Understanding
the
crucial
interactions
between
Cytochrome
P450s
and
non-ribosomal
peptide
synthetases
during
glycopeptide
antibiotic
biosynthesis
Madeleine
Peschke
1,
Melanie
Gonsior
2,
Roderich
D
Su¨ssmuth
2and
Max
J
Cryle
1,3,4TheimportanceofCytochromeP450-catalyzedmodifications
ofnaturalproductsproducedbynon-ribosomalpeptide
synthetasemachineriesismostapparentduringglycopeptide
antibioticbiosynthesis:specifically,theformationofessential
aminoacidsidechainscrosslinksinthepeptidebackboneof
theseclinicallyrelevantantibiotics.Thesecyclizationreactions
takeplacewhilstthepeptidesubstrateremainsboundtothe
non-ribosomalpeptidesynthetaseinaprocessmediatedby
aconserveddomainofpreviouslyunknownfunction—the
X-domain.Thisreviewaddressesrecentadvancesin
understandingP450recruitmenttonon-ribosomalpeptide
synthetase-boundsubstratesandhighlightstheimportanceof
bothcarrierproteinsandtheX-domainindifferent
P450-catalyzedreactions.
Addresses
1DepartmentofBiomolecularMechanisms,MaxPlanckInstitutefor Medical Research, Jahnstrasse 29, 69120 Heidelberg, Germany 2Institut fu¨r Chemie, Technische Universita¨t Berlin, 10623 Berlin, Germany
3
EMBLAustralia,MonashUniversity,Clayton,Victoria3800,Australia 4TheMonashBiomedicineDiscoveryInstitute,Departmentof Biochemistry and Molecular Biology and ARC Centre of Excellence in Advanced Molecular Imaging, Monash University, Clayton, Victoria 3800,Australia
Correspondingauthors:Su¨ssmuth,RoderichD ([email protected]) and Cryle, Max J ([email protected])
CurrentOpinioninStructuralBiology2016,41:46–53
ThisreviewcomesfromathemedissueonCatalysisandregulation EditedbyDavidWChristiansonandNigelSScrutton
For a complete overview see theIssueand theEditorial
Available online 9th June 2016
http://dx.doi.org/10.1016/j.sbi.2016.05.018
0959-440X/#2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creative-commons.org/licenses/by-nc-nd/4.0/).
The
role
of
Cytochrome
P450s
in
glycopeptide
antibiotic
biosynthesis
Theuseofglycopeptideantibiotics(GPAs)asantibiotics of last resort against methicillin-resistant Staphylococcus aureus(MRSA)togetherwith theirlong clinicallifetime
has madethesehighly importantcompounds forhuman health[1,2].From astructural viewpoint GPAs havean intriguinglycomplexarchitecture,specificallytheirhighly crosslinked state:GPAs possessthree or four biaryl and biaryl-ethercross-linksbetweenthesidechainsofaromatic amino acids present in their heptapeptide backbone (Figure1)[1,2].Thesecross-linksleadtothecharacteristic three-dimensionalcupshape ofthe GPAaglycone.This structureisessentialfortheiractivity,asitconfersbinding towardstheLys-D-Ala-D-Alamotifpresentinvariousstages
ofthebacterialcellwallbiosynthesis,forexample,LipidII anditsprecursors,viafivehydrogenbonds[2,3].Previously it has been shown that GPAs are assembled by a non-ribosomalpeptidesynthetase(NRPS)[4].Gene inacti-vationstudiesrevealedthatP450monooxygenaseenzymes installthecrosslinkspresentinbothbalhimycin[5,6]and intheteicoplanin-typeGPAA47934[7]withastrictorder ofcyclizationsteps.Thiswasdeterminedbyinactivation ofeachoftheP450s,whichresultedintheaccumulationof intermediatesatadistinctstageof side-chaincyclization duetostallingofthebiosyntheticmachinery.Hence,for thetype-IGPAbalhimycinitcouldbeshownthatthefirst ring-systemto beformedis theC-O-D diaryletherring catalyzedbyOxyB,followedbytheOxyA-catalyzedD-O-E diaryletherring-systemandfinallythebiarylABsystem, which is catalyzed by OxyC (Figure 1) [5,6]. Similar experiments were performed for the teicoplanin-like GPAA47934, throughwhichtheorderofcyclizations
C-O-D!F-O-G(performedbytheenzymeOxyE)!D-O -E!AB could be established (Figure 1) [7]. Further-more,theseresultsindicatedthatP450 monooxygenases wouldfunctionincloseinteractionwiththeNRPS assem-blyline[8].Thisknowledgesetthestageforsubsequent experimentsontheinvitroreconstitutionofthe vancomy-cin P450 monooxygenase OxyB with peptide substrates [9–11,12,13].Here,theconversionoflinearhexa-peptide peptidylcarrierproteinandhepta-peptidepeptidylcarrier protein(PCP)thioesterscould beachieved,whilstlinear heptapeptidesnotpresentedonaPCPdomainwere not suitablesubstrates[11,12].
