Kudrin Supplemental Material - Antimicrobial Agents and...

18
1 Supplemental Material Sub-inhibitory concentrations of bacteriostatic antibiotics induce relA- dependent and relA-independent tolerance to β-lactams Pavel Kudrin 1,⌘ , Vallo Varik 1,2,3⌘ , Sofia Raquel Alves Oliveira 1 , Jelena Beljantseva 1 , Teresa Del Peso Santos 2,3 , Ievgen Dzhygyr 2,3 , Dominik Rejman 4 , Felipe Cava 2,3 , Tanel Tenson 1 , Vasili Hauryliuk 1,2,3,* University of Tartu, Institute of Technology, Tartu, Estonia 1 ; Department of Molecular Biology, Umeå University, Umeå, Sweden 2 ; Laboratory for Molecular Infection Medicine Sweden (MIMS), Umeå University, Umeå, Sweden 3 ; Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences v.v.i., Flemingovo nám. 2, 166 10 Prague 6, Czech Republic 4 Running head: relA-dependent and relA-independent β-lactam tolerance * address correspondence to Vasili Hauryliuk, [email protected] P.K. and V.V. contributed equally to this work

Transcript of Kudrin Supplemental Material - Antimicrobial Agents and...

1

SupplementalMaterialSub-inhibitory concentrations of bacteriostatic antibiotics induce relA-dependentandrelA-independenttolerancetoβ-lactamsPavel Kudrin1,⌘, Vallo Varik1,2,3⌘, Sofia Raquel Alves Oliveira1, JelenaBeljantseva1, Teresa Del Peso Santos2,3, Ievgen Dzhygyr2,3, Dominik Rejman4,FelipeCava2,3,TanelTenson1,VasiliHauryliuk1,2,3,*University of Tartu, Institute of Technology, Tartu, Estonia1; Department ofMolecularBiology,UmeåUniversity,Umeå, Sweden2; Laboratory forMolecularInfectionMedicineSweden (MIMS),UmeåUniversity,Umeå,Sweden3; Instituteof Organic Chemistry and Biochemistry, Czech Academy of Sciences v.v.i.,Flemingovonám.2,16610Prague6,CzechRepublic4Runninghead:relA-dependentandrelA-independentβ-lactamtolerance*addresscorrespondencetoVasiliHauryliuk,[email protected]⌘P.K.andV.V.contributedequallytothiswork

2

SupplementaryMethodsPreparationofthiostreptonstocksolutionsSeveral crystals of thiostrepton (Tocris Bioscience, validated by massspectrometry) dissolved in ≈300 μl of 100% DMSO (Sigma Aldrich) and theconcentration(≈600mM)ismeasuredbyabsorbanceat280nmwithε=0.027µM-1cm-1. DMSO stock is then used to prepare the working stock in 1xHepes:polymixbuffer(25mMHepes-KOHpH7.5,15mMMgCl2,0.5mMCaCl,95mMKCl,5mMNH4Cl,8mMputrescine,1mMspermidine,5mMK3PO4pH7.3and 1 mM DTT (1)) supplemented 0.1% Pluronic F-127 (Sigma Aldrich) asfollows.ThiostreptonstockinDMSOwasmixedwith20%(v/w)PluronicF-127inDMSO and 1xHepes:polymix bufferwas gradually added to final volume toachievethefinalconcentrationof12.5µM.ThefinalDMSOconcentrationshouldbekeptbelow3%andallthepreparationsshouldbedoneatroomtemperaturetoavoidprecipitation.Workingstockwasbrieflycentrifuged(21,000rcf,5min)atroomtemperatureinordertoremoveanytracesofprecipitatedthiostrepton,supernatant transferred into a new tube and the final concentration ofthiostreptonre-measuredusingabsorbanceat280nmusing1xHepes:polymixsupplementedwith3%DMSOand0.1%PluronicF-127asablankreference.PreparationofppGppppGpp was produced enzymatically using either RelSeq enzyme fromStreptococcus equisimilis (2) or RelQ enzyme from Enterococcus faecalis (3)essentiallyasdescribedinMecholdetal.(4).Reactionmixturecontaining5mMGDP,10mMATP,15mMMgCl2,30mMTris-HCl(pH8.0),100mMNaClinMilliQH2Owaspreincubatedfor5minat37°Cfollowedbytheadditionof50µMRelSeqor5µMRelQ.SinceRelSeq isstrongly inhibitedby freeMg2+ ions, therefore thetotalnucleotideconcentration(GDP+ATP)shouldbeequaltothatofMgCl2;inthecaseofusingRelQfreeMg2+isnotaconcernsinceduetoRelQ’sinsensitivity.After 2 hours of incubation at 37 °C with constant shaking, nucleotides wereextracted by the addition of acidic phenol (pH 4.5) (Amresco); directly afterphenol addition to the reaction (1:1), the mixture was subjected to vigorousvortexingfollowedbycentrifugation(16krcf,4°C,30min)inordertoseparatethephases.To completely remove the tracesofphenol this stepwas repeated.UpperphasewascollectedandappliedontoMonoQTM5/50GL(GEHealthcare)columnonequilibratedinbufferA(2.5mMTris-HCl(pH8.0),0.5mMEDTAand0.5 mM LiCl in MilliQ H2O). Nucleotides were separated by gradient elutionincreasing concentration of LiCl up to 2M.Wehave collected the ppGpppeakthatelutsat≈250mMLiCl,andthenucelotidewasprecipitatedbytheadditionof1MLiCl(f.c.)and3volumesofcold96%EtOHfollowedincubationeitherat-20 °C (overnight) or at -80 °C (2 hours). The precipitate was vortexed,transferred to centrifugation tubes and pelleted at 16k rcf, 4 °C for 30 min.Supernatantwasdiscarded,thepelletwashedwith-20°C70%EtOHinordertoremove traces of LiCl and the pellet (ppGpp) dissolved in MilliQ H2O. TheconcentrationofppGppwasdeterminedwithextinctioncoefficientEa=13,600μM-1cm-1at252nm.

