Ion mobility-mass spectrometry reveals a … cross section growth curves for various types of...

SUPPLEMENTARY INFORMATIONdoi: 10.1038/nchem.945

nature chemistry | www.nature.com/naturechemistry 1

- S1 -

Ion mobility-mass spectrometry reveals a

conformational conversion from random assembly to

beta-sheet in amyloid fibril formation

Christian Bleiholder, Nicholas F. Dupuis, Thomas Wyttenbach, Michael T. Bowers

Department of Chemistry and BiochemistryUniversity of California

Santa Barbara, CA 93106 - 9510Correspondence to: [email protected]

Supporting Information

© 2010 Macmillan Publishers Limited. All rights reserved.

2 nature chemistry | www.nature.com/naturechemistry

SUPPLEMENTARY INFORMATION doi: 10.1038/nchem.945

- S2 -

S1. Collision cross section growth curves for various types of self-assembly

S1.1. Spatially isotropic self-assemblyThe collision cross section nσ for a spatially isotropic peptide cluster with oligomer size n is given by the radius nr of the oligomer:

(1.1) 2n nrσ π=

For a spherical oligomer, the radius nr is related to the volume nV of the oligomer with oligomer size n:

(1.2)1/3

34 33 4n n n nV r r Vπ

π = ⇔ =

Substituting rn in eq. (1.1) with (1.2) gives:

(1.3)2/33

4n nVσ ππ

=

The volumes nV and mV of two spatially isotropic oligomers with cluster sizes n and mare related by (1.4):

(1.4) n

m

V nV m

=

Equation (1.4) holds, because 1nV n V= ⋅ for any size n and 1V is the monomer volume.Using eq (1.4) the volume nV of any arbitrary cluster size n can be computed if the collision cross section mσ of oligomer with size m is known:

(1.5)3234 4

3 3m

n m mn n nV V rm m m

σπ ππ

= = ⋅ = ⋅

Here, the oligomer with size m really can be any reference point. Typically, the monomer (m = 1) is used as reference. Inserting (1.5) into (1.3) relates the collision cross section

nσ for spatially isotropic self-assembly to the oligomer size n as:

(1.6) 2/3 * 2/32 /3m

n mn nmσσ σ= ⋅ = ⋅




- S2 -

S1. Collision cross section growth curves for various types of self-assembly

S1.1. Spatially isotropic self-assemblyThe collision cross section nσ for a spatially isotropic peptide cluster with oligomer size n is given by the radius nr of the oligomer:

(1.1) 2n nrσ π=

For a spherical oligomer, the radius nr is related to the volume nV of the oligomer with oligomer size n:

(1.2)1/3

34 33 4n n n nV r r Vπ

π = ⇔ =

Substituting rn in eq. (1.1) with (1.2) gives:

(1.3)2/33

4n nVσ ππ

=

The volumes nV and mV of two spatially isotropic oligomers with cluster sizes n and mare related by (1.4):

(1.4) n

m

V nV m

=

Equation (1.4) holds, because 1nV n V= ⋅ for any size n and 1V is the monomer volume.Using eq (1.4) the volume nV of any arbitrary cluster size n can be computed if the collision cross section mσ of oligomer with size m is known:

(1.5)3234 4

3 3m

n m mn n nV V rm m m

σπ ππ

= = ⋅ = ⋅

Here, the oligomer with size m really can be any reference point. Typically, the monomer (m = 1) is used as reference. Inserting (1.5) into (1.3) relates the collision cross section

nσ for spatially isotropic self-assembly to the oligomer size n as:

(1.6) 2/3 * 2/32 /3m

n mn nmσσ σ= ⋅ = ⋅

- S3 -

Here, *mσ is a constant. For the common case that the monomer (m = 1) is taken as the

reference point it holds *1m mσ σ σ= ≡ and it follows:

(1.7)

S1.2. Fibrillar self-assemblyIf self-assembly proceeds in a highly directional manner as a fibril, the relationship between the collision cross section nσ and the oligomer number n follows a linearcorrelation (1.8).

(1.8) n a n kσ = ⋅ +

Here, a denotes the difference in cross section ,i jσ∆ between two oligomer states with sizes n = i and n = j, i.e. the slope a in the linear progression curve (see Figure S1):

(1.9) , const ai j j iσ σ σ∆ = − = ≡

For an infinite fibril this is the only contribution, i.e. k = 0. This is not the case for a finite fiber with caps, where the intercept k is generally non-zero (see Figure S1).




- S4 -

Figure S1: Fibrilar self-assembly is schematized. (a) The collision cross section a n kσ = ⋅ + depends linearly on the oligomer size n (number n of monomer units in the aggregate), and is fully described by the two parameters a and k . (b) The slope a is the change in cross section generated by adding or removing one monomer unit along the fibril axis. The intercept k is the cross section that arises from the fiber caps.

S2. Influence of charge interactions on measured cross sectionThe peptide oligomers identified in this work are typically highly charged. It is known that the collision-cross section increases with increasing charge on an analyte due to coulombic repulsion between the multiple charges. This work relates an increase in cross section with oligomer size n with a conformational conversion. Because charge state also increases with increasing oligomer size n, it must be shown that the resulting increase in cross section does not result from charge repulsion but a change in conformation.

