Magic Angle Spinning Frequencies Beyond 300 kHz are Necessary to

Jun 19, 2018 - In the last decade, proton detection in MAS solid-state NMR became a popular strategy for biomolecular structure determination. In part...
0 downloads 0 Views 985KB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

C: Physical Processes in Nanomaterials and Nanostructures

Magic Angle Spinning Frequencies Beyond 300 kHz are Necessary to Yield Maximum Sensitivity in Selectively Methyl Protonated Protein Samples in Solid State NMR Kai Xue, Riddhiman Sarkar, Carina Motz, Sam Asami, Venita Decker, Sebastian Wegner, Zdenek Tosner, and Bernd Reif J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b05600 • Publication Date (Web): 19 Jun 2018 Downloaded from http://pubs.acs.org on June 25, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Magic Angle Spinning Frequencies beyond 300 kHz are Necessary to Yield Maximum Sensitivity in Selectively Methyl Protonated Protein Samples in Solid State NMR

Kai Xue,a Riddhiman Sarkar,a,b,* Carina Motz,b Sam Asami,b Venita Decker,c Sebastian Wegner,c Zdenek Tosner,b,d* and Bernd Reifa,b*

June 05, 2018

a

Helmholtz-Zentrum München (HMGU), Deutsches Forschungszentrum für

Gesundheit und Umwelt, Ingolstädter Landstr. 1, 85764 Neuherberg, Germany b

Munich Center for Integrated Protein Science (CIPS-M) at Department Chemie, Technische Universität München (TUM), Lichtenbergstr. 4, 85747 Garching, Germany c

d

Bruker BioSpin GmbH, Silberstreifen 4, 76287 Rheinstetten, Germany

Department of chemistry, Faculty of Science, Charles University, Hlavova 8, 12842 Praha 2, Czech Republic

For submission to: Journal of Physical Chemistry C To whom correspondence should be addressed: [email protected], [email protected], [email protected]

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT In the last decade, proton detection in MAS solid-state NMR became a popular strategy for biomolecular structure determination. In particular, probe technology has experienced tremendous progress with smaller and smaller diameter rotors achieving ever higher MAS frequencies. MAS rotation frequencies beyond 100 kHz allow to observe and assign protons in fully protonated samples. In these experiments, resolution is however compromised as homogeneous proton-proton dipolar coupling interactions are not completely averaged out. Using a combination of experiments and simulations, we analyze the MAS frequency dependent intensities of the 1H,13C methyl correlation peaks of a selectively methyl protonated (CH3) microcrystalline sample of the chicken α-spectrin SH3 domain (αSH3). Extensive simulations involving 9 spins employing the program SIMPSON allow the prediction of MAS frequency dependence on the proton intensities. The experimental results are used to validate the simulations. As quantitative measure, we determine the characteristic MAS frequency which is necessary to obtain >50 % of the maximum achievable sensitivity. Our results show that this frequency is sitespecific and strongly depends on the local methyl density. We find that the characteristic MAS frequency ranges from as low as 20 kHz up to 324 kHz with the average value of (135 ± 88) kHz for this particular sample at a magnetic field strength of 11.7 T. Inclusion of side chain dynamics in the analysis reduces the average characteristic MAS frequency to (104 ± 68) kHz within the range of 11-261 kHz. In case, >80 % of the maximum sensitivity shall be achieved, a MAS rotation frequency of (498 ± 370) kHz and (310 ± 227) kHz is required with and without including side chains dynamics in the analysis, respectively.

2 ACS Paragon Plus Environment

Page 2 of 14

Page 3 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

INTRODUCTION Solid-state NMR for biological samples has seen tremendous progress in the last decade

1-6

. It is now possible to extract high-resolution structural information of

protein samples using proton-detected experiments, which are significantly more sensitive in comparison to

13

C or

15

N detected experiments. Resolution of proton

spectra in solids has always been an Achilles heel. In the recent years, two strategies have been applied to overcome this issue. On one hand, partial sitespecific deuteration

7-13

, on the other hand design of probes that yield increased

rotation frequencies (up to 111 kHz as of today)

14-16

. Despite the success of proton

detected experiments at very high MAS rotation frequencies, it is not clear which gains in sensitivity are to be expected if it would be possible to increase the MAS rotation frequency even further. The general benefits for proton detection have been discussed extensively in the literature

8, 17

. The detection sensitivity is proportional to

the full width at half height (∆ν1/2) and to the transfer efficiency of the experiment. Both parameters depend on the MAS frequency employed in the experiments. Given the complexity of probe design, performance of pulse sequences, rf inhomogeneities and the network of anisotropic interactions, the maximum achievable sensitivity for a given sample is hard to predict. The standards for resolution are set by comparison to spectra obtained from deuterated samples. The expected gains in terms of sensitivity are not well understood, as also perdeuterated samples yield an increase in sensitivity upon increase of the MAS frequency. We have shown recently that deuterated samples yield higher sensitivity and better resolution in 1H,13C correlation experiments at 111 kHz spinning, compared to protonated protein samples

18

.