3

PreparationofRelAandEF-GC-terminally6His-taggedRelA(5)wasoverexpressedinE.coliBL21(DE3)frompET28a plasmid; a colony from overnight grown on LB/Kanamycin agar platewasinoculatedinto3mlofLBmediaandgrownshakingoverdayat37°Cinthepresenceof25µg/mlKntoOD600=0.5.Afterthattheculturewastransferredinto200mloffreshpre-warmed2xYTmedia(with25µg/mlKn)andgrownshakingat37°CtoOD600=0.5followedbytheinductionofRelAoverexpressionbyadding1mMIPTG(Sigma).Straightlyafterinductionthetemperaturewasdecreasedto30°Candbacteriawas left togrowfor1.5hours.Afterwards thebacterialcellmass was collected by centrifugation. For the lysis step bacteria wereresuspended inabuffer, consistingof25mMTris (pH7.5),2mMMgCl2and1mM2-mercaptoethanol (βME)with the addition of 100 μM PMSF and 1 U/mlDNase I (Thermo Scientific). Cells were lysed with Stansted SPCH-10homogenizer and the lysate centrifugedwith following application for Ni-NTAaffinitychromatographyinGEHealthcareHisTrapHPcolumnonÄktaPrimePlus(GEHealthcare).Therunningbufferconsistedof25mMTris(pH7.5-8),1MKCl,350mMNaCl, 2mMMgCl2, 5mM imidazole, 1mM βMEwhile in the elutionbuffer the concentration of imidazole was increased to 300 mM. RelA-correspondent peakwas collected and the protein concentrated using AmiconUltra-15 (Millipore) centrifugal filter devices with a 50 kDa cut-off. The finalpreparations were aliquoted by 15 μl, shock-frozen with liquid nitrogen andstoredat-80 °C inbuffer containing25mMHEPES-KOH,pH7.5,0.7MKCl, 2mMDTT,15mMMgCl2and10%glycerol.N-terminally6His-taggedEF-Gproteinwasoverexpressed inE.coliBL21(DE3)from pQE30 plasmid provided by J. Remme (6). The overnight-grown culturewas transferred into 200ml of fresh 2x YTmedia (with 100 µg/mlAmp) andgrown shaking at 37 °C to OD600=0.5 followed by the induction of EF-Goverexpressionbyadding1mMIPTG(Sigma).After the induction thebacteriawere left to grow shaking for 2more hours at 37 °Cwith the cellmass beingcollectedby centrifugation in the end.Cellswere lysedwith StanstedSPCH-10homogenizerinbufferA(50mMTris-HClpH7.5,10mMMgCl2,200mMPMSFandDNaseI)andclarifiedbycentrifugationat40krpminTi50rotorfor40min(Beckman&Coulter).ClarifiedlysatewasloadedonHisTrapHPNi-NTAaffinitychromatography column (GE Healthcare) and the protein was eluted by agradient of buffer B (A supplemented with 300 mM imidazole). EF-G wasconcentratedandstored-80°Cinbuffercontaining25mMHEPES-KOH,pH7.5,0.7MKCl,2mMDTT,15mMMgCl2and10%glycerol.fMet-tRNAfMetpurificationonBDsepharoseion-exchangecolumnPreparation was done essentially as described in (7). The starting material,deacylatedE.colitRNAifMetwaspurchasedfromChemblock.Theaminoacylationand formylation reactionmixture containing 100 μMdeacylated tRNAifMet, 300μM 3H-methionine (Perkin Elmer), 1 μM E.coli MetRS, 1 μM E.coli formylase(FMT),1mMTHF,2mMATP,3mMPEP,0.05g/lphosphokinase,0.01g/lyeastpyrophosphatase (100 U/mg, Sigma-Aldrich), 0.002 g/l myokinase, 3 mM β-mercaptoethanol,50mMHEPES,50mMKCland10mMMgCl2wasincubatedat