Figure S2 shows that with the addition of one charge to an oligomer there is a corresponding increase in cross section. Regardless of the oligomer size, n = 1 or n = 9, the increase in cross section is ~ 5% with the addition of a single extra charge. It is noteworthy, that the monomer of NNQQNY was recorded in the charge states +1 and +2, effectively doubling the charge density and having two charges on just one unit, but the collision cross section merely increased from 185.54 Å2 (n/z = 1/1) to 192.93 Å2 (n/z =1/2). A similar effect is observed for the dimer of YGGFL (245 Å2 for n/z = 2/1 and 255Å2 for n/z = 2/2) and the trimer of NNQQNY (368 Å2 for n/z = 3/2 and 411 Å2 for n/z =




- S4 -

Figure S1: Fibrilar self-assembly is schematized. (a) The collision cross section a n kσ = ⋅ + depends linearly on the oligomer size n (number n of monomer units in the aggregate), and is fully described by the two parameters a and k . (b) The slope a is the change in cross section generated by adding or removing one monomer unit along the fibril axis. The intercept k is the cross section that arises from the fiber caps.

S2. Influence of charge interactions on measured cross sectionThe peptide oligomers identified in this work are typically highly charged. It is known that the collision-cross section increases with increasing charge on an analyte due to coulombic repulsion between the multiple charges. This work relates an increase in cross section with oligomer size n with a conformational conversion. Because charge state also increases with increasing oligomer size n, it must be shown that the resulting increase in cross section does not result from charge repulsion but a change in conformation.

Figure S2 shows that with the addition of one charge to an oligomer there is a corresponding increase in cross section. Regardless of the oligomer size, n = 1 or n = 9, the increase in cross section is ~ 5% with the addition of a single extra charge. It is noteworthy, that the monomer of NNQQNY was recorded in the charge states +1 and +2, effectively doubling the charge density and having two charges on just one unit, but the collision cross section merely increased from 185.54 Å2 (n/z = 1/1) to 192.93 Å2 (n/z =1/2). A similar effect is observed for the dimer of YGGFL (245 Å2 for n/z = 2/1 and 255Å2 for n/z = 2/2) and the trimer of NNQQNY (368 Å2 for n/z = 3/2 and 411 Å2 for n/z =

- S5 -

3/3). For these small oligomeric states, the charges carried on the cluster equals (or exceeds) the number of monomer units, and any charge – charge repulsion should therefore have a much stronger influence on the collision cross section than on larger oligomers where the charge density per monomer unit is much less.

Figure S2: (a) Increase in collision cross section σ with the addition of a single extra charge as a function of the oligomer size n for the peptides studied in this work (red squares). (b) relative difference in cross section between an ideal fibril piece and an ideal spherical oligomer for NNQQNY (red) and VEALYL (green). Note the difference in cross section due to conformational conversion (b) is much larger than the increase in cross section from coulombic repulsion (a), and this difference continues to increase with n.Data recorded on the high-pressure instrument.

The modest change in collision cross section due to coloumbic repulsion is clearly smaller than the difference in cross section between a fibril and a spherical oligomer,where changes of 10% - 20% are expected up to n = 9. Additionally, the difference in cross section increases with increasing oligomer number n. Therefore, it is clear the observation of extended oligomer structures results from an increased cluster size and is not due to charge-charge repulsion.

S3. Confirmation of fibrillar growth by Atomic Force Microscopy (AFM)Atomic force microscopy (AFM) has been utilized to probe the self-assembly mechanisms deduced by MS-IMS. For AFM images aliquots of the same peptide sample solutions used for IMS-MS analysis were drop cast onto freshly cleaved mica slides after several weeks of incubation and imaged on a MFP-3D-BIO instrument (Asylum Research, Santa Barbara, USA), as described elsewhere 1. AFM results clearly support the described self-assembly mechanisms. Only granular macroscopic aggregates are found for YGGFL (Figure 1 and Figure S3), while NNQQNY (Figure S4) and SSTNVG (Figure S6) show coexisting isotropic and fibrous peptide aggregates. AFM analysis of the VEALYL solution indicates only fibrous macroscopic aggregates (Figure 1 and Figure S5). AFM analysis is thus completely consistent with the self-assembly pathwaysdeduced from our IMS-MS analysis. This demonstrates that the oligomer growth observed by IMS-MS (Figure 3 in the main text) represent early events and oligomer states en route to fibril formation. AFM images depicted for YGGFL (Figure S3) and




- S6 -

VEALYL aggregates (Figure S5) are identical to those depicted in the main text (Figure 1) and are reproduced here for completeness only.

Figure S3: AFM image of YGGFL (600μM). Only isotropic aggregates (circles) are seen.

Figure S4: AFM images of NNQQNY (300μM). The fibril structures have a height of ca. 6nm. Please note small isotropic aggregates (circles) as well as protofibrillar assemblies.




