Table Of Contentfmicb-07-01733 October5,2017 Time:15:56 #1
ORIGINALRESEARCH
published:08November2016
doi:10.3389/fmicb.2016.01733
Cross-Talk between Staphylococcus
aureus and Other Staphylococcal
Species via the agr Quorum Sensing
System
JaimeCanovas1†,MaraBaldry1†,MartinS.Bojer1†,PaalS.Andersen1,2,BengtH.Gless3,
PiotrK.Grzeskowiak3,MarcStegger2,PeterDamborg1,ChristianA.Olsen3and
HanneIngmer1*
1DepartmentofVeterinaryDiseaseBiology,FacultyofHealthandMedicalSciences,UniversityofCopenhagen,
Frederiksberg,Denmark,2DepartmentofMicrobiologyandInfectionControl,StatensSerumInstitut,Copenhagen,
Denmark,3CenterforBiopharmaceuticalsandDepartmentofDrugDesignandPharmacology,FacultyofHealthand
MedicalSciences,UniversityofCopenhagen,Copenhagen,Denmark
Staphylococci are associated with both humans and animals. While most are non-
pathogeniccolonizers,Staphylococcusaureusisanopportunisticpathogencapableof
causing severe infections. S. aureus virulence is controlled by the agr quorum sensing
Editedby: system responding to secreted auto-inducing peptides (AIPs) sensed by AgrC, a two
SusanneFetzner,
componenthistidinekinase.agr lociarefoundalsoinotherstaphylococcalspeciesand
UniversityofMünster,Germany
forStaphylococcusepidermidis,theencodedAIPrepressesexpressionofagrregulated
Reviewedby:
MattiasCollin, virulencegenesinS.aureus.Inthisstudyweaimedtobetterunderstandtheinteraction
LundUniversity,Sweden
betweenstaphylococciandS.aureus,andshowthatthisinteractionmayeventuallylead
ShinyaWatanabe,
JichiMedicalUniversity,Japan totheidentificationofnewanti-virulencecandidatestotargetS.aureusinfections.Here
*Correspondence: weshowthatculturesupernatantsof37outof52staphylococcalisolatesrepresenting
HanneIngmer 17differentspeciesinhibitS.aureusagr.Thedogpathogen,Staphylococcusschleiferi,
[email protected]
expressedthemostpotentinhibitoryactivityandwasactiveagainstallfouragr classes
†Theseauthorshavecontributed
equallytothiswork. found in S. aureus. By employing a S. aureus strain encoding a constitutively active
AIP receptor we show that the activity is mediated via agr. Subsequent cloning and
Specialtysection:
heterologous expression of the S. schleiferi AIP in S. aureus demonstrated that this
Thisarticlewassubmittedto
InfectiousDiseases, moleculewaslikelyresponsiblefortheinhibitoryactivity,andfurtherproofwasprovided
asectionofthejournal when pure synthetic S. schleiferi AIP was able to completely abolish agr induction
FrontiersinMicrobiology
of an S. aureus reporter strain. To assess impact on S. aureus virulence, we co-
Received:01August2016
inoculated S. aureus and S. schleiferi in vivo in the Galleria mellonella wax moth larva,
Accepted:17October2016
Published:08November2016 and found that expression of key S. aureus virulence factors was abrogated. Our data
Citation: show that the S. aureus agr locus is highly responsive to other staphylococcal species
Canovas J,Baldry M,Bojer MS,
suggesting that agr is an inter-species communication system. Based on these results
Andersen PS,GlessBH,
Grzeskowiak PK,Stegger M, we speculate that interactions between S. aureus and other colonizing staphylococci
Damborg P,Olsen CAandIngmer H will significantly influence the ability of S. aureus to cause infection, and we propose
(2016)Cross-Talkbetween
that other staphylococci are potential sources of compounds that can be applied as
StaphylococcusaureusandOther
StaphylococcalSpeciesviatheagr anti-virulencetherapyforcombatingS.aureusinfections.
QuorumSensingSystem.
Front.Microbiol.7:1733. Keywords:Staphylococcusaureus,Staphylococcusschleiferi,quorumsensing,agr,quorumsensinginhibition,
doi:10.3389/fmicb.2016.01733 auto-inducingpeptide,cross-talk,anti-virulencetherapy
FrontiersinMicrobiology|www.frontiersin.org 1 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #2
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
INTRODUCTION MATERIALS AND METHODS
Atleast40differentStaphylococcusspecieshavebeendescribed Bacterial Strains and Growth Conditions
to date (Harris et al., 2002). While a large number of these are Staphylococcus aureus strains used in this study include:
foundtocolonizehumans,manyalsocolonizeanimals(Nagase Strain 8325-4 (Novick, 1967) was used as a source of AIP-I
etal.,2002).Staphylococcusaureusisbyfarthebestcharacterized containing supernatant. For the β-galactosidase plate assays
staphylococcal species. This opportunistic pathogen resides on PC203 (S. aureus 8325-4, spa::lacZ), PC322 (S. aureus 8325-4,
skin and mucosal membranes in humans and animals, and it hla::lacZ), SH101F7 (S. aureus 8325-4, rnaIII::lacZ) (Chan
can cause a variety of infections ranging from mild skin and and Foster, 1998) and strain MOZ53 S. aureus agr type III
soft tissue infections to severe conditions such as septicemia (Horsburgh et al., 2002) were used. For the β-lactamase liquid
(Lowy, 1998). Expression of the majority of S. aureus virulence assay strain RN10829 WT and RN10829 Const (Geisinger
factors is controlled by the accessory gene regulator (agr) et al., 2009) a β-lactamase reporter strain substituting the
quorum sensing system. agr is composed of two divergent native agr locus with a chromosomal integration of P2-agrA
transcripts:one(P2)encodingtheAgrACtwocomponentsignal and P3-blaZ and a plasmid from which a constitutive active
transduction system that responds to autoinducing peptides variant of AgrC (agrC-I-R238H) is expressed, was used to
(AIPs) encoded and secreted by the products of agrBD; and assess AgrC-dependent effects of staphylococcal supernatants.