4

37°Cfor30minandthereactionwasstoppedbyadding10mLofcoldbufferA(20mMNaOAcpH5.1,1mMEDTA,5mMBME).100mlBDSepharosecolumnwas equilibrated with 5 CV of buffer A at room temperature. The reactionmixture diluted in 10 mL of buffer A was loaded at speed of 1 ml/min,equilibratedwith bufferA until the baseline, and 3H-fMet-tRNAifMetwas elutedwitha two-stepgradientof thebufferB (B=A+1MNaCl):1.5CV0-50%(B)followed by 2.5 CV 50-100% (B). The fractions containing fMet-tRNAifMet peak(≈95%B) (identified by scintillation counting) was collected and precipitatedwith 3 volumes of EtOH (95%) and 300 mM KOAc at -20°C overnight. Theprecipitatewascentrifugedat20000rpminTi45rotor(Backman&Coulter)for40min,thepelletdissolvedin4mlinKOAc5mM(pH5.1)andre-precipitatedagainwithKOAcandEtOHasstatedabovefor30minat-80°C.Precipitatewascentrifugedat21,000rcffor15mininthebench-topcentrifuge.Thepelletwaswashedwith500μl EtOH70%,dried at room tepmeratrue anddissolved in5mMKOAc(pH5.1)toyielda100μMstocksolution.Concentrationandcharginglevels of fMet-tRNAifMetwere determined spectrophotometrically (1 AU260 = 40ng/μl RNA, Mw tRNA = 25 kDa) and by scintillation counting using 3H-methionineusedforchargingreactionasastandard.PreparationofribosomalinitiationcomplexesWild typeMRE600E.coli andA1067U (providedby J.Remme)70S ribosomeswere prepared as per (7). Model mRNA(MF) 5'-GGCAAGGAGGUAAAAAUGUUCAAA-3' was purchased from Sigma-Aldrich. Toformtheinitiationcomplexes,8μM70Sribosomes,3.1μMIF1,4.4μMIF2,3.1μM IF3, 12 μM mRNA, 12 μM [3H]fMet-tRNAfMet, 1 mM GTP were mixed inHepes:polymixbuffer(finalreactionvolume500μl;2mMDTTand5mMMg2+).Themixturewasincubatedat37°Cfor30minutes,chilledonicefor5minutes,Mg2+concentrationadjustedto15mMandthereactionmixturewasloadedon400 μl sucrose cushion (1.1 M sucrose, 1x Hepes:polymix, 15 mM Mg2+) andcentrifuged at 45,000 rpm in TLS-55 rotor (Beckman & Coulter) for 2 hours,supernatantdiscardedandthepelletdissolvedin200μlofpolymixbuffer(5mMMg2+), aliquoted, snap-frozen with liquid nitrogen and stored at -80 °C.Ribosomal concentrationwas determined by diluting 2 μl of IC 2000 times inwaterandmeasuringtheabsorbanceat260nm:1AU260=23nM70S.[3H]fMet-tRNAfMet concentrationwas determined by scintillation counting. IC occupancywascalculatedasratiobetweenfMet-tRNAifMetand70Sconcentrations.IsolationofpeptidoglycanandUPLCanalysisPeptidoglycan of E. coli was isolated as described previously (8), with minormodifications.E.colistrainsweregrowninMOPSmedia(9),supplementedwith0.4%glucoseand25µg/mlof20aminoacids.600mlofmediawasinoculatedfrom1mlofovernightcultureandgrownat37°CwithvigorousaerationuntilOD600 = 0.5, then the corresponding antibiotics were added for furtherincubation. Cells were harvested by 10 min at 5000 g at room temperature,washed with 10 ml PBS, harvested again and snap frozen in liquid nitrogen.Pelletswerere-suspendedin3mlofPBS,addedtoanequalvolumeof10%SDSinaboilingwaterbathandvigorouslystirredfor3h,thenstirredovernightat37

