Measurement of amyloid fibril mass-per-lengthby tilted-beam transmission electron microscopyBo Chena, Kent R. Thurbera, Frank Shewmakerb, Reed B. Wicknerb, and Robert Tyckoa,1

aLaboratory of Chemical Physics and bLaboratory of Biochemistry and Genetics, National Institute of Diabetes and Digestive and Kidney Diseases, NationalInstitutes of Health, Bethesda, MD 20892

Contributed by Reed B. Wickner, July 13, 2009 (sent for review June 9, 2009)

We demonstrate that accurate values of mass-per-length (MPL),which serve as strong constraints on molecular structure, can bedetermined for amyloid fibrils by quantification of intensities indark-field electron microscope images obtained in the tilted-beammode of a transmission electron microscope. MPL values for fibrilsformed by residues 218–289 of the HET-s fungal prion protein, for2-fold- and 3-fold-symmetric fibrils formed by the 40-residue�-amyloid peptide, and for fibrils formed by the yeast prion proteinSup35NM are in good agreement with previous results fromscanning transmission electron microscopy. Results for fibrilsformed by the yeast prion protein Rnq1, for which the MPL valuehas not been previously reported, support an in-register parallel�-sheet structure, with one Rnq1 molecule per 0.47-nm �-sheetrepeat spacing. Since tilted-beam dark-field images can be ob-tained on many transmission electron microscopes, this workshould facilitate MPL determination by a large number of researchgroups engaged in studies of amyloid fibrils and similar supramo-lecular assemblies.

Alzheimer’s disease � molecular structure � prions � solid state NMR

The mass-per-length (MPL) of an amyloid fibril is an impor-tant constraint on its molecular structure. Given that amyloid

fibrils contain �-sheets with cross-� alignment relative to thelong fibril axis (1, 2) and that the spacing between �-strands in�-sheets is always 0.47 � 0.01 nm, one expects the MPL to benearly ��MW/0.47 kDa/nm, where MW is the amyloid-formingpolypeptide’s molecular weight and � is the number of moleculesin each �-sheet spacing. MPL measurements were the firstindication that fibrils formed by the 40-residue �-amyloid pep-tide (A�1–40) might have both 2-fold-symmetric (� � 2) andthree-fold-symmetric (� � 3) polymorphs (3–6). MPL measure-ments support a two-fold-symmetric structural model for amylinfibrils with a specific morphology (7) and provide evidence forpolymorphism similar to that of A�1–40 fibrils (8). MPL mea-surements (9) provide support for a �-helix-like structure (10,11) for fibrils formed by residues 218–298 of the HET-s prionprotein (HET-s218–289), with each molecule spanning two turnsof the �-helix (� � 0.5). MPL measurements on fibrils formedby the prion domains of the yeast prion proteins Ure2p (12) andSup35 (13) support in-register parallel �-sheet structures (14–16), with backbone hydrogen bonds in the �-sheets being purelyintermolecular (� � 1). Molecular models corresponding tothese various values of � are given in the relevant papers (4, 7,10, 16).

MPL data in these cases comes from quantification of inten-sities in images obtained with scanning transmission electronmicroscopy (STEM). STEM images of unstained samples aredark-field images, in the sense that the background has lowintensity (arising from weak electron scattering when the ras-tering electron beam strikes the thin carbon film on which fibrilsare adsorbed), while the fibrils have higher intensity (arisingfrom stronger electron scattering when the beam strikes thefibrils themselves). As shown by earlier studies (17–20), STEMimage intensities are proportional to mass densities (per unitarea), so that MPL values can in principle be determined

quantitatively in either of two ways: (i) by measurement ofincident electron beam flux, knowledge of electron scatteringcross-sections and detector geometry, and calibration of detectorsensitivity; (ii) by comparison of fibril image intensities withimage intensities of reference objects with known mass densities,typically tobacco mosaic virus (TMV) rods codeposited with thefibrils. In practice, the second method has been used in studiesof amyloid fibrils. Although dedicated STEM instruments havebeen used in most MPL studies, MPL determination using theSTEM mode of a commercial transmission electron microscopeat 300 kV has also been demonstrated (21). As an alternative toSTEM, energy-filtered transmission electron microscopy (EF-TEM) has been used to measure mass densities (22, 23) andamyloid fibril MPL values (24) in the same manner. EF-TEMalso produces dark-field images of unstained samples, but byuniform illumination and image formation from inelasticallyscattered electrons using electron optics, rather than by rasteringof a narrow electron beam and collection of primarily elasticallyscattered electrons as in STEM.