PCP-driven
P450
recruitment
AminoacidhydroxylationAlongwiththecyclizationreactionsperformedbyP450sin GPAbiosynthesis,someGPAs requireb-R -hydroxytyro-sineasaprecursorforNRPS-catalyzedpeptideassembly [4,14,15]. This hydroxylation is P450 catalyzed and
involvesaseparateNRPSmodulethatgenerates aminoa-cyl-PCP (Tyr-PCP)substrate for this P450, OxyD [4]. ThestructureofOxyD,whilstconformingtothecanonical P450fold,displayssomeunusualfeatures:inparticulara resolved B-C loop, which is typically unfolded in the absenceofsubstrate[16].Biochemicaldatashowingthat thePCP-domainwasthemainsourceofOxyDselectivity ledtothehypothesisthatthisstructuralarrangementwas generating the required PCP-binding surface to allow complexformation(Figure2)[16].Sequencealignments withotheraminoacyl-PCPoxidizingP450sconfirmedthat many residues in this hypothesized interaction surface were highlyconserved,supportingthishypothesis[16].
Structural comparison of a related P450 responsible for
aminoacyl-PCP oxidation from skyllamycin biosynthesis showedahighlysimilartertiarystructuretoOxyDinspite ofthelowidentitiesofthePCPbindingpartners(Figure2) [17]. In contrast to OxyD, P450sky selectively oxidizes
specificPCP-boundAAswithinthemainNRPS machin-ery: despitethismajormechanisticdifference,the PCP-bindinginterfaceremainsessentiallythesameasforOxyD [17,18].StructuralcharacterizationofP450skyintrapped
complex with a skyllamycin PCP-domain was a further
importantstepintounderstandinghowsuchPCPbinding is mediated (Figure 2), which revealed a different CP binding site tothatof theonlyotherexampleof sucha complex solved to date (the P450BioI-ACP complex)
[19,20].IncontrasttotheP450BioI-ACPcomplex,the
majority of interactions are hydrophobic in nature, with hydrophobicpocketsfrombothPCPandP450servingto bind complementary residues from the protein partner [19,20].Also, as the only PCP-bound P450 complex solvedtodate,thisstructurehasalsoprovedhighly valu-abletomodelPCPbindingtoGPAcyclizationP450s.
Peptidecyclization
OxyBvan is the initial peptide cyclization enzyme from
vancomycin biosynthesis and was the first structurally and functionally characterized GPA cyclization enzyme [12,21]. In the structure of OxyBvan, the active site is
highly solvent exposed as the F and Ghelices are held openabovetheactivesite;thestructureisevenmoreopen than those of the aminoacyl-PCP oxidases [21]. Flexible loops around the active site were not resolved in the structure of OxyBvan and thus the question of peptide
orientation is difficult to assess, with peptide soaking experiments not successful to date. Following on from
Figure1
Current Opinion in Structural Biology CI CI CI CI CI CI CI C C C C C C PCP PCP PCP PCP PCP OxyB OxyA OxyC OxyE O O O O O O O O O O O O NH NH sulfatation NH NH NH NH NH O O O O O O O O O OH O H H H A A B B F F G E E D CI NH2 NH2 NH2 NH2 NH2 NH2 HO2C NH2 NH2 H2N H3C HO2C CH3 CH3 CH2OH OH OH OH OH OH B OH OH OH G O F Br Br O OH G OH OH OH OH OH OH OH OH OH OH OH
A. balhimycina: fed with bromine salts
A. balhimycina: insertion of an additional
Hpg module OH HO HO A OH HO HO HO HO O O O O D E CI CI CI E O O O O F O O O S O O O O F G G HO HO HO A47934 HO OH A B OH OH OH OH OH OH HO HO HO HO HO HO HO H N H N H H H H H H H H H H H H H H H H H H H N H N HN HN H N HN H N H N CI C C O D O O O O O O O O O O O OO O O O O A B E OH OH HO HO HO HO HO HO H H H H H N H N H N H N HN H N HN HN H N C C O D E O D A B H H HO H N H N O H H H N HN H N N H N H N H CI C NH NH O OH O D O O O O O O O O OO O HO HO HO HO HO H H H H H N H N HN N H N H NH NH NH N H N H N H O O O O H F G CI NH2 OH OH HO HO H N HN N H NH (a) (b) (c)
Structuraloverviewofglycopeptideantibiotics.(a)OrderofcyclizationreactionsandenzymesinvolvedduringthebiosynthesisofA47934.Altered GPA structures of (b) bromobalhimycin, produced by supplementation of the media with bromine salts, and (c) an octapeptide, produced by an engineered NRPS Amycolatopsisbalhimycinamutant.