theother(P3)encodingaregulatoryRNA,RNAIII,theeffector Fluorescent reporters AH1677, AH430, AH1747, and AH1872
moleculeofagr.AIPsbindtotheagrC-encodedhistidinekinase (Halletal.,2013)wereusedtoevaluateagrinductionofthefour
(AgrC),andviaphosphorylationoftheAgrAresponse-regulator, different S. aureus agr groups. We cloned the Staphylococcus
stimulate the expression of RNAIII (Wang et al., 2014). At schleiferi agrBD genes into the BglII/EcoRI sites of expression
highcelldensities,AIPaccumulationresultsinup-regulationof vector pRAB12-lacZ (Helle et al., 2011) using primers 5(cid:48)-
exoproteinexpressionincludingthehla-encodedvirulencefactor GATACAAGATCTGTTAAGGAGGAGGGCTATTTG-3(cid:48) and
α-hemolysin,anddown-regulationofsurface-associatedproteins 5(cid:48)-GATACAGAATTCCGCTCTCTAAACATTATTTTATTATTC-
such as spa-encoded protein A (Queck et al., 2008; Shoham, 3(cid:48) and chromosomal DNA from S. schleiferi strain 2898 as
2011). template generating plasmid pRAB12-agrBD . This construct
Ss
Staphylococcusaureusstrainscanbedividedintoatleastfour or the vector itself was expressed in the S. aureus agr deletion
agr classes(groupsI–IV),withdistinctAIPsinducingvirulence strain 8325-4(cid:49)agr [constructed by transduction (ϕ11) from
gene expression within each group, but repressing expression strain RN6911 (Novick et al., 1993) into strain 8325-4] by
in strains of other groups by acting as competitive inhibitors growing the strains overnight under induction (0.2 µg/ml
to AgrC binding (Ji, 1997; Otto et al., 2001; Geisinger et al., anhydrotetracycline) generating AIP containing or AIP
Ss
2012). The agr specificity is determined by AgrB, AgrC, and negative supernatants respectively. All other staphylococcal
AgrD, which vary among the four groups. agr has also been strains used are listed in Table 1. Unless otherwise stated,
identified in other staphylococcal species although with much bacteriaweregrowninTryptoneSoyaBroth(TSB),Oxoid(1:10
higher sequence divergence. Despite this diversity, functional volume/flaskratio),at37◦Cwithshakingat200rpm.
agr loci have been demonstrated in Staphylococcus lugdunensis
andStaphylococcusepidermidis(Vandeneschetal.,1993;Wamel β-Galactosidase Plate Assay
et al., 1998). S. epidermidis is another clinically important
ThereporterassaywasconductedasdescribedbyNielsenetal.
opportunistic pathogen whose virulence is largely controlled
(2010). Test supernatants, and control supernatants of strains
via the agr system (Olson et al., 2014), and the S. epidermidis
8325-4(AIP-I)andM0Z53(AIP-III)wereused.Incubationuntil
AIP is a potent inhibitor of the S. aureus agr system (Otto
bluecolorappearedinplatesvariedfrom9to48h.
et al., 2001). Although, the biological importance of this cross-
inhibition of quorum sensing remains unknown, it has been
In vitro Competition As Assessed Using
suggested that it contributes to niche competition between
S. epidermidis and S. aureus, resulting in the predominance the β-Galactosidase Liquid Assay
of S. epidermidis on the skin and in indwelling device- Theassaywasbasedontheliquidβ-galactosidaseassay(Miller,
associated chronic infections (Otto et al., 2001; Thoendel 1972)withsomemodifications.Briefly,Overnight(ON)cultures
et al., 2011). The frequent isolation of other staphylococci of our SH101F7 S. aureus RNAIII reporter strain and our
together with S. aureus also point to a possible interaction S. schleiferi strains (grown at 37◦C while shaking at 200 rpm)
betweenstaphylococcalspecieswitharoleinnichecompetition werediluted100xin15mLTSBandallowedtoreachanOD
600
and colonization (Lina et al., 2003; Kozioł-Montewka et al., of0.5andafteradjustingeachstraintoanOD of0.1inTSB
600
2006). the competition was started at a ratio of 1:1 and followed over
The aim of this study was to examine the extent to which time. From each culture, 1 mL was taken at each time interval
agr cross-inhibitory activity occurs between S. aureus and and the samples were sonicated for 10 s to disrupt aggregates
other staphylococcal species, and to elucidate the mechanism formed. The OD was measured and serial dilutions for each
600
behind this cross-talk. Understanding this interaction between sampleweremadeandplatedonTSAwithX-gal(150µg/mL).
staphylococcimayhelpintheidentificationofnewanti-virulence Theremainingsampleswerecentrifugedfor3minat8000rpm
strategiestargetingS.aureusinfections. and 4◦C. After centrifugation, the supernatants were removed
FrontiersinMicrobiology|www.frontiersin.org 2 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #3
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
n
o
ati
ul
ownreg X + X + + + ++++ ++++ ++++ X X X + + + ++++ ++++ ++++ ++ ++ ++++ X X + +++ +++
d
aIII
rn ct.