Dark-field images of unstained samples can also be obtainedwith a conventional transmission electron microscope (TEM) bytilting the incident electron beam by a small angle so that it isblocked by the objective aperture after passing through thesample. The image is then formed from scattered electrons thatpass through the aperture, using the same electron optics as inbright-field TEM imaging of stained samples. This is a dark-fieldmode available on many TEM instruments, which we calltilted-beam TEM (TB-TEM). TB-TEM images of amyloid fibrilsare superficially quite similar to STEM and EF-TEM images,suggesting that MPL values might also be obtainable fromTB-TEM images. In this paper, we demonstrate that accurateMPL values for amyloid fibrils can indeed be obtained byquantification of intensities in TB-TEM images, so that MPLmeasurements can be carried out with widely available instru-mentation and relative ease.

ResultsFig. 1 shows examples of bright-field TEM and dark-fieldTB-TEM images of amyloid fibrils formed by HET-s218–289 (Fig.1A), A�1–40 (Fig. 1 B and C), and Sup35NM (Fig. 1D). A�1–40fibrils have predominantly two-fold-symmetric (2f-A�1–40, Fig.1B) or three-fold-symmetric (3f-A�1–40, Fig. 1C) structures,depending on growth conditions as previously described (3, 4,25). Sup35NM (residues 1–253 of Sup35p) includes the N-terminal prion (N) and middle (M) domains (26) and is oftenused in studies of the [PSI�] prion (27). Bright-field images inFig. 1 are negatively stained with uranyl acetate. Dark-fieldimages are unstained and also contain TMV rods. All imageswere obtained at 80 kV electron beam energy.

Author contributions: B.C., K.R.T., F.S., R.B.W., and R.T. designed research, performedresearch, analyzed data, and wrote the paper.

The authors declare no conflict of interest.

1To whom correspondence should be addressed. E-mail: robertty@mail.nih.gov.

This article contains supporting information online at www.pnas.org/cgi/content/full/0907821106/DCSupplemental.

www.pnas.org�cgi�doi�10.1073�pnas.0907821106 PNAS � August 25, 2009 � vol. 106 � no. 34 � 14339–14344

BIO

PHYS

ICS

AN

DCO

MPU

TATI

ON

AL

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

10, 2

021

MPL values were extracted from TB-TEM images as previ-ously described for STEM (8, 28): (i) image intensities wereintegrated over rectangular areas centered on fibril segments(IF) and over equal areas of background on either side of eachfibril segment (IB1 and IB2); (ii) similarly, image intensities wereintegrated over rectangular areas centered on TMV segments(ITMV) and over equal areas of background on either side of each

TMV segment (IB3 and IB4); (iii) for each image, the quantity�ITMV� was calculated as the average of the quantities ITMV–(IB3 � IB4)/2 within the image; (iv) MPL values were calculatedas MPL � 131�[IF � (IB1 � IB2)/2]/�ITMV�, based on the known131 kDa/nm MPL of TMV (29). All rectangles were 80 nm inlength. Rectangle widths were adjusted to include the fibril orTMV width in each image. Except as noted below, all intensities

A B

C D

Fig. 1. TEM images of amyloid fibrils. Bright-field images of negatively stained samples are in the upper left and dark-field TB-TEM images are in the remainingthree sections of each panel. (Scale bars, 100 nm and 200 nm for bright-field and dark-field images, respectively.) Single-headed arrows indicate examples offibrils analyzed for MPL determinations. Double-headed arrows indicate TMV rods. (A) HET-s218–289 fibrils. Lower two sections show examples of MPL values(kDa/nm) determined for segments enclosed in rectangles. (B) 2f-A�40 fibrils. Lower left section shows examples of MPL values along the length of a single fibril.Upper right section shows examples of MPL values for segments of TMV rods, calibrated by assuming the average TMV MPL value to be 131 kDa/nm. (C) 3f-A�40fibrils. Lower left and upper right sections show examples of MPL values for 3f-A�40 fibrils and TMV rods, respectively. (D) Sup35NM fibrils. Lower left sectionshows examples of MPL values along the length of a single fibril.

14340 � www.pnas.org�cgi�doi�10.1073�pnas.0907821106 Chen et al.

Dow

nloa

ded

by g

uest

on

Janu

ary

10, 2

021

used to calculate a given MPL value were taken from the sameimage, as intensities from different images were not necessarilydirectly comparable due to variations in incident beam intensityand other factors.

