Health & Lifestyle

A natively unfolded yeast prion monomer adopts an ensemble of collapsed and rapidly fluctuating structures

Description
A natively unfolded yeast prion monomer adopts an ensemble of collapsed and rapidly fluctuating structures
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
  A natively unfolded yeast prion monomer adoptsan ensemble of collapsed and rapidlyfluctuating structures Samrat Mukhopadhyay*, Rajaraman Krishnan † , Edward A. Lemke*, Susan Lindquist †‡ , and Ashok A. Deniz* ‡ *Department of Molecular Biology, The Scripps Research Institute, La Jolla, CA 92037; and  † Whitehead Institute for Biomedical Research,Cambridge, MA 02142Contributed by Susan Lindquist, December 22, 2006 (sent for review December 5, 2006) The yeast prion protein Sup35 is a translation termination factor,whoseactivityismodulatedbysequestrationintoaself-perpetuatingamyloid.Theprion-determiningdomain,NM,consistsoftwodistinctregions: an amyloidogenic N terminus domain (N) and a chargedsolubilizing middle region (M). To gain insight into prion conversion,we used single-molecule fluorescence resonance energy transfer(SM-FRET) and fluorescence correlation spectroscopy to investigatethe structure and dynamics of monomeric NM. Low protein concen-trations in these experiments prevented the formation of obligateon-pathway oligomers, allowing us to study early folding interme-diatesinisolationfromhigher-orderspecies.SM-FRETexperimentsonadual-labeledamyloidcorevariant(N21C/S121C,retainingwild-typeprionbehavior)indicatedthattheNregionofNMadoptsacollapsedform similar to ‘‘burst-phase’’ intermediates formed during the fold-ing of many globular proteins, even though it lacks a typical hydro-phobiccore.Themeandistancebetweenresidues21and121was  43Å.Thisincreasedwithdenaturantinanoncooperativefashionto  63Å, suggesting a multitude of interconverting species rather than asmall number of discrete monomeric conformers. Fluorescence cor-relation spectroscopy analysis of singly labeled NM revealed fastconformational fluctuations on the 20- to 300-ns time scale. Quench-ing from proximal and distal tyrosines resulted in distinct fast andslowerfluctuations.OurresultsindicatethatnativemonomericNMiscomposed of an ensemble of structures, having a collapsed andrapidly fluctuating N region juxtaposed with a more extended Mregion. The stability of such ensembles is likely to play a key role inprion conversion. amyloid    conformational fluctuation    single-molecule fluorescence   yeast prion T he prion hypothesis (1), involving self-replicating or infectiousprotein conformations, has attracted broad interest in recenttimesduetoitsroleinthebiologyofdebilitatingneurodegenerativediseases (1, 2), protein-based inheritance of novel phenotypes in yeast (3–5), and (potentially) long-term memory (6, 7). The  Sac- charomycescerevisiae translationalterminationfactor,Sup35,isonesuchproteincapableofswitchingtoaself-perpetuatingstate.Intheprion state [  PSI   ], the glutamine/asparagine (Q/N)-rich priondomain of Sup35 is sequestered into an amyloid conformer, reduc-ingtheefficienciesoftranslationtermination(8).Thisswitchcausesribosomes to read through stop codons at biologically significantrates, changing a multitude of phenotypes (9).The NM segment (253 residues) of Sup35 determines the prionstate and comprises two distinct regions. The N-terminal region(residues 1–123) is abundant in uncharged polar amino acids(glutamines, asparagines, and tyrosines), and forms the major partof the amyloid core that directs the protein into the [  PSI   ] prionstate. The highly charged middle region M (residues 124–250)confers solubility  in vitro  and  in vivo , allowing the protein to existin the non-prion [  psi  ] state. In the prion state, the N region adoptsa   -sheet-rich conformation, whereas the M region remains rela-tively unstructured (10).Structural studies on NM amyloids have provided severalinsights into the molecular basis of prion nucleation (11–14). Anearly step is the establishment of an equilibrium between pre-dominantly unstructured NM polypeptide monomers and mol-ten oligomeric intermediates that are obligate on-pathway spe-cies (14–16). The structure and dynamics of early monomericintermediates are of considerable interest in deciphering themolecular mechanism of amyloid formation (17, 18). However,the heterogeneity and transient nature of these partially foldedintermediates impede structural characterization by steady-statebulk measurements, which provide ensemble-averaged