DTIC ADA548930: Interaction of the Disordered Yersinia Effector Protein YopE with Its Cognate Chaperone SycE

0.3 MB·English

by  Defense Technical Information Center

11158 Biochemistry2009,48,11158–11160
DOI: 10.1021/bi9017347
Interaction of the Disordered Yersinia Effector Protein YopE with
Its Cognate Chaperone SycE†
XinHu,‡MichaelS.Lee,‡,§andAndersWallqvist*,‡
‡BiotechnologyHPCSoftwareApplicationsInstitute,TelemedicineandAdvancedTechnologyResearchCenter,USArmyMedical
ResearchandMaterielCommand,FortDetrick,Maryland21702and §ComputationalSciencesandEngineeringBranch,
USArmyResearchLaboratory,AberdeenProvingGround,Maryland21005
ReceivedOctober7,2009;RevisedManuscriptReceivedOctober29,2009
ABSTRACT:Wedescribeanefficientapproachtomodelthe specific interactions that are required between the CBD of the
bindinginteractionofthedisorderedeffectorproteintoits Yersinia effector protein YopE (YopE ) and its cognate
CBD
cognatechaperoneinthetypeIIIsecretionsystem(T3SS). chaperone SycE to form a T3SS-competent complex. The
Startingfromdenovomodels,wegeneratedensemblesof structuresoftheunboundchaperoneSycEaswellasthebound
unfolded conformations of the Yersinia effector YopE YopE /SycEcomplexhavebeenexperimentallysolved(9,10),
CBD
usingREMDsimulationsanddockedthemtothechaper- whereasthefreeCBDofYopE(residues20-78)isconsideredto
one SycEusinga multistep protein docking strategy. The beunstructuredinsolution(7).Giventhattheunboundeffector
predictedYopE/SycEcomplexwasingoodagreementwith is in a globular, yet partially unfolded state, is it possible to
the experimental structure. The ability of our computa- computationallysampletheunfoldedconformationssufficiently
tional protocol to mimic the structural transition upon and obtain the near-native conformers suitable for complex
chaperonebindingopensupthepossibilityofstudyingthe prediction?
underlying specificity of chaperone/effector interactions To address this issue, we generated three-dimensional struc-
anddevisingstrategiesforinterferingwithT3SStransport. turesofunboundYopE usingtheRosettadenovofragment
CBD
assembly program (11). The structural models were refined by
ThetypeIIIsecretionsystem(T3SS)utilizedbymanyGram-
energy minimization, rescored with an implicit solvent model,
negative bacteria plays an important role in the pathogenic
and clustered using a pairwise hierarchical method with a
invasion of the host cell by delivering effector proteins directly
structural similarity metric in order to identify a small set of
into the cytosol of the host (1, 2). The delivery mechanism
unique models (12, 13). Analysis showed that the predicted
dependsonaninitialassociationstepoftheeffectorproteinwith
solution structures of YopE were molten globules with a
specific cognate chaperones in the bacterial cytosol (3). This CBD
hydrophobiccore.Moleculardynamics(MD)simulationsindi-
association is characterized by a binding mode whereby the
catedthatthestructureremainedcompactatroomtemperature.
unfolded effector chaperone binding domain (CBD) wraps
WealsocalculatedtheNOEintensitiesofthesestructuralmodels.
around its chaperone in an extended, nonglobular form (4).
The results confirmed that these conformations were qualita-
Though this binding is typically thought of as disordered,
tivelycompatiblewiththepartiallydisorderedstructureindicated
structuralanalysisofanumberofeffectorproteinswithsecretion
by the experimental NOE data (7) (Figure S1, Supporting
chaperonesacrossmanydifferentspecieshasrevealedacommon
structuralbindingpattern,termedtheβ-motif(5).Thenatureof Information).