Fig. 2 shows MPL histograms, extracted from 30 or moreTB-TEM images of each fibril sample. MPL peak positions weredetermined by fitting the histograms to one or more Gaussianfunctions. For HET-s218–289 fibrils, the histogram in Fig. 2 Ashows a single peak at 8.3 kDa/nm, in good agreement with thevalue 9.2 kDa/nm determined previously by STEM (9) and thevalue 9.20 kDa/nm predicted by the �-helix-like molecularstructural model of Meier and coworkers (10, 11) (MW � 8.65kDa, � � 0.5). For 2f-A�1–40, the histogram in Fig. 2B shows amajor peak at 17.4 kDa/nm, in good agreement with the value ofapproximately 21 kDa/nm determined previously by STEM (3)and the value 18.4 kDa/nm predicted by the two-fold-symmetricstructural model of Petkova et al. (25) (MW � 4.33 kDa, � � 2).For 3f-A�1–40, the histogram in Fig. 2C shows a major peak at27.7 kDa/nm, in good agreement with the value of 26–29 kDa/nmdetermined previously by STEM (3, 4) and the value 27.6kDa/nm predicted by the three-fold-symmetric structural modelof Paravastu et al. (3, 4) (MW � 4.33 kDa, � � 3). Goldsburyet al. have also reported that MPL values for A�1–40 fibrilsderived from STEM images are either 17 � 2 kDa/nm or 28 �4 kDa/nm (5).

The histogram in Fig. 2B shows a minor peak at 28.0 kDa/nm,attributable to a minor population of 3f-A�1–40 fibrils in the2f-A�1–40 sample. The histogram in Fig. 2C shows minor peaksat 20.1 kDa/nm and 34.9 kDa/nm, attributable to minor popu-lations of single and paired 2f-A�1–40 fibrils in the 3f-A�1–40sample. Both histograms show minor peaks near 9 kDa/nm,arising from rare occurrences of fibrils with � � 1 (see Fig. S1).Such fibrils were not observed in previous STEM studies (3–6,28) and may represent ‘‘immature’’ structures, especially sinceA�1–40 fibril samples for our TB-TEM measurements wereincubated for 24 h or less after seeding.

Our assignment of minor peaks in Fig. 2 B and C to specificminority structures is supported by the observation of multiple,nearly equal MPL counts along the lengths of individual fibrils

(such as fibrils in Fig. S1 A and C), as well as by the inability tofit the MPL histogram to single Gaussian function, as in Fig. 2B and C.

For Sup35NM fibrils, the histogram in Fig. 2D shows a singlepeak at 60.8 kDa/nm, consistent with the value 62.6 kDa/nmpredicted for MW � 29.4 kDa if � � 1, as in the in-registerparallel �-sheet structure indicated by solid state NMR mea-surements (14, 16). Although we are not aware of STEMmeasurements on Sup35NM fibrils, STEM measurements forfibrils formed by residues 1–65 of Sup35p fused to greenfluorescent protein (GFP-Sup35p1–65) indicate MPL valuesranging from 72.5 kDa/nm to 95.4 kDa/nm, depending on theexact fibril morphology (13), in approximate agreement with thevalue 79.2 kDa/nm predicted for MW � 37.2 kDa and � � 1.Thus, the TB-TEM results in Fig. 2D suggest that Sup35NMfibrils and certain classes of GFP-Sup35p1–65 fibrils have acommon structural organization. Interestingly, fibrils formed invitro by both Sup35NM (30) and GFP-Sup35p1–65 (31) have beenshown to be capable of inducing the [PSI�] prion phenotype inyeast.

As an initial application to a system for which no MPL dataare available from STEM or other techniques, we performed themeasurements shown in Fig. 3 on fibrils formed by the priondomain of Rnq1, the protein that determines the [PIN�] prionof yeast. As shown by Liebman and coworkers, the Rnq1 priondomain is contained in residues 153–405 (32). Our samplecontained a mixture of residues 153–405 and residues 216–405(see Materials and Methods). The MPL histogram in Fig. 4Bshows a peak centered at 46.2 kDa/nm, in good agreement withthe value 46.6 kDa/nm predicted for a cross-� structure formedby residues 216–405 (MW � 21.9 kDa) with � � 1. A smallnumber of counts were also observed near MPL approximately59.8 kDa/nm, the value predicted for residues 153–405 (MW �28.1 kDa) with � � 1. These results are consistent with the earlierdemonstration by solid state NMR that Rnq1 prion domainfibrils have an in-register parallel �-sheet structure (33).