e
eff
e e
stisolationsourc Horsenose Dogear Birdtrachea Birdtrachea Notreported Caturine Pigjoint Notreported Notreported Minkskin MinkSkin Minkskin Horsetrachea Horsewound Horsewound Minkskin Dogear Dogskin Cat Dogear Minkskin Notreported Notreported Notreported Dog Oxmilk ++++:verysever
o d
H n
a
ct
e
us us us eff
equorum haemolytic haemolytic haemolytic hominis hominis hyicus hyicus hyicus lentus lentus lentus vitulinus vitulinus vitulinus schleiferi schleiferi schleiferi schleiferi schleiferi schleiferi sciuri sciuri sciuri simulans simulans ++:severe
us us us us us us us us us us us us us us us us us us us us us us us us us us +
Name Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc Staphylococc derateeffect;
o
m
Id. 450 1312 2181 28993 9525 1084 218 2808 4948 6948 6949 aC-30966-5 62 96 128 30689-20 30743 30219 2319 2862 2898 9468 9470 9505 1457 312 ++hteffect;:
g
sli
n +
o
ati ct;
ownregul ++ ++ ++ ++ + ++ ++ + + + X X ++ X X ++ +++ ++ ++ X X + X +++ +++ X X:noeffe
d e:
naIII wher
ctivity. cer thewell
RNAIII-regulationa Hostisolationsour Dogwound Dogskin Dogwound Dognose Dogurine Dogskin Dogwound Dogwound Dogurine Dogskin Cowear Horseabscess Notrecorded Dogear Dogtonsil Horseuterus Horseuterus Cowwound Cowmilk Dogurine Oxmilk Notrecorded Dogkindney Horsewound Minkskin Minkwound nhibitionhaloaround
d ei
n h
hylococcalstrains,theirorigin,a Name Staphylococcusintermedius Staphylococcuspseudintermedius Staphylococcuspseudintermedius Staphylococcuspseudintermedius Staphylococcuspseudintermedius Staphylococcuspseudintermedius Staphylococcuspseudintermedius Staphylococcuswarneri Staphylococcuswarneri Staphylococcuswarneri Staphylococcusxylosus Staphylococcusxylosus Staphylococcuscapitis Staphylococcuscapitis Staphylococcuscapitis Staphylococcuscapitis Staphylococcuschromogenes Staphylococcuschromogenes Staphylococcuschromogenes Staphylococcusepidermidis Staphylococcusepidermidis Staphylococcusepidermidis Staphylococcusepidermidis Staphylococcusdelphini Staphylococcusdelphini Staphylococcusdelphini wasratedaccordingtothesizeoft
TABLE1|Stap Id. 27472 30755 C-31106 C-31304 30510 30665 30703 30106 29927 30331 28351 L136 9529 2877 6994 52 53 2890 313 1069 352 389 30575 29886 aC-30966-8 C-31232-2 Down-regulation
FrontiersinMicrobiology|www.frontiersin.org 3 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #4
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
andthepelletsresuspendedin1mLofTRIS50mM,pH8and TABLE2|RealtimePCRprimersusedinthisstudyfortheassessmentof
3µLoflysostaphin.Themixwasincubatedat37◦Cfor30min reference(ileS,pyk)andtargetgene(rnaIII)expression.
toallowforcelllysis.Z-buffer(400µL)wasthenaddedtoeach
Gene Sequenceforward Sequencereverse
samplewhichwerefurtherincubatedfor5minat28◦C.Lastly,
100µLofONPG(ortho-Nitrophenyl-β-galactoside)(4mg/mL) rnaIII GCACTGAGTCCAAGGAAACTAAC AAGCCATCCCAACTTAATAACC
wasaddedtothemixandthetimenecessaryforthesolutionto ileS ACATACAGCACCAGGTCACG CGCCTTCTTCAGTAAATACACC
turnyellowwascontrolled.TheODat420and550nmforeach pyk AGGTTGAACTCCCCAAACAA GCAGCCCAAGATTACAAAAA
samplewasmeasured.Theactivityofthesampleswascalculated
inMillerunitsusingthefollowingformulaasdescribedbyMiller
supernatant alone was added and the other with an additional
(1972):
10% of S. schleiferi 2898 supernatant. Cultures were allowed to
grow until sample withdrawal at 30 and 60 min. Samples were
1000∗(OD420−(1.75∗OD550)) centrifuged and the pellet was immediately frozen at −80◦C.
MillerUnits:
OD600∗T∗V The disruption of the cell membranes was performed with
FastPrep and the the QIAGEN RNeasy kit was used to purify
Where:T=timeofreactioninmin;
theRNAaspermanufacturer’sinstructions.GenomicDNAwas
V=mlcellsaddedtotheassaytubes.
thenremovedfromthesamplesusingDNase-I,RNase-freefrom
β-Lactamase Assay and Inhibitory Fermentas.Thesampleswerefirstincubatedfor60minat37◦C,
followed by 10 min incubation with 50 mM of EDTA at 65◦C.
Concentration (IC )
50 To generate cDNA from the RNA samples, the High Capacity
ThemethodusedisdescribedbyNielsenetal.(2014).Brieflythe cDNARTkitfromAppliedBiosystemswasused.10µLofRNA
RN10829 (P2-agrA:P3-blaZ)/pagrC-I (WT) and RN10829(P2- and 10 µL of RT-master mix were added to each reaction. The
agrA:P3-blaZ)/pagrC-I-R23H(AgrCconst.)reporterstrainswere master mix contained RT Buffer, RT random primers, dNTP
grown to an OD600 of 0.4–0.5 where a 1/10 volume of AIP- mix,Nuclease-freewater,andreversetranscriptase.Thenegative
I containing supernatant (obtained from strain 8325-4) and controlsconsistedofsampleswithnoreversetranscriptaseadded.