invivoexperiments,PCPsubstratesweretestedandshown to be greatly preferred substratesfor OxyBvan over free
peptides [9–11,12]: this established the paradigm that GPAcyclization P450 catalysts accept peptidyl-PCP sub-strates,althoughworkonothersystemscombinedwiththe lackofprogressinestablishingsubsequentstepsinpeptide cyclizationwereacluethatour understandingofpeptide cyclizationin GPAbiosynthesiswasnotcomplete[22].
X-domain
driven
recruitment
TheX-domainA unique feature of the terminal module of all GPA producingsystemsisanadditionaldomain betweenthe
terminalPCPandTe-domains,knownastheX-domain [1,23].PhylogeneticanalysisshowedthattheX-domainis relatedtoanLCLtypecondensationdomain,althoughno
function in peptide synthesis could be assigned [24]. Furthermore,theconservedcatalyticmotifofC-domains isalteredfromHHxxxDGtoHRxxxDD,likely render-ingthisdomaininactive.ThestructureoftheX-domain demonstratestheeffectsofthesemutations[25],whilst confirming that the X-domain indeed conforms to a C-domain type fold (Figure 3): this is described by two structurally similar subdomains that belong to the chloramphenicol-acetyltransferase superfamilyfold con-nectedatthebottomoftheirV-shapedarrangement.In
(a) OxyD P450sky P450sky PCP7sky (b) (c) (d) C C H G G F F I F F E E M J D D L J I I I H G G K K L A A B B A’ A’ K2 K2 B2 β2 β2 β3 β0 β3 β1 β1 α1 α1 α4 α2 α2 α3 α3
Current Opinion in Structural Biology
Structures of the P450s OxyD (a) and P450sky(b) involved in hydroxylation of PCP-bound amino acids, with residues postulated to be involved in thePCP-bindinginterfaceshownincolor;structureofthecomplexofaninhibitorboundPCPdomainandP450skyshowingthepositioningofthe PCPdomain(c)andacloseupviewfocusingonthemajorhydrophobicinteractioninterfacebetweenthetwoproteins(d).
comparison toother C-domainandE-domain structures [26,27,28,29,30,31],atleastfiveadditionalresidues ex-tendthisconnection,whilstcontactbetweensubdomains is established bytwocrossover regionsoriginating from theC-terminalsubdomain: regionI(knownasthefloor) andII(knownasthelatch).Thealteredactivesiteofthe X-domainislocatedintheV-shapedcleft,withresidues R141 and D146 blocking the expected binding site of thePCPdonor-Ppantarmduringpeptidebondformation. In addition, R141 forms a salt bridge to E391 in the C-terminal subdomain, which closes off theactive site. Comparison to theC-domainoftheAB3403NRPS ter-minal module—in which the C-domain interacts with the holo-PCP domain—demonstrates that R23 from
a1-helixoftheX-domainalso wouldblocktheentryof thePpant-linkerfromtheacceptorPCPside[26]. Struc-tural characterization therefore demonstrates that the X-domainisunabletoplayarolein peptideelongation: thus,theroleofthisconserveddomaincontinuedtobea mystery.
P450recruitmentmediatedbytheX-domain
OncebindinginteractionsbetweenGPAcyclizingP450s and theX-domain wererevealed[25],apotentialrole for the X-domain in peptide cyclization became clear. ActivityassaysindicatedthatallOxyenzymestested— except OxyBvan—required the X-domain to catalyze
evenmoderate levelsof peptidyl-PCP cyclization[25]:
Figure3
Current Opinion in Structural Biology
(a) (c) (d) (b) Crossover-II Crossover-II Insertion C-terminal subdomain N-terminal subdomain
X-domain OxyBtei
Cross over-I Cross over-I G F E D β2 β0 α4 α4 α4 α2 β4 β3 β1 C G G H H F F E E D D L L J J I I B K K K2 75º F F P169 E170 A158 R167 D161 D161 D162 A172 R64 R171
StructureoftheX-domainfromteicoplaninbiosynthesisconformstothatofacondensationdomain(a);acomparisonofthestructuresofOxyB from teicoplanin biosynthesis (b) both isolated (green) and in complex with the X-domain (gray with altered regions shown in magenta); structure of the complex of the X-domain and OxyB from teicoplanin biosynthesis (c); and crucial interactions between the X-domain and the aspartate residuesoftheOxyBPRDDmotif(d).