- S6 -

VEALYL aggregates (Figure S5) are identical to those depicted in the main text (Figure 1) and are reproduced here for completeness only.

Figure S3: AFM image of YGGFL (600μM). Only isotropic aggregates (circles) are seen.

Figure S4: AFM images of NNQQNY (300μM). The fibril structures have a height of ca. 6nm. Please note small isotropic aggregates (circles) as well as protofibrillar assemblies.

- S7 -

Figure S5: AFM images of VEALYL (1.6mM). The fibril structures have a diameter of ca. 500nm.

Figure S6: AFM image of SSTNVG (1.3mM). The fibril structures have a diameter of ca. 6nm. Note the existence of several granular, spatially isotropic peptide assemblies of varying size (circles).

S4. Mass spectrometry and ion mobility data used to compile the growth curves in Figure 2 in the main text

S4.1. Methods and Materials All peptide samples were purchased from Sigma (St. Louis, MO) or custom synthesized by Genscript Corp. (Piscataway, NJ). The samples were dissolved in 50:50 Water:Methanol to give stock solutions at concentrations ranging from 1 to 3 mM. The samples were then diluted to the desired concentration (typically 10-100μM) and loaded into gold coated nanoESI capillaries for experiments.




- S8 -

The samples were electrosprayed on home built instruments described in detail elsewhere2,3. Briefly, the ions are focused and stored in a source ion funnel and are subsequently pulsed into a drift tube filled with 13 torr of He. They are pulled down the length of the drift tube under the influence of a weak electric field. At the end of the drift tube the ions are collected by a second ion funnel where they are steered and focused into a quadrupole mass filter for mass selection and subsequently are detected as a function of time in the arrival time distribution (ATD). Collision cross section in He gas are measured by mass-selecting peaks in the mass spectrum by their m/z value, and recording ATDs at different drift voltages V (with V ~ 1800, 2300, 2800, 3300, and 4300 kV) for each selected m/zvalue. The center of each peak in a particular ATD is taken as arrival time ta and plotted as a function of the p/V ratio (where V is the drift voltage; see Eq. 4.3). The slope of this correlation is related to the mobility K0 of the corresponding feature, which in turn is inversely proportional to the collision cross-section σ 4. All factors in these equations are either universal constants or parameters that were accurately measured in the experiment. The collision cross-sections are determined with an accuracy of ±1% 2. This instrument is denoted as the high-pressure instrument.

Several samples were also studied on a different IMS/MS instrument, described in detail elsewhere 3,5. Briefly, ion mobility experiments were conducted on a home built instrument comprised of a nano-electrospray ionization (nano-ESI) source, an ion funnel, a temperature controlled drift cell, a quadrupole mass analyzer and an electron multiplier detector. Samples were loaded into gold coated borosilicate capillaries that had been pulled to a fine point on an in-house micropipette puller (Sutter Instrument Co., Novato, CA). A positive voltage was applied to the tip and ions were extracted and entered the instrument through a small (0.010” id) orifice. The ions were collected, desolvated and focused in the ion funnel where they were stored immediately preceding the drift cell. A 10 μs pulse of ions was injected into the drift cell filled with ~5 torr of He gas where they gently drift through under the influence of a weak electric field (5-20 V/cm). The injection energy was usually kept as low as possible to minimize thermal heating of the ions during the injection process. Following the drift cell the ions were mass analyzed with a quadrupole mass filter and detected with a traditional conversion dynode/electron multiplier arrangement. This instrument has a mobility resolving power of about 15 and is referred to as the low-pressure instrument in the following. Arrival times ta were recorded at different drift voltages V (V ~ 90V, 50V, 40V, 30V) in analogy to the description above. Again, arrival times were plotted as a function of the ratio p/V, and the resulting mobility K0 was used to obtain the collision cross section σ in Helium gas.

Ions that are injected into the drift cell experience a constant force from the electric field, E, which is balanced by a retarding frictional force due to collisions with the buffer gas. Under these conditions the velocity, υd, at which the ions move through the buffer gas is proportional to the magnitude of the electric field where the mobility, K is the proportionality constant:

(4.1)d KEυ =




- S8 -

The samples were electrosprayed on home built instruments described in detail elsewhere2,3. Briefly, the ions are focused and stored in a source ion funnel and are subsequently pulsed into a drift tube filled with 13 torr of He. They are pulled down the length of the drift tube under the influence of a weak electric field. At the end of the drift tube the ions are collected by a second ion funnel where they are steered and focused into a quadrupole mass filter for mass selection and subsequently are detected as a function of time in the arrival time distribution (ATD). Collision cross section in He gas are measured by mass-selecting peaks in the mass spectrum by their m/z value, and recording ATDs at different drift voltages V (with V ~ 1800, 2300, 2800, 3300, and 4300 kV) for each selected m/zvalue. The center of each peak in a particular ATD is taken as arrival time ta and plotted as a function of the p/V ratio (where V is the drift voltage; see Eq. 4.3). The slope of this correlation is related to the mobility K0 of the corresponding feature, which in turn is inversely proportional to the collision cross-section σ 4. All factors in these equations are either universal constants or parameters that were accurately measured in the experiment. The collision cross-sections are determined with an accuracy of ±1% 2. This instrument is denoted as the high-pressure instrument.

