac amase-p oducing Esche ichia coli causing bac e emia in he Ro e dam
A ea, Ne he lands. An imic ob Agen s Chemo he 55: 3576–3578.
16. La igne JP, Blanc-Po a d AB, Bou g G, Mo eau J, Chanal C, e al. (2006)
Vi ulence geno ype and nema ode-killing p ope ies o ex a-in es inal Esche ichia
coli p oducing CTX-M be a-lac amases. Clin Mic obiol In ec 12: 1199–1206.
17. Cle mon O, La ollay M, Vimon S, Deschamps C, Fo es ie C, e al. (2008)
The CTX-M-15-p oducing Esche ichia coli di using clone belongs o a highly
i ulen B2 phylogene ic subg oup. J An imic ob Chemo he 61: 1024–1028.
18. La igne JP, Ve guns AC, Go e L, So o A, Combescu e C, e al. (2012)
Vi ulence po en ial and genomic mapping o he wo ldwide clone Esche ichia coli
ST131. PLoS One 7: e34294.
19. Cha lson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new me hod o
classi ying p ognos ic co-mo bidi y in longi udinal s udies: de elopmen and
alida ion. J Ch on Dis 40: 373–383.
20. Clinical and Labo a o y S anda d Ins i u e (CLSI) (2009) Pe o mance
S anda ds o An imic obial Suscep ibili y Tes ing; 19 h In o ma ional Supple-
men . CLSI documen M100-S19. CLSI, Wayne, PA, USA.
21. Cle mon O, Dhanji H, Up on M, Gib eel T, Fox A, e al. (2009) Rapid
de ec ion o he O25b-ST131 clone o Esche ichia coli encompassing he CTX-M-
15-p oducing s ains. J An imic ob Chemo he 64: 274–277.
22. Cle mon O, Bonaco si S, Bingen E (2000) Rapid and simple de e mina ion o
he Esche ichia coli phylogene ic g oup. Appl En i on Mic obiol 66: 4555–4558.
23. Le Bouguenec C, A chambaud M, Labigne A (1992) Rapid and speci ic
de ec ion o he pap, a a, and s a adhesin-encoding ope ons in u opa hogenic
Esche ichia coli s ains by polyme ase chain eac ion. J Clin Mic obiol 30: 1189–
1193.
24. Ka kkainen UM, Kauppinen J, Ikaheimo R, Ka ila ML, Sii onen A (1998)
Rapid and speci ic de ec ion o h ee di e en G adhesin classes o P- imb iae in
u opa hogenic Esche ichia coli by polyme ase chain eac ion. J Mic obiol Me hods
34: 23–29.
25. Blanco M, Blanco JE, Alonso MP, Mo a A, Balsalob e C, e al. (1997) De ec ion
o pap, s a and a a adhesin-encoding ope ons in u opa hogenic Esche ichia coli
s ains: ela ionship wi h exp ession o adhesins and p oduc ion o oxins. Res
Mic obiol 148: 745–755.
26. Yamamo o S, Teai A, Yu i K, Ku azono H, Takeda Y, e al. (1995) De ec ion o
u o i ulence ac o s in Esche ichia coli by mul iplex polyme ase chain eac ion.
FEMS Immunol Med Mic obiol 12: 85–90.
27. Ananias M, Yano T (2008) Se og oups and i ulence geno ypes o Esche ichia coli
isola ed om pa ien s wi h sepsis. B az J Med Biol Res 41: 877–883.
28. Nakano M, Yamamo o S, Te ai A, Ogawa O, Makino SI, e al. (2001)
S uc u al and sequence di e si y o he pa hogenici y island o u opa hogenic
Esche ichia coli which encodes he USP p o ein. FEMS Mic obiol Le 205: 71–76.
29. Russo TA, Ca lino UB, Johnson JR (2001) Iden i ica ion o a new i on- egula ed
i ulence gene, i eA, in an ex ain es inal pa hogenic isola e o Esche ichia coli.
In ec Immun 69: 6209–6216.
30. Gi a deau JP, Lalioui L, Said AM, De Champs C, Le Bouqie´nec C (2003)
Ex ended i ulence geno ype o pa hogenic Esche ichia coliiIsola es ca ying he
a a-8 ope on: e idence o simila i ies be ween isola es om humans and animals
wi h ex ain es inal in ec ions. J Clin Mic obiol 41: 218–226.
31. Bide P, Me ais A, Mahjoub-Messai F, Du and L, Dehem M, e al. (2007)
De ec ion and iden i ica ion by PCR o a highly i ulen phylogene ic subg oup
among ex ain es inal pa hogenic Esche ichia coli B2 s ains. Appl En i on
Mic obiol 72: 2373–2377.
32. Nicolas-Chanoine MH, Blanco J, Le lon-Guibou V, Dema y R, Alonso MP, e
al. (2008) In e con inen al eme gence o Esche ichia coli clone O25:H4-ST131
p oducing CTX-M-15. J An imic ob Chemo he 61: 271–281.
33. Pi ou JDD, G egson DG, Campbell L, Laupland KB (2009) Molecula
cha ac e is ics o ex ended-spec um-b-lac amase-p oducing Esche ichia coli
isola es causing bac e emia in he Calga y Heal h Region om 2000 o 2007:
eme gence o clone ST131 as a cause o communi y-acqui ed in ec ions.
An imic ob Agen s Chemo he 53: 2846–2851.
34. B isse S, Diancou L, Laoue´nan C, Vigan M, Ca o V, e al. (2012) Phylogene ic
dis ibu ion o CTX-M and non-ESBL p oducing Esche ichia coli isola es: g oup
Vi ulence o Bac e emic ESBL-E. coli
PLOS ONE | www.plosone.o g 8 Sep embe 2012 | Volume 7 | Issue 9 | e44238
B2 isola es, excep clone ST131, a ely p oduce CTX-M enzymes. J Clin
Mic obiol (in p ess).
35. Be F, Johnson JR, Oua a a B, Le lon-Guibou V, Johns on B, e al (2010)
Gene ic di e si y and i ulence p o iles o Esche ichia coli isola es causing
spon aneous bac e ial pe i oni is and bac e emia in pa ien s wi h ci hosis. J Clin
Mic obiol 48: 2709–2714.
36. Johnson JR, O’B yan TT, Kuskowski M, Maslow JN (2001) Ongoing ho izon al
and e ical ansmission o i ulence genes and papA alleles among Esche ichia coli
blood isola es om pa ien s wi h di e se sou ce bac e emia. In ec Immun 9:
5363–5374.
Vi ulence o Bac e emic ESBL-E. coli
PLOS ONE | www.plosone.o g 9 Sep embe 2012 | Volume 7 | Issue 9 | e44238