informa-tion for monomers and oligomers.In recent years, single-molecule measurements have proven veryuseful in studying such complex systems, which are expected todemonstrate static and/or dynamic conformational heterogeneity(19–25). These experiments permit direct observations and quan-tificationofintermediates,theirstructuraldistributionsanddynam-ics during assembly. Furthermore, these studies can be carried outat subnanomolar protein concentrations, minimizing oligomeriza-tion or aggregation.Single-molecule FRET (SM-FRET) can be used to estimateintramolecular distances (  r  ) of individual molecules, fluorescentlylabeled with donor and acceptor dyes, from the FRET-efficiency(  E ), by using Fo¨rster’s equation  E   1    r   /   R 0  6   1 ,  [1]  where  R 0  is the Fo¨rster’s distance (at which  E    0.5) for a givendye pair (20, 21). Although very powerful as commonly practiced, SM-FRETdoes not probe conformational dynamics occurring faster than  50  s. Fluorescence correlation spectroscopy (FCS) is a closelyrelated single-molecule or small-ensemble method that can beused to measure such faster fluctuations via an autocorrelationanalysis of conformationally coupled fluorescence intensity fluc-tuations (26–28). In this work, we describe single-moleculestudies on a monomeric amyloidogenic protein, employing bothSM-FRET and FCS to provide insights into the structure anddynamics of monomeric NM. Results Ensemble FRET Measurements.  Our previous work indicates that theregion from amino acids   20 to 120 encompasses the amyloid core Author contributions: S.M., R.K., S.L., and A.A.D. designed research; S.M., R.K., and E.A.L.performed research; S.M., R.K., E.A.L., S.L., and A.A.D. contributed new reagents/analytictools;S.M.,R.K.,E.A.L.,S.L.,andA.A.D.analyzeddata;andS.M.,R.K.,S.L.,andA.A.D.wrotethe paper.The authors declare no conflict of interest.Abbreviations:SM-FRET,single-moleculeFRET;FCS,fluorescencecorrelationspectroscopy. ‡ To whom correspondence may be addressed. E-mail: deniz@scripps.edu or lindquist    admin@wi.mit.edu.This article contains supporting information online at www.pnas.org/cgi/content/full/ 0611503104/DC1.© 2007 by The National Academy of Sciences of the USA www.pnas.org  cgi  doi  10.1073  pnas.0611503104 PNAS    February 20, 2007    vol. 104    no. 8    2649–2654       B      I      O      P      H      Y      S      I      C      S  intheprionstateofNMfibersassembledatroomtemperature(13).Therefore,toprobethestructureanddynamicsofthisregioninthemonomeric non-prion form, a double cysteine variant (N21C,S121C) was generated by site-directed mutagenesis to provide sitesforattachmentoffluorescentlabels(Fig.1  a ).Thismutantwasusedto replace the NM portion of the wild-type  SUP35  gene  in vivo ,resulting in a single, full length functional  SUP35  gene. This variantretained its capacity to support both the prion [  PSI   ] and thenon-prion [  psi  ] states. Next, the dual cysteine variant was ex-pressed in and purified from  Escherichia coli . This variant sponta-neously assembled into amyloid at the same rate as wild-type NMunder reducing conditions, and labeling with a variety of fluores-centprobeshadnoeffectonitsassemblykinetics(datanotshown).Having established that the cysteine variant behaves like the wild-type protein both  in vivo  and  in vitro , we labeled it with AlexaFluor 488 (donor) and Alexa Fluor 594 (acceptor) for FRETexperiments (Fig. 1  a ). As a prelude to SM-FRET experiments, weperformed ensemble FRET measurements on the protein. Steady-state fluorescence spectra were recorded using donor excitation at488 nm under denaturing and native conditions during the lagphase,beforeamyloidassembly.Underdenaturingconditions(6MGdmCl), the acceptor emission was very weak (Fig. 1  b ), consistent with a low FRET efficiency expected for a dye-pair with a Fo¨rsterdistance of    54 Å, separated by 100 amino acid residues in apolypeptide chain. Immediately upon dilution into buffer, emissionfrom the acceptor increased significantly with a concomitant de-crease in donor signal, indicative of a significant increase in theFRET-efficiency and a decrease in the inter-dye distance in thenative protein.This observation was consistent with our previous investigationofassemblykineticsusingsteady-statefluorescencemeasurements, wherein acrylodan-labeled core mutants (amino acids   20–120)showed an immediate increase in fluorescence intensity and aconcomitantblueshiftinwavelengthbeforeanyamyloidformation,suggestive of a collapsed intermediate (13). However, under boththese experimental conditions (0.1   M to low micromolar concen-trations), NM undergoes partial conversion to oligomeric speciesthatareon-pathwayforassembly.Ensembledatacannotdistinguish whether the blue shift in acrylodan fluorescence in our previousexperiments or the increased FRET signal in our present experi-ments arise from intermolecular interactions (due to the formationof oligomers), or intramolecular energy transfer (due to proteincompaction), or both. Furthermore, they provide no informationabout the possible existence of conformational subpopulations andthe dynamics of their interconversion. Hence, we next used single-molecule techniques, which allow us to distinguish between thesepossibilities and to observe the conformational distributions of proteins at very low concentrations (50–200 pM). Under theseconditions, most proteins are not expected to form any aggregates,or to form aggregates only very slowly. Single-Molecule Fluorescence Coincidence: NM Is Monomeric at LowConcentration.  To determine whether aggregation of NM takesplace at subnanomolar concentrations, single-molecule fluores-cence coincidence experiments were performed (29, 30). In theseexperiments, we monitored fluorescence bursts from individualproteinspecies,labeledwitheitheroneortwodifferentdyes,astheyfreely diffused through a confocal volume. Overlapped two-color 500 525 550 575 600 625 6500.250.500.751.00    R  e   l  a   t   i  v  e   f   l  u  o  r  e  s  c  e  n  c  e   i  n   t  e  n  s   i   t  y Wavelength (nm) Donor emission Acceptor emission 0.00.20.40.60.81.0020406080100   Zero-peak FRET-peak     N   u   m    b   e   r   o    f   e   v   e   n   t   s FRET-efficiency 0.00.20.40.60.81.0060120180     N   u   m    b   e   r   o    f   e   v   e   n   t   s 0408012016050050     C   o   u   n   t   s   p   e   r    0 .    5   m   s 05010015050050     C   o   u   n   t   s   p   e   r    0 .    5   m   s time (ms) Coincident bursts Non-coincident bursts 0.00.20.40.60.81.00150300     N   u   m    b   e   r   o    f   e   v   e   n   t   s Stoichiometry factor (S) X a bcd Fig.1.  Aminoacidsequence,ensemble,andsingle-moleculedataforNM.( a )Sequenceofthepriondomain(NM)ofSup35showingresiduesmutatedtocysteineingreenforsinglefluorescencelabeling(AlexaFluor488).Underscores(21and121)indicatethedualcysteinemutationfordonor(AlexaFluor488)andacceptor(Alexa Fluor 594) labeling for FRET studies. Tyrosines are shown in red. ( b ) Steady-state ensemble fluorescence spectra of double-labeled NM (0.1   M) showingenergy transfer under denatured (6 M GdmCl) conditions (blue) and under native condition (red) obtained by exciting the donor (  ex  488 nm). ( c  ) Two-colorsingle-moleculefluorescencecoincidence:Coincidentburstsandstoichiometricfactorhistogramfordual-labeledDNA(non-FRET)standardsample( Upper  )andfor a mixture of two single-labeled NM (100 pM each) under native condition showing no intermolecular association under this condition ( Lower  ). ( d  )Single-moleculeFRET-efficiencyhistogrammeasuredratiometricallyforNMundernativeconditions.BlackcurvesarethebestfitsusingGaussianfunctions.See SI Text   for details. 2650    www.pnas.org  cgi  doi  10.1073  pnas.0611503104 Mukhopadhyay  et al.  (blueandred)laserexcitationandtwo-channeldetectionwereusedtosimultaneouslyexciteanddetectfluorescencefromthetwodyes.In equimolar mixtures of proteins individually labeled with the twodyes,fluorescenceburstswillbeobservedoneitherofthedetectionchannelsastheseproteinsdiffusethroughtheconfocalvolume,butsimultaneousburstsonthetwochannelswillbeabsentiftheproteinremains monomeric. However, if the proteins oligomerise to formdimers or higher order species, a significant fraction of diffusingprotein will contain both dyes, and hence two-channel (coincident)burstswillbeobserved.Fordataanalysis,astoichiometryfactor( S ) was calculated from the signals obtained in the two channels(blue-excited and red-excited fluorescence) using S    I  red  /    I  red   I  blue  .  [2] Using calculated  S  values for individual bursts,  S  histograms can beplotted. In such a histogram,  S  can vary from 0 to 1, with  S    0 and1indicatingtheabsenceofcoincidence(andhencenoaggregation), whereasan S of0.5indicatesmaximumcoincidence[correspondingto aggregates with a 1:1 stoichiometric detection of the dyes, usingappropriate relative excitation intensities; see supporting informa-tion (SI)  Text ].Coincidence experiments were first performed with a standardsingle-strandedDNAsamplelabeledwithAlexaFluor488andCy5positioned 40 bases apart, so that the dyes were beyond FRETrange. A prominent coincidence peak (at 0.5  S ) was observed, asexpected for such a dual-dye sample (Fig. 1  c Upper  ). In sharpcontrast,underthesameconditions,amixtureoftwosinglylabeledNM proteins (Alexa Fluor 488- and Cy5-labeled NM), each at 100pM concentration showed only peaks close to  S    0 and 1 (Fig. 1  c Lower  ). The absence of a coincidence peak for NM demonstratesthat the protein is predominantly monomeric under these condi-tions. Indeed, NM remained monomeric even at twice this con-centration. We used a 100 pM protein concentration for allsubsequent SM-FRET experiments. SM-FRET: Native NM Adopts a Compact State.  Having establishedconditions under which NM polypeptides remain monomeric,SM-FRETwasnextusedtomonitortheconformationalpropertiesof dual-labeled NM. In these experiments, after donor excitation,fluorescence bursts were separated into donor (  I  D ) and acceptor(  I   A  ) signals. We estimated the FRET efficiency for each moleculeusing the relationship  E    I       I  D  /   I   A    1 ,  [3]  where     is a factor correcting for differences in quantum yieldsof the dyes and the detection efficiencies of the two optical paths(see  SI Text ).The FRET histogram showed a compact Gaussian distribution with a FRET-peak centered around 0.8  E , indicative of a uniformpopulation of molecules (Fig. 1  d ). The additional peak at 0  E (zero-peak)observedintheseexperimentsisduetononfluorescentacceptor (due to photobleaching or other causes) in some of theobserved molecules. From the mean FRET efficiency, the averagedistance between the donor and the acceptor in this monomericstate was estimated to be   43 Å (For distance estimation fromthese experiments, see  SI Text ). Similar experiments done in thepresenceof6MGdmClshowedasignificantlyunfoldedpopulationof polypeptides. We estimate that this shift in the peak positioncorresponds to an   20 Å increase in the distance between the twodyesupongoingfromnativetodenaturedconditions,andthemeandistance in the denatured state was estimated to be    63 Å. Thus,upondilutionintobuffer,NMadoptedacompactmonomericform.The SM-FRET data could be indicative of one of the twopossibilities, that native monomeric NM ( i ) adopts a single nativeconformationwithwelldefinedstructuralelements(andhenceveryfew or no conformational fluctuations), or ( ii ) fluctuates eitherslowly or quickly between multiple FRET-distinguishable confor-mations. To distinguish among these alternatives, we first investi-gated the conformational behavior of NM in the presence of adenaturant. Progressive Noncooperative Expansion Observed Upon Denaturation. In previous equilibrium SM-FRET unfolding experiments withglobular proteins (19, 23), two FRET peaks were observed. Asdenaturant concentrations increased, a lower FRET peak corre-sponding to an unfolded species appeared and grew. This wasmirroredbyadecreaseinthehighFRETpeakcorrespondingtothefolded protein. A plot of the fractional population of the foldedstate showed a rather sharp sigmoidal transition as a function of denaturant concentration indicative of cooperative unfolding. A  very different transition was observed with SM-FRET measure-ments of NM. As GdmCl concentrations increased from 0 to 6 M, we observed only a single nonzero FRET peak (Fig. 2). This peaksmoothly and gradually transitioned from a high FRET value (atlow concentrations of denaturant) to a low FRET value (at highconcentrations of denaturant). Conversion to the low-FRET formappeared complete only at   3 M denaturant.For NM, the progressive shift in the FRET-peak coupled withthe broad non-sigmoidal shape of the transition (Fig. 3  a ) providesevidence for a continuous expansion of the protein in the presenceof denaturant, rather than a distinct two-state transition (23, 31).Thus, the observed compact monomeric NM species represents anensemble of states that are characterized by surprisingly strong, butnon-cooperative interactions, rather than a single or a small groupof distinct structural entities. 020406080 0.0 0.2 0.4 0.6 0.8 1.0020406080100 0.0 0.2 0.4 0.6 0.8 1.0 020406080100050100150200     N   u   m    b   e   r   o    f   e   v   e   n   t   s FRET efficiency0.0M GdmCl0.5M GdmCl2.0M GdmCl4.0M GdmCl Fig. 2.  FRET-efficiency histograms of NM at various concentrations of Gd-mCl. Dotted lines are drawn to show the progressive shift in the FRET-peakfrom 0.8 to 0.3. Mukhopadhyay  et al.  