Several samples were also studied on a different IMS/MS instrument, described in detail elsewhere 3,5. Briefly, ion mobility experiments were conducted on a home built instrument comprised of a nano-electrospray ionization (nano-ESI) source, an ion funnel, a temperature controlled drift cell, a quadrupole mass analyzer and an electron multiplier detector. Samples were loaded into gold coated borosilicate capillaries that had been pulled to a fine point on an in-house micropipette puller (Sutter Instrument Co., Novato, CA). A positive voltage was applied to the tip and ions were extracted and entered the instrument through a small (0.010” id) orifice. The ions were collected, desolvated and focused in the ion funnel where they were stored immediately preceding the drift cell. A 10 μs pulse of ions was injected into the drift cell filled with ~5 torr of He gas where they gently drift through under the influence of a weak electric field (5-20 V/cm). The injection energy was usually kept as low as possible to minimize thermal heating of the ions during the injection process. Following the drift cell the ions were mass analyzed with a quadrupole mass filter and detected with a traditional conversion dynode/electron multiplier arrangement. This instrument has a mobility resolving power of about 15 and is referred to as the low-pressure instrument in the following. Arrival times ta were recorded at different drift voltages V (V ~ 90V, 50V, 40V, 30V) in analogy to the description above. Again, arrival times were plotted as a function of the ratio p/V, and the resulting mobility K0 was used to obtain the collision cross section σ in Helium gas.

Ions that are injected into the drift cell experience a constant force from the electric field, E, which is balanced by a retarding frictional force due to collisions with the buffer gas. Under these conditions the velocity, υd, at which the ions move through the buffer gas is proportional to the magnitude of the electric field where the mobility, K is the proportionality constant:

(4.1)d KEυ =

- S9 -

The absolute mobility of the ion packet is dependent on the temperature and the pressure of the buffer gas so it is customary to use the reduced mobility K0.

(4.2)

After the ions exit the drift cell they are mass analyzed and detected as a function of the arrival time, tA. The reduced mobility can be determined from instrumental parameters by converting equation (4.1) to equation (4.3) and plotting tA versus p/V.

(4.3)

In equation 4.3, l is the length of the drift cell, p is the gas pressure in torr, V is the voltage across the cell, and t0 is the time the ions spend outside the drift cell before hitting the detector. Collision theory 4 relates K0 to σ, the collision cross section of the ion of interest:

(4.4)

Here, N is the buffer gas number density, μ is the reduced mass of the collision system (ion + He) and kb is Boltzmann’s constant. The collision cross sections are reported as the average of multiple measurements, typically more than three, and all measurements are reproducible within 1%. All data presented in the manuscript and SI are given as raw data, with no smoothing and no post-processing.The measured reduced mobility (K0) and subsequently the collision cross section (σ)contain information about the three dimensional configuration of the ion.

The flux of ions exiting the drift tube can be calculated 4 and is used to fit the experimental arrival time distributions. It is assumed that the ion packet takes the form of a periodic delta function and the flux is given by equation 4.5.

(4.5)

0273.16K

760torrpK K

T = ⋅ ⋅

2

00

273.16K760torrA

l pt tK T V

= ⋅ ⋅ +⋅

12

03 2 1

16 b

qKN k T

πµ σ

=

( )( )

( )220

T 44

12

0, , 14

dL

z v tD tD t

rt

d

L

s a e zz t v e etD t

α

π

− − ⋅− − ⋅

⋅ ⋅ Φ = ⋅ + ⋅ − ⋅




- S10 -

Here z is the ion charge, r0 is the radius of the initial ion packet, a is the area of the exit aperture, DL and DT are the longitudinal and transverse diffusion coefficients, s is the initial ion density and α is the loss of ions due to reactions in the drift tube.

Several of the samples were also studied on a prototype of the commercial Waters Synapt instrument 6. High resolution mass spectra were obtained on a quadrupole-time-of-flight (qTOF) instrument (Waters UK) equipped with a nano-ESI source. The structures of fibers and steric zippers were taken from the reported x-ray structures 7 and hydrogens were added by the InsightII (Accelrys, San Diego, USA) suite of programs. This instrument was used only to obtain high resolution mass spectra, and no arrival time information obtained with this instrument was used to compile structural information information discussed in this work.

Theoretical collision-cross sections for molecular geometries were computed as follows.Briefly, molecular geometries for the fibril pieces of NNQQNY, VEALYL and SSTNVG were obtained from x-ray fibril structures published by Eisenberg and co-workers 8. The Mercury software suite9 was used to extract single fibril pieces with oligomer numbers n=1 up to n=20, or steric zippers (double-strand fibril pieces) where appropriate. The InsightII software suite10 was used to include hydrogen atoms at appropriate positions. Collision cross sections σ in Helium at T=300K were then computed by the trajectory method as described elsewhere 11.




