Essays

Cys-Cys crosslinking shows direct contact between the N-terminus of lethal factor and Phe427 within the anthrax toxin pore

Description
Electrophysiological studies of wild-type and mutated forms of anthrax protective antigen (PA) suggest that the Phe clamp, a structure formed by the Phe427 residues within the lumen of the oligomeric PA pore, binds the unstructured N-terminus of the
Categories
Published
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/50806989 Cys − Cys Cross-Linking Shows Contactbetween the N-Terminus of Lethal Factor andPhe427 of the Anthrax Toxin Pore  Article   in  Biochemistry · March 2011 DOI: 10.1021/bi1017446 · Source: PubMed CITATIONS 8 READS 65 3 authors , including:Blythe E JanowiakSaint Louis University 17   PUBLICATIONS   364   CITATIONS   SEE PROFILE Robert John CollierHarvard Medical School 301   PUBLICATIONS   18,851   CITATIONS   SEE PROFILE All content following this page was uploaded by Blythe E Janowiak on 11 October 2014. The user has requested enhancement of the downloaded file.  Published:  March 22, 2011 r 2011 American Chemical Society  3512  dx.doi.org/10.1021/bi1017446 | Biochemistry   2011, 50, 3512 – 3516 ARTICLEpubs.acs.org/biochemistry Cys  Cys Cross-Linking Shows Contact between the N-Terminus of Lethal Factor and Phe427 of the Anthrax Toxin Pore Blythe E. Janowiak, † Laura D. Jennings-Antipov, ‡ and R. John Collier* Department of Microbiology and Molecular Genetics, Harvard Medical School, 200 Longwood Ave., Boston, Massachusetts 02115,United States  A  nthrax toxin is an ensemble of three proteins that causesymptoms of anthrax. One of the three proteins, protectiveantigen (PA; 83 kDa), serves to transport the other two, lethalfactor (LF; 90 kDa) and edema factor (EF; 89 kDa), to thecytosol, where these factors enzymatically modify intracellularsubstrates, eliciting symptoms of the disease. PA binds to cellularreceptors and, after being proteolytically converted to an active63 kDa form (PA  63 ), oligomerizes, generating heptameric andoctameric pore precursors (prepores). 1,2 The enzymatic  “ cargoproteins ”  LF and EF bind competitively to the prepores, gen-erating complexes that are then endocytosed and tra ffi cked to anacidic compartment. There, under the in fl uence of acidic pH, theprepores transform into mushroom-shaped transmembranepores (channels) capable of translocating bound cargo proteinsto the cytosol. 3 LF and EF bind to PA prepores via their homologousN-terminal domains (LF N  and EF N  , respectively). 4,5  An unstruc-tured segment, containing a high density of charged residues and with an overall positive charge, is present at the extreme Nterminus of both LF N  and EF N  , and there is evidence that thissegment initiates N- to C-terminal translocation through PA pores. 6  8  A crystallographic model of LF N  bound to the octa-meric prepore shows helix   R 1, the  fi rst secondary structureelement of the domain, bound at the entrance to the lumen ata site (the  R  clamp) formed from two adjacent PA  63  subunits. 9 The unstructured segment was not resolved in the crystallo-graphic model, however.Translocation of cargo proteins by the PA  63  pore has beenstudied extensively in planar lipid bilayer systems. 10 PA  63  formsion-conducting pores in these bilayers, and the cargo proteins ortheir N-terminal domains (LF N  is commonly used) bind to thecap of the pores. At symmetrical pH 5.5 and a low ( δ 20 mV)  cis -positive potential, LF N  binds to the pore and blocks ionconductance through it, an activity dependent upon the highly charged N-terminal segment of LF N . Blockage is released upontranslocation, which may be initiated by either introducing a pHgradient or increasing the  cis -positive potential to g 50 mV.Mutation of PA residue F427, which lies within the porelumen, to Ala or many other amino acids causes severe defects inPA-mediated toxicity and in PA-dependent translocation acrossplanar bilayers. 