Eme ging Mic obes & In ec ions
ISSN: 2222-1751 (Online) Jou nal homepage: www. and online.com/jou nals/ emi20
Real- ime moni o ing o Pseudomonas ae uginosa
biofilm g ow h dynamics and pe sis e cells’
e adica ion
Miglė Žiemy ė, Miguel Ca da-Diéguez, Juan C. Rod íguez-Díaz, Ma ia P.
Ven e o, Alex Mi a & Ma ía D. Fe e
To ci e his a icle: Miglė Žiemy ė, Miguel Ca da-Diéguez, Juan C. Rod íguez-Díaz, Ma ia P.
Ven e o, Alex Mi a & Ma ía D. Fe e (2021) Real- ime moni o ing o Pseudomonas ae uginosa
biofilm g ow h dynamics and pe sis e cells’ e adica ion, Eme ging Mic obes & In ec ions, 10:1,
2062-2075, DOI: 10.1080/22221751.2021.1994355
To link o his a icle: h ps://doi.o g/10.1080/22221751.2021.1994355
© 2021 The Au ho (s). Published by In o ma
UK Limi ed, ading as Taylo & F ancis
G oup, on behal o Shanghai Shangyixun
Cul u al Communica ion Co., L d
Published online: 10 No 2021.
Submi you a icle o his jou nal
A icle iews: 6428
View ela ed a icles
View C ossma k da a
Ci ing a icles: 20 View ci ing a icles
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= emi20
ORIGINAL ARTICLE
Real- ime moni o ing o Pseudomonas ae uginosa biofilm g ow h dynamics
and pe sis e cells’e adica ion
MiglėŽiemy ė
a
, Miguel Ca da-Diéguez
a
, Juan C. Rod íguez-Díaz
b
, Ma ia P. Ven e o
b
, Alex Mi a
a,
c
and Ma ía D. Fe e
a,c
a
Genomics & Heal h Depa men , FISABIO Founda ion, Valencia, Spain;
b
Se icio de Mic obiología, Hospi al Gene al Uni e si a io de
Alican e, ISABIAL, Alican e, Spain;
c
CIBER o Epidemiology and Public Heal h, Mad id, Spain
ABSTRACT
Biofilm o ma ion and he appea ance o pe sis e cells wi h low me abolic a es a e key ac o s affec ing con en ional
ea men ailu e and an ibio ic esis ance. Using impedance-based measu emen s, c ys al iole s aining and adi ional
cul u e we ha e s udied he biofilm g ow h dynamics o 13 Pseudomonas ae uginosa s ains unde he effec o se en
con en ional an ibio ics. Real- ime g ow h quan ifica ions e ealed ha he exposu e o es ablished P. ae uginosa
biofilms o ce ain concen a ions o cip ofloxacin, ce azidime and ob amycin induced he eme gence o pe sis e
cells, ha showed diffe en mo phology and pigmen a ion, as well inc eased an ibio ic esis ance. Whole-genome
sequencing o wild ype and pe sis e cells iden ified se e al SNPs, a genomic in e sion and a genomic duplica ion in
one o he s ains. Howe e , hese mu a ions we e no uniquely associa ed wi h pe sis e s, sugges ing ha he
pe sis en pheno ype may be ela ed o me abolic and ansc ip ional changes. Gi en ha manni ol has been
p oposed o ac i a e bac e ial me abolism, he syne gis ic combina ion o manni ol and cip ofloxacin was e alua ed
on clinical 48 h P. ae uginosa biofilms. When adminis e ed a doses ≥320 mg/L, manni ol was capable o p e en ing
pe sis e cell o ma ion by efficien ly ac i a ing do man bac e ia and making hem suscep ible o he an ibio ic.
These esul s we e confi med using iable colony coun ing. As he es ed cip ofloxacin-manni ol combina ion
appea ed o ully e adica e ma u e biofilms, we conclude ha impedance-based biofilm diagnos ics, which pe mi s
an ibio ic suscep ibili y es ing and he iden ifica ion o pe sis e cells, is o g ea po en ial o he clinical p ac ice
and could aid in es ablishing ea men b eakpoin s o eme ging biofilm- ela ed in ec ions.
ARTICLE HISTORY Recei ed 8 July 2021; Re ised 16 Sep embe 2021; Accep ed 12 Oc obe 2021
KEYWORDS P. ae uginosa; xCELLigence; biofilms; manni ol; pe sis e cells; an ibio ic esis ance; cip ofloxacin
In oduc ion
Biofilms can be desc ibed as bac e ial communi ies
immobilized in a sel -sec e ed biopolyme ic ma ix.
This ma ix is mainly composed o mac omolecules
including DNA, p o eins and polysaccha ides and
has a e y selec i e pe meabili y o nu ien s and an i-
mic obial compounds [1,2]. Thus, biofilms ha e
al eady been desc ibed o be up o 1000 imes mo e
esis an o an ibio ics compa ed o hei plank onic
o ms and exhibi a la ge h ea o human heal h [3].