5

°C.Theinsolublefraction(peptidoglycan)waspelletedat60000rpm,15min,20°CusingaTL100BeckmanrotorinanOptimaMAX-TLultracentrifuge(BeckmanCoulter),andthenwashed3timeswithMQ-water.Samplesweredigestedfor1hat60°CwithpronaseE(100µg/ml)in10mMTris-HClpH7.5andNaCl0.06%(w/v), to remove Braun’s lipoprotein. The reaction was heat-inactivated byaddingSDStoafinalconcentrationof1%(w/v)andboilingfor10mininawaterbath. SDS was then removed by washing 3 times with MQ-water. Purifiedpeptidoglycanwas re-suspended in100µlof50mMNaPO4buffer,pH4.9anddigested overnight at 37 °C with 100 µg/ml muramidase. The digestion wasstoppedby15min incubation inaboilingwaterbath,andcoagulatedproteinswere removed by 10 min centrifugation at 14,000 rpm. The supernatants,containing the digested muropeptides, were reduced by adding 15 µl 0.5 Msodiumborate pH9 and5 µl 2M sodiumborohydride and incubating at roomtemperature for 30 min. Finally, samples were adjusted to pH 3.5 withphosphoricacid.Muropeptideswereseparatedinalineargradientof50mMNaPO4,pH4.35to50mMNaPO4,pH4.95,15%methanol(v/v)in20minusinga0.25ml/minflowonC18 column (150x21, 100Å pore size) column using Acquity UPLC (Waters)(Supplementary Figure 9A). Peptides were detected at Abs 204 nm. Therelativeabundanceoftheindividualmuropeptideswascalculatedasthe%ofthetotalarea.Thevaluesare themeansof twoor three independentexperiments.The analysis resolves monomeric (M), dimeric (D) and trimeric (T)muropeptides, referred to as per Glauner and colleagues (1988) (10): M3:disaccharide tripeptide (GlcNAc-MurNAc-L-Ala-D-Glu-g-mesoDAP); M4:disaccharide tetrapeptide (GlcNAc-MurNAc-L-Ala-D-Glu-g-mesoDAP-D-Ala);D33: LD-crosslinked (DAP-DAP) dimer tetrasaccharide hexapeptide; D34: LD-crosslinked (DAP-DAP) dimer tetrasaccharide heptapeptide; D43: DD-crosslinked (D-Ala-DAP) dimer tetrasaccharide heptapeptide; T444: DD-crosslinked trimer hexasaccharide dodecapeptide. Braun’s lipoprotein boundmonomer(M3-Lpp):GlcNAc-MurNAcL-Ala-D-Glu-g-mesoDAP-e-L-Lys-L-Arg;anddimer (D33L):GlcNAc-MurNAcL-Ala-D-Glu-g-mesoDAP-(e-L-Lys-L-Arg)-g-D-Glu-L-Ala-MurNAc-GlcNAc. The Thr-Arg dipeptide corresponds to the C-terminaldipeptideofE.coliBraun's lipoprotein. IdentitiesofE.coliLCmuropeptidesonUPLCprofilewereidentifiedasper(10)andconfirmedgenetically(11,12).

6

SupplementaryResultsAmpicillin’s molecular target is penicillin-binding proteins, PBPs (13) – DD-transpeptidasesgenerating crosslinksbetweenDaminoacids inpeptidoglycansacculus (14). While the DD crosslinks form the majority of the crosslinks,another type – LD, i.e. between L and D stereocenters of diaminopimelic acid(DAP)–areformedbyampicillin-insensitiveLD-transpeptidases,Ldts(14).HPLCanalysisofpeptidoglycancompositionrevealedthatLDcrosslinksbetweenDAPmoieties in the third position are significantly enriched in bothwild typeand relaxed strains upon antibiotic pre-treatment, especially in the case ofmupirocin and trimethoprim combination (Supplementary Figure 9AB).Remodeling the cell wall depends on both denovo synthesis of peptidoglycanandremodelingoftheexistingone.ThereforeweperformedakineticanalysisofDAP-DAPaccumulationinrelaxedstrainexposedtoeithertomupirocinaloneormupirocintogetherwithtrimethoprim.InthelattercasethefractionofDAP-DAPcrosslinkincreasesmorethansixtimesfrom1.8%priortoantibioticchallengeto11%afterfiveandahalfhours(SupplementaryFigure9C).However,asignificantincreaseintheDAP-DAPcrosslink(twoandahalftimes)isobserveduponmupirocinchallengeofrelaxedstrainwithouttheconcomitantincreaseinampicillintolerance(SupplementaryFigure9D),indicatingthattherelationship between peptidoglycan remodeling and ampicillin tolerance isunlikely to be direct. To directly disprove the causal connection betweenaccumulation of LD crosslinks and antibiotic-induced ampicillin tolerance wetook advantage of bactericidal β-lactam antibiotic imipenemwhich is a potentLdtinhibitor(15):ifaccumulationoftheLDcrosslinksis,indeed,essentialforE.colisurvivaluponPBPinhibitionbyampicillin,thenacombinationofimipenemand ampicillin should result in bacterial lysis. However, trimethoprim andmupirocinshowedthesamerelA-independentprotectiveeffectinkillingassaysemployingsimultaneouschallengebyimipenemandampicillin(SupplementaryFigure9E). This suggests that, first, the pretreatment leads to tolerance to β-lactams in general rather than specifically to ampicillin, and, second, thataccumulationoftheLDcrosslinksisnotthecauseoftheeffect.Finally,weusedan E. colistrain lacking functional Ldt genes ycbB (ldtD) and ynhG (ldtE) andtherefore unable to form DAP-DAP crosslinks (16). Pre-treatment withtrimethoprim combined with either chloramphenicol or tetracycline protectsthis strain from ampicillin, directly confirming that LD crosslinks are notnecessaryfortheeffect(SupplementaryFigure9E).