11 Results of site-directed spin-labeling studiesindicate that the F427 side chains of PA oligomers move towardthe axis of symmetry during prepore-to-pore conversion andcome into close proximity with each other, forming a structure,called the Phe clamp, which plays a key role in translocation. 12 IthasbeenhypothesizedthattheLF N Nterminusinteractsdirectly  with the Phe clamp during translocation initiation, 13  but directcontact has not been demonstrated. To test for such contact, weconducted disul fi de trapping experiments, in which we intro-duced a single Cys into the Phe clamp of PA and another on Received:  October 31, 2010 Revised:  March 22, 2011  ABSTRACT:  Electrophysiological studies of wild-type and mutated forms of anthrax protectiveantigen(PA)suggestthatthePheclamp,astructureformedbythePhe427residueswithinthelumenof the oligomeric PA pore, binds the unstructured N-terminus of the lethal factor and the edemafactorduringinitiationoftranslocation.Wenowshowbyelectrophysiologicalmeasurementsandgelshift assays that a single Cys introduced into the Phe clamp can form a disul fi de bond with a Cysplaced at the N-terminus of the isolated N-terminal domain of LF. These results demonstrate directcontact of these Cys residues, supporting a model in which the interaction of the unstructuredN-terminus of the translocated moieties with the Phe clamp initiates N- to C-terminal threading of these moieties through the pore.  3513  dx.doi.org/10.1021/bi1017446 | Biochemistry   2011, 50,  3512–3516 Biochemistry ARTICLE either the N-terminus or the C-terminus of LF N  (Figure 1). Wethen tested the ability of these variants to form an interspeciesdisul fi de bond in planar lipid bilayer measurements and gel-shiftanalyses. ’ EXPERIMENTAL PROCEDURES Materials.  Biochemical reagents were purchased from Sigmaunless indicated otherwise. Oligonucleotides for mutagenesis were synthesized by Integrated DNA Technologies (Coralville,IA).  E. coli  BL21 (DE3) used for expression of proteins wasgrown in ECPM1 medium. 14 Expression and Purification of Proteins.  Recombinant WTPA and PA F427C were overexpressed in the periplasm of   E. coli BL21 (DE3) and purified by anion-exchange chromatography. 15 The PA F427C monomer contained a C-terminal hexa-Hisaffinity tag and was stored in 5 mM tris(2-carboxyethyl)pho-sphine (TCEP) in order to maintain the thiol in the reducedstate.WTLF N  ,LF N  A1C,andLF N R263Cwereoverexpressedin  E. coli  BL21 (DE3) as N-terminal Hexa-His tagged SUMO-fusion proteins and purified by Ni-NTA chromatography fol-lowed by SUMO cleavage, resulting in untagged proteins. 6 Formationand PurificationofHomoheptamericand Het-eroheptamericPA 63 Prepore.  WT PA  63  prepore ([WT] 7 ) wasformed by limited trypsin digestion followed by anion-exchangechromatography. 16 Heteroheptamers with predominantly onemutant subunit containing F427C (Figure 1A,B) were prepared by a similar method as previously reported. 17,18 Briefly, WT PA monomers were mixed with PA F427C (containing a hexa-Hisaffinity tag) monomers in a 40:1 molar ratio. The mixture wasthen nicked with trypsin, and the resulting homoheptamer/heteroheptamer mixture was purified by anion-exchange chro-matography. The preparation was then passed over Ni-NTA resin, and bound heptamers were eluted with a gradient of 40  500 mM imidazole. The product was desalted to removeimidazole and passed over a second Ni-NTA column to ensurethe removal of [WT] 7 . ModificationofMutantProteins. LF N  A1C was alkylated by  bromoacetamide. 