1/10S. schleiferisupernatantswereaddedtothereporterstrain Thesampleswererunfor10minat25◦C,120minat37◦C,5min
culture. In assays using heterologously expressed AIPSs 1/20 at 85◦C in a standard PCR machine. Finally, 1.5 µL of cDNA
volumeofAIP-Icontainingsupernatantwaschallengedwith1/5 wasassayedbyreal-timePCRonaLightCycler(cid:13)R 96Instrument
volumesupernatantfromexpressioncultures.Samplesobtained using FastStart Essential DNA Green Master FastStart Essential
at 30 min time intervals after addition of test solutions were DNA Green Master (Roche) and primers for the reference
analyzed for β-lactamase activity by nitrocefin conversion. The genes (ileS, pyk) and the target gene (rnaIII) as presented in
IC50 of the selected S. schleiferi supernatants was also tested Table2.
using the β-lactamase assay, where a 1/10 volume (0.5 mL)
of supernatant was added to the total volume of 5 mL of In vivo Competition Assay with Galleria
the reporter strain culture (RN10829-WT) representing the mellonella
undilutedsupernatant(100%).Then,80,60,40,20,10,5,2.5,and
The Galleria mellonella infection was carried out as described
2%oftheinitialvolumeoftheselectedsupernatantwasaddedto
by Pollitt et al. (2014) with minor modifications. Briefly, fifth-
obtaintheIC curve.
50 instar G. mellonella larvae were inoculated (in a proleg) with a
total of 2 × 107 CFU/mL of S. aureus (SH101F7 rnaIII::lacZ),
Assessment of agr Inhibition Across agr
S.schleiferi(2898erythromycinresistant)oraco-cultureofthe
Groups twostrains(toacombinedfinalCFU/mLof2×107)andsplit
FluorescentS.aureusreporterstrainsofagrtypesI-IV(P3-yfp) intogroupsforCFUcounting(35pergroup)andsurvivalbenefit
were used to evaluate the inhibitory potential of staphylococcal observation(20pergroup).Twocontrolgroupswereincluded;
supernatantsonrespectiveagrtypes.Individualreporterstrains oneinjectedwithphosphatebufferedsaline(PBS)andtheother
were grown to early exponential phase (non-fluorescent, OD not handled at all. The larvae were then incubated at 37◦C. At
approximately0.01),thenadded20%S.schleiferi2898stationary 24,48,and72hpost-inoculation,thehemolymphofthelarvae
phase supernatant or TSB as a control and grown to late was collected for CFU determination. Colonies were counted
stationary phase to allow for full agr induction. P3 promoter afterO/Nincubationat37◦ConTSAcontainingerythromycin
activityofthebacterialpopulationwasmonitoredasaccumulated (5 µg/mL) and X-gal (150 µg/mL). Survival of the larvae was
YFP by flow cytometry on a BD AccuriTM C6 flow cytometer monitoredatthesametimepointsashemolymphcollection.This
usingtheFL1channel. experiment was repeated three times using different batches of
larvapurchasedfromalocalpetstore(HPReptiles,Copenhagen)
Reverse Transcriptase-Quantitiative PCR withsimilarresults.
Overnight culture of the strain 8325-4 was diluted 100x in
AIP Sequencing
15 mL of TSB in a 300 mL flask. Three replicate cultures were
prepared and once the cultures reached an OD of 0.35 they SeveralStaphylococcusspp.weresequencedtolookfortheagrD
600
were split into two different conditions; one with 10% 8325-4 andtheaminoacidsequenceoftheAIP.DNAfromtheselected
FrontiersinMicrobiology|www.frontiersin.org 4 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #5
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
staphylococciisolateswasextractedusingtheDNeasyBloodand Statistics
Tissue kit as described by the manufacturer (Qiagen, Valencia, Where applicable for the β-lactamase assays (Figures 2B and
CA, USA). For sequencing preparation fragment libraries were 4A) statistical analysis was performed using the multiple t-test
constructedusingtheNexterakit(Illumina,SanDiego,CA,USA) allowing unequal variance. For the qPCR data (Figure 2A)
followedby251-bppaired-endsequencingonaMiSeqsequencer an unpaired t-test was performed on normalized and log
(Illumina) according to the manufacturer’s instructions. The transformeddata.Forsurvivalcurveanalysis,theKaplan–Meier
sequencing reads were assembled using CLC Genomics Work- methodwasappliedandstatisticalanalysiswascarriedoutusing
bench 8.0 (Qiagen, Aarhus, Denmark) with default parameters the Log Rank (Mantel–Cox) Chi square test. Any value above
to include only contigs of 500 nucleotides and with over 30- P = 0.05 was considered as not significant. All statistical tests
fold average coverage. The sequence of S. schleiferi 2898 has wereperformedwithGraphPadPrismv.7.0.
beendepositedinGenBank(accessionnumberPRJEB15874).All
othersequencecontigsareavailableuponrequest.ExtractedAIP
sequenceswerealignedusingMuscleasimplementedinMEGAv RESULTS
6.06,wherethephylogenywasconstructedusingthemaximum
parsimony approach with 100 bootstraps and represented with S. aureus Virulence Gene Expression is
midpointrooting. Modulated by Staphylococcal Culture
Supernatants
Chemical Synthesis of AIP A total of 52 staphylococcal isolates representing 17 species
obtainedfromavarietyofdifferentanimalhosts(Table1)were
TheAIPwassynthesizedbyadoptingaprotocolbasedonlinear
examined for their ability to interfere with the agr quorum
peptide hydrazides, which was previously reported by Liu and
sensingsystemofS.aureususingapreviouslyestablishedreporter
coworkers(Zhengetal.,2013)Thelinearpeptidehydrazidewas
assay (Nielsen et al., 2010). Staphylococcal strains to be tested
synthesized by standard 9-fluorenylmethyloxycarbonyl (Fmoc)
weregrownovernighttostationaryphaseintryptonesoybroth
solid-phase peptide synthesis (SPPS) using HCTU/i-Pr EtN
2 (TSB), and cell-free supernatants were added to wells formed
for amino acid activation on an automated peptide synthesizer
in tryptone soy agar (TSA) plates containing reporter strains
(SyroWave XP, Biotage). A hydrazide-2-chlorotrityl resin was
carryinglacZfusionstoeitheroneoftheagrcontrolledvirulence
applied to afford the desired C-terminal peptide hydrazide
geneshla,spa,orrnaIII aswellastheβ-galactosidasesubstrate,
[YPFCIAYF-NHNH ] after nine coupling/deprotection
2 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside(X-Gal).For
cycles, followed by concomitant deprotection of side chain
the majority (37 out of 52) of the staphylococcal supernatants
functionalities and cleavage from the solid support with
tested we observed a reduced expression of hla and rnaIII but
CF COOH-i-Pr SiH-water (95:2.5:2.5, 5 mL, 3 h, room
3 3 increasedspaexpression(Figure1;Table1).The37supernatents
temperature). The crude peptide was obtained by trituration
exhibiting this expression pattern represents 14 of the 17
with cold ether and used without any further purification.