7

SupplementaryFigure1|Experimentalstrategyforbactericidalantibiotickillings.At OD600 of 0.5 bacteriostatic antibiotics were added at concentrations thatreducethegrowthratetwotimesandincubatedforadditional30minutes.Afterthat bactericidal antibiotic ampicillin was added to final concentration of 200μg/mlandthetimecourseofantibiotickillingwasfollowedbydeterminingthesurvivingfractionusingLBplatingandcolonycount.Theslowerkillingphaseorplateauisdefinedaspersistence,andslowerkillingofthebacterialpopulationasa whole is defined as antibiotic tolerance. If not stated otherwise, theexperimentswereperformedat37°C inMOPSmedia supplementedwith0.4%glucoseand25μg/mlaminoacidsusingBW25113E.colistrains.Theantibioticswereusedatconcentrationscausinggrowthratereductionoftwotimes.

0.5

1.0O

D 600

Time, h

100 x dilutionfrom O/N

0.0

30 min

split the culture forantibiotic pretreatment

causing 50% growth inhibition

10-5

10-4

10-3

10-2

10-1

100

Surviving fraction

543202- 1--3

Ampi

cillin

at 2

00 µ

g/m

l

rapid killing

tolerance

persistence

1

8

Supplementary Figure 2 | Instead of gradual inhibition of growth,mupirocininducesapronouncedlagphaseincultureofB.subtilis.The growth experiments were performed at 37°C in MOPS mediumsupplementedwith0.4%glucoseand25μg/mlofeachaminoacidusingBSB1B.subtiliswildtypestrain.Cellswerepre-growntoOD600of0.5priortoadditionofmupirocinandtheconsequentgrowthwasfollowed.AntibioticconcentrationsinnMareindicatedonthefigures,e.g.mup70correspondsto70nMmupirocin.

mup300 mup140 mup70 control

0.5

0.6

0.7

0.8

0.9

1

OD 6

00

3.02.01.00.0Time, h

9

SupplementaryFigure3|PluronicF-127solubilizesthiostreptonanddoesnotinterferewithEF-GandRelAactivity.(A)EffectTFE(filledblackcircles),DMSO(emptyblackcircles),F-127(filledredcircles)on thiostreptonsolubilityas judgedby theparticlesizemeasuredwithDLS. Analysis has been prepared in triplicates and the error bars indicatestandarddeviation.(B)EF-GGTPhydrolysisactivityisinsensitivetotheadditionof0.1%ofPluronicF-127,eitherinthepresence(redcircles)orabsence(blackcircles)ofvacantwildtype70Sribosomes.(C)RelAppGppsynthesisactivityisinsensitivetotheadditionof0.1%ofPluronicF-127,eitherinthepresence(redcircles)orabsence(blackcircles)ofppGpp.

A B C2500

2000

1500

1000

500

0

Pa

rtic

le s

ize

, n

m

1612840

Thiostrepton, µM

no additives

+ 0.1% F-127

+ 3% DMSO

+ 3% TFE

0.4

0.3

0.2

0.1

0.0

GT

P t

o G

DP

co

nve

rsio

n

121086420

Time, min

- 70S - F-127

- 70S + F-127

+ 70S + F-127

+ 70S - F-127

0.5

0.4

0.3

0.2

0.1

0.0

GD

P t

o p

pG

pp

co

nve

rsio

n

20151050

Time, min

+F-127 + ppGpp

+ F-127 - ppGpp

- F-127 + ppGpp

- F-127 - ppGpp

10

SupplementaryFigure4|TetracyclineandchloramphenicolarepoorandA-site tRNA insensitive inhibitors of RelA’s enzymatic activity in the testtube.Enzymatic activity of 30 nM RelAwas assayed in the presence of 0.5 μM 70Sribosomes, either vacant (empty circles) or programmed with 2 μM poly(U)mRNA and 2 μM deacylated tRNAPhe (filled circles). All experiments wereperformedinHepes:polymixbuffer(15mMMg2+and0.1%PluronicF-127),0.3mM 3H-GDP, 1 mM ATP and 100 μM ppGpp. (A) tetracycline and (B)chloramphenicol. Enzymatic activities (turnovers, ppGppperRelAperminute)were normalized to that of the corresponding system in the absence ofantibiotics.Errorbars represent standarddeviationsof the turnoverestimatesbylinearregression,eachexperimentwasperformedatleastthreetimes.