DiscussionReliability and Precision of MPL Determinations by TB-TEM. Results inFigs. 1 and 2 demonstrate empirically that accurate MPL valuesfor amyloid fibrils can be determined from TB-TEM images over

Fig. 2. MPL histograms extracted from TB-TEM images. Solid curves are fitsto one or more Gaussian functions. Vertical dashed lines indicate ideal MPLvalues predicted by experimentally-based structural models, as discussed inthe text. (A) HET-s218–289 fibrils, fit to one peak at 8.3 kDa/nm, with 4.7 kDa/nmfull width at half maximum (FWHM). (B) 2f-Ab40 fibrils, fit to three peaks at6.6, 17.4, and 28.0 kDa/nm, with 1.9, 9.3, and 11.0 kDa/nm FWHM, respectively.Relative areas are 0.03, 1.00, and 0.21. (C) 3f-Ab40 fibrils, fit to four peaks at8.2, 20.1, 27.7, and 34.9 kDa/nm, with 6.4, 6.3, 6.4, and 5.8 kDa/nm FWHM,respectively. Relative areas are 0.18, 0.13, 1.00, and 0.09. (D) Sup35NM fibrils,fit to one peak at 60.8 kDa/nm, with 18.2 kDa/nm FWHM.

A B

C D

Fig. 3. MPL determination for Rnq1 prion domain fibrils. (A) Bright-fieldimage of negatively stained sample. (B) MPL histogram determined fromTB-TEM images, fit to one Gaussian peak at 46.2 kDa/nm, with 13.2 kDa/nmFWHM. (C and D) Examples of TB-TEM images, with TMV rods indicated bydouble-headed arrows. MPL values (kDa/nm) are shown for fibril segmentsenclosed in rectangles.

Chen et al. PNAS � August 25, 2009 � vol. 106 � no. 34 � 14341

BIO

PHYS

ICS

AN

DCO

MPU

TATI

ON

AL

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

10, 2

021

an MPL range from 10 kDa/nm (or possibly less) to 60 kDa/nm(or probably more). These experimental results allay concernsthat MPL determination might be precluded by damage to thefibrils induced by the electron beam, non-uniform illuminationof the image field by the electron beam, nonlinearities in theelectron detector or camera system, imperfections in the elec-tron optics, multiple scattering, or other considerations. Aspectsof our experimental protocol that we believe to be essential foraccurate MPL determination are described below (see Materialsand Methods).

The precision of the MPL values, estimated from the uncer-tainty in the major Gaussian peak positions in Fig. 2, is not ashigh as in some STEM studies of amyloid fibrils (5, 9, 13, 28) butis in all cases sufficient to determine �, the important structuralparameter, to within � 0.3 (e.g., � � 1.9 � 0.3 for 2f-A�40fibrils). In the specific context of amyloid fibrils, the demon-strated precision is sufficient to distinguish among plausiblecandidate structural models. For HET-s218–289, the precision isapparently limited by the signal-to-noise of the TB-TEM images,that is, the low fibril image intensity relative to fluctuations inbackground intensity. Fluctuations in background intensity inmeasurements on HET-s218–289 fibrils, measured as 131� IB1 �IB2/�ITMV�, have a Gaussian distribution with FWHM � 5.6kDa/nm (Fig. S2 A). These f luctuations should produce aFWHM of (3/2) � 5.6 kDa/nm in the MPL histogram forHET-s218–289 fibrils when background subtraction is carried outas described above, in reasonable agreement with the 4.7kDa/nm FWHM in Fig. 2 A.

Fig. 1 B and C (lower left) show series of MPL values alongthe lengths of single 2f-A�40 and 3f-A�40 fibrils. The randomscatter among these values indicates that non-uniform illumina-tion by the electron beam within a single image is not a majorcause of variations in the apparent MPL values, as non-uniformillumination would be expected to produce monotonic varia-tions. Fig. 1 B and C (upper right) show MPL values for TMVrods within a single image, derived by setting the average of thesevalues to 131 kDa/nm. Again, the random scatter argues against

non-uniform illumination as a source of imprecision. A histo-gram of TMV MPL values (Fig. S3) indicates that structuralvariations in TMV rods (34) contribute less than 10% toimprecision in MPL.

The MPL histogram for Sup35NM fibrils (Fig. 2D) shows thebroadest peak. Background intensity f luctuations in TB-TEMimages of Sup35NM fibrils have FWHM � 10.8 kDa/nm (Fig.S2B), greater than in TB-TEM images of HET-s218–289 fibrils duethe presence of more extraneous material on the sample gridsand the use of wider rectangular integration areas. Backgroundintensity f luctuations may then contribute roughly 13 kDa/nm tothe observed 18.2 kDa/nm FWHM in Fig. 2D. Additionalcontributions to MPL variations apparently arise from structuralnonuniformity of the fibrils, as shown in Fig. 1D where MPLvalues along the length of a single fibril have relatively largescatter. Structural nonuniformity may include unresolved breaksin the fibrils and adhesion of nonfibrillar material to the fibrils.We conclude that the most important limitation on the precisionof MPL determination is sample quality, at least for MPL valuesabove 10 kDa/nm. Similar issues can affect STEM measurements.