PNAS    February 20, 2007    vol. 104    no. 8    2651         B      I      O      P      H      Y      S      I      C      S  Simple Simulations Support Fast Polypeptide Chain Dynamics in NM. To understand whether fluctuating multiple FRET-distinguishablestates could be accessible to natively unfolded NM, we modeled itsenergy-transferdistributionusingsimulationswithinasimpleGaus-sian-chain approximation (23, 31–33) (see  Methods  and  SI Text ).Fig. 3  b  shows (dashed line) the broad FRET-histogram expected if the resulting conformational distribution is frozen on the timescaleof the FRET experiments (23). Regardless of the precise details of the structural model used, it is expected that a similarly broaddistribution of conformations might be generally accessible to andobserved for an ‘‘unstructured’’ and flexible protein chain. Forexample, an excluded volume-limit model would also show a broaddistribution, with some deviations due to correlated chain repul-sions(34).However,becausearelativelynarrowpeakwasobservedfor native NM, any multiple FRET-distinguishable conformationsmust not be stable on the experimental timescale (  50–100   s).The solid line in Fig. 3  b  shows the relatively narrow simulatedFRET-histogram expected if averaging over multiple conforma-tions is complete within the experimental time scale, showing veryreasonable agreement with the experimental data. Thus, our SM-FRET data, taken together with the simple simulations, are con-sistentwithnativeNMoccupyinganensembleofrapidlyfluctuatingor interconverting conformations. FCS: Observation of Fast Fluctuations in NM.  To directly test for thepresence of such fast fluctuations in native NM, we took advantageof the ability of aromatic amino acids to quench Alexa Fluor 488fluorescence (35, 36) and the fact that the N domain is unusuallyabundant in tyrosines. In FCS measurements, the fluctuations of the fluorescence signals from freely diffusing molecules are re-corded for a small-ensemble of molecules with low nanomolarconcentrations in a confocal geometry as for the SM-FRET ex-periments (27, 28). The autocorrelation function of this fluctuatingfluorescence signal can provide information about ( i ) conforma-tional fluctuations on nanosecond to microsecond time scales(37–39)thatariseduetosidechainmovementsand( ii )translationaldiffusion (tens of microseconds to several milliseconds, dependingon molecular size). We first established that the fast fluorescencefluctuations (sub-microsecond) in Alexa Fluor 488-labeled NM were not due to dye photo-physics, rotational motion, an imperfectconfocal geometry, or from oligomers, but rather must arise due tomonomeric polypeptide chain dynamics (see  SI Text ).Next, we carried out FCS experiments on several preparations of NM molecules, each individually labeled with Alexa Fluor 488 at adifferentposition(aminoacids21,38,51,77,96,121,137,and184).Significant fast decay amplitudes were observed on the 20–300 nstimescale for residue positions 21, 38, 51, 77, 96, and 121, which arelocated in the region that becomes the core of the amyloid fiber when NM acquires its prion state (Fig. 4  a ). The amplitude waslargest for position 51 (0.6, compared with 0.3–0.5 for most otherpositions). In the case of position 137, which lies at the interface of the core and unstructured regions, only a small amplitude (  0.1) was observed. This amplitude variation correlates well with thenumber of nearby tyrosine residues (see SI Table 1), with position 51havingsixtyrosineswithina10-residuesequenceseparation,andposition 137 having none within this range. For position 184, withno tyrosines within 71 residues, no sub-microsecond decay ampli-tude was observed (Fig. 4  b ).The FCS data on monomeric NM also revealed the presence of  01234560.20.40.60.8    F   R   E   T  -  e   f   f   i  c   i  e  n  c  y [GdmCl]/M250200150100501.00.80.60.40.20.0 543210 FRET-efficiency    N  u  m   b  e  r  o   f  e  v  e  n   t  s     N  o  r  m  a   l   i  z  e   d   P  r  o   b  a   b   i   l   i   t  y ab Fig. 3.  Progressive  E  -shifts and Gaussian simulations support fast dynamics.( a ) Plot of FRET efficiency vs. denaturant concentration. The solid line is theexponential fit to guide the eye. ( b ) Expected FRET-efficiency distributioncalculated by taking shot-noise into account for the two limiting cases of fast(solid line) and slow (dashed line) conformational averaging with respect totheobservationtime(0.5ms)ofaGaussianchain(formean E  of0.8).Observedhistogram is shown in gray. 0123450.100.150.200.25     N   o   r   m   a    l    i   z   e    d   a   u   t   o   c   o   r   r   e    l   a   t    i   o   n Delay time ( µ s) Delay time ( µ s) 1E-5 1E-3 0.1 10 10000.00.10.2Delay time (ms) 012345 0.100.150.200.25   1E-51E-30.11010000.00.10.2Delay time (ms) Fast component Slower component Conformational fluctuation Diffusion a b Fig.4.  