A B957134tu

rnov

er 2166166tu

rnov

er

wt 70S wt 70S polyU tRNAPhe

1.2

0.8

0.6

0.4

0.2

0.0

Rel

A ac

tivity

10008006004002000Chloramphenicol, µM

1.2

0.8

0.6

0.4

0.2

0.0

Rel

A ac

tivity

5004003002001000Tetracycline, µM

wt 70S wt 70S polyU tRNAPhe

11

Supplementary Figure 5 | Antibiotics targeting translation causeconcurrent inhibition of ppGpp production and increase in the GTP/ATPratioinB.subtilisartificiallystarvedforisoleucinebymupirocin.Bacterialculturesweretreatedbyadditionof70nMmupirocin(designatedbymup70) combined with increasing concentrations of thiostrepton (A),chloramphenicol (B), tetracycline (C) or trimethoprim (D). 30 minutes afteradditionofantibioticssampleswerecollectedandnucleotidelevelsdeterminedby HPLC. Experiments were performed with BSB1 B. subtiliswild type straingrownat37°CinMOPSmediumsupplementedwith0.4%glucoseandafullsetof20aminoacidsat25μg/ml.ppGpplevelsarecalculatedasappGppfractionofcombinedGTP,GTPandppGppnucleotidepool,dashedred trace indicates theppGpplevelsinunstressedcells.DashedblacktraceindicatestheGTP/ATPratioinunstressedcells.Errorbarsindicatestandarderrorofthemean(3-5biologicalreplicates).Asindicatedbybrackets,theP-valueswerecalculatedusingthetwo-tailedWelch’st-testeitherinrelationshiptotheuntreatedcultureorwithinthetitrationseries.

A B

C D

0.001

0.01

0.1pp

Gpp

/ (pp

Gpp

+ GT

P +

GDP)

0 0.1 1 10 100Thiostrepton, nM

2.0

1.0

0.60.4

0.2

GTP/ATP

0.001

0.01

0.1

ppGp

p / (

ppGp

p +

GTP

+ GD

P)

0 1 10 100Trimethoprim, µM

2.0

1.0

0.60.4

0.2

GTP/ATP

0.001

0.01

0.1

ppGp

p / (

ppGp

p +

GTP

+ GD

P)

0 1 10 100Chloramphenicol, µM

2.0

1.0

0.60.4

0.2

GTP/ATP

0.001

0.01

0.1

ppGp

p / (

ppGp

p +

GTP

+ GD

P)

0 0.1 1 10 100Tetracyline, µM

2.0

1.0

0.60.4

0.2

GTP/ATP

GTP/ATP ppGpp unstressed ppGpp unstressed GTP/ATP

*** ****** ***

ns***

** p ≤ 0.01*** p ≤ 0.001

** ***mup70mup70

mup70mup70

12

SupplementaryFigure6|Growthwildtype,ΔrelAandppGpp0E.coliinLBandMOPSmedia.E.coliwild type BW25113 aswell as isogenic ΔrelA and ppGpp0 (ΔrelAΔspoT)strainsweregrownon96-wellplateformat(100µlperwellwithfastshakingat37°C) in filtered LB (A), andMOPSmedium supplementedwith 0.4%glucoseandaminoacidsateither25µg/ml (B)orSerineat400and the restatof theaminoacidsat40µg/ml(C).

A B C

LB

OD 60

0

MOPSaa 40 Ser 400

wt ΔrelA ppGpp0

0.001

0.01

0.1

12108642Time, h

OD 60

0

0.001

0.01

0.1

12108642Time, h

0.001

0.01

0.1

12108642Time, h

OD 60

0

MOPSaa 25

wt ΔrelA ppGpp0

wt ΔrelA ppGpp0

13

SupplementaryFigure7|EffectsofantibioticcombinationsonBW25113E.coligrowth.The growth experiments were performed at 37°C in MOPS mediumsupplemented with 0.4% glucose and 25 μg/ml of each amino acid usingBW25113 E. coli wild-type strain (A and B) and an isogenic relA knock-out(ΔrelA, C and D). Cells were pre-grown to OD600=0.5, antibiotics (mup =mupirocin, cam = chloramphenicol, tet = tetracycline, trim = trimethoprim, +designatesantibioticcombinations)wereadded,andtheconsequentgrowthwasfollowed.AntibioticconcentrationsinμMareindicatedonthefigures,e.g.mup70+ tet2 designates treatment with 70 μM mupirocin and 2 μM tetracycline.Experiments were performed at least three times and error bars indicatestandarderrorofthemean.