Perhaps surprisingly, damage to the fibrils and TMV by theelectron beam plays no discernible role in our TB-TEM mea-surements. Previous studies indicate that a total electron dose of104 e/nm2 at 80 keV produces a mass loss of roughly 30% fortypical protein assemblies (including TMV) in STEM measure-ments at room temperature (18, 34). Similar mass losses havebeen measured in EF-TEM studies (22, 23). We estimate thetypical electron dose for a single TB-TEM image, acquired in 10 sunder our experimental conditions, to be 0.6–2.5 � 104 e/nm2

(see Materials and Methods). Fig. 4A shows TB-TEM images of3f-A�1–40 and TMV obtained at the beginning and end of a 200-speriod of continuous exposure to the electron beam. No deg-radation of the image is seen. Fig. 4B and C shows MPL valuesand TMV gray scale intensities determined from images of asingle field during a 200-s period. No systematic changes in thesequantities are observed (other than a possible 5% reduction in3f-A�1–40 fibril MPL over 200 s). Comparison of TMV imagesrecorded with total beam exposure times as short as 1 s (i.e.,roughly 2.5 � 103 e/nm2) with images at longer exposure timesshows no evidence for rapid mass loss.

Theoretical Justification for Quantification of TB-TEM Images. As forSTEM and EF-TEM imaging (18, 21, 23), the integrated inten-sity within an area A of the TB-TEM image is proportional to thenumber of collected, scattered electrons Ns from that area. Afterbackground subtraction, this quantity can be expressed as

NS � Ne �k

nk�k fk

A,

where Ne is the number of incident electrons on A, nk is thenumber of atoms of type k within A, �k is the scatteringcross-section, and fk is the fraction of scattered electrons that arecollected and focused to form the image. This expressionassumes that no multiple scattering events occur, that unscat-tered electrons are entirely blocked by the objective aperture,that the microscope’s camera system is linear, and that diffrac-tion effects (i.e., interference among scattering amplitudes fromdifferent atoms in the sample) are negligible. The mass per areaof the sample is

� � �k

nkMk

A

where Mk is the atomic mass. Consequently, NS � �NeQ, with

Fig. 4. Assessment of electron beam damage during MPL measurements.Electron flux density is approximately 2.4 � 103 nm�2s�1. (A) TB-TEM imagesof a 3f-A�40 fibril, obtained after 15 s of exposure to the electron beam (firstimage in a series at the same location on the sample grid) and after 203 s ofexposure (final image in a series). (B) Fractional variations in the apparent MPLof the fibril in panel A, calibrated against the TMV rod in the same set ofimages, with MPL0 � 27 kDa/nm. (C) Average gray scale value for a single TMVrod in a series of images, under continuous exposure to the electron beam.

14342 � www.pnas.org�cgi�doi�10.1073�pnas.0907821106 Chen et al.

Dow

nloa

ded

by g

uest

on

Janu

ary

10, 2

021

Q �

�k

nk�kf k

�k

nkMk

being the effective scattering area per mass within A. The valueof Q depends on the elemental composition and elementaldensities as well as the microscope geometry. Q is expected tobe approximately the same for all proteinaceous materials (18,23). Therefore, TB-TEM image intensities can be used todetermine �/A, and hence MPL.

Both elastic and inelastic scattering contribute to �k. Given a1.5-mm distance from the sample grid to the objective aperture,a 1.2° (21 mrad) beam tilt, and a 50-�m objective aperturediameter, electrons with scattering angles in the 4.3–37.6 mradrange contribute to the TB-TEM image. Although the totalinelastic scattering cross-section is greater than the total elasticscattering cross-section, the majority of inelastically scatteredelectrons have scattering angles less than 4 mrad under ourexperimental conditions, while the majority of elastically scat-tered electrons have scattering angles greater than 8 mrad (35).Our TB-TEM images are therefore formed primarily fromelastically scattered electrons, with a minor contribution frominelastic scattering. Since previous work has shown that MPLvalues can be determined from either elastic scattering, as inSTEM studies, or inelastic scattering, as in EF-TEM studies,images that result from a combination of elastic and inelasticscattering should also permit MPL determination.

Materials and MethodsSample Preparation. HET-s218–289 was expressed in E. coli with a C-terminalhexa-His tag and purified as described by Dos Reis et al. (36). Briefly, cells werelysed by sonication and the protein was extracted from the insoluble pelletfraction with 8 M guanidine-HCl, purified on a Talon (Clontech) column, andeluted in 8 M urea. Denaturant was removed using a HiTrap column GEHealthcare) in 175 mM acetic acid (37). The eluate containing 1.4 mM proteinwas neutralized with Tris base and incubated at 4 °C without agitation for 7days to allow fibril formation.