bicyclizationcatalyzedbyanOxyAhomologue[25 ,32, 33]. OxyBvan is clearly an outlier in GPA cyclizing
P450s:recentresults haveshownthatalthoughOxyBvan
can catalyze peptide crosslinking when substrates are presentedsolelyby aPCP-domain, thebindingaffinity and reaction velocity is dramatically increased in the presenceoftheX-domain[11,13,25,32,33,34].Whilst theroleoftheX-domainislikelyconservedinvivo,the propertiesofOxyBvanmakeitanimportanttargetforuse
asaninvitrocatalyst.
ThenatureoftheP450/X-domaininteractionwas identi-fied once a structure of OxyB in complex with the X-domain(bothfromtheteicoplaninsystem)wasresolved (Figure 3) [22,25]. This structure demonstrated that hydrogenbondingandsaltbridgesdominatethe interac-tionbetweentheproteins,andthatonly minorchanges occurtotheirstructuresincomplex(RMSD0.9A˚ — X-domain alone,1.1A˚ —OxyBtei alone).OxyBtei bindsto
theX-domainviaresiduesintheF-,G-,D-andE-helices, thuspresenting itsactivesite towardsthedonor sideof the X-domain. This orientation leaves space for the
peptidyl-PCP substrate, which would be located at the X-domaindonorsideduetoitspositionN-terminalofthe X-domainwithintheNRPSmodule.
TheX-domaininteractioninterfaceinvolvesboth cross-overregionsbetweensubdomainsaswellasthea-4helix and subsequent loop region. To date, only X-domain mutants with substitutions in the loop region (R167A, R171A)have been shownto beessentialfor bothP450 interaction and peptide crosslinking (Figure 3) [25, 33].ThisregioninteractswiththePRDDmotiffound attheN-terminalpartoftheOxyBteiF-helix;suchamotif
isalsopresentin thesamepositioninother structurally characterized Oxy enzymes [21,35–38], which is con-servedamongsttheGPAcrosslinkingP450sandappears tobeafingerprintforX-domainmediatedrecruitment.
Comparison of thestructure of OxyBtei isolated and in
complexwiththeX-domainshowsthatthemainchanges intheP450arelocatedin theF-Ghelices,whichmove slightlydowntowardstheI-helixasaconsequenceof X-domainbinding(Figure3)[22,25].Similartothe Oxy-Bteistructure,themajorityoftheB-C-regioninvolvedin
substratebindingremainsunresolved:thisservesto dem-onstratethattheOxy/X-domaincomplexisarecruitment complex and not a substrate-bound complex, which wouldrequirethepresence ofpeptidyl-PCP.Althougha PCP/Oxystructurehasnotyetbeenachieved,theOxyB/ X-domain complex combined with the PCP/P450sky
structure allows a model of the expected PCP-domain positionfortheGPAcrosslinkingP450stobegenerated [25]. However, alterations in this model for different GPA crosslinkingP450s are to be expected due to the different positions of the crosslinks that these P450s
AteialsoindicatedthattheactivesiteoftheseP450scan
vary significantly between the different catalysts, with OxyAdisplayingfarmorehydrophilicactivesite environ-mentstothoseof OxyB/C(Figure4)[21,22,35,37].
WhyanX-domain?
InGPAbiosynthesis,theX-domainefficientlyconnects NRPS-catalyzedpeptideassemblywithpeptide modifi-cationbyuptofourtailoringenzymesintrans,allofwhich must occur prior to cleavage of the peptide from the NRPSmachinery.Inthiscontext,theconserved‘PRDD’ motifintheF-helixofallGPAcrosslinkingP450enzymes but absent in other related P450s, for example, OxyD, may function as a preselection motif during X-domain recruitmentthatimproves theefficiencyof thepeptide modificationprocess.Incontrastto otherP450enzymes with PCP-bound substrates, the GPA crosslinking cas-caderequirestheefficient selectionof thepeptide sub-strate on the same PCP-domain by multiple, highly related P450s. Recent results indicate that whilst GPA crosslinking state is selected for by the P450s, the X-domainprovidestheplatformfor aconstantshufflingof theGPAcyclizationP450s,whichappearsto maintaina highenoughrateofpeptidecyclizationsoasnottoimpede peptidebiosynthesisbytheNRPS[33].AnalysisofGPA terminalmodulesindicatethattheX-domainwaslikelyto havebeenaC-domainofthesubsequentmoduleinthe ancestorNRPStothatproducingGPAs[24,39],whichhas evolved into aP450 recruitment platform for GPAs (an intact,accessibleC-domainactivesitecouldbothinhibit P450bindingandcausepeptidehydrolysisfromthe PCP-domain;thissuggeststheX-domainactivesitealterations are highly important). As no other NRPSsystems have beenreportedtocontainanX-domain,itappearstobean essential part of the unique GPA cyclization process in coordinating multiple trans interacting proteins towards onePCP-boundpeptidesubstrate.