FCSautocorrelationofNMlabeledwithAlexaFluor488at21( a )and184( b )undernativeconditionshowingthefluctuationanddiffusioncomponents.( Insets ) The complete autocorrelation curves. 2652    www.pnas.org  cgi  doi  10.1073  pnas.0611503104 Mukhopadhyay  et al.  at least two well separated components in the fast decays (SI Table1), a faster component in the range of 20–40 ns, and a slowercomponent in the range of 150–250 ns. The presence or absence of these two decay components for different labeling positions pro- videsinsightsintotheirsrcins.AlexaFluor488atallpositionsfrom21 to 121 (all of which have one or more tyrosines within a10-residue separation) exhibited both faster and slower dynamics, whereas the fluorophore at position 137 (no tyrosines closer thana 24-residue separation) exhibits only fast dynamics. As notedabove,thefluorophoreat184(notyrosinescloserthana71residuesseparation) does not sense any sub-microsecond fluctuations.These observations suggest that the faster decay component src-inates from quenching due to relatively proximal (short-range)tyrosines, whereas the slower component srcinates from distant(long-range)tyrosines.Wenotethat,althoughweonlyresolvedtwocomponents in the decays, additional unresolved components dueto quenching from multiple tyrosines and specific transient struc-tures may also be present. Finally, in the presence of GdmClconcentrations of 3 M or higher, these fluctuation componentsessentially disappear, either because the average quenching effi-ciency drops as a result of protein expansion or because of anabsence of fluctuations within the timescale range that we areexamining. These results establish that the monomeric form of NMadopts an ensemble of relatively unordered states with rapidconformational fluctuations. Discussion We have used single-molecule and small-ensemble techniques todirectlystudytheconformationalpropertiesofNMinphysiologicalbuffers in the non-prion monomeric state. Our results demonstratethat natively unfolded NM is not a completely denatured randomcoil under these conditions. Instead, it occupies an ensemble of rapidly interconverting compact conformations. The mean inter-residue distance between amino acid residues 21 and 121 is   43 Å based on our SM-FRET data. This is significantly lower than a value of    72 Å calculated from a reported correlation betweenmean radii of gyration and number of amino acids for denaturedpolypeptides (40), and may also be somewhat lower than the valueexpected for a polymer chain in Flory’s theta solvent (47 Å), in which intrachain and chain-solvent interactions counterbalance ( SI Text ). From the FCS diffusion time, we estimate a value of 45    5 Å for the hydrodynamic radius of native NM ( SI Text ), which is inagreement with previous dynamic light scattering results (15).Interestingly, this size is very similar to   50 Å estimated for adenatured 250-residue random coil polypeptide (41). Therefore,our results support the view that two distinct structural units (N,compact; M, expanded) are juxtaposed in NM. Upon addition of denaturant, the compact native N undergoes a progressive nonco-operative transition to a more expanded form with a   63 Å interdye separation, presumably moving the conformational distri-bution of the entire NM region closer to that which would beoccupied by a denatured globular protein. This transition is notcomplete until 3 M GdmCl, indicating that relatively strong forcesmay govern the collapsed state.NM dynamics comprising short- and long-range conformationalfluctuations of the polypeptide chain were revealed in our FCSmeasurements. The time scale of the short-range conformationaldynamics is similar to previous estimates of short-range contactformation in unfolded polypetides (42, 43). However, it should benoted that a more detailed molecular interpretation of NM chaindynamics is complicated by the existence of a number of tyrosinesat varying length scales, varying relative stiffness around differentamino acid residues and excluded-volume effects (44). Our resultsand analysis also establish evidence that native NM does not existasasingleconformer,butratherpopulatesanensembleofmultiplerapidly interconverting conformers.Nativelyunfoldedorintrinsicallyunstructuredproteins(45)havebeen classified as either less structured (random-coil like) ormore-structured (premolten globule like), based on a ‘‘double- wavelengthplot’’of[   ] 222  vs.