staphylococcal species, indicating that these secrete substances
ESI-MS m/z calcd for C H N O S 1037.5, found 1037.2
[M+H+]. 53 68 10 10 thatinterferewiththeagrsystemofS.aureus.
Thecrudelinearpeptide(10mg,0.01mmol)inDMF(20µL)
The Dog Pathogen Staphylococcus
wasaddedtoasolutionofNaNO (6mg,0.09mmol)insodium
2 schleiferi is a Potent Inhibitor of the S.
phosphate buffer (0.1 mM, pH 3.0) containing guanidinium
chloride (6 M) at −20◦C. The reaction mixture was kept at aureus agr Quorum Sensing System
−20◦C for 20 min and was then quenched by addition of The magnitude of agr-interference caused by staphylococcal
DMF(0.2mL)andcyclizationbuffer[sodiumphosphatebuffer supernatants was monitored in S. aureus 8325-4 by RT-qPCR
(0.6 mL, 0.1 mM, pH 6.5) containing thiophenol (44 mg, using previously described primers (Nielsen et al., 2012). Using
0.4 mmol)]. This reaction mixture was agitated on a rocking thesupernatantofthenotablyactiveS.schleiferistrain2898we
table for 16 h at room temperature. The desired AIP was observedthattherelativeRNAIIIexpressioninS.aureus8325-4
then purified by preparative reversed-phase HPLC on a C18 decreased by 300 and 3000-fold at 30 and 60 min respectively,
Phenomenex Luna column (250 mm × 20 mm, 5 µm, 100 Å) when compared to unexposed cells (Figure 2A). Furthermore,
using an Agilent 1260 LC system equipped with a diode array the 50% inhibitory concentration (IC ) was reached with
50
UV detector, applying a gradient of eluent I (water-MeCN- S.schleiferisupernatantconstitutingonly6%ofthetotalS.aureus
TFA, 95:5:0.1) and eluent II (0.1% TFA in acetonitrile) with culturevolume.
a flow rate of 20 mL/min. Lyophilization of the fractions To address if the observed effect of staphylococcal
containingproduct,providedawhitefluffysolid(∼0.5mg,5%) supernatants on virulence gene expression in S. aureus is
at >98% homogeneity as determined by UPLC–MS analysis mediatedviadirectinterferencewiththeS.aureusagrregulatory
at 254 nm. The compound was reconstituted in DMSO and system, we examined if the effect could be mitigated in a
accurate concentration was determined by UV spectroscopy S. aureus strain encoding a constitutively active AgrC sensor
(5 mM) before use. ESI-MS m/z calcd for C H N O S histidinekinase(Geisingeretal.,2009).Usingthereporterstrains
53 64 8 10
1005.5, found 1005.2 [M+H+]. MALDI-TOF MS found 1005.6 RN10829 WT (expressing the WT AgrC) and RN10829 Const.
[M+H+]. (isogenic mutant expressing the constitutively active AgrC)
FrontiersinMicrobiology|www.frontiersin.org 5 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #6
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
FIGURE1|ModulationofStaphylococcusaureusvirulencegeneexpressionbystaphylococcalculturesupernatants.TSAagarplates(with
erythromycinandX-gal)containing(A)thehla-lacZ(PC322;Eryr),(B)thernaIII-lacZ(SH101F7;Eryr),or(C)thespa-lacZ(PC203;Eryr)reporterstrainofS.aureus
wereexposedto20µL(inpre-drilledwells)ofsupernatantsfromcentrifugation(8000rpmfor60s)ofovernightculturesofstrains27472(Staphylococcus
intermedius),28993(Staphylococcushaemolyticus),30755(Staphylococcuspseudintermedius),30743(Staphylococcusschleiferi),29886(Staphylococcus
delphini),30106(Staphylococcuswarneri)and2898(Staphylococcusschleiferi).H2Owasusedasacontrol.Zonesappearedbetween9and36hofincubationat
37◦C.Thisfigureisrepresentativeofonesetofscreeningplates.