Structural
implications
for
the
role
of
GPA
P450s
in
vivo
Previously,attemptsweremadetoalterthestructureofthe GPAaglyconesviabiosynthesis(Figure1):thisincluded thereplacementofchlorinesubstituentswithbromineby precursor-directed biosynthesis [40] or the feeding of fluorinatedb-hydroxytyrosinesandvariousphenylglycines toGPAmutantstrainsemployingmutasynthesis[41,42]. Gene inactivation experiments also afforded GPA-like compounds with distorted aromatic side-chain bridging: suchcompounds lost antibacterialactivity [8].Finally, manipulationofthebalhimycinNRPSfroma heptamod-ulartoanoctamodularassemblylinebymoduleinsertion [43] was able to successfully generate an octapeptide: however,concomitantlytheP450monooxygenase proces-singwasalmostcompletelyabolished.Furthermore,while the efforts to synthesize precursor compounds and to reprogramthebiosynthesiswereenormous,inmanycases
asignificantdropinyieldsofthecorrespondingGPAswas observed. The structural findings onthe X-domainas a platformforGPAcross-linkingbyP450shaveshownthat NRPSprocessingisacomplexprocessrequiringa sophis-ticatedinterplaybetweenenzymedomains[25].Afuture goal, thatis,reprogramming ofthe biosynthesisin order to generate new GPA backbones, will require adeeper understanding of theinteraction andinterplayof NRPS domains.Alternatively,restoringtheNRPS-independent oxidative power of P450 monooxygenases from NRPS could bring researchers much closer to generating new glycopeptideantibiotics.
Conclusions
The structuralcharacterizationoftheCytochromeP450 enzymes responsible for GPA cyclization and related processes has greatly improved our understanding of
how the protein-bound substrates of these P450s are selected andtherolethat theseprotein-protein interac-tions play in this process. The characterization of sub-stratecomplexesoftheseP450sisnowahighpriority,as this will not only illuminate the mechanism of these enzymesbutwillalsorevealtheoriginoftheirsubstrate specificity: this information will be vital to the future successofreprogrammingeffortstoproducenovelGPAs.
Conflict
of
interest
statement
Nothingdeclared.
Acknowledgements
ThisworkwassupportedbyMonashUniversity,theEMBLAustralia program,theDeutscheForschungsgemeinschaft(Emmy-NoetherProgram, CR392/1-1),theClusterofExcellence‘UnifyingConceptsinCatalysis UniCat’fundedbytheDFGandcoordinatedbytheTUBerlinandbythe Figure4 C-O-D C-O-D D-O-E D-O-E C-O-D C-O-D AB F-O-G F-O-G F-O-G 7 1 7 7 1 7 1 1
Current Opinion in Structural Biology
StructuralcomparisonsoftheaminoacidsinthevicinityoftheactivesitehememoietiesofOxyBtei(green),OxyAtei(red),OxyEtei(blue)and OxyCvan(purple), revealing significant differences between these P450s.
References
and
recommended
reading
Papersofparticularinterest,publishedwithintheperiodofreview, have been highlighted as:
of special interest of outstanding interest
1. YimG,ThakerMN,KotevaK,WrightG:Glycopeptideantibiotic biosynthesis.JAntibiot2014,67:31-41.
2. ButlerMS,HansfordKA,BlaskovichMAT,HalaiR,CooperMA:
Glycopeptideantibiotics:backtothefuture.JAntibiot2014,
67:631-644.
3. SchneiderT,SahlH-G:Anoldiebutagoodie—cellwall biosynthesisasantibiotictargetpathway.IntJMedMicrobiol
2010,300:161-169.
4.