- S10 -

Here z is the ion charge, r0 is the radius of the initial ion packet, a is the area of the exit aperture, DL and DT are the longitudinal and transverse diffusion coefficients, s is the initial ion density and α is the loss of ions due to reactions in the drift tube.

Several of the samples were also studied on a prototype of the commercial Waters Synapt instrument 6. High resolution mass spectra were obtained on a quadrupole-time-of-flight (qTOF) instrument (Waters UK) equipped with a nano-ESI source. The structures of fibers and steric zippers were taken from the reported x-ray structures 7 and hydrogens were added by the InsightII (Accelrys, San Diego, USA) suite of programs. This instrument was used only to obtain high resolution mass spectra, and no arrival time information obtained with this instrument was used to compile structural information information discussed in this work.

Theoretical collision-cross sections for molecular geometries were computed as follows.Briefly, molecular geometries for the fibril pieces of NNQQNY, VEALYL and SSTNVG were obtained from x-ray fibril structures published by Eisenberg and co-workers 8. The Mercury software suite9 was used to extract single fibril pieces with oligomer numbers n=1 up to n=20, or steric zippers (double-strand fibril pieces) where appropriate. The InsightII software suite10 was used to include hydrogen atoms at appropriate positions. Collision cross sections σ in Helium at T=300K were then computed by the trajectory method as described elsewhere 11.

- S11 -

S4.1. ESI-Quadrupole mass spectra

Figure S7: ESI-Q mass spectra (positive ion mode) obtained on the high-pressure drift tube instrument for NNQQNY, SSTNVG, VEALYL and YGGFL. Note the abundant formation of oligomers for all four peptides investigated. The peaks are identified by their n/z value where n is the oligomer number and z the overall charge.

Figure S7 depicts ESI-Quadrupole mass spectra of the four systems investigated obtained on the high-pressure instrument. In all four systems YGGFL, NNQQNY, VEALYL and SSTNVG intensive aggregation is visible. ESI-Q mass spectra for VEALYL and YGGFL depicted here are identical to those presented in the main text (Figure 1) and are reproduced here only for completeness. Oligomers comprised of up to 30 monomer units can be identified on the ESI-qTOF instrument (data not shown). Please note that many peaks are comprised of several distinct oligomers (see Figures S8, S9, S11, S14).




- S12 -

S4.3. Mobility data for YGGFL

Figure S8: Representative ATDs recorded for the peptide YGGFL that have been used to generate the oligomer growth curves in Figure 2 in the main text. The oligomer state is given in n/z notation along with the resulting collision cross section in Å2. The data are in black while peak shapes calculated for a single conformation by equation (4.5) are given in color. Data recorded on the high-pressure instrument.

Figure S8 depicts representative ATDs for several oligomeric states of YGGFL recorded at ca. 2.8kV on the high-pressure instrument. Oligomeric states are denoted in the n/znotation with n being the oligomeric number, and z the overall charge on the cluster. It is of note that with the exception of the pentamer (n/z=5/3) only one feature can be unambiguously identified for each oligomeric state. Included in the figures are fits of the ATDs to equation 4.5. It is noted that for n/z ≥ 5/3 the peak widths are much larger than expected, indicating the existence of several conformations belonging to one structural family. The full-width at half maxixum (FWHM) of these peaks are in the range of ±2%. This shows that a conformational conversion is not observed for YGGFL. This data was used to compile Figure 2a in the main text.




- S12 -

S4.3. Mobility data for YGGFL

Figure S8: Representative ATDs recorded for the peptide YGGFL that have been used to generate the oligomer growth curves in Figure 2 in the main text. The oligomer state is given in n/z notation along with the resulting collision cross section in Å2. The data are in black while peak shapes calculated for a single conformation by equation (4.5) are given in color. Data recorded on the high-pressure instrument.

Figure S8 depicts representative ATDs for several oligomeric states of YGGFL recorded at ca. 2.8kV on the high-pressure instrument. Oligomeric states are denoted in the n/znotation with n being the oligomeric number, and z the overall charge on the cluster. It is of note that with the exception of the pentamer (n/z=5/3) only one feature can be unambiguously identified for each oligomeric state. Included in the figures are fits of the ATDs to equation 4.5. It is noted that for n/z ≥ 5/3 the peak widths are much larger than expected, indicating the existence of several conformations belonging to one structural family. The full-width at half maxixum (FWHM) of these peaks are in the range of ±2%. This shows that a conformational conversion is not observed for YGGFL. This data was used to compile Figure 2a in the main text.

- S13 -

S4.4. Mobility data for NNQQNYFigure S9 depicts representative ATDs for several oligomeric states of NNQQNYrecorded at ca. 2.8kV on the high-pressure instrument. The only ATD depicted that was recorded on the low-pressure instrument is the 16/7 oligomer. Oligomeric states are denoted in the n/z notation with n being the oligomeric number, and z the overall charge on the cluster. Impurities of the pentapeptide [NNQQNY - Asn] in the sample of about 2% results in satellite peaks in the mass spectra, which were useful to unambiguously assign oligomeric states of NNQQNY. Peaks in the ATDs that are assigned to an oligomeric assembly including one [NNQQNY - Asn] monomer unit are denoted in the Figure as n#/z, i.e. 6#/4 denotes a 6/4 oligomer with one Asn less than 6/4. These assignments were done by comparing the reduced mobilities K0 with the parent oligomer.