6 Briefly, 500  μ M LF N  A1C was fully reduced by incubation with 10 mM dithiolthreitol (DTT) for 15 min atroom temperature (RT) and, after desalting to remove the DTT, was reacted with 50 mM 2-bromoacetamide for 15 min (RT).Sodium 2-mercaptoethanesulfonate (100 mM) was added tostopthereaction, andtheproteins wereagain desaltedto removefree bromoacetamide and sodium 2-mercaptoethanesulfonate.The resulting protein is referred to as LF N -AlC-alkylated(Figure 1C). As a negative control for translocation, LF N R263C was biotinylated and bound to streptavidin, a nontran-slocatable protein, as described 17 (Figure 1C). Electrophysiology.  Planar phospholipid bilayer experiments were performed in a Warner Instruments Planar Lipid Bilayer Workstation (BC 525D, Hamden, CT). Planar bilayers werepainted 19 onto a 200  μ m aperture of a Delrin cup in a Lucitechamber,with3%1,2-diphytanoyl- sn -glycerol-3-phosphocholine(DPhPC)in n -decane(Avanti PolarLipids,Alabaster,AL).1mLaliquots of buffer were added to the cup and the chamber, and both compartmentswere stirred continuously.  Cis (side to whichPA prepore and LF N  were added) and  trans  compartmentscontained 100 mM KCl, 1 mM ethylenediaminetetraacetic acid(EDTA), and 10 mM each of sodium oxalate, potassiumphosphate, and 2-(  N  -morpholino)ethanesulfonic acid (MES),pH 5.5.Once a membrane was formed in the planar lipid bilayersystem, PA prepore (25 pM) was added to the  cis  compartment, which was held at a  V  m  =  þ 20 mV with respect to the  trans compartment. After appropriate current increase, the  cis  com-partment was perfused with  ∼ 10 mL of non-PA-containing bu ff  er at a  fl ow rate of   ∼ 3 mL/min to remove any free PA.Once the current was constant, LF N  was added to the  cis compartment (1 mg/mL), and binding to PA channels wasmonitored by the decreasein conductance. The  cis  compartment was perfused again after LF N  addition to eliminate free ligand.The total time between LF N  addition and initiation of transloca-tion was kept constant at 10 min, unless otherwise noted.Translocation was initiated by raising the pH of the  trans compartment to pH 7.2 with 2 M KOH, while maintaining the cis  compartment at pH 5.5. Experiments for each PA proteintested were normalized to control experiments where KOH wasaddedtothe trans compartmentintheabsenceofLF N inordertoadjust for current changes due to salt addition alone. Transloca-tion was monitored at  V  m  = þ 20 mV by the rise in current. Inparallel experiments to translocation, PA channels blocked by LF N  or an LF N  mutant construct were unblocked by reversingthe membrane potential from  V  m  = þ 20 mV to  V  m  =  20 mV.Thefraction ofLF N  that wasreversedwas monitored over 1min. Gel-Shift Analysis.  WT or [WT] 6 [F427C] 1  PA heptamer(100  μ g) was mixed with an equimolar amount (7.3  μ g) of LF N  WT,LF N  A1C,LF N  A1C-alkylated,orLF N R263Cinavolumeof 50  μ L. The mixtures were incubated for 30 min at pH 8.5 with Figure 1.  Schematic of proteins used in this study. (A) Space- fi llingmodel ofheteroheptameric PAprepore,basedonthecrystal structureof [WT] 7  prepore (PDB 1TZN 21 ), showing a single subunit (dark gray)containing F427C ( f ) and a C-terminal hexa-His tag (green circles).(B) Membrane-inserted model of heteroheptameric PA pore based onthe 3D reconstruction of negatively stained EM particles 22 showing asingle mutated F427C (yellow star) at the predicted location of the Phe-clamp. Domain 4 was not resolved in the EM reconstruction, and adotted oval is shown where one domain 4, with its C-terminal hexa-Histag (green circles) attached, is expected to be located. (C) Schematic of the four mutated LF N  constructs used in the study. The cysteinesubstitution is shown as a yellow star, alkylation of the cysteine isrepresented by a green circle surrounding the yellow star, and biotin  streptavidin linkage is shown as a pink circle and a cyan ribbon diagram,respectively. The LF N  models are based on the crystal structure of LF(PDB1J7N 5 ),andthestreptavidinmodelisbasedonitscrystalstructure(PDB 1MEP 23 ).  