Al hough bac e ial biofilms ha e been deeply in es i-
ga ed du ing he pas decades, a majo limi a ion is
ha mos exis ing me hods o biofilm esea ch p o-
ide in o ma ion a a single endpoin , he e o e miss-
ing ele an ea u es abou he g ow h and pe sis ence
dynamics [4–6]. In addi ion, classical biofilm mass
quan ifica ion usually p o ides high s anda d de i-
a ions, and can esul in lack o ep oducibili y due
o manipula ion s eps. Fo his eason, me hods o
con inued moni o ing o biofilm g ow h whe e he
biofilm does no need o be labelled o manipula ed
ha e been p oposed as a powe ul al e na i e o a-
di ional s aining p ocedu es [7,8].
Pseudomonas ae uginosa is an oppo unis ic pa ho-
genic bac e ium ha causes bo h acu e and ch onic
in ec ions in se e e wounds, u ina y ac , o in
pa ien s unde going chemo he apy o any o he medi-
cal condi ion ela ed o a weakened immune sys em
like cys ic fib osis [9,10]. The pa hogenic success o
his bac e ium is closely ela ed o quo um sensing
sys ems and he abili y o syn hesize diffe en me ab-
oli es and i ulence ac o s such as pyo e din, exo-
oxin A, phospholipase C and elas ase [11–13].
Mo eo e , P. ae uginosa in ec ions a e ex emely
difficul o ea , as his bac e ium is esis an o
many an imic obial compounds and can easily su i e
on abio ic and bio ic su aces such as medical equip-
men e en a e disin ec ion [14]. In addi ion,
biofilm o ma ion capaci y o P. ae uginosa con ib-
u es eno mously o in ec ion de elopmen because
biofilms p o ec his bac e ium om immune sys em
a ack and con en ional an ibio ic ea men [15–19].
© 2021 The Au ho (s). Published by In o ma UK Limi ed, ading as Taylo & F ancis G oup, on behal o Shanghai Shangyixun Cul u al Communica ion Co., L d
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.
CONTACT Alex Mi a [email p o ec ed] Genomics & Heal h Depa men , FISABIO Founda ion, A da. Ca aluña 21, 46020 Valencia, Spain
Supplemen al da a o his a icle can be accessed a h ps://doi.o g/10.1080/22221751.2021.1994355.
Eme ging Mic obes & In ec ions
2021, VOL. 10
h ps://doi.o g/10.1080/22221751.2021.1994355
Ano he impedimen o ea Pseudomonas- ela ed
biofilm in ec ions is he eme gence o pe sis e cells,
ha a e cha ac e ized as subpopula ions exhibi ing a
non-di iding s a e, educed ansla ion, low p o on
mo ing o ce and dec eased ATP le els [20–22]. The
appea ance o hese cells can be igge ed by con en-
ional an ibio ics o en i onmen al s ess, including
spa ial and nu ien cons ains wi hin bac e ial
biofilms [20,23]. P e ious epo s ha e sugges ed
ha pe sis e cells exhibi a diffe en pheno ype com-
pa ed o wild ype (w ) s ains including smalle colony
size o changes in pigmen a ion [24–26]. Gi en ha
pe sis e cells can en e a non-g owing s a e, ole a e
high concen a ions o bac e icidal an ibio ic ( o
example by ac i a ing efflux pumps) and eg ow
once he ea men is ceased, hese cells ha e al eady
been linked o he elapse and ecalci ance o ch onic
in ec ions [25,27,28]. Al hough a ious s a egies o
comba pe sis e cells such as u iliza ion o DNA
c osslinking agen s, use o colis in-based an ibio ic
combina ions o addi ion o glycolysis in e media es
ha e been sugges ed, he e a e s ill many limi a ions
o s udy and comba he eme gence o pe sis e cells
wi hin biofilms [29,30].
Thus, he objec i e o he cu en wo k was o s udy
P. ae uginosa biofilm o ma ion capaci y and g ow h
dynamics, in o de o gain insigh s on pe sis e cell
o ma ion and e adica ion. We es ed he biofilm o -
ma ion capaci y o 10 clinical isola es, wo e e ence
s ains, and he de i a i e mu an PaO1▵ hl▵las by
eal- ime impedance measu emen s and con en ional
biofilm s aining me hodology. Fu he , we desc ibed
he in i o he apeu ic efficacy o se en con en ional
an ibio ics on biofilm o ma ion and de elopmen as
well as he effec o ou o hem on biofilm e adica ion
once he biofilm was al eady o med. Using Real-Time
impedance moni o ing, we we e able o iden i y and
s udy pe sis e subpopula ions wi hin ma u e
P. ae uginosa biofilms ea ed wi h subinhibi o y
cip ofloxacin concen a ions, and o e alua e he
effec o manni ol alone and in combina ion wi h
cip ofloxacin o e adica e he biofilms. Finally, we
assessed gene ic and genomic diffe ences be ween do -
man P. ae uginosa cells ea ed wi h cip ofloxacin and
con ols, in o de o iden i y po en ial mu a ions
associa ed wi h pe sis ence.