RecktenwaldJ,ShawkyR,PukO,PfennigF,KellerU, WohllebenW,PelzerS:Nonribosomalbiosynthesisof vancomycin-typeantibiotics:aheptapeptidebackboneand eightpeptidesynthetasemodules.Microbiology2002,
148:1105-1118.
Analyzed the NRPS machinery involved in the biosynthesis of vancomy-cin-type GPAs, including the specificities of four of the NRPS adenylation domains and assigning the NRPS module BpsD to b-hydroxytyrosine formation.
5. BischoffD,PelzerS,HoltzelA,NicholsonGJ,StockertS, WohllebenW,JungG,Su¨ssmuthRD:Thebiosynthesisof vancomycin-typeglycopeptideantibiotics-newinsightsinto thecyclizationsteps.AngewChemIntEd2001,40:1693-1696.
6.
BischoffD,PelzerS,BisterB,NicholsonGJ,StockertS,SchirleM, WohllebenW,JungG,Su¨ssmuthRD:Thebiosynthesisof vancomycin-typeglycopeptideantibiotics—theorderofthe cyclizationsteps.AngewChemIntEd2001,40:4688-4691.
ThefirstassignmentoftherolesandtimingofthethreeOxyenzymesfrom GPAbiosynthesisusingproductanalysisfromgenedisruptionmutantsof abalhimycinproducerstrain.
7.
HadatschB,ButzD,SchmiedererT,SteudleJ,WohllebenW, Su¨ssmuthRD,StegmannE:Thebiosynthesisof teicoplanin-typeglycopeptideantibiotics:assignmentofP450 mono-oxygenasestosidechaincyclizationsofglycopeptideA47934.
ChemBiol2007,14:1078-1089.
Detailed invivogene disruption study assigning the roles of the four Oxy enzymes from the biosynthesis of a teicoplanin-type GPA A47934. 8.
BischoffD,BisterB,BertazzoM,PfeiferV,StegmannE, NicholsonGJ,KellerS,PelzerS,WohllebenW,Su¨ssmuthRD:The biosynthesisofvancomycin-typeglycopeptideantibiotics— amodelforoxidativeside-chaincross-linkingbyoxygenases coupledtotheactionofpeptidesynthetases.ChemBioChem
2005,6:267-272.
FirstevidencegainedfrominvivostudiesthattheOxyenzymesinGPA biosynthesisactonNRPS-boundpeptidesubstrates.
9. WoitheK,GeibN,MeyerO,WortzT,ZerbeK,RobinsonJA:
ExploringthesubstratespecificityofOxyB,aphenolcoupling P450enzymeinvolvedinvancomycinbiosynthesis.OrgBiomol Chem2008,6:2861-2867.
10. GeibN,WoitheK,ZerbeK,LiDB,RobinsonJA:Newinsightsinto thefirstoxidativephenolcouplingreactionduringvancomycin biosynthesis.BioorgMedChemLett2008,18:3081-3084.
11. WoitheK,GeibN,ZerbeK,LiDB,HeckM,Fournier-RoussetS, MeyerO,VitaliF,MatobaN,Abou-HadeedKetal.:Oxidative phenolcouplingreactionscatalyzedbyOxyB:aCytochrome P450fromthevancomycinproducingorganism.implications forvancomycinbiosynthesis.JAmChemSoc2007,
129:6887-6895.
12.
ZerbeK,WoitheK,LiDB,VitaliF,BiglerL,RobinsonJA:An oxidativephenolcouplingreactioncatalyzedbyOxyB,a cytochromeP450fromthevancomycin-producing microorganism.AngewChemIntEd2004,43:6709-6713.
First invitroevidence for Oxy enzyme activity against PCP-bound peptide substratesratherthanfreepeptides.
phenolcouplingreactioncatalyzedbythevancomycin biosyntheticP450enzymeOxyB.AngewChemIntEd2012,
51:11468-11472.
14. PukO,BischoffD,KittelC,PelzerS,WeistS,StegmannE, Su¨ssmuthRD,WohllebenW:Biosynthesisof chloro-beta-hydroxytyrosine,anonproteinogenicaminoacidofthe peptidicbackboneofglycopeptideantibiotics.JBacteriol
2004,186:6093-6100.
15. PukO,HuberP,BischoffD,RecktenwaldJ,JungG,Su¨ssmuthRD, vanPeeK-H,WohllebenW,PelzerS:Glycopeptidebiosynthesis inAmycolatopsismediterraneiDSM5908.Functionofa halogenaseandahaloperoxidase/perhydrolase.ChemBiol
2002,9:225-235.
16.