Included in the Figure are fits of the ATDs to equation 4.5. The FWHM indicates that each feature in most ATDs is significantly broader than the model peak width indicating it is actually comprised of many similar conformations.

It is noted that the ATD for the formal monomer peak (m/z 780) is quite complex with several individual features for the dimer and trimer. Starting with the hexamer 6/4 more than one feature is observed for most oligomeric states with collision cross sections often differing by more than 10%. These data was used to compile Figure 2b in the main text.




- S14 -

Figure S9: Representative ATDs recorded for the peptide NNQQNY that have been used to generate the oligomer growth curves in Figure 2 in the main text. The oligomer state is given in n/z notation along with the resulting collision cross section in Å2. Features that are assigned to have one Asparagin unit less are indicated by ‘#’. Included also are the theoretically expected peak widths assuming the peak represented only one conformational family. The width of the experimental peaks indicate that a number of conformations are observed for an individual feature. Asterisks (*) denote noise peaks. Data recorded on the high-pressure instrument, expect for m/z 1782, which was recorded on the low-pressure instrument.




- S14 -

Figure S9: Representative ATDs recorded for the peptide NNQQNY that have been used to generate the oligomer growth curves in Figure 2 in the main text. The oligomer state is given in n/z notation along with the resulting collision cross section in Å2. Features that are assigned to have one Asparagin unit less are indicated by ‘#’. Included also are the theoretically expected peak widths assuming the peak represented only one conformational family. The width of the experimental peaks indicate that a number of conformations are observed for an individual feature. Asterisks (*) denote noise peaks. Data recorded on the high-pressure instrument, expect for m/z 1782, which was recorded on the low-pressure instrument.

- S15 -

Figure S10: Growth curve for NNQQNY with all data not assigned as fibril-like omitted from the plot(red) as well as the ideal fibril growth as extracted from x-ray data (blue). A linear regression is indicated for the IMS data (r2=0.98,k=122 Å2, a = 90 Å2). Note the close agreement between the fibril growth observed in the MS-IMS experiment and that derived from x-ray data (k=130 Å2, a = 99 Å2).




- S16 -

S4.5. Mobility data for VEALYL

Figure S11: Representative ATDs recorded for the peptide VEALYL that have been used to generate the oligomer growth curves in Figure 2 in the main text. The oligomer state is given in n/z notation along with the resulting collision cross section in Å2. Asterisks (*) denote noise peaks. The data are in black while peak shapes calculated for a single conformation by equation (4.5) are given in color. Data recorded on the high-pressure instrument.

Figure S11 depicts representative ATDs for several oligomeric states of VEALYL recorded at ca. 2.8kV on the high-pressure instrument. Included in the Figure are fits of the ATDs to equation 4.5, indicating a distribution of conformers in each structural family. Only a single broad feature is observed for all oligomers except n = 8, 9.




- S16 -

S4.5. Mobility data for VEALYL

Figure S11: Representative ATDs recorded for the peptide VEALYL that have been used to generate the oligomer growth curves in Figure 2 in the main text. The oligomer state is given in n/z notation along with the resulting collision cross section in Å2. Asterisks (*) denote noise peaks. The data are in black while peak shapes calculated for a single conformation by equation (4.5) are given in color. Data recorded on the high-pressure instrument.

Figure S11 depicts representative ATDs for several oligomeric states of VEALYL recorded at ca. 2.8kV on the high-pressure instrument. Included in the Figure are fits of the ATDs to equation 4.5, indicating a distribution of conformers in each structural family. Only a single broad feature is observed for all oligomers except n = 8, 9.

- S17 -

Figure S12: Growth curve for VEALYL from n = 1 to n = 5. The n/z = 3/3 feature has been omitted from the regression for oligomer states n = 2 to n = 5. Note the close resemblance between the single-strand fiber growth as expected from x-ray data and the linear regression to the experimental MS-IMS data.

Figure S12 represents a close-up of the growth curve obtained for VEALYL for the oligomer sizes n = 1 to n = 5. It is clearly seen that the experiment gives cross sections much larger than expected for compact, spatially isotropic growth. Also, a close resemblance between the experimental and the single strand fiber cross sections derived from x-ray 7 is apparent. It is also of note that the slope a and especially the intercept k of the fibril growth (a = 91 Å2 and k = 118 Å2) and the linear regression for n = 2 to n = 5 (a= 96 Å2 and k = 120 Å2) are in excellent agreement with each other. This clearly supportsgrowth as a single strand fiber.




- S18 -

Figure S13: Growth curve for VEALYL from n = 5 to n = 14. Note the excellent agreement between the fibril growth as expected from x-ray data for a steric zipper (a = 71 Å2 and k = 265 Å2) and the experimental MS-IMS data (a = 62 Å2 and k = 265 Å2).