A�1–40 was synthesized and purified as previously described (3, 4). Approx-imately 1 mg of lyophilized peptide was dissolved initially in dimethyl sulfox-ide at 8 mM concentration, then diluted to 230 �M in 10 mM phosphatebuffer, pH 7.4. For seeded fibril growth, preexisting 2f-A�1–40 or 3f-A�1–40

fibrils were added in a 1:20 molar ratio of fibrillar peptide to fresh peptide, thesolution was sonicated to break the preexisting fibrils into short fragments(Branson model 250 sonifier, lowest power, 10% duty cycle, 30 s), and fibrilswere allowed to grow at 24 °C without mixing or agitation of the solution.Sup35NM was expressed with a C-terminal hexa-His tag and purified underdenaturing conditions as previously described (14). Sup35NM was then dia-lyzed into non-denaturing buffer (5 mM phosphate, pH 7.4, and 150 mM NaCl)and incubated for 7 days at 4 °C to permit fibril formation. Fibrils werecollected by centrifugation and resuspended in deionized water.

The prion domain of Rnq1p was expressed with a C-terminal hexa-His tagand partially purified under denaturing conditions by binding to a Ni-NTAcolumn and eluting with imidazole (33). Electrospray-ionization mass spec-trometry showed our sample to contain proteins with MW � 28.102 kDa(corresponding to residues 153–405 of Rnq1 with the hexa-His tag) and MW �21.931 kDa (corresponding to residues 216–405 of Rnq1p with the hexa-Histag). Proteins were precipitated with cold methanol, dried, and dissolved in 4M urea, 5 mM potassium phosphate, pH 7.4, and 150 mM NaCl. Fibrils thenformed at room temperature with gentle agitation (38).

For TB-TEM, fibrils were adsorbed to carbon films on lacey carbon supportson 300 mesh copper grids. Films were 4–10 nm thick, estimated from thevolume of evaporated carbon and geometry of our carbon deposition cham-

ber. Grids were glow-discharged immediately before use. A 5-�L aliquot offibril solution and a 1-�L aliquot of TMV solution were applied simultaneouslyto the grid. Fibril solutions were diluted as required to produce suitablecoverages in TB-TEM images. TMV concentration was 0.08–0.23 mg/mL. Aftera 5-min adsorption period, solutions were blotted, washed three times withdeionized water (5 �L aliquot, 10 s each), blotted, and dried in air. Grids of2f-A�1–40 and 3f-A�1–40 fibrils were prepared 8 h and 24 h after seeding,respectively, to minimize lateral association of fibrils.

For TEM images of negatively stained samples, grids were prepared by thesame procedure, but without adsorption of TMV and with a 60-s period ofstaining with 5 �L of 3% uranyl acetate before final blotting and drying.

Image Acquisition and Processing. Images were acquired with an FEI MorgagniTEM, operating at 80 kV, equipped with a side-mounted, 1 megapixel AMTAdvantage HR CCD camera. Bright-field images were acquired at 140,000�magnification. TB-TEM images were acquired at 56,000� magnification, usinga beam tilt angle of 1.2°, a 50-�m diameter objective aperture, and a 300-�mcondenser aperture. Smaller tilt angles did not produce adequate blocking ofthe unscattered electron beam. Other microscope settings included spot 5,bias 3, and filament current in the 10–20 �A range. Before recording TB-TEMimages, condenser stigmators were carefully adjusted to give a circular beamprofile when the beam was viewed on the carbon film in TB-TEM mode (atlower magnification), and the beam was carefully centered and spread toproduce uniform illumination over the field of view, as indicated by uniformbackground intensity from the carbon film in the final images. The sample gridwas scanned manually for promising areas at lower magnifications and lowerbeam intensities (�100 e/nm2-s), using the survey mode of the AMT softwarewith a camera gain value of 8. Once an area that apparently contained bothTMV rods and amyloid fibrils was identified, the magnification and beamintensity were increased (to �600–2,500 e/nm2-s), the focus was quicklyadjusted to maximize the clarity of the TMV rods, and final TB-TEM imageswere recorded. Each image was the average of 8 acquisitions, each with a 1.2-sexposure time and a camera gain value of 1. Images were stored either as 8-bittiff files, after automatic linear rescaling of image intensities by the camerasoftware to fill the 8-bit range (threshold � 10 and tail � 5 in the AMTsoftware), or as raw 16-bit tiff files without rescaling (threshold � 0 and tail �0). Equivalent MPL results were obtained in the two cases. Electron doses wereestimated from our own calibration of the camera gray scale in images of anempty sample grid with no beam tilt against direct measurements of currentfrom the TEM viewing screen. For our system, 1 gray scale unit in a 16-bit imagecorresponds to roughly 0.040 electrons per pixel.