3514  dx.doi.org/10.1021/bi1017446 | Biochemistry   2011, 50,  3512–3516 Biochemistry ARTICLE 10 mM DTT to limit nonspecific disulfide bond formation.Samples were then desalted using Zeba (Thermo Scientific)desaltingspincolumnsintobuffer(20mMTris,pH8.5,150mMNaCl)withorwithout15mMDTT.ThepHwasdroppedtopH5.5withtheadditionof6.8  μ Lof100mMHCl(samplevolume=60  μ L), andthesamples wereincubatedfor1hatRTtoallowforoxidation. 2-Bromoacetamide was then added to a final concen-tration of 5 mM to quench any unreacted cysteines, and thereaction was allowed to proceed for 10 min at RT. Trichloroa-cetic acid (TCA) was then added to precipitate the proteins, andthe samples were incubated on ice for 10 min. Protein pellet wascollected by centrifugation and washed twice with acetone toremove TCA. Following the final acetone wash step, samples were incubated for 5  10 min in a 95   C heat block to evaporateany remaining acetone. The protein pellets were then resus-pended in SDS sample buffer and boiled for 10 min before beingloaded onto an SDS-PAGE gel. Results were visualized by Coomassie staining of the resulting gel.To con fi rm the identity of the bands observed in the gel-shiftanalysis, we performed a parallel analysis in which we probed the bands with an antibody speci fi c to LF N . Speci fi cally, we mixedeither WT or [WT] 6 [F427C] 1  PA heptamer (50  μ g) with anequimolaramount(3.8  μ g)ofeitherWTLF N orLF N  A1C,inthepresence of 10 mM DTT, in a total volume of 100  μ L. Afterallowing time for the formation of the PA prepore  LF N  com-plex, the samples were desalted against bu ff  er containing 20 mMTris, pH 8.5, 150 mM NaCl, with or without 15 mM DTT. After bu ff  er exchange, the pH of the solutions was dropped to pH 5.5 by addition of 100 mM HCl to mimic the conditions of theendosome, and the samples were incubated for 1 h at RT topromote oxidation. After oxidation, a 20  μ L aliquot of eachsample was applied to a 4  20% acrylamide SDS-PAGE gel.Following electrophoresis, the proteins were transferred to anitrocellulosemembraneusinga25Velectrophoresisfor90min.The membranes were then probed with a polyclonal antibody raised in goats against the N-terminal domain of LF. ’ RESULTS AND DISCUSSION To test for direct contact between the Phe clamp of the PA  63 pore and the LF N  N terminus, we  fi rst created a heterohepta-mericformoftheporecontainingtheF427Cmutationinasinglesubunit ([WT] 6 [F427C] 1  , Figure 1A,B). We then tested whether this variant could form a disul fi de bond with eitherLF N  A1C or LF N  R263C, residue 263 representing the Cterminus (Figure 1C). Earlier reports suggested that the LF N N-terminus enters the pore lumen during translocation initia-tion, whiletheLF N  C-terminusremains outside. 8  Ascontrols, weconstructed an LF N  A1C variant in which the Cys thiol wasalkylated to prevent disul fi de bond formation (LF N  A1C-al-kylated) and an LF N  R263C variant in which a biotin wasattached to the Cys. Binding of streptavidin to the biotinylatedCys yielded the complex, LF N  R263C-biotin:streptavidin, whichis known not to translocate 8,17 (Figure 1C).To test for disul fi de bond formation between PA [WT] 6 -[F427C] 1 andtheLF N  variants,we fi rstusedelectrophysiologicalassays in planar lipid bilayers. Brie fl  y, channels were formed withPA [WT] 7  or PA [WT] 6 [F427C] 1  at pH 5.5 and  V  m  =  cis -positive 20 mV, and an LF N  variant was then added to the  cis compartment, causing a decrease in ion conductance. Oxidation wasallowedtoproceedfor10minatRTbeforeintroducingapH Figure 2.  