A B

C Dwt

ΔrelA

control

mup70

mup70 + cam

mup70 + tet2 mup70 + trim

control

mup70

mup70 + cam

mup70 + tet2 mup70 + trim16

control

trim19

trim16 + cam8

trim16 + tet0.75

trim16 + mup70

control

trim19

trim16 + cam8

trim16 + tet0.75

trim16 + mup705

6

7

8

91

2

3

4

OD

60

0

543210

Time, h

5

6

7

8

9

1

2

OD

60

0

3210

Time, h

OD

60

0

5

6

7

8

91

2

3

4

543210

Time, h

OD

60

0

5

6

7

8

91

2

3

4

543210

Time, h

20

20

16

14

Supplementary Figure 8 | Effects of pre-treatment with individualantibioticsatgrowthsuppressingconcentrationsonampicillintoleranceofwildtypeBW25113E.coli.The antibiotics were used at concentrations causing complete inhibition ofgrowthandmaximumreductionintheppGpplevels(20μMchloramphenicol,2μMtetracyclineand16μMtrimethoprim),exceptformupirocin(70μM),whichwasusedataconcentrationcausingatwo-foldreductionofthegrowthrateandserves as a comparison for the results on Figures 6A and 6B. Antibioticconcentrations in μM are indicated on the figures, e.g. mup70 designates pre-treatmentwith 70μMmupirocin. The antibiotic pre-treatmentwasperformedfor30minutesat37°C inMOPSmediumsupplementedwith0.4%glucoseand25 μg/ml of each amino acid, followed by addition of ampicillin to finalconcentration of 200 μg/ ml. The surviving fraction was determined by LBplatingandcolonycount.Experimentswereperformedatleastthreetimesanderrorbarsindicatestandarderrorofthemean.

10-5

10-4

10-3

10-2

10-1

100

Surv

ivin

g fra

ctio

n

543210Length of ampicillin treatment, h

neg mup70 cam20 tet2 trim16

15

Figure 9 | Accumulation of DAP-DAP crosslinks upon mupirocin andtrimethoprimchallengeisnotresponsibleforβ-lactamtolerance.(A)HPLCanalysis of E. coli peptidoglycan composition was used to resolve monomeric(M), dimeric (D) and trimeric (T) muropeptides. DD-transpeptidases (PBPs)catalyze crosslinking between α-carboxyl of D-Ala in fourth position of thepeptide moiety of the donor monomer and the ε-amine of mesoDAP of theacceptor monomer (D43, D44 and T444). Ampicillin-insensitive LD-