For Rnq1 prion domain fibrils, we had difficulty finding grid areas thatcontained both fibrils and TMV rods within the same field at 56,000� mag-nification. Therefore, images of many non-overlapping fields within a prom-ising grid area were recorded under identical beam conditions and saved as16-bit tiff files without rescaling. We found that TMV intensities in theseimages were identical to within the usual uncertainty (e.g., as in Fig. S2), andthat the average TMV intensity across these images could be used to deriveMPL values for fibrils within the same set of images, even though most imagescontained only fibrils or only TMV rods.

TB-TEM images were analyzed with ImageJ software (available at http://rsbweb.nih.gov/ij/). Only fibrils that appeared to be single filaments, ratherthan pairs or higher-order bundles, were selected for MPL measurements.Thus, as with STEM measurements, the relative areas of peaks in MPL histo-grams do not necessarily reflect the true populations of fibrils with differentMPL values in the sample.

The 80-nm length of rectangular areas was chosen to allow several hundredMPL counts for each sample in Fig. 2 and for consistency among samples. Inprinciple, longer rectangles might produce more precise final MPL values,subject to restrictions imposed by fibril curvature and overlap of fibrils in theTB-TEM images. The same considerations apply in STEM measurements.

ACKNOWLEDGMENTS. This work was supported by the Intramural ResearchProgram of the National Institute of Diabetes and Digestive and KidneyDiseases (NIDDK) of the National Institutes of Health. We thank Dr. D. EricAnderson of NIDDK for mass spectrometry of Rnq1 prion domain fibrils, andProf. Gerald Stubbs and Dr. Amy Kendall of Vanderbilt University for provid-ing TMV.

1. Sunde M, Blake CCF (1998) From the globular to the fibrous state: Protein structure andstructural conversion in amyloid formation. Q Rev Biophys 31:1–39.

2. Tycko R (2006) Molecular structure of amyloid fibrils: Insights from solid state NMR. QRev Biophys 39:1–55.

3. Petkova AT, Leapman RD, Guo ZH, Yau WM, Mattson MP, Tycko R (2005) Self-propagating, molecular-level polymorphism in Alzheimer’s �-amyloid fibrils. Science307:262–265.

4. Paravastu AK, Leapman RD, Yau WM, Tycko R (2008) Molecular structural basis forpolymorphism in Alzheimer’s �-amyloid fibrils. Proc Natl Acad Sci USA 105:18349–18354.

5. Goldsbury C, Frey P, Olivieri V, Aebi U, Muller SA (2005) Multiple assembly pathwaysunderlie amyloid-� fibril polymorphisms. J Mol Biol 352:282–298.

6. Goldsbury CS, et al. (2000) Studies on the in vitro assembly of A�1–40: Implications forthe search for A� fibril formation inhibitors. J Struct Biol 130:217–231.

Chen et al. PNAS � August 25, 2009 � vol. 106 � no. 34 � 14343

BIO

PHYS

ICS

AN

DCO

MPU

TATI

ON

AL

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Janu

ary

10, 2

021

7. Luca S, Yau WM, Leapman R, Tycko R (2007) Peptide conformation and supramolecularorganization in amylin fibrils: Constraints from solid state NMR. Biochemistry46:13505–13522.

8. Goldsbury CS, et al. (1997) Polymorphic fibrillar assembly of human amylin. J Struct Biol119:17–27.

9. Sen A, et al. (2007) Mass analysis by scanning transmission electron microscopy andelectron diffraction validate predictions of stacked �-solenoid model of HET-s prionfibrils. J Biol Chem 282:5545–5550.

10. Wasmer C, Lange A, Van Melckebeke H, Siemer AB, Riek R, Meier BH (2008) Amyloidfibrils of the HET-s(218–289) prion form a �-solenoid with a triangular hydrophobiccore. Science 319:1523–1526.

11. Ritter C, et al. (2005) Correlation of structural elements and infectivity of the HET-sprion. Nature 435:844–848.

12. Baxa U, et al. (2003) Architecture of Ure2p prion filaments: The N-terminal domainsform a central core fiber. J Biol Chem 278:43717–43727.

13. Diaz-Avalos R, King CY, Wall J, Simon M, Caspar DLD (2005) Strain-specific morphol-ogies of yeast prion amyloid fibrils. Proc Natl Acad Sci USA 102:10165–10170.

14. Shewmaker F, Ross ED, Tycko R, Wickner RB (2008) Amyloids of shuffled prion domainsthat form prions have a parallel in-register �-sheet structure. Biochemistry 47:4000–4007.

15. Baxa U, et al. (2007) Characterization of �-sheet structure in Ure2p1–89 yeast prionfibrils by solid state nuclear magnetic resonance. Biochemistry 46:13149–13162.