Ma e ials and me hods
Bac e ial s ains and g ow h condi ions
Clinically ele an P. ae uginosa s ains we e isola ed
a he Mic obiology Depa men o he Alican e Gen-
e al Hospi al (Spain). Model s ains ATTC27853 and
PaO1, and he de i a e s ain PaO1▵ hl▵las
(PaO1▵) wi h a educed abili y o o m biofilms,
we e also used (Table S1). PaO1 was conside ed as
s ong, ATCC2753 mode a e and PaO1▵weak
biofilm- o ming s ains, espec i ely. P. ae uginosa
s ains we e cul u ed on T yp ic Soy Aga (TSA),
Lysogeny B o h Aga (LBA) o B ain Hea In usion
(BHI) aga pla es a 37°C o 24 h and hen inocula ed
in T yp ic Soy B o h (TSB), Lysogeny B o h (LB) o
B ain Hea In usion (BHI) media a 37°C wi h igo -
ous o bi al shaking (120 pm).
Real- ime biofilm g ow h assay
Real- ime biofilm g ow h moni o ing expe imen s
we e pe o med using xCELLigence RTCA SP equip-
men (Agilen ) acco ding o manu ac u e ’s ins uc-
ions as p e iously desc ibed [8,31].
To de e mine he mos sui able condi ions o
P. ae uginosa biofilm g ow h in his sys em, BHI, LB
and TSB media supplemen ed wi h o wi hou 0.5%
o 1% glucose we e es ed. Fi s ly, 100 µl o each med-
ium we e used o backg ound measu emen s. A e
ha , o e nigh cul u es o P. ae uginosa we e dilu ed
wi h BHI, LB and TSB g ow h media supplemen ed
wi h o wi hou 0.5 o 1% o glucose and 100 µl o
hese bac e ial cell suspensions we e added in o E-
pla e wells, eaching a final OD
600
=0.15 and 0.3,
espec i ely. Biofilm g ow h was hen measu ed
e e y 10 min a 37°C o 72 h.
To e alua e an ibio ic efficacy o inhibi biofilm o -
ma ion, se en an ibio ics commonly used in clinical
p ac ice we e es ed: ce azidime (CAZ, No mon),
cip ofloxacin (CIP, Ke n Pha ma), colis in (CST,
Acco d), ob amycin (TOB, B own), imipenem (IPM,
F esenius Kabi), me openem (MEM, Ke n Pha ma)
and pipe acillin- azobac am (TZP, Acco d). Biofilm
o ma ion capaci y in p esence o he an ibio ics was
moni o ed o 72 h, in he o m o Cell Index (CI)
alues, which co ela e wi h o al biofilm mass [32,33].
Fo ea men o ma u e biofilms, an ibio ics wi h
highes an imic obial ac i i y in p e en ion o biofilm
o ma ion (TZP, CAZ, CIP and TOB) we e es ed. In
sho , 100 µl o bac e ial suspension (OD
600
=0.2625)
we e used o backg ound measu emen s. A e ha ,
75 µl o LB we e added ( eaching he final OD
600
=0.15)
and biofilms we e g own o 24 h (PaO1) and 48 h
(MF120) espec i ely,dependingon hebiofilmg ow h
a e o each s ain. A his ime, co esponding o he
highes CIs o bo h PaO1 and MF120, 25 µl o he co -
esponding an ibio ic we e added, eaching final con-
cen a ions om 128 o 0.0625 mg/L. Biofilm
inhibi ion/e adica ion capaci y o he an ibio ics was
hen quan ified o u he 120 h.
C ys al iole s aining
Biofilm o ma ion assay in mic o i e pla es o c ys al
iole s aining was pe o med as p e iously desc ibed
[5]. B iefly, o e nigh cul u es o P. ae uginosa s ains
EMERGING MICROBES & INFECTIONS 2063
we e dilu ed in LB medium o OD
600
=0.1 and 300 µl o
bac e ial suspensions we e ans e ed in o he co e-
sponding wells o 96 well fla -bo om pla es 89626
(Ibidi, Ge many). Biofilms we e g own o 8, 24, 30,
48, 56 and 72 h a 37°C. A e ha , he cul u e supe -
na an was disca ded, cells we e insed using Phos-
pha e Buffe Saline (PBS, pH=7.4) and a ached
biomass was s ained wi h 0.1% c ys al iole (CV)
o 20 mins. Subsequen ly, CV was emo ed,
biofilm-embedded cells we e insed wi h PBS in
o de o emo e una ached cells and esidual CV
and esuspended wi h 300 µl o 96.5% / E OH.
A e ha , op ical densi y o eleased CV was
measu ed by means o he abso bance pla e eade
Infini e M200 (Tecan, Du ham NC) a 610 nm.
Effec o manni ol on plank onic P. ae uginosa
g ow h
The o e nigh cul u es o P. ae uginosa isola es PaO1
and MF120 we e dilu ed in LB medium o OD
600
=0.1
and 100 µl o hese suspensions we e added in o co e-
sponding wells o 96-well mic opla es (The mo Fishe
Scien ific). A e ha , 100 µl o manni ol we e added
in o he co esponding wells eaching final concen-
a ions o 3200, 320, 64, 32, 16, 8, 4, 2, 1, 0,5, 0.25,
0.125 and 0.0625 mg/L. The ea e , mic opla es we e
incuba ed a 37°C wi h o bi al shaking a 120 pm
and he effec o manni ol on plank onic bac e ial
g ow h was e alua ed o 24 h by he means o an
abso bance pla e eade Infini e M200 (Tecan, Du -
ham NC).