containing a blaZ gene fused to the P3 promoter as previously hypothesis that the inhibition was due to the presence of non-
described by Nielsen et al. (2014), we examined the culture nativeAIPinthesupernatantofS.schleiferi.S.aureusAIPwas
supernatantsobtainedfromS.schleiferistrains2898and30743. previouslybiosynthesizedinS.aureusbycloningtheagrBDgenes
As predicted, in the presence of the WT AgrC in the reporter inanotherwiseagrnegativebackground(ThoendelandHorswill,
strain,supernatantsofbothoftheS.schleiferistrainssignificantly 2009). Here, after genome sequencing of the S. schleiferi strain
down-regulated RNAIII expression (Figure 2B) but when the 2898,weclonedthe2898agrBDgenesintotheexpressionvector
S. schleiferi supernatants were added to the strain expressing pRAB12-lacZ (Helle et al., 2011) generating plasmid pRAB12-
the constitutively active AgrC (RN10829 Const.) we observed agrBD and expressed it in the S. aureus agr deletion strain
Ss
no difference in RNAIII expression (Figure 2C). These results 8325-4(cid:49)agr. In contrast to the inactive vector control strain
suggestthattheeffectisAgrCmediated. thesupernatantoftheS.schleiferi-AIP(AIP )producingstrain
Ss
Under the presumption that S. schleiferi produces AIPs, clearlyinhibitedRNAIIIexpressionasmeasuredbyβ-lactamase
which cross-inhibit the agr system of S. aureus, we assessed activity from the P3-blaZ reporter strain both 30 and 60 min
the specificity with respect to the different S. aureus agr types. after addition of the supernatant (Figure 4A). These results
Hence,weexaminedexpressionfromapreviouslydescribedP3- confirmthattheAIPproducedbyS.schleiferiisinhibitingagrof
yfpreporterconstructpresentinS.aureusstrainsofknownagr S.aureus.TofurthersupportthatitisinfacttheS.schleiferiAIP
types(Halletal.,2013)thathadbeengrowninthepresenceor that is responsible for inhibition of S. aureus RNAIII via AgrC
absence of supernatant from S. schleiferi strain 2898. Evidently, agonist activity, we synthesized the proposed S. schleiferi AIP
the supernatant confered a considerable degree of repression and tested the synthetic material in the same P3-blaZ reporter
across all agr groups of S. aureus (Figure 3) giving rise to a strain. Our results show indisputably that the S. schleiferi AIP
10–100 fold reduction in observed fluorescence intensity of all is a potent inhibitor of S. aureus RNAIII expression and that
reporters.Wenotethattheagrsystemappearsalmostcompletely it acts antagonistically on the reporter strain at low nanomolar
repressed in the types I and III reporters even at the late time concentrations(Figure4B).
pointoftheassay(24h),whiletherepressionofthetypesIIand
IVreportersdisplayadelayintheresponse.Thisdifferencemay S. aureus agr Repression by S. schleiferi
stem from different cross-inhibitory affinities of the suspected is Observed during Niche Competition
AIP molecule of S. schleiferi against the different AIP-AgrC
Both In vitro and In vivo
receptor pairs in S. aureus, or simply reflect a difference in the
To address if the agr inhibitory activity of S. schleiferi also is
agractivationkineticsand/orauto-fluorescenceofthereporters
expressedwhenbothstrainssharethesamenicheenvironment
employed.
we co-cultured S. aureus SH101F7 rnaIII::lacZ reporter strain
togetherwiththeS.schleiferi2898or30743strainsata1:1ratio
S. schleiferi Inhibition of S. aureus agr is
inTSB.After4hwherebacterialgrowthhadreachedstationary
AIP-Mediated
phase and the agr quorum sensing system in S. aureus is
Having demonstrated that S. schleiferi supernatant is a potent normallyfullyactivated,weobservedalmostcompleteinhibition
inhibitor of RNAIII expression in S. aureus we investigated the of S. aureus RNAIII expression in cells co-cultured with either
FrontiersinMicrobiology|www.frontiersin.org 6 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #7
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
FIGURE2|Continued
andRNAIIIlevelswerequantified.InterferenceofS.schleiferisupernatanton
S.aureusagrwasmonitoredusingthestrains(B)RN10829(P2-agrA:
P3-blaZ)/pagrC-I(WT)and(C)RN10829(P2-agrA:P3-blaZ)/pagrC-I-R23H
(AgrCconst.)ReporterstrainsweregrowntoanOD600of0.4–0.5wherea
1/10volumeofAIP-Icontainingsupernatantfromstrain8325-4and1/10
S.schleiferisupernatantwereaddedtothereporterstrainculture.Samples
obtainedat30mintimeintervalsafteradditionoftestsolutionswereanalyzed
forβ-lactamaseactivitybynitrocefinconversion(Nielsenetal.,2014).Each
barrepresentstheaverageof3biologicalreplicatesandtheerrorbars
representthestandarddeviation.
FIGURE3|Staphylococcusschleiferisupernatantaffectstheactivity
ofallfourS.aureusagrtypes.ActivityoffourS.aureusagrgroup
reporters(agr-I,agr-II,agr-III,andagr-IV)assessedbyP3expression(P3-yfp)
measuredasYFPaccumulationbyflowcytometry.Non-fluorescent,
exponentialphasecellsweregrownwith(blue)orwithout(red)20%
S.schleiferi2898supernatantuntilstationaryphase.Depictedfluorescence
histogramsoriginatefromcellsanalyzedatthe24htimepoint.
of the S. schleiferi strains (Figure 5A). Within the time frame
of the experiment, the growth proportion of S. aureus to
S.schleifericellsremainedessentiallyunchangedsuggestingthat
neitherstrainexertagrowthinhibitoryeffecttowardeachother
(Figure5B).Thus,duringco-culturethepresenceofS.schleiferi
efficientlyrepressestheS.aureusagrquorumsensingsystem.