16. Shewmaker F, Wickner RB, Tycko R (2006) Amyloid of the prion domain of Sup35p hasan in-register parallel �-sheet structure. Proc Natl Acad Sci USA 103:19754–19759.

17. Wall JS, Hainfeld JF (1986) Mass mapping with the scanning transmission electronmicroscope. Ann Rev Biophys Biophys Chem 15:355–376.

18. Engel A (1982) Mass determination by electron scattering. Micron 13:425–436.19. Engel A (1978) Molecular weight determination by scanning transmission electron

microscopy. Ultramicroscopy 3:273–281.20. Zeitler E, Bahr GF (1962) Photometric procedure for weight determination of submi-

croscopic particles by quantitative electron microscopy. J Appl Phys 33:847–853.21. Sousa AA, Leapman RD (2007) Quantitative STEM mass measurement of biological

macromolecules in a 300 kV TEM. J Microsc-Oxf 228:25–33.22. Feja B, Aebi U (1999) Molecular mass determination by STEM and EFTEM: A critical

comparison. Micron 30:299–307.23. Feja B, Durrenberger M, Muller S, Reichelt R, Aebi U (1997) Mass determination by

inelastic electron scattering in an energy-filtering transmission electron microscopewith slow-scan CCD camera. J Struct Biol 119:72–82.

24. Paravastu AK, Qahwash I, Leapman RD, Meredith SC, Tycko R (2009) Seeded growth of�-amyloid fibrils from Alzheimer’s brain-derived fibrils produces a distinct structure.Proc Natl Acad Sci USA 106:7443–7448.

25. Petkova AT, Yau WM, Tycko R (2006) Experimental constraints on quaternary structurein Alzheimer’s �-amyloid fibrils. Biochemistry 45:498–512.

26. Teravanesyan MD, Dagkesamanskaya AR, Kushnirov VV, Smirnov VN (1994) The Sup35omnipotent suppressor gene is involved in the maintenance of the non-Mendeliandeterminant [PSI�] in the yeast Saccharomyces cerevisiae. Genetics 137:671–676.

27. Glover JR, Kowal AS, Schirmer EC, Patino MM, Liu JJ, Lindquist S (1997) Self-seededfibers formed by Sup35, the protein determinant of [PSI�], a heritable prion-like factorof S. cerevisiae. Cell 89:811–819.

28. Antzutkin ON, Leapman RD, Balbach JJ, Tycko R (2002) Supramolecular structuralconstraints on Alzheimer’s �-amyloid fibrils from electron microscopy and solid statenuclear magnetic resonance. Biochemistry 41:15436–15450.

29. Namba K, Stubbs G (1986) Structure of tobacco mosaic virus at 3.6 Å resolution:Implications for assembly. Science 231:1401–1406.

30. Sparrer HE, Santoso A, Szoka FC, Weissman JS (2000) Evidence for the prion hypothesis:Induction of the yeast [PSI�] factor by in vitro-converted Sup35 protein. Science289:595–599.

31. King CY, Diaz-Avalos R (2004) Protein-only transmission of three yeast prion strains.Nature 428:319–323.

32. Vitrenko YA, Pavon ME, Stone SI, Liebman SW (2007) Propagation of the [PIN�] prionby fragments of Rnq1 fused to GFP. Curr Genet 51:309–319.

33. Wickner RB, Dyda F, Tycko R (2008) Amyloid of Rnq1p, the basis of the [PIN�] prion, hasa parallel in-register �-sheet structure. Proc Natl Acad Sci USA 105:2403–2408.

34. Muller SA, Engel A (2001) Structure and mass analysis by scanning transmission electronmicroscopy. Micron 32:21–31.

35. Wall J, Isaacson M, Langmore JP (1974) Collection of scattered electrons in dark fieldelectron microscopy. 2. Inelastic scattering. Optik 39:359–374.

36. Dos Reis S, Coulary-Salin B, Forge V, Lascu I, Begueret J, Saupe SJ (2002) The HET-s prionprotein of the filamentous fungus Podospora anserina aggregates in vitro into amy-loid-like fibrils. J Biol Chem 277:5703–5706.

37. Benkemoun L, et al. (2006) Methods for the in vivo and in vitro analysis of [HET-s] prioninfectivity. Methods 39:61–67.

38. Sondheimer N, Lindquist S (2000) Rnq1: An epigenetic modifier of protein function inyeast. Mol Cell 5:163–172.

14344 � www.pnas.org�cgi�doi�10.1073�pnas.0907821106 Chen et al.

Dow

nloa

ded

by g

uest

on

Janu

ary

10, 2

021