Comparing RAP (Reduced Adjoining Protonation) labelled and selectively methyl protonated samples, we found that even MAS rotation frequencies of 111 kHz are insufficient to efficiently average proton-proton dipolar interactions in the selectively methyl protonated samples. In this paper, we aim to predict the MAS rotation frequencies that are necessary to yield optimum sensitivity in proton detected experiments using a selectively methyl protonated micro-crystalline sample of the chicken α-spectrin SH3 domain. We employ a combination of experiments and simulations to estimate the sensitivity at MAS frequencies beyond 110 kHz that are not amenable with current probe technology. Selectively methyl protonated protein samples are suited for this purpose, as we restrict ourselves to a 9-proton spin system that still can be calculated with reasonable effort. In a recent study to predict proton line widths for a

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

fully protonated dipeptide at MAS frequencies up to 100 kHz by Sternberg et. al.19, a 48 spin simulation has been implemented.

RESULTS AND DISCUSSION The work described in this manuscript was driven by the observation that the cross peak intensities of 1H,13C correlation spectra of a selectively methyl protonated microcrystalline sample of the chicken α-spectrin SH3 domain increase as a function of the MAS frequency (Figure 1A). While only a few cross peaks can be observed at 40 kHz MAS, the number of cross peaks increases significantly at 70 kHz MAS. At 111 kHz, it is possible to observe all the expected valine and leucine cross peaks. We find that cross peak intensities vary significantly among different methyl groups at a given MAS frequency. We suspect that this is due to variations of the magnitude of the 1H-1H dipolar couplings in the core of the protein.

1

13

Figure 1. (A) H, C correlation spectra for a selectively methyl protonated micro-crystalline sample of the chicken α-spectrin SH3 domain. The protein was produced by adding α20 ketoisovalerate to the perdeuterated M9 minimal medium yielding CH3 isotopomers . Experiments were recorded at MAS frequencies of 40, 70 and 111 kHz using INEPT for magnetization transfer. The measurements were performed at 11.7 T (500 MHz) using a 0.7 mm MAS probe. All spectra were recorded and processed using the same parameters and plotted with the same contour levels (setting the maximum contour level to the cross peak of

4 ACS Paragon Plus Environment

Page 4 of 14

Page 5 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

L8δ2). (B) Simulated 1H spectra for every methyl group in the protein assuming a MAS rotation frequency of 110 kHz (black: without side chain dynamics; red: with side chain dynamics using experimental order parameter values 12). Calculations are performed using the program SIMPSON 21-22. (C) In the simulations, a 9-proton spin system is considered using the atomic coordinates from the PDB file of the α-spectrin SH3 domain (2NUZ). Intermethyl interactions were treated using methyl pseudo-atoms.

We have chosen selectively methyl protonated protein samples, as this kind of labelling allows to reduce the complexity of the numerical simulations and restrict the number of involved proton spins to < 10. We first numerically simulated the proton line shapes (Figure 1B), considering the explicit geometry of the atoms from the αSH3 structure, PDB ID: 2NUZ

23

. The numerical simulations were carried out using 9

proton spins employing the program SIMPSON

21-22

. Thus, two neighbouring valine

or leucine residues are accounted for (Figure 1C). Typical dipolar couplings are on the order of -10.5 kHz and < 5.7 kHz for intra- and inter-methyl dipolar interactions, respectively. As the protein was re-crystallized from D2O, no additional exchangeable protons had to be taken into account. Additional simulations for isolated methyl groups, as well as for two methyl groups that are dipolar coupled are presented as part of the Supporting Information (Figs. S1-S7). In the simulations, 1H,1H dipolar interactions are scaled due to the three-fold rotation around the C-C axis. We have found recently, that there is significant motion in the protein hydrophobic core parameters

25

11-12, 24

. We have employed REDOR derived order

, and repeated the simulations to find out how dynamics affects the

sensitivity of the MAS dependent proton spectra. In these simulations, the order parameters have been used to re-scale the amplitude of the intra-methyl 1H,1H dipolar interactions. Dynamics of inter-methyl dipolar interactions are not taken into account. The peak intensity of the simulated spectra increases according to the degree of dynamics (Figure 1B). Experimentally we find that INEPT based methods perform significantly better than CP based experiments for the selectively methyl protonated micro-crystalline sample investigated here, in particular at very fast spinning. This is due to the fact that the apparent transverse relaxation rate constant of protons T2’ increases at higher MAS frequencies. Therefore, all experiments shown in this work are recorded using INEPT based magnetization transfer steps. We expect very similar results for CP based experiments. The increase of the MAS dependent signal intensities is more pronounced in case of the scalar coupling based experiments, as the efficiency of the magnetization transfer steps is strongly dependent of the MAS frequency, and is particularly low at slow MAS.

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

To validate whether the simulation can accurately describe the experiments, we correlated the simulated and the experimental peak intensities. The site-specific variations in signal-to-noise ratio (SNR) in the INEPT based correlation can be attributed to differences in the 1H and 13C line widths, and to differences in the proton T2’ relaxation times resulting in different transfer efficiencies. The site-specific proton T2’ relaxation times are shown in Figure 2. We find that proton T2’ relaxation times vary significantly presumably as a function of the proton density around a particular methyl group. As expected, 1H T2’ increases with faster MxAS. A)

B)

Figure 2. (A) Representative site-specific 1H T2’ decay curves for V53γ2 and L34δ1 in αketoisovalerate CH3 labelled SH3. (B) 1H T2’ as a function of the amino acid sequence. Experiments were recorded at a MAS frequency of 100 kHz at a magnetic field strength of 11.8 T (500 MHz).

To normalize the simulated spectra, we adopted the following protocol: We find that at a MAS frequency of 10 MHz and above the intensities of the simulated spectra do not improve any further. The simulated FID was then folded with an exponential line broadening of 20 Hz. The maximum intensity of the respective Fourier transformed spectrum was set arbitrarily to 100. To compare the