Given the obvious interaction between the two bacterial
speciesduringco-cultureinlaboratorymediumweinvestigated
ifasimilarphenomenonisobservedinG.mellonellawaxmoth
larvae, a model previously reported for studies of S. aureus
FIGURE2|QuantificationofinterferencebyS.schleiferionRNAIII
virulence(DesboisandCoote,2011).Applyingthesamegrowth
expressioninS.aureus.(A)RT-qPCRquantitativeverificationofRNAIII
downregulationintheS.aureus8325-4strain.S.aureusin15mLTSBwas conditionsasfortheinvitroco-cultureexperimentweinoculated
growntoanOD600=0.35andsupplementedwitha1:10volumeof 2 × 107 colony forming units (CFUs) into the prolegs of fifth-
S.schleiferi2898supernatant.Sampleswerecollected30and60min instar larvae and followed the CFU counts and larval survival
post-exposuretoS.schleiferisupernatant,RNAwasisolated,cDNAprepared
over a period of 3 days. When single species were inoculated,
(Continued)
S. aureus was recovered at 1 × 108 cells 24 h post-inoculation
FrontiersinMicrobiology|www.frontiersin.org 7 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #8
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
FIGURE5|StaphylococcusschleiferiinhibitsS.aureusagrduring
co-culturewithoutinfluencinggrowth.(A)S.aureusrnaIII:lacZreporter
strainSH101F7grownaloneorinco-culture(1:1)withS.schleiferistrain
30743or2898inTSB.β-galactosidaseactivityundereachconditionwas
determinedtomonitoragrinduction.Eachbarrepresentstheaverageofthree
replicatesandtheerrorbarsrepresentthestandarderrorofthemean.(B)
Growthofthebacterialculturesmeasuredascolonyformingunits(CFUs)/mL
monitoredinparallelforeachtimepoint.Distinctionbetweenthetwospecies
FIGURE4|StaphylococcusschleiferiAIPinterfereswithS.aureusagr. intheco-culturewasmadebyplatingonTSAsubstitutedwithX-gal,resulting
(A)RNAIIIexpressionwasrecordedasβ-lactamaseexpressedfromthe intheSH101F7reporterstraingrowingasbluecolonies,whiletheS.schleiferi
P3-blaZreporterfusioninS.aureusRN10829(P2-agrA:P3-blaZ)/pagrC-I(WT) growingaswhite.
withadditionof5%AIP-Isupernatantand20%supernatantfromeitherstrain
8325-4(cid:49)agr/pRAB12-agrBDSs(AIPSs)orthecontrolstrain8325-4(cid:49)agr/
pRAB12-lacZ(vectorcontrol)thathadbeengrownandinduced(0.2µg/ml
anhydrotetracycline)overnight.Eachbarrepresentstheaverageofthree were present in equal numbers at 24 h, followed by a decline
biologicalreplicatesandtheerrorbarsrepresentthestandarddeviation.(B) at 48 h and complete eradication of both species after 72 h
P3-blaZexpressionrecordedfromS.aureusRN10829(P2-agrA:P3-blaZ)/ (Figure 6A). The presence of S. schleiferi 24 and 48 h after
pagrC-I(WT)whentheinducingAIP-Icontainingsupernatant(10%)is
inoculation with S. aureus suggests that factors produced by
challengedfor45minwithdifferentconcentrationsofsyntheticS.schleiferi
S. aureus allow maintenance of both species in the larva, while
AIPatindicatedconcentrations.NoinductionandAIP-Icontaining
supernatantalonewasincludedascontrols.Eachbarrepresentstheaverage inverselytheclearanceofbothspeciesat72hmaybesuggestive
ofthreebiologicalreplicatesandtheerrorbarsrepresentthestandard that there is a S. schleiferi-mediated repression of S. aureus
deviation. virulencefactorsviaagrdown-regulation,andthatthisrepression
allows the larvae to eradicate the combined staphylococcal
population. Despite the clearance of both species at 72 h, the
andonlydeclinedslightlyat72h.Wewereunabletodetectany co-culturedidnotofferasignificantlarvalsurvivalbenefitover
S. schleiferi in the larval hemolymph at any of the time points the single S. aureus inoculant group (Figure 6B). Nevertheless,
(Figure 6A), where survival of this group was comparable to our data show that in the presence of S. aureus, S. schleiferi is
the PBS control larvae group (Figure 6B). In contrast, when maintainedforanextendedperiodoftimeinthelarvaeandthat
S. schleiferi was inoculated together with S. aureus both species ultimatelytheirpresenceleadstoeradicationofbothspecies.
FrontiersinMicrobiology|www.frontiersin.org 8 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #9
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
FIGURE6|DualpresenceofS.schleiferiandS.aureusinfluencecolonizationcapabilitiesinaGalleriamellonellainfectionmodel.Fifth-instar
G.mellonellalarvaewereinoculatedwithatotalof2×107CFU/mLofS.aureus(SH101F7),S.schleiferi(2898)oraco-cultureofthetwostrains(toacombined
finalCFU/mLof2×107)andsplitintogroupsforCFUcounting(35pergroup)andsurvivalbenefitobservation(20pergroup).(A)At24,48,and72h
post-inoculation,thehemolymphofthelarvaewascollectedforCFUdetermination.ColonieswerecountedafterO/Nincubationat37◦ConTSAcontaining
erythromycinandX-gal,asbothstrainsareerythromycinresistant.(B)Survivalofthelarvaewasmonitoredatthesametimepointsashemolymphcollection.This
experimentwasrepeatedthreetimeswithsimilarresults.
AIP Homology in Staphylococci is not required for the antagonistic or the activation activity
The AIP amino acid composition, positioning and properties (Wright et al., 2004). Also they retained the central cysteine
havebeenreportedtoplayanimportantroleinthetypeofAgrC necessary for AIP cyclization (Thoendel and Horswill, 2009).