ToexploretheunfoldedconformationalspaceofYopE ,
thebindingprocessitselfwiththechaperoneistypicallypostu- CBD
weperformedreplicaexchangeMD(REMD)simulationsforthe
latedasanorder-to-disordertransitionsincetheCBDisinitially
modeledstructuresinimplicitsolvent.Thebestpredictedmodels
required to unfold in order to effectively bind to the chaper-
withthelowestscoresinthetoptwomajorclusterswereselected
one(6).Analternativedisorder-to-orderbindingmechanismhas
as representatives for REMD. Ten replicas were used with
been proposed that the native CBD exists in an intrinsically
temperature windows ranging from 270 to 600 K. The results
disordered state but transitions into an ordered state upon
showedthattheglobularconformationdenaturedabove400K,
bindingtoitscognatechaperone(7).
resultinginawiderangeofunfoldedconformations(FigureS2,
Thedynamicsofthedisorderedeffectorproteinandstructural
Supporting Information). These temperature-denatured struc-
transitions associated with chaperone binding are challenging
tures constituted a large ensemble of unfolded conformations
problemstostudybothexperimentallyandcomputationally(8).
of YopE that likely include conformations similar to those
Thougha largenumberof T3SSeffector/chaperonecomplexes CBD
known to bind the chaperone. To examine this hypothesis, we
have been identified, only a few have been structurally deter-
minedsofar(5).Here,weexplorecomputationalapproachesto measuredtheroot-meansquaredeviation(rmsd)comparedtothe
model the disordered effector proteins and predict the binding experimentalstructureofSycE-boundYopECBDandtheradiusof
complex with their cognate chaperones. Our focus is on the gyration (Rg) of the denatured conformations (Figure 1). Re-
markably,asubstantialnumberofstructuralmodelsobtainedat
†ThisworkwassponsoredbytheU.S.DepartmentofDefenseHigh highertemperatureswerefoundtocloselyresemblethewrapped
PerformanceComputingModernizationProgram(HPCMP),underthe YopE conformationofthecrystalstructure.
CBD
HighPerformanceComputingSoftwareApplicationsInstitutes(HSAI) The existence of near-native conformations of unfolded
initiative.
YopE fromREMDindicatedthattemperaturedenaturation
*Correspondingauthor.Phone:[phone]Fax:[phone]. CBD
E-mail:[email protected]. alonecanproduceconformationscommensuratewiththeknown
pubs.acs.org/Biochemistry PublishedonWeb10/30/2009 r2009AmericanChemicalSociety
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it

does not display a currently valid OMB control number.
1. REPORT DATE 3. DATES COVERED
OCT 2009 2. REPORT TYPE 00-00-2009 to 00-00-2009
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Interaction of the Disordered Yersinia Effector Protein YopE with Its
5b. GRANT NUMBER
Cognate Chaperone SycE
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
U.S. Army Medical Research and Materiel Command,Biotechnology REPORT NUMBER
High Performance Computing Software Applications
Institute,Telemedicine and Advanced Technology Research Center,Fort
Detrick,MD,21702
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT
We describe an efficient approach to model the binding interaction of the disordered effector protein to its
cognate chaperone in the type III secretion system (T3SS). Starting from de novo models, we generated
ensembles of unfolded conformations of the Yersinia effector YopE using REMD simulations and docked
them to the chaperone SycE using a multistep protein docking strategy. The predicted YopE/SycE complex
was in good agreement with the experimental structure. The ability of our computational protocol to mimic
the structural transition upon chaperone binding opens up the possibility of studying the underlying
specificity of chaperone/effector interactions and devising strategies for interfering with T3SS transport.
15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE Same as 3
unclassified unclassified unclassified Report (SAR)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
RapidReport Biochemistry,Vol.48,No.47,2009 11159
chaperone-boundstructure.TheseresultssuggestthatREMDis werethendockedtotheunboundformofchaperoneSycEusing
apromisingadjuncttoprotein-proteindockingalgorithmsthat ZDOCK,followedbyrescoringwithZRANK(15).Toimprove
can be used for predicting the interactions of the denatured hitselection,thetopZRANK-scoringdecoyswereassessedon
effector protein binding to its chaperone (14). However, the thebasisofthepresenceoftheconservedβ-motifinteraction(5).
probabilityofoccurrenceofthenear-nativeconformationsinthe Two β-motif interacting residues (hotspots) from the CBD-
simulations was low (<0.1%). Given the large number of binding pocket of chaperone SycE were selected (Leu15 and
denatured conformations, the challenge remains as to whether Val31ofSycEsubunitsAandB).Thepresenceofaninteraction
we can discriminate near-native binding conformations from between these residues and the YopE CBD in the docked
nonbinders. complexwasbasedondistance.Onlydockeddecoysthathave
Hence,wedevelopedastepwisehierarchicalproteinensemble these interactions were retained. This hotspot filtering was
docking strategy to predict the disordered CBD/chaperone required to identify the near-native binding conformations
bindingcomplex.Theoverallcomputationalprocedureissum- from the large pool of docked decoys because the scoring
marized in Figure 2 (for further details, see Supporting Infor- function alone was insufficient. Consequently, a number of
mation).First,weperformedastructuralclusteringanalysisto decoysthatlackedhotspotinteractionswerefilteredout,leaving
extract representative protein structures from the multiple asmallersetofrepresentativeYopE clustersinthetophitlist
CBD
REMDtrajectories.Therepresentativestructuresofeachcluster (Figure2D).