MIC de e mina ion
The minimum inhibi o y concen a ions (MICs) o
CIP, CAZ, COL, TOB, IMI, MEM and TZP we e
assessed using he CLSI b o h mic odilu ion me hod
as p e iously desc ibed [34]. UMIC mic odilu ion
es was used o CST (Biocen ic, Bucke ) and E-
es s (Biomé ieux) we e used o he es o an ibio ics.
MBICs (minimum biofilm inhibi o y concen a ions)
we e calcula ed o PaO1, and o he clinical isola es
wi h he s onges biofilm o ma ion capaci y (i.e.
MF120 and MF124), conside ing he ime poin s
whe e an ibio ic- ee cell con ol eached he highes
CIs (24 h o PaO1, 48 h o MF120 and 72 h o
MF124, espec i ely), ollowing Fe e e al. [8].
Iden ifica ion o pe sis e s and eg ow h assay
To confi m he p esence o bac e ial pe sis en cells
and no CIP esis an mu an s, P. ae uginosa MF120
biofilms we e cul i a ed in he xCELLigence sys em
o 48 h as desc ibed abo e. A e ha , CIP was
added, eaching a final concen a ion o 0.25 mg/L.
A 140 h o biofilm g ow h, he expe imen was
s opped, and he biofilms exposed o CIP we e pla ed
on LBA pla es con aining 0.25 mg/L o CIP, in o de
o main ain he pe sis e pheno ype, while MF120
con ol cells we e pla ed on LBA pla es wi hou
addi ional an ibio ics.
Besides he changes in g ow h, he abili y o whi e
pe sis e cells (MF120 ea ed wi h CIP) o e e o
g eenish w pheno ype was in es iga ed by wo inde-
penden obse e s. Fi e pe sis e colonies and fi e
con ols wi hou an ibio ic we e selec ed and esus-
pended in 200 µL o LB media con aining manni ol
(3200 mg/L) and LB alone and g own a 37°C o
120h o obse e he swi ch om pe sis e popula ion
in o ac i ely g owing cells.
Effec o manni ol alone and in combina ion
wi h cip ofloxacin on P. ae uginosa pe sis e s
To desc ibe he effec o manni ol on P. ae uginosa
biofilm o ma ion, manni ol was se ially dilu ed in
LB medium eaching final concen a ions o 3200–
0.0625 mg/L. One hund ed mic oli e s o each
dilu ion in iplica e we e used o backg ound
measu emen s. A e ha , 100 µl o PaO1 and
MF120 bac e ial suspensions we e added in o he co -
esponding E-pla e wells eaching final bac e ial cell
op ical densi y o OD
600
= 0.15 and con inuous
biofilm g ow h was obse ed o 72h in xCELLigence
equipmen .
The abili y o manni ol o po en ia e he effec o
CIP and/o p e en pe sis e bac e ia o ma ion
wi hin P. ae uginosa MF120 biofilms was hen in es i-
ga ed using manni ol concen a ions o 3200, 320, 64,
32 and 16 mg/L. Fi s ly, manni ol was added oge he
wi h bac e ial inoculum (OD
600
=0.15). A e 48 h o
biofilm g ow h, 25 µl o CIP we e added (final concen-
a ion: 0.25 mg/L), and biofilm g ow h was obse ed
o addi ional 140 h. Simila expe imen s we e pe -
o med adding 25 µl o mixed suspensions o manni ol
and CIP on ma u e MF120 biofilms a 48 h. Biofilm
g ow h a e addi ion o bo h compounds was quan-
ified o 140 h.
Colony o ming uni (CFU) coun ing
To assess he numbe o iable cells a e biofilm ea -
men wi h manni ol, CIP o hei combina ion,
P. ae uginosa biofilms we e collec ed a 140 h o
g ow h in he RTCA sys em and sonica ed o 5 min
in o de o elimina e bac e ial agg ega es and dis up
ex acellula biofilm ma ix. A e sonifica ion, se ial
dilu ions we e p epa ed, pla ed in iplica es on LB
pla es and incuba ed a 37°C o e nigh . CFUs hen
we e coun ed, a e aged, and exp essed as log10
CFUs. Each expe imen included h ee echnical epli-
ca es and was epea ed h ee imes (biological epli-
ca es). S a is ical significance was assessed using
2064 M. ŽIEMYTĖET AL.
S uden ’s - es , whe e p- alue 0.05 was conside ed as
significan .