CryleMJ,MeinhartA,SchlichtingI:Structuralcharacterization ofOxyD,acytochromeP450involvedinb-hydroxytyrosine formationinvancomycinbiosynthesis.JBiolChem2010,
285:24562-24574.
Characterization of the GPA Oxy enzyme responsible for hydroxylation of PCP-boundtyrosinefromvancomcyinbiosynthesis:thisrevealedaPCP interactioninterfaceontheOxyprotein.
17.
UhlmannS,Su¨ssmuthRD,CryleMJ:CytochromeP450sky interactsdirectlywiththenonribosomalpeptidesynthetaseto generatethreeaminoacidprecursorsinskyllamycin biosynthesis.ACSChemBiol2013,8:2586-2596.
P450skyfrom skyllamycin biosynthesis was demonstrated to interact with substrates bound to three different PCP domains, the first such P450 shown to oxidizes amino acids directly during NRPS-catalyzed peptide synthesis.
18. KokonaB,WinesettES,NikolaivonKrusenstiernA,CryleMJ, FairmanR,CharkoudianLK:Probingtheselectivityofb -hydroxylationreactionsinnon-ribosomalpeptidesynthesis usinganalyticalultracentrifugation.AnalBiochem2016,
495:42-51.
19.
HaslingerK,BriekeC,UhlmannS,SieverlingL,Su¨ssmuthRD, CryleMJ:Thestructureofatransientcomplexofa nonribosomalpeptidesynthetaseandacytochromeP450 monooxygenase.AngewChemIntEd2014,53:8518-8522.
Structural characterization of the first complex of a P450 and PCP-domain, which was obtained using PCP-bound P450 inhibitor-based substrate mimics.
20.
CryleMJ,SchlichtingI:StructuralinsightsfromaP450carrier proteincomplexrevealhowspecificityisachievedinthe P450Biol-ACPcomplex.ProcNatlAcadSciUSA2008,
105:15696-15701.
ThefirstexampleofaP450/CPcomplextobestructurallycharacterized: theP450BioI/fattyacyl-ACPcomplexfromtheB.subtilisbiotin biosynth-esispathway.
21. ZerbeK,PylypenkoO,VitaliF,ZhangW,RousetS,HeckM, VrijbloedJW,BischoffD,BisterB,Su¨ssmuthRDetal.:Crystal structureofOxyB,aCytochromeP450implicatedinan oxidativephenolcouplingreactionduringvancomycin biosynthesis.JBiolChem2002,277:47476-47485.
22. HaslingerK,MaximowitschE,BriekeC,KochA,CryleMJ:
CytochromeP450OxyBteicatalyzesthefirstphenolic couplingstepinteicoplaninbiosynthesis.ChemBioChem2014,
15:2719-2728.
23. StegmannE,FraschH-J,WohllebenW:Glycopeptide biosynthesisinthecontextofbasiccellularfunctions.
CurrOpinMicrobiol2010,13:595-602.
24. RauschC,HoofI,WeberT,WohllebenW,HusonD:Phylogenetic analysisofcondensationdomainsinNRPSshedslightontheir functionalevolution.BMCEvolBiol2007,7:78.
25.
HaslingerK,PeschkeM,BriekeC,MaximowitschE,CryleMJ:
X-domainofpeptidesynthetasesrecruitsoxygenasescrucial forglycopeptidebiosynthesis.Nature2015,521:105-109.
This study revealed that the role of the X-domain — which is conserved in the final NRPS module in GPA biosynthesis — is Oxy recruitment to the PCP-bound heptapeptide substrate.
26.
DrakeEJ,MillerBR,ShiC,TarraschJT,SundlovJA,LeighAllenC, SkiniotisG,AldrichCC,GulickAM:Structuresoftwodistinct
conformationsofholo-non-ribosomalpeptidesynthetases.
Nature2016,529:235-238.
DescribesthestructuresofNRPSmodulesindifferentstates,including oneinwhichthePCPdomaininteractswiththecondensationdomain. 27. KeatingTA,MarshallCG,WalshCT,KeatingAE:Thestructure
ofVibHrepresentsnonribosomalpeptidesynthetase condensation,cyclizationandepimerizationdomains.
NatStructMolBiol2002,9:522-526.
28. BloudoffK,RodionovD,SchmeingTM:Crystalstructuresofthe firstcondensationdomainofCDAsynthetasesuggest conformationalchangesduringthesyntheticcycleof nonribosomalpeptidesynthetases.JMolBiol2013,
425:3137-3150.
29.
TanovicA,SamelSA,EssenL-O,MarahielMA:Crystalstructure oftheterminationmoduleofanonribosomalpeptide synthetase.Science2008,321:659-663.