Figure S13 shows the dominant features of the VEALYL growth curve from n = 5 to n =14. The compact, clearly isotropic oligomers for n=8 and n=9 have been omitted from this plot for clarity. The linear regression gives values for a (62 Å2) and k (265 Å2) are in excellent agreement to those expected for a ‘steric zipper’ growth (a = 71 Å2 and k = 265Å2).




- S18 -

Figure S13: Growth curve for VEALYL from n = 5 to n = 14. Note the excellent agreement between the fibril growth as expected from x-ray data for a steric zipper (a = 71 Å2 and k = 265 Å2) and the experimental MS-IMS data (a = 62 Å2 and k = 265 Å2).

Figure S13 shows the dominant features of the VEALYL growth curve from n = 5 to n =14. The compact, clearly isotropic oligomers for n=8 and n=9 have been omitted from this plot for clarity. The linear regression gives values for a (62 Å2) and k (265 Å2) are in excellent agreement to those expected for a ‘steric zipper’ growth (a = 71 Å2 and k = 265Å2).

- S19 -

S4.6. Mobility data for SSTNVG

Figure S14: Depicted are representative ATDs recorded for the peptide SSTNVG that have been used to generate the oligomer growth curves in Figure 2. The oligomer state is given in n/z notation along with the resulting collision cross section in Å2. The data are in black while peak shapes calculated for a single conformation by equation (4.5) are given in color. Data recorded on the high-pressure instrument.

Figure S14 depicts representative ATDs for several oligomeric states of SSTNVG recorded at ca. 2.8kV on the high-pressure instrument. Included in the Figure are fits of the ATDs to equation 4.5. The FWHM of typically ±2-3% indicates that also here eachfeature in the ATD (i.e. one major structural family) is actually comprised of many similar conformations. For n ≥ 12 multiple features start to emerge, which include extended structures.




- S20 -

S4.7. Tabulated collision cross sections for the peptides YGGFL, SSTNVG , NNQQNY and VEALYL

Table S1. Collision cross section data for the peptides YGGFL, SSTNVG, NNQQNY, and VEALYL tabulated according to oligomer number (n) and number of positive charges (z). Values are given in Å2. Measurements were performed on the high-pressure instrument unless noted (see Supplementary Information section S4.1).

YGGFL SSTNVG

n z

collision cross section

[Å2] n z


[Å2]1 1 160 1 1 1482 1 245 2 2 2402 2 255 3 2 3113 2 335 4 3 3864 2 407 5 2 4325 2 466 6 4 5235 3 480 7 3 5466 3 540 7 4 5707 3 591 9 5 7178 4 668 10 4 6999 4 710 12 5 916

11 4 797 12 5 81112 5 863 12 5 75813 5 904 13 5 81714 5 935 13 5 88017 6 1075 14 5 819

14 5 89314 5 92214 5 1004




- S20 -

S4.7. Tabulated collision cross sections for the peptides YGGFL, SSTNVG , NNQQNY and VEALYL

Table S1. Collision cross section data for the peptides YGGFL, SSTNVG, NNQQNY, and VEALYL tabulated according to oligomer number (n) and number of positive charges (z). Values are given in Å2. Measurements were performed on the high-pressure instrument unless noted (see Supplementary Information section S4.1).

YGGFL SSTNVG

n z


[Å2] n z


[Å2]1 1 160 1 1 1482 1 245 2 2 2402 2 255 3 2 3113 2 335 4 3 3864 2 407 5 2 4325 2 466 6 4 5235 3 480 7 3 5466 3 540 7 4 5707 3 591 9 5 7178 4 668 10 4 6999 4 710 12 5 916

11 4 797 12 5 81112 5 863 12 5 75813 5 904 13 5 81714 5 935 13 5 88017 6 1075 14 5 819

14 5 89314 5 92214 5 1004

- S21 -

NNQQNY VEALYL

n z


[Å2] n z


[Å2]1 1 186 1 1 1901 2 193 2 2 3102 2 295 2 2 2853 3 411 3 2 4053 3 390 3 3 4284 4 491 4 2 4785 3 517 4 3 5005 4 543 6 3 6326 4 599 6 4 6587 5 686 7 3 6958 5 731 7 4 7209 7 1040 8 4 7689 7 962 8 4 7209 5 900 b 9 4 8249 5 783 b 9 4 769

10 6 875 9 5 891 a

10 6 947 9 5 853 a

10 7 1015 11 4 95410 7 1110 14 6 111911 7 988 a

11 7 1008 a

11 7 1057 a

12 7 122512 7 101013 7 102613 7 109915 7 1185 a, b

16 7 1183 b

16 7 1442 b

19 13 190220 13 1961

a ATD not shownb data recorded on the low-pressure instrument




- S22 -

S5. Dynamic aspects of the structural evolution of peptide oligomers investigated in this study Figures S14 and S15 show the time dependency of UV absorption for various concentrations of the peptide VEALYL. Prior to UV analysis, the incubated peptide solution was centrifuged at 13.000rpm for 10min to separate soluble peptide oligomers from any insoluble peptide aggregates, and only the supernatant was used for UV analysis. No drop in UV absorbance was observed for the ca. 125 µM peptide solution within 12 days. In contrast, formation and precipitation of insoluble aggregates was observed for the 360 and 825 µM peptide solutions. We note a concentration-dependent lag time, consistent with an aggregation mechanism involving the formation of a nucleus.The lag-time for the 360 µM solution was > 1 day and the 825 µM solution showed a lag time in the minutes time-scale.