Whole genome-sequencing and bioin o ma ic
analysis
DNA om indi idual colonies g own in he p esence o
CIP (0.25 mg/L), manni ol (3200 mg/L) o hei combi-
na ion (exhibi ing non-iden ical pheno ypes) was
ex ac ed using MagNA Pu e LC DNA Isola ion Ki III
o Bac e ia and Fungi (Roche Diagnos ics) ollowing
manu ac u e ’s ins uc ions. Fo each colony ype, wo
colonies om he same aga pla e we e selec ed as epli-
ca es. A e ha , Illumina lib a ies we e cons uc ed a
he Sequencing Pla o m in FISABIO (Valencia, Spain)
and he genomes we e sequenced wi h Nex Seq echnol-
ogy (Illumina) using he 500/550 High Ou pu 75 cycles
Ki (single-ends 75 bp eads).
Resul an sequenced eads (75 bp long) we e
immed o low quali y eads (<20) and leng h
(<50) using p inseq [35]. SPAdes [36] was used o
genome assembly and ORFs we e de ec ed using p o-
digal [37] and anno a ed using hmmsea ch [38]. Fo
genomic mu a ions, he so wa e flex2 (h ps://
gi hub.com/asie za agoza/flex2) was used o isually
de ec genomic inse ions, duplica ions o in e sions.
Single-nucleo ide polymo phisms (SNPs) we e
de ec ed using he so wa e snippy and snippy-co e
h ough he Galaxy se e (h ps://gi hub.com/
seemann/snippy). The assembled con igs and he
o iginal eads ha e been deposi ed in SRA da abase
unde he accession numbe PRJNA753320.
S a is ical analysis
Biofilm inhibi ion/induc ion a e ea men wi h
diffe en an ibio ics we e conside ed o be significan ly
diffe en om con ols using linea models om he
lm lib a y in he R S a is ical Package e sion 1.0.1.7
wi h p- alue <0.05 (h ps://c an. -p ojec .o g/web/
packages/glmul i) (accessed in Janua y, 2021). S u-
den ’s - es s we e pe o med o e eal s a is ical
diffe ences be ween ea men s a e CFUs coun s.
p- alues < 0.05 we e conside ed significan .
Resul s
Influence o cul u e condi ions on P. ae uginosa
biofilms
Biofilm assembly can be influenced by many diffe en
ac o s including cul u ing condi ions and ini ial
inoculum size [8,31]. We in es iga ed he effec o
ini ial op ical densi y (OD) and cul u e g ow h med-
ium on bac e ia a achmen and biofilm o ma ion
on E-pla e su aces. Figu e S2 shows biofilm o ma ion
dynamics o model s ain PaO1 and clinical isola es
MF117, MF118 and MF124 when g own in TSB
media using ini ial ODs
600
o 0.15 and 0.3. Resul s
showed only sligh diffe ences in o al biofilm mass
(CIs) o all he es ed s ains. In addi ion, we es ed
how biofilm g ow h dynamics migh be influenced
by se e al g ow h media: LB, TSB and BHI. Since
CIs eached wi h BHI media wi h o wi hou
addi ional glucose we e e y low (be ween 0.01 and
0.1) o mos es ed s ains compa ed o o he cul u e
media, his g ow h medium was conside ed as no
sui able (da a no shown). On he con a y, bo h
TSB and LB media pe mi ed P. ae uginosa o o m
obus biofilms o e ime, eaching high biofilm mass
(CI alues up o 10 imes highe when compa ed o
hose obse ed using BHI medium) (Figu e S1). The
use o TSB medium (wi h and wi hou addi ional
suga s) esul ed in biofilm g ow h delay o all es ed
s ains, excep clinical isola e MF123 (Figu e S1ab).
Fo example, s ain PaO1 eached he maximum CI
in LB medium a 24 h, while simila CI was obse ed
only a 55 h in TSB-0.5%-glucose. A simila end in
biofilm g ow h delay was obse ed when LB medium
was supplemen ed wi h 0.5% o glucose (Figu e
S1d). Gi en i s highes biofilm o ma ion capaci y,
LB wi hou addi ional glucose was chosen o
P. ae uginosa biofilm g ow h in he xCELLigence
sys em.
Compa ison o impedance-based measu es o
CV s aining
Cul u ing pla es ma e ials ha e an impac on he
adhesion capaci y o P. ae uginosa and es ima es o
biofilm mass a e also influenced by he quan ifica ion
me hod used [39]. Figu e 1(a) ep esen s biofilm g ow h
dynamics o diffe en P. ae uginosa s ains when g own
in96-wellE-pla es,as measu edbyimpedancedu ing72
h; panel b co esponds o he quan ifica ion o eleased
c ys al iole measu ed as OD
610
a 8, 24, 36, 48, 56
and72 h, while panel c depic s P. ae uginosa biofilmbio-
mass s aining wi h c ys al iole a 36 h o g ow h o
each es ed s ain. The esul s indica e ha mos es ed
s ains had a simila biofilm o ma ion and g ow h
dynamics in bo h xCELLigence and Ibidi pla es,
sugges ing ha hese me hodologies a e compa able.
Fo example, s ains PaO1, MF120 and MF124 we e
he s onges biofilm o me s in bo h se ings. Howe e ,
as i can be seenin panel b, some o he isola es we e able
o adhe e and o m biofilms as e o s onge in Ibidi
pla es.