ReportsthestructureofthefirstcompleteNRPSmodule(the4-domain terminalmodulefromthesurfactinNRPS)tobecharacterized. 30. SamelSA,CzodrowskiP,EssenL-O:Structureofthe
epimerizationdomainoftyrocidinesynthetaseA.
ActaCrystD2014,70:1442-1452.
31. SamelSA,SchoenafingerG,KnappeTA,MarahielMA,EssenL-O:
Structuralandfunctionalinsightsintoapeptidebond-forming bidomainfromanonribosomalpeptidesynthetase.Structure
2007,15:781-792.
32.
BriekeC,PeschkeM,HaslingerK,CryleMJ:Sequentialinvitro cyclizationbyCytochromeP450enzymesofglycopeptide antibioticprecursorsbearingtheX-domainfrom
nonribosomalpeptidebiosynthesis.AngewChemIntEd2015,
54:15715-15719.
Revealed the specificity of the second Oxy enzyme, OxyA, using sub-strates containing the X-domain.
33.
PeschkeM,HaslingerK,BriekeC,ReinsteinJ,CryleM: RegulationoftheP450oxygenationcascadeinvolvedin glycopeptideantibioticbiosynthesis.JAmChemSoc2016
http://dx.doi.org/10.1021/jacs.6b00307.
DeterminedthattheOxyenzymescompeteforX-domainbindingandthat thecorrectpeptidestate/Oxypairinducesahigheraffinitycomplexto allowefficientprocessingofthepeptide.
34. BriekeC,KratzigV,HaslingerK,WinklerA,CryleMJ:Rapid accesstoglycopeptideantibioticprecursorpeptidescoupled withcytochromeP450-mediatedcatalysis:towardsa
biomimeticsynthesisofglycopeptideantibiotics.OrgBiomol Chem2015,13:2012-2021.
35. HaslingerK,CryleM:StructureofOxyAtei:completingour
pictureoftheglycopeptideantibioticproducingCytochrome P450cascade.FEBSLett2016,590:571-581.
36. CryleMJ,StaadenJ,SchlichtingI:Structuralcharacterizationof CYP165D3,acytochromeP450involvedinphenoliccoupling inteicoplaninbiosynthesis.ArchBiochemBiophys2011,
507:163-173.
37. PylypenkoO,VitaliF,ZerbeK,RobinsonJA,SchlichtingI:Crystal structureofOxyC,acytochromeP450implicatedinan oxidativeC–Ccouplingreactionduringvancomycin biosynthesis.JBiolChem2003,278:46727-46733.
38. LiZ,RupasingheSG,SchulerMA,NairSK:Crystalstructureofa phenol-couplingP450monooxygenaseinvolvedinteicoplanin biosynthesis.Proteins:StructFunctBioinf2011,79:1728-1738.
39. ChiuH-T,HubbardBK,ShahAN,EideJ,FredenburgRA, WalshCT,KhoslaC:Molecularcloningandsequenceanalysis ofthecomplestatinbiosyntheticgenecluster.ProcNatlAcad SciUSA2001,98:8548-8553.
40. BisterB,BischoffD,NicholsonGJ,StockertS,WinkJ,BrunatiC, DonadioS,PelzerS,WohllebenW,Su¨ssmuthRD:
Bromobalhimycinandchlorobromobalhimycins— illuminatingthepotentialofhalogenasesinglycopeptide antibioticbiosyntheses.ChemBioChem2003,4:658-662.
41.
WeistS,KittelC,BischoffD,BisterB,PfeiferV,NicholsonGJ, WohllebenW,Su¨ssmuthRD:Mutasynthesisofglycopeptide antibiotics:variationsofvancomycin’sAB-ringamino acid3,5-dihydroxyphenylglycine.JAmChemSoc2004,
126:5942-5943.
DemonstratedthattheOxyenzymescancyclizepeptideswithalternate amino acid substituents invivo.
42. WeistS,BisterB,PukO,BischoffD,PelzerS,NicholsonGJ, WohllebenW,JungG,Su¨ssmuthRD:Fluorobalhimycin—anew chapteringlycopeptideantibioticresearch.AngewChemInt Ed2002,41:3383-3385.
43. ButzD,SchmiedererT,HadatschB,WohllebenW,WeberT, Su¨ssmuthRD:Moduleextensionofanon-ribosomalpeptide synthetaseoftheglycopeptideantibioticbalhimycin producedbyAmycolatopsisbalhimycina.ChemBioChem2008,