[   ] 200 obtainedfromcirculardichroism(CD) experiments (46–48). Although our previous CD data ob-tained during the lag-phase indicated that NM is largely unfolded(15), there are experimental limitations in interpreting such CDdata. First, small oligomers would always be present at the micro-molar concentrations required for CD experiments. Secondly, thelarge random coil signals from the highly charged unstructured Mregion would mask signals from any small ordered region in the Nregion. Natively unfolded proteins have also been classified basedon their dimensions. Random-coil like natively unfolded proteinshave dimensions very similar to those of denatured globular pro-teins, whereas those of premolten globule, like natively unfoldedproteins, are significantly smaller. Our finding that the N region of native NM is significantly more compact than a correspondingdenaturedprotein(40)thenidentifiesitwiththepremoltenglobuleclass of natively unfolded proteins. It has been noted that acombination of low mean hydrophobicity and high charge is aprerequisite for maintaining a flexible, unstructured state underphysiologicalconditions(49).TheamyloidcoreofSup35(Nregion)contains only 8% hydrophobic and 4% charged amino acids. OurresultsindicatethatthecoreNregionremainscompactandflexibledespite lacking both hydrophobic and charged residues. When N isseparatedfromManddilutedintobuffer,itimmediatelyassemblesinto amyloid fibers with no lag phase. Thus, the highly chargedextended M region appears essential for allowing N to maintain amonomeric state.The compact, unstructured form of NM is reminiscent of early(burst-phase) collapsed intermediates observed during proteinfolding from denaturant (50). Given the low occurrence of hydro-phobicresiduesinNM,itislikelythatitscollapseoccurswhenpolarresiduesintheN-segment(QandN)formhydrogenbondsthroughbackbone and side chain interactions, and possibly by    -stackinginteractions among the many tyrosine residues. In support of sucha mechanism, the hydrogen bond disrupting solvent DMSO signif-icantly reduced the FRET efficiency (and hence increased thedimensions) of NM (data not shown). Our hypothesis of hydrogenbonding-mediated chain collapse is consistent with recent FRETresults from unstructured Gly–Ser repeat peptides (51), as well asa recent FCS study of scaling behavior (52) and simulations of poly(Q) peptides (53). The conformational flexibility in thepolypeptide is probably governed by the interplay between hydra-tion of polar side chains of Gln, Asn, and Tyr (54, 55) andintramolecular hydrogen bonds. Additionally, in light of a recent viewpoint on the nature of protein unfolded states, transientinterconverting polyproline type II and   -sheet structures mightalso be present in native NM (56).Our finding that the N domain undergoes conformational fluc-tuations on a time scale well below a millisecond explains why anoligomeric species is required for nucleation of NM prion assembly(14, 16). The likelihood of independent monomers finding thecorrect prion conformation at the same time and nucleating as-sembly via such preformed structures (as suggested for anotherprion) (57) would seem infinitesimally small. In the context of anoligomeric assembly, rapidly fluctuating conformations would fa-cilitate chance encounters between the critical sequence elements whose interaction has been shown to be sufficient to drive nucle-ation (13). These oligomers would then reorganize themselves toform the nucleating species that is capable of rapid assembly intoamyloid fibers. Thus, relative stability of the collapsed state in themonomeric species and the ease with which the monomers transitto on-pathway oligomeric intermediates may determine how amy-loidogenic a protein is.In conclusion, we have investigated the structural ensemble of a natively unfolded yeast prion, which forms amyloid underphysiological conditions. Our single-molecule experiments pro- vide a picture of the structure and dynamics of the monomer,from which the amyloid state ensues. Future work with addi- Mukhopadhyay  et al.  PNAS    February 20, 2007    vol. 104    no. 8    2653       B      I      O      P      H      Y      S      I      C      S

C

Mar 16, 2019
Search
Similar documents
View more...
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