Translocation of various LF N  constructs through either (A) WT PA channels or (B) heteroheptamer F427C channels. Macroscopicconductance was measured at symmetrical pH of 5.5 and  V  m  =  þ 20 mV. At time 0, translocation was initiated by adding 2 M KOH to the  trans compartmenttoraisethepHto7.2;therewasan ∼ 20smixingdelayinthissystemwithbothcompartmentscontinuouslystirred.Representativedataareshownfrom n g 3trials.The fi  veLF N constructswereasfollows:WTLF N (black),LF N  A1C(blue),LF N  A1C-alkylated(green),LF N R263C(red),andLF N  R263C   biotin  streptavidin (orange). Figure 3.  Kinetics of interactions between PA  63  [WT] 6 [F427C] 1  andLF N  A1C. Three parameters were measured over time: occlusion of thePA [WT] 6 [F427C] 1  channel by LF N  A1C (triangles), translocation of LF N  A1C through the PA [WT] 6 [F427C] 1  channel after the pHgradient was applied (squares), and relief of blockage of the PA [WT] 6 [F427C] 1  channel by LF N  A1C after reversal of the membranepotential (circles). At each time point, the conductance was measuredover a period of 1 min and normalized to account for total PA channels.Each point represents an average of three independent experiments, with standard deviation shown as error bars.  3515  dx.doi.org/10.1021/bi1017446 | Biochemistry   2011, 50,  3512–3516 Biochemistry ARTICLE gradient to initiate translocation, detected by restoration of current. With channels formed from WT PA  63  , all of the LF N  variants translocated at essentially the same rate and e ffi ciency as WT LF N  , except for the LF N  R263C   biotin  streptavidincomplex,anegativecontrol(Figure2A).However,withchannelsformed from [WT] 6 [F427C] 1  , translocation of LF N  A1C was virtually nil, whereas WT LF N  and LF N  R263C retained theability to translocate (Figure 2B). LF N  A1C in which the thiolhad been alkylated also translocated e ffi ciently through F427C-containing channels. These  fi ndings supported the hypothesisthat the loss of translocation resulted from the formation of adisul fi de bond between the cysteine of LF N  A1C and that of thePhe clamp of [WT] 6 [F427C] 1  (Figure 2B). Curiously, additionofDTTdidnotalleviatetheblockageoftranslocation,suggestingthat the disul fi de is protected from reduction, perhaps by shielding within the hydrophobic environment of the Phe clamp. While cross-linked dimers of LF N  are known not to translocatethrough PA channels, 20 the decreased translocation rate seen with[WT] 6 [F427C] 1  and LF N  A1C was almost certainly not caused by LF N  dimerization because both LF N  A1C and LF N  R263C translo-cated e ffi ciently through WT channels (Figure 2A). Presumably,disul fi de formation did not occur with LF N  R263C duringtranslocation because contact of this residue with F427C wastransient and brief. As shown in Figure 3, LF N  A1C lost the ability to translocateunder the in fl uence of a pH gradient within seconds to a few minutes after the protein bound to [WT] 6 [F427C] 1  channels.The ability to relieve channel blockage by reversing the polarity of the transmembrane potential was also lost with approximately the same kinetics. The regain of conductance upon polarity reversal is rapid and thought to result from electrophoretic withdrawal of the unstructured, positively charged N terminusof LF N  via the mouth of the pore. These  fi ndings support thehypothesisofadisul fi debridgeformingbetweentheCysresiduesof the mutated LF N  and the heteroheptameric pore. As an orthogonal test of disul fi de bond formation between thePA [WT] 6 [F427C] 1  pore and LF N  A1C, we conducted gel shiftassays. Brie fl  y, [WT] 7  or [WT] 6 [F427C] 1  PA  63  heptamers weremixedwithWTLF N  ,LF N  A1C,LF N  A1C-alkylated,orLF N R263CatpH8.5inthepresenceofDTTtolimitnonspeci fi cdisul fi debondformationduringthebindingstep.DTTwasthenremoved,andthepH wasdroppedto pH5.5to promoteconversionto thepore. Thesamples were allowed to incubate 1 h at RT before being treated with 2-bromoacetamide, precipitated with TCA, and analyzed by SDS-PAGE (Figure 4). Two distinct high molecular weight bands were observed with [WT] 6 [F427C] 1  þ  LF N  A1C (Figure 4,lane1).Wehypothesizedthatthesebandscorrespondedtodisul fi decross-linked PA  63  LF N  and disul fi de cross-linked PA  63  dimers,respectively. In support of this hypothesis, both high molecular weight bands were eliminated by the addition DTT (Figure 4,lane 2), and only the upper band (corresponding to cross-linkedPA   PA) waspresentwith[WT] 6 [F427C] 1 þ LF N  WT(Figure4,lane 3); this band was also eliminated by the addition of DTT(Figure 4, lane 4). As expected, and in agreement with our planarlipid bilayer data, the PA   LF N  cross-link was only observed with[WT] 6 [F427C] 1  and LF N  A1C; no PA   LF N  cross-link was seen with WT LF N  , LF N  A1C-alkylated, or LF N  R263C (Figure 4, lanes5, 6, and 7).To con fi rm the identities of the bands, we transferred theproteins to nitrocellulose and performed Western blots using anantibody speci fi c to LF N  (Figure 5). As predicted, the proposedPA   LF N  band was present in only in the lane that contained bothPA [WT] 6 [F427C] 1  and LF N  A1C without DTT (Figure 5A). Additionally,abandconsistentwithadisul fi deformingbetweentwomoleculesofLF N  A1C(LF N  LF N )wasonlyobservedinlanesthatcontained LF N  A1C (Figure 5A,C), and that band was reduceddramatically in the presence of 15 mM DTT. Together, these dataindicate that a disul fi de cross-link forms selectively between PA [WT] 6 [F427C] 1  pore and LF N  A1C.In summary, our results in arti fi cial membranes and by gel-shiftanalysis support the hypothesis that there is a direct interaction between the N-terminus of LF N  and the Phe clamp of the PA poreduring initiation of translocation and thereby validate the conceptthat the N-terminus of LF N  plays an important role in initiatingtranslocationthroughthePApore.Theseresultsarecomplemented by two recent reports: (i) site-directed spin labeling studiesdemonstrating proximity of the N terminus of bound LF to thePhe clamp of the PA pore, 13 and (ii) the crystal structure of LF N  bound to the octameric prepore showing, as mentioned above, thathelix  R 1 of the bound cargo domain is bound within the mouth of the pore. The latter report localizes the unstructured N terminus tothe vicinity of the Phe clamp. 9 ’ AUTHOR INFORMATION Corresponding Author *Tel: (617) 432-1930. Fax: (617) 432-0115. E-mail: jcollier@hms.harvard.edu. Figure 4.  Coomassie-stained gel shift analysis shows formation of adisul fi de cross-link between PA [WT] 6 [F427C] 1  and LF N  A1C. Lanes:1,PA[WT] 6 [F427C] 1 þ LF N  A1C;2,PA[WT] 6 [F427C] 1 þ LF N  A1C þ DTT; 3, PA [WT] 6 [F427C] 1 þ LF N  WT; 4, PA [WT] 6 [F427C] 1 þ LF N  WT þ DTT;5,PAWT þ LF N  A1C;6,PA[WT] 6 [F427C] 1 þ LF N  A1C-alkylated; 7, PA [WT] 6 [F427C] 1 þ LF N  R263C. DTT was added,as shown, to 15 mM in all samples that included DTT. Figure 5.  Western blot probed with an anti-LF N  antibody to revealLF N -containing bands resulting from a gel shift analysis. Lanes: A, PA [WT] 6 [F427C] 1 þ LF N  A1C; B, PA [WT] 6 [F427C] 1 þ LF N  WT; C,PAWT þ LF N  A1C;D,PAWT þ  WTLF N .DTTwasadded,asshown,to 15 mM in all samples that included DTT.
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x