interaction(Mayvilleetal.,1999;Lyonetal.,2000;Wrightetal., Interestingly, despite their strong inhibitory activity toward
2004;Tal-Ganetal.,2013a).Accordingly,weanalyzedsequence S. aureus agr, neither S. schleiferi nor Staphylococcus delphini
homology of AIPs from a selection of staphylococci for which AIP contain an alanine residue, which has been reported to
supernatantactivityrangedfromnoeffecttosevereinhibitionof play a role in non-native AIP-AgrC binding antagonism (Lyon
S.aureusagr(Figure7A).AlignmentofAIPsequencesbetween et al., 2000; Tal-Gan et al., 2013a). Inferred relationships of
differentstaphylococcalspeciesrevealedlessthan30–40%amino the various AIPs revealed that they clustered according to
acidconservation.Withinthesamespecies,theAIPswerehighly species and for the most part also according to AIP group
conserved;aphenomenonthathasalsobeenobservedforalready (Figure7B).Thedifferencesinaminoacidconservationandthe
reported staphylococcal AIP sequences (Dufour et al., 2002; difficultiesinlocatingspecificaminoacidsequencesresponsible
Thoendel and Horswill, 2009). For our sequenced S. schleiferi for non-cognate AIP-AgrC interactions, highlights that spatial
AIPs there was 100% conservation and we noted that the most arrangement and secondary structure of the AIPs play an
highly conserved amino acids between species were those with importantroleintheinteractionsbetweenthenon-cognateAIPs
specific properties needed for correct interaction or folding of andtheS.aureusAgr-C(Mayvilleetal.,1999;Lyonetal.,2000;
AIP. For example, the C-terminal amino acid of most of the Wright et al., 2004). We speculate, that AgrC is promiscuous
ascertained AIP sequences is tyrosine or phenylalanine which in terms of susceptibility to AIP inhibitory structures and
provides the hydrophobicity essential for binding of the AIP this promiscuity allows inhibition by multiple staphylococcal
to a putative hydrophobic pocket in the AgrC receptor, but species.
FrontiersinMicrobiology|www.frontiersin.org 9 November2016|Volume7|Article1733
fmicb-07-01733 October5,2017 Time:15:56 #10
Canovasetal. StaphylococcalSpeciesCross-Talkviatheagr
FIGURE7|agrDhomologyofselectedStaphylococci.IsolateswereanalyzedforagrDhomologybyIlluminasequencing;(A)TableshowingAIP-alignment.The
AIPregionishighlightedinred.S.I.G.:StaphylococcusintermediusGroup;S.A.G.:S.aureusGroup;Strains8325-4,RN6607,MOZ53,andRN4580wereincluded
asS.aureusAIPsequencereferences.(B)PhylogeneticanalysisbasedonaMusclealignmentoftheAIPsequenceswithmidpointrooting.Scalebarindicate
substitutionspersite.
DISCUSSION withS.aureusagr,buteveninfluencevirulenceandcolonization
capabilities of S. aureus when forced to share a niche, as was
Thisstudylooksattheecologicalaspectsofregulatorycross-talk shownintheco-infectionofG.mellonellawithS.schleiferiand
andnichesharingbetweenstaphylococciasmediatedviatheagr S. aureus. Cross-communication involving agr interference has
quorumsensingsystem,anddelvesintothepossibleeffectsthis previouslybeenobservedasaresultofco-habitualcompetition
cross-talkmayhaveonaspectsofvirulenceandcolonizationfor within the same ecological niche, and we speculate that this
thehumanpathogenStaphylococcusaureus. also is a reason for the observed interference in our study.
Cell to cell communication via quorum sensing is very A few known examples are the interactions between S. aureus
common amongst bacteria and promotes both inter and intra- and S. epidermidis on human skin and between S. aureus and
species interactions (LaSarre and Federle, 2013). The agr QS Pseudomonasaeruginosainthelungsofpatientssufferingfrom
system of S. aureus is especially sensitive to pheromones cystic fibrosis (Otto et al., 2001; Qazi et al., 2006). In the first
produced by S. aureus strains of other agr groups (Thoendel example,S.epidermidisAIPpheromoneiscapableofinhibiting
andHorswill,2009;Thoendeletal.,2011)orproducedbyother agr activity of S. aureus groups I, II and III, but not that of
staphyloccocalspecies,namelyS.epidermidis(Ottoetal.,2001), group IV, which interestingly is the only S. aureus AIP capable
and has even displayed sensitivity to compounds produced by of inhibiting S. epidermidis agr activity. It has been suggested
other microorganisms of unrelated niche environments such as that this cross-talk is one of the reasons why S. epidermidis
the marine Photobacterium halotolerans-derived QS inhibitor predominatesoverS.aureusonhumanskin(Ottoetal.,2001).
solonamide B (Nielsen et al., 2010, 2014). Despite this well- Inthesecondexample,thecross-talkbetweenP.aeruginosaand
documentedagrcross-inhibitionbetweenthedifferentS.aureus S.aureusisacaseofcross-communicationbetweentwounrelated
agr groups and between S. aureus and other staphylococcal microorganisms. It has been shown that N-acylhomoserine
species,weaimedtoexplorefurtherthecommunicationbetween lactone 3-oxo-C -HSL produced by Pseudomonas is capable
12
S. aureus and staphylococci with little niche overlap, and what of inhibiting S. aureus growth, and that this compound at
repercussions this communication might have in terms of sub-inhibitory concentrations interferes with S. aureus agr and
virulence adaptation. While our findings corroborate those of sarA expression in a manner involving cytoplasmic membrane
others with regards to staphylococcal cross-talk via agr (Otto saturation (Qazi et al., 2006). In cystic fibrosis patients, the
et al., 2001; Lina et al., 2003; Thoendel et al., 2011), we interaction between them in addition to host factors tends to
alsoshowextensiveAIP-mediatedcross-talkbetweenpreviously favor the predominance of S. aureus colonization in young
untestedstaphylococciandS.aureus,with82%ofthepreviously patientsandP.aeruginosainadultpatients,thoughco-isolation
untested staphylococci exerting ability to interfere with agr of of both organisms is still found in 50% of adult CF patients
S.aureus.Importantly,weshowthatstaphylococcifromvarying (Qazi et al., 2006). The co-existence of these two organisms
environmental niches not only have the capacity to interfere in CF patients suggests an evolutionary role of cross-talk on
FrontiersinMicrobiology|www.frontiersin.org 10 November2016|Volume7|Article1733
Description:S. aureus virulence is controlled by the agr quorum sensing system responding to . template generating plasmid pRAB12-agrBDSs. This construct.