A

B

D

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

4 6 8 10 12 14 16 18 20 22

M3

M4

M3L

D34

D43

D33

D44

D33L T4

44

Retention time, min

A 204

Anhydro muropeptides

D-Ala

GlucNac-MurNAc

L-Ala

D-GluD-DAP-L

GlucNac-MurNAc

L-AlaD-Glu

D--DAP-L

donor muropeptide

acceptor muropeptide

D34DAP-DAP dimer

E

5

4

3

2

1

0

DAP-

DAP

cros

slink

ed d

imer

, %

wt

wt m

70

ΔrelA

ΔrelA

m70

ΔrelA

m70

+ t 16

12

10

8

6

4

2

0

DAP-

DAP

cros

slink

ed d

imer

, %

3002001000 Time, min

ΔrelA m70 ΔrelA m70 + t16

C

neg trim16 + cam8 trim16 + mup70

ΔrelAwt

****

**

**

**

***

**** **

*** p ≤ 0.001** p ≤ 0.01

10-510-410-310-210-1100101

Surv

iving

frac

tion

543210Length of IMP+ AMP treatment, h

10-510-410-310-210-1100101

Surv

iving

frac

tion

543210Length of ampicillin treatment, h

neg trim16 + cam8 trim16 + mup70

ΔldtEΔldtD

*** ************

**

**

******

**

*****

** ******

16

transpeptidases (Ldts) catalyze DAP-DAP between the L- and D-stereogeniccentersof twomesoDAP (D33andD34, see insert). (B)A30minute-longpre-treatmentwithmupirocin(m)aloneand incombinationwith trimethoprim(t)resulted in accumulation of DAP-DAP crosslinks. (C) Longer exposure toantibiotics resulted in DAP-DAP crosslinked dipeptide becoming one of thedominant species. To probe the role of DAP-DAP crosslinked dipeptide inampicillin resistance induced by mupirocin combined with trimethoprim,ampicillin(AMP,200μg/ml)killingwas followed inconditionswhenDAP-DAPformation was abrogated either chemically by addition of saturatingconcentrationsofbactericidalantibioticLdt inhibitor imipenem(IMP,4μg/ml)(D) or genetically by using ΔldtE ΔldtD E. coli strain (E). Antibioticsconcentrations in μM are indicated on the figures, e.g. mup70 designates pre-treatmentwith70μMmupirocin.Errorbarsindicatestandarddeviationofthreeindependentexperiments.Asindicatedbybrackets,theP-valueswerecalculatedusingtwo-tailedWelch’st-testinrelationshiptokillingtimecoursesofuntreatedculture.

17

Supplementaryreferences:1. JelencPC,KurlandCG.1979.Nucleosidetriphosphateregeneration

decreasesthefrequencyoftranslationerrors.ProcNatlAcadSciUSA76:3174-3178.

2. HoggT,MecholdU,MalkeH,CashelM,HilgenfeldR.2004.ConformationalantagonismbetweenopposingactivesitesinabifunctionalRelA/SpoThomologmodulates(p)ppGppmetabolismduringthestringentresponse[corrected].Cell117:57-68.

3. GacaAO,KudrinP,Colomer-WinterC,BeljantsevaJ,LiuK,AndersonB,WangJD,RejmanD,PotrykusK,CashelM,HauryliukV,LemosJA.2015.From(p)ppGppto(pp)pGpp:CharacterizationofRegulatoryEffectsofpGppSynthesizedbytheSmallAlarmoneSynthetaseofEnterococcusfaecalis.JBacterioldoi:10.1128/JB.00324-15.

4. MecholdU,MurphyH,BrownL,CashelM.2002.Intramolecularregulationoftheopposing(p)ppGppcatalyticactivitiesofRel(Seq),theRel/SpoenzymefromStreptococcusequisimilis.JBacteriol184:2878-2888.

5. KnutssonJenvertRM,HolmbergSchiavoneL.2005.CharacterizationofthetRNAandribosome-dependentpppGpp-synthesisbyrecombinantstringentfactorfromEscherichiacoli.FEBSJ272:685-695.

6. PulkA,MaivaliU,RemmeJ.2006.IdentificationofnucleotidesinE.coli16SrRNAessentialforribosomesubunitassociation.RNA12:790-796.

7. AntounA,PavlovMY,TensonT,EhrenbergMM.2004.Ribosomeformationfromsubunitsstudiedbystopped-flowandRayleighlightscattering.BiolProcedOnline6:35-54.

8. CavaF,dePedroMA,LamH,DavisBM,WaldorMK.2011.Distinctpathwaysformodificationofthebacterialcellwallbynon-canonicalD-aminoacids.EMBOJ30:3442-3453.

9. NeidhardtFC,BlochPL,SmithDF.1974.Culturemediumforenterobacteria.JBacteriol119:736-747.

10. GlaunerB,HoltjeJV,SchwarzU.1988.ThecompositionofthemureinofEscherichiacoli.JBiolChem263:10088-10095.

11. MagnetS,BellaisS,DubostL,FourgeaudM,MainardiJL,Petit-FrereS,MarieA,Mengin-LecreulxD,ArthurM,GutmannL.2007.IdentificationoftheL,D-transpeptidasesresponsibleforattachmentoftheBraunlipoproteintoEscherichiacolipeptidoglycan.JBacteriol189:3927-3931.

12. MagnetS,DubostL,MarieA,ArthurM,GutmannL.2008.IdentificationoftheL,D-transpeptidasesforpeptidoglycancross-linkinginEscherichiacoli.JBacteriol190:4782-4785.

13. ZapunA,Contreras-MartelC,VernetT.2008.Penicillin-bindingproteinsandbeta-lactamresistance.FEMSMicrobiolRev32:361-385.

14. TypasA,BanzhafM,GrossCA,VollmerW.2012.Fromtheregulationofpeptidoglycansynthesistobacterialgrowthandmorphology.NatRevMicrobiol10:123-136.

15. MainardiJL,HugonnetJE,RusconiF,FourgeaudM,DubostL,MoumiAN,DelfosseV,MayerC,GutmannL,RiceLB,ArthurM.2007.

18

UnexpectedinhibitionofpeptidoglycanLD-transpeptidasefromEnterococcusfaeciumbythebeta-lactamimipenem.JBiolChem282:30414-30422.

16. SandersAN,PavelkaMS.2013.PhenotypicanalysisofEschericiacolimutantslackingL,D-transpeptidases.Microbiology159:1842-1852.