Effec o con en ional an ibio ics on biofilm
o ma ion
Fo u he analysis, we chose P. ae uginosa s ains
ha exhibi ed he highes biofilm- o ming capaci y
as measu ed by impedance (PaO1, MF120 and
EMERGING MICROBES & INFECTIONS 2065
MF124) and e alua ed he effec in eal ime o se en
con en ional an ibio ics commonly used in clinical
p ac ice (cip ofloxacin (CIP), ob amycin (TOB), ce -
azidime (CAZ), colis in (CST), pipe acillin- azobac-
am (TZP), imipenem (IPM) and me openem
(MEM)).
Figu e 2 summa izes he effec i eness o an ibio ics
on biofilm o ma ion in PaO1 and MF120 s ains,
while Figu e S3 depic s he effec o hese an ibio ics
on MF124. CIP showed a high efficiency and sup-
p essed biofilm o ma ion o all es ed s ains in a
dose-dependen manne . Howe e , he exposu e o
PaO1 and MF120 biofilms o 0.0625 mg/L o his an i-
bio ic esul ed in biofilm g ow h induc ion and
changes in biofilm g ow h dynamics (Figu e 2). TOB
and TZP also inhibi ed biofilm o ma ion when
added a high concen a ions. In addi ion, lowe con-
cen a ions we e no able o inhibi biofilm o ma ion
o any o he es ed s ains, al hough mos o hem
esul ed in biofilm g ow h delay o dec eased biomass.
Al hough none o he es ed concen a ions o CAZ
could ully inhibi biofilm o ma ion in any o he
es ed s ains, he inhibi o y capaci y o his an ibio ic
was concen a ion-dependen , and mos concen-
a ions o CAZ esul ed in biofilm g ow h delay.
Nei he CST no IPM o MEM could comple ely inhi-
bi o delay biofilm g ow h in PaO1 and MF120. In
con as , he exposu e o MF124 biofilms o 32 mg/L
o CST esul ed in comple e biofilm g ow h inhibi ion,
while lowe concen a ions o CST showed a dose-
dependen effec (Figu e S3). These esul s indica e
ha each biofilm- o ming s ain should be analyzed
indi idually, aking in o accoun mo e han one end-
poin in o de o assu e he bes clinical ou come.
Finally, Table 1 shows an ibio ic suscep ibili ies o
plank onic b o h cul u es (MIC) compa ed o he
Figu e 1. Biofilm o ma ion capaci y o se en P. ae uginosa s ains as quan ified by impedance-based measu emen s in xCELLi-
gence (a) and by C ys al Viole (CV) s aining (b, c). Panel b ep esen s he abso bance o eleased CV a 0, 8, 24, 36, 56 and 72 h,
whe e highe abso bance indica es la ge biofilm g ow h. Panel cdepic s biofilm o ma ion in ibiT ea 96-well pla es a 36 h,
whe e mo e in ense colou eflec s la ge biofilm mass. Bac e ial s ains used in he expe imen a e indica ed in he legend
and include a mu an o PAO1 s ain wi h impai ed biofilm o ma ion. Da a a e he means o h ee biological eplica es.
2066 M. ŽIEMYTĖET AL.
same isola es when hey we e g own in biofilms
(MBIC). MBICs we e up o 32 imes highe compa ed
o adi ional MICs ob ained by E- es and mic odilu-
ion (Table 1 and Table S2).
Capabili y o an ibio ics o e adica e
p e- o med biofilms
P e- o med biofilms a e no o iously difficul o e adi-
ca e, as an ibio ics mus c oss he biofilm ma ix o
each biofilm-embedded bac e ia [40]. Thus, o ob ain
u he insigh s on biofilm e adica ion dynamics, we
chose he an ibio ics ha showed he highes efficacy
when added a he beginning o biofilm g ow h
(TZP, CAZ, CIP, TOB) and es ed hei abili y o e a-
dica e P. ae uginosa PAO1 and MF120 biofilms a 24
and 48 h, espec i ely. Impedance-based measu e-
men s showed ha he es ed concen a ions o TZP
(128–0.0625 mg/L) could no dis up biofilms o he
examined s ains (Figu e 3). CAZ, which has a simila
mechanism o ac ion o TZP and inhibi s bac e ial cell
wall syn hesis, had no biofilm inhibi o y effec on
MF120 biofilms, bu was able o dis up PAO1
biofilms when added a he highes dose es ed (32
mg/L). Howe e , he inc ease in CI a 90 h sugges s
ha his an ibio ic is no able o kill all biofilm-
embedded bac e ia, which could consequen ly esul
in elapsed biofilm in ec ion (Figu e 3). On he o he
Figu e 2. Effec o cip ofloxacin (CIP), ob amycin (TOB), ce azidime (CAZ), colis in (CST), pipe acillin- azobac am (TZP), imipenem
(IPM) and me openem (MEM) on P. ae uginosa PaO1w and MF120 biofilm o ma ion. G aphs show es ima es o o al biofilm mass
as measu ed a 37°C using impedance-based measu emen s. Black lines indica e un ea ed con ols. Each line ep esen s he
mean o h ee biological eplica es. All an ibio ics we e added a he beginning o he expe imen oge he wi h bac e ial inocu-
lum om 0.625 o 32 mg/L o all an ibio ics excep TPZ (0.625 o 128 mg/L). SDs a e no shown o cla i y.
Table 1. Minimum inhibi o y concen a ion (MIC) s minimum biofilm inhibi o y concen a ion (MBIC) alues o cip ofloxacin
(CIP), ob amycin (TOB), ce azidime (CAZ), colis in (CST), pipe acillin- azobac am (TZP), imipenem (IPM), and me openem
(MEM) in diffe en P. ae uginosa s ains.
Minimum Inhibi o y Concen a ion/Minimum Biofilm Inhibi o y Concen a ion
S ain CIP TOB CAZ CST TZP IPM MEM
PAO1 0.094S/0.125 1.5S/2 1S/2 1S/>32 3S/16 1.5S/>32 0.38S/>32
MF120 0.125S/0.250 <=2S/4 2S/16 1S/>32 <=8S/32 <=1S/>32 <=1S/>32
MF124 0.19S/2 1S/8 2S/>32 1S/32 8S/32 2S/>32 0.75S/>32
No es: MICs we e measu ed by s anda d p o ocols (UMIC mic odilu ion es o colis in and E- es s o es o an ibio ics) and exp essed as (mg/L), while
MBIC we e calcula ed acco ding o impedance-based measu emen s in xCELLigence sys em. S and R in he able indica e i a s ain is conside ed sus-
cep ible o esis an acco ding o EUCAST guidelines [34]. MBICs we e assessed by he e alua ion o no malized biofilm g ow h cu es using impedance
measu es (Cell Index measu emen s) a 24 h o PaO1, 48 h o MF120 and 72 h o MF124, espec i ely, ollowing Fe e e al. [8].
EMERGING MICROBES & INFECTIONS 2067
hand, when ma u e P. ae uginosa biofilms we e
exposed o CIP, high concen a ions o his an ibio ic
supp essed new biofilm accumula ion and comple ely
e adica ed he p e- o med biofilm in bo h PaO1 and
MF120 s ains. On he con a y, low concen a ions
o his an ibio ic (0.0625 mg/L in PaO1 and 0.125–
0.0625 mg/L in MF120) we e ineffec i e. In addi ion,
we obse ed ha ce ain concen a ions o CIP
(0.125 mg/L in PaO1 and 0.25 mg/L in MF120),
al hough ini ially seemed effec i e, esul ed in a sud-
den inc ease in CIs a app oxima ely 80 and 100 h,
espec i ely. This second peak o g ow h obse ed
a e ea men wi h hese CIP concen a ions
sugges ed he eme gence o a do man cell ac ion
wi hin PaO1 and MF120 biofilms (Figu e 3), which
was la e in es iga ed in mo e de ail.
Compa ably o CIP, TOB comple ely e adica ed
ma u e MF120 biofilm when added a ela i ely
high concen a ions (32–4 mg/L), sugges ing ha
TOB can pene a e h ough he exopolysaccha ide
ma ix and efficien ly kill biofilm-embedded bac e ial
cells. On he o he hand, TOB was no efficien on
ma u e PaO1 biofilm, gi en ha none o he es ed
concen a ions could dis up he biofilm, indica ing
a s ain-dependen effec . In addi ion, simila ly o
CIP, TOB a 1 and 2 mg/L esul ed in he ise o a
second biofilm peak a 90 and 120 h, espec i ely,
sugges ing ha bo h an ibio ics can only kill a pa
o bac e ial cells embedded in he biofilms and induce
he eme gence o do man cells wi hin P. ae uginosa
biofilms.
Manni ol effec on plank onic and biofilm
g ow h
A ange o concen a ions o manni ol we e es ed o
cha ac e ize i s effec on bo h plank onic and adhe -
en P. ae uginosa g ow h. Abso bance measu emen s
showed ha he es ed concen a ions did no affec
PAO1 o MF120 plank onic g ow h (da a no
shown), indica ing ha his compound does no
ha e an imic obial ac i i y. Gi en ha many com-
pounds wi hou an imic obial effec s ha e been
linked o biofilm disassembly [41–43]weha ealso
es ed i manni ol had an effec on biofilm disagg e-
ga ion. Impedance measu emen s showed ha he
es ed concen a ions did no affec biofilm mass
compa ed o he un ea ed con ol (Figu e 4(a)).
This sugges s ha manni ol alone is no a biofilm dis-
pe sal agen . Thus, conside ing ha manni ol has
been shown o be in ol ed in he swi ching o pe sis-
e and ole an pheno ypes o me abolically ac i e
cells [44], we u he e alua ed how he combina ion
o manni ol and CIP could affec biofilm o ma ion
and ma u a ion o p e- o med MF120 biofilms in
eal- ime.
Manni ol enhances he effec o cip ofloxacin
The addi ion o manni ol alone a he beginning o
biofilm g ow h (3200 mg/L) did no esul in any
changes in o al biofilm mass compa ed o he
un ea ed con ol (Figu e 4(a)). Howe e , when man-
ni ol p e- ea men (3200 mg/L) was combined wi h a
CIP concen a ion ha a ou ed he appea ance o
do man cells (0.250 mg/L) added on a 48 h biofilm,
he biofilm was comple ely e adica ed. Gi en ha
lowe manni ol concen a ions we e no efficien and
esul ed in he appea ance o a second biofilm peak,
we conclude ha manni ol may be me abolized o e -
ime. Fo his eason, we ha e es ed whe he manni ol
could po en a e CIP effec mo e efficien ly when
added oge he wi h he an ibio ic a 48 h. Figu e 4
(b) shows ha he addi ion o 3200 o 320 mg/L o
manni ol, in combina ion wi h CIP a 48 h esul ed
in a comple e dis up ion o ma u e MF120 biofilms.
Howe e , he combina ion o CIP wi h lowe manni-
ol concen a ions was no sufficien o e e all pe s-
is e cells in o ac i ely g owing bac e ia. Ne e heless,
a concen a ion-dependen effec was obse ed, wi h a
delay in he second g ow h peak a manni ol concen-
a ions o 16, 32 and 64 mg/L. These esul s we e
confi med by colony coun s (Figu e 4(c)), whe e CIP
alone educed cell iabili y 2.5 o de s o magni ude,
whe eas 3200 mg/L o manni ol combined wi h CIP
comple ely killed all bac e ia embedded in he
biofilm ma ix while 320 mg/L esul ed in almos
se en o de s o magni ude educ ion in iable cell
numbe (p- alue<0.0001).
Iden ifica ion o pe sis e subpopula ions
Bac e ial cells g own wi h CIP a e he second CI peak
we e cul u ed on LBA pla es wi h CIP in o de o
main ain he pe sis e pheno ype. Pe sis e cells had
adiffe en colony mo phology, we e less mucoid and
we e no able o p oduce pyocyanin (whi e colonies).
In addi ion, hey we e opaque , compa ed o he
g eenish un ea ed con ol colonies pla ed on LBA
pla es wi hou addi ional an ibio ics (Figu e 5(a)).
I is known ha pe sis e s e e o a no mal w
pheno ype once an ibio ic p esence is ceased [45].
Thus, we u he de e mined he e e sibili y o his
pheno ype a e CIP emo al. Pe sis e s and un ea ed
con ols we e einocula ed in o esh g ow h media
supplemen ed wi h manni ol (3200 mg/L) and wi hou
manni ol (LB alone) and obse ed bac e ial g ow h
and colou changes. A e 72 h o g ow h, pe sis e
cells had a pheno ype simila o ha o un ea ed con-
ols in media which con ained manni ol bu no in LB
alone, concluding ha pe sis e s we e e e ed in o
ac i ely g owing cells. In addi ion, pe sis e cells ha
we e inocula ed in LB media wi hou manni ol
es o ed pyocyanin p oduc ion a 96 h (Figu e 5(b))
2068 M. ŽIEMYTĖET AL.
concluding ha manni ol accele a ed he e e sal o
pe sis e s o he w pheno ype. We ha e also assessed
MICs o pe sis e cells subcul u ed in LB aga pla es
wi h CIP and ound ha he esis ance o his an i-
bio ic was inc eased a leas 10 imes. Pe sis e cells
had also inc eased esis ance o IMI and MEM
(Table S3). Finally, we einocula ed pe sis e s o
esh LB media, LB media supplemen ed wi h manni-
ol and LB media supplemen ed wi h CIP. MICs pe -
o med a 96 h o inocula ion showed ha pe sis e
cells g own in he p esence o manni ol became
mo e suscep ible o mos o he es ed an ibio ics,
while in LB alone hey showed inc eased esis ance
o CIP and IMI. This indica es ha , in addi ion o phe-
no ypic changes (Figu e 5), pe sis e s g own in he
absence o an ibio ic also e e hei esis ance
p ofile. On he con a y, pe sis e s einocula ed in
LB media supplemen ed wi h CIP main ained he
whi e pheno ype and showed inc eased esis ance o
all he es ed an ibio ics (Table S3).
Gene ic changes associa ed wi h he pe sis e
pheno ype
We ex ac ed DNA om colonies g own in he p es-
ence o CIP (0.25 mg/L), manni ol (3200 mg/L) o
Figu e 3. Effec o pipe acillin- azobac am (TPZ), ce azidime (CAZ), cip ofloxacin (CIP) and ob amycin (TOB) on ma u e biofilms
o PaO1 (24 h a e inocula ion) and MF120 (48 h) s ains. A e he addi ion o he an ibio ics, biofilm g ow h was obse ed o
addi ional 100 h. Black lines ep esen an ibio ic- ee con ols while black a ows indica e he ime when an ibio ics we e added.
Each line ep esen s he mean o h ee eplica es. The eme gence o do man , pe sis e cells, which appea ed a e he biofilm
e adica ion using diffe en an ibio ics, is ma ked by black as e isks.
EMERGING MICROBES & INFECTIONS 2069