Figure S14: UV absorption at λ=280nm (A280) normalized to A280 at t=0h as a function of incubation time in 50:50 MeOH:H2O for different initial concentrations of VEALYL (125, 360 and 825 µM, respectively).

Figures S15 shows the time evolution of mass spectra taken from the same sample with an initial concentration of 825 µM VEALYL followed for 2500 minutes starting with the time of sample preparation. The major effect visible in Fig. S15 is the increasing abundance of monomer relative to that of dimer or other oligomers. This is consistent with increased formation of an insoluble fraction over time that is in steady state with the monomer, and with a stable steady state soluble oligomer distribution. No significant change in collision cross sections was observed (i.e. less than 1%) over the course of this study (i.e. several weeks).




- S22 -

S5. Dynamic aspects of the structural evolution of peptide oligomers investigated in this study Figures S14 and S15 show the time dependency of UV absorption for various concentrations of the peptide VEALYL. Prior to UV analysis, the incubated peptide solution was centrifuged at 13.000rpm for 10min to separate soluble peptide oligomers from any insoluble peptide aggregates, and only the supernatant was used for UV analysis. No drop in UV absorbance was observed for the ca. 125 µM peptide solution within 12 days. In contrast, formation and precipitation of insoluble aggregates was observed for the 360 and 825 µM peptide solutions. We note a concentration-dependent lag time, consistent with an aggregation mechanism involving the formation of a nucleus.The lag-time for the 360 µM solution was > 1 day and the 825 µM solution showed a lag time in the minutes time-scale.

Figure S14: UV absorption at λ=280nm (A280) normalized to A280 at t=0h as a function of incubation time in 50:50 MeOH:H2O for different initial concentrations of VEALYL (125, 360 and 825 µM, respectively).

Figures S15 shows the time evolution of mass spectra taken from the same sample with an initial concentration of 825 µM VEALYL followed for 2500 minutes starting with the time of sample preparation. The major effect visible in Fig. S15 is the increasing abundance of monomer relative to that of dimer or other oligomers. This is consistent with increased formation of an insoluble fraction over time that is in steady state with the monomer, and with a stable steady state soluble oligomer distribution. No significant change in collision cross sections was observed (i.e. less than 1%) over the course of this study (i.e. several weeks).

- S23 -

Figure S15: ESI-qTOF mass spectra of VEALYL in 50:50 MeoH:H2O as a function of time. Initial concentration ca. 825µM.




- S24 -

References

1 Dong, M. et al., AFM-based force spectroscopy measurement of mature amyloid fibrils of the epeptide glucagon. Nanotechnology 19, 384013 (2008).

2 Kemper, P.R., Dupuis, N.F., & Bowers, M.T., A new, higher resolution, ion mobility mass spectrometer. Int. J. Mass Spectrom., in press (2009).

3 Wyttenbach, T., Kemper, P.R., & Bowers, M.T., Design of a new electrospray ion mobility mass spectrometer. Int. J. Mass Spectrom. 212 (1-3), 13-23 (2001).

4 Mason, E.A. & McDaniel, E.W., Transport Properties of Ions in Gases. (Wiley, 1988).

5 Wyttenbach, T. & Bowers, M.T., Gas-phase conformations: The ion mobility/ion chromatography method. Top Curr Chem 225, 207-232 (2003).

6 Pringle, S.D. et al., An investigation of the mobility separation of some peptide and protein ions using a new hybrid quadrupole/travelling wave IMS/oa-ToF instrument. Int. J. Mass Spectrom. 261, 1-12 (2007).

7 Sawaya, M.R. et al., Atomic structures of amyloid cross-β spines reveal varied steric zippers. Nature 447, 453-457 (2007).

8 Sawaya, M.R. et al., Atomic structures of amyloid cross-beta spines reveal varied steric zippers. Nature 447 (7143), 453-457 (2007).

9 Macrae, C.F. et al., Mercury: visualization and analysis of crystal structures. J. Appl. Cryst. 39, 453-457 (2006).

10 Accelrys InsightII (San Diego, CA, 2000).11 Mesleh, M.F., Hunter, J.M., Shvartsburg, A.A., Schatz, G.C., & Jarrold, M.F.,

Structural Information from Ion Mobility Measurements: Effects of the Long Range Potential. J. Phys. Chem. A 100, 16082-16086 (1996).


Ion mobility-mass spectrometry reveals a … cross section growth curves for various types of...

Documents

Transcript of Ion mobility-mass spectrometry reveals a … cross section growth curves for various types of...