Inthesecondstep,weselectedallstructuralmembersassoci-
ated with the identified top cluster representatives and docked
themtothechaperone.Atotalof350ensembleconformationsof
YopE weredockedtoSycEusingamoreextensivesearching
CBD
byZDOCK,followedbyZRANKrescoringandhotspotfilter-
ing. Inspection of the top decoys showed that the unfolded
effectorYopE boundtochaperoneSycEinthesamemanner
CBD
asthatobservedinthecrystalstructure.Thetop-scoringbind-
ing conformation of YopE had the lowest rmsd of 7.6 A˚
CBD
comparedtothatoftheexperimentalstructure.Interestingly,a
number of binding complexes adopted a symmetric binding
modeinwhichtheeffectorwrappedthedimerchaperonefrom
thebackofthebindingsurface(Figure2E).Thisbindingmodelis
essentially the same as the front binding of the experimental
structurebecausethetwoSycEsubunitsAandBinthedimerare
symmetric.Inaddition,areversebindingmodewasalsofoundin
FIGURE1: Scatterplotoftheradiusofgyration(Rg)vsrmsdforthe whichtheeffectorwrappedaroundthesamechaperonesurface,
unfoldedconformationsofYopE generatedatdifferenttempera-
CBD but the N-terminus bound to the B subunit instead of the A
tures. The rmsd’s (backbone) were calculated with respect to the
X-raystructureofSycE-boundYopE . subunit(Figure2E).Bindingfreeenergycalculationsindicated
CBD
FIGURE2: REMD-based protein ensemble docking approach for effector-chaperone binding prediction. (A) Sequence of the N-terminal
chaperonebindingdomain(CBD)ofYersiniaYopEandpredictedsecondarystructureproperties.(B)3DstructuresofYopE fromtheab
CBD
initio prediction. (C) Unfolded conformation of YopE generated from REMD simulations. (D) Docked complexes ofYopE /SycE
CBD CBD
obtainedfromcluster-baseddocking.Thechaperoneisshownasaribboningreen(subunitA)andcyan(subunitB).Thehitselectedfromhotspot
filteringisshownintheredbox.(E)RepresentativebindingmodesofYopE /SycEcomplexesobtainedfromensembledocking.(F)Refined
CBD
bindingcomplexofYopE /SycEfromMDsimulations.ThepredictedconformationofYopE isshowninmagenta,andtheX-raystructure
CBD CBD
isshowninyellow.
11160 Biochemistry,Vol.48,No.47,2009 Huetal.
simulation matching the bound conformation seems remote.
Furthermore,thenumberofconformationsthatmightneedto
be examined by protein-protein docking would appear to be
incomprehensiblylarge.Whileextensivetestingonabroadrange
ofprotein-proteincomplexeswilldefinitivelyanswerthisques-
tion,therearetwotheoreticalargumentswhichcanprovidesome
optimism.First,therelativefreeenergyofthecorrectunfolded
conformationcannotbelargerthanthecompensatingfreeenergy
duetocomplex formation. Second,theconformationsactually
sampled by an unfolding protein at low temperatures must be
greatlyreducedcomparedtothespaceofallpossibleconforma-
tions (i.e.,nature’ssolution toLevinthal’s paradox).Neverthe-
less,theREMD-basedproteindockingapproachopensupthe
possibility of computationally mimicking and probing inter-
actions between bacterial effector proteins and their cognate
FIGURE3: Interactionsoftheconservedβ-motifofYersiniaeffector chaperones to understand their underlying specificity and to
YopE(stickrepresentationinmagentaandyellow)withitscognate devisepossiblestrategiesforinterferingwithT3SStransport.
chaperoneSycE(ribbonincyanandgreen).Thebindingresiduesof
thechaperonehydrophobicpocketareshownassticksinorange.
ACKNOWLEDGMENT
thatthisreversebindinginteractionwaslessfavoredbecauseof
ComputationaltimewasprovidedbytheU.S.ArmyResearch
different binding interface properties across the chaperone
Laboratory Department of Defense (DoD) Supercomputing
(FigureS3,SupportingInformation).
Resource Center and the Maui High Performance Computing
Finally,werefinedthepredictedbindingcomplexeswithMD
Center.
simulations. This allowed for local side chain and backbone
rearrangements to improve the binding interface contacts. The
SUPPORTINGINFORMATIONAVAILABLE
bestpredictedbindingconformationofYopE wasrefinedto
CBD
armsdof∼5A˚ (Figure2F).Themostsignificantvariationwas Detailed computational methods, REMD simulations, and
located at the terminal regions, which exhibited the largest dockingresults.Thismaterialisavailablefreeofchargeviathe
fluctuations in the simulations. In fact, if the flexible termini Internetathttp://pubs.acs.org.
were excluded, the binding conformation of YopE had a
CBD
rmsd of 2 A˚ . Most importantly, Figure 3 shows that the
REFERENCES
conservedCBDbindingmotifinteractionswerewellreproduced.
Our REMD-based conformational sampling and protein 1.Galan,J.E.,andCollmer,A.(1999) Science284,1322–1328.
2.Cornelis,G.R.,andVanGijsegem,F.(2000) Annu.Rev.Microbiol.
ensemble docking approach were able to mimic the disorder-
54,735–774.
to-order transition seen between an effector protein and its
3.Matsumoto,H.,andYoung,G.M.(2009) Curr.Opin.Microbiol.12,
cognatechaperone.Thistransitionrequiresthatthedisordered 94–100.
hydrophobic core of the compact CBD structure ruptures to 4.Ghosh,P.(2004) Microbiol.Mol.Biol.Rev.68,771–795.
5.Lilic,M.,Vujanac,M.,andStebbins,C.E.(2006) Mol.Cell21,653–
form an extended contact state with the cognate chaperone,
664.
rendering the effector protein transportable through the T3SS. 6.Stebbins,C.E.,andGalan,J.E.(2001) Nature414,77–81.
Fromthetheoryofconformationalselectionuponbinding(16), 7.Rodgers, L., Gamez, A., Riek, R., and Ghosh, P. (2008) J. Biol.
Chem.283,20857–20863.
itislikelythatthefreeCBDoftheeffectorproteinexistsasan
8.Thirumalai, D., and Lorimer, G. H. (2001) Annu. Rev. Biophys.
ensemble of partially disordered conformations in dynamic
Biomol.Struct.30,245–269.
equilibrium. The cognate chaperone binds preferentially to 9.Birtalan,S.,andGhosh,P.(2001) Nat.Struct.Biol.8,974–978.
weakly populated, higher-energy conformations, shifting the 10.Birtalan,S.C.,Phillips,R.M.,andGhosh,P.(2002) Mol.Cell9,
971–980.
equilibrium in favor of effector conformations conducive to
11.Simons, K. T., Kooperberg, C., Huang, E., and Baker, D. (1997)
complex formation. Exactly how this unfolding and binding J.Mol.Biol.268,209–225.
processoccursinnaturecannotbeaddressedhere,butitisworth 12.Feig,M.,Karanicolas,J.,andBrooks,C.L.(2004) J.Mol.Graphics
Modell.22,377–395.
noting that the REMD-based temperature denaturation did
13.Lee, M. S., Bondugula, R., Desai, V., Zavaljevski, N., Yeh, I. C.,
generatethenear-nativeconformationsofeffectorproteinthat Wallqvist,A.,andReifman,J.(2009) PLoSOne4,e6254.
can be docked to the chaperone to closely reproduce the 14.Keskin,O.,Gursoy,A.,Ma,B.,andNussinov,R.(2008) Chem.Rev.
experimentally determined complex. An important concern is 108,1225–1244.
15.Pierce,B.,andWeng,Z.P.(2007) Proteins67,1078–1086.
thegeneralapplicabilityoftheproposedapproachsincenaively,
16.Tsai,C.J.,Ma,B.,andNussinov,R.(1999) Proc.Natl.Acad.Sci.
the possibility of an unfolded conformation from the REMD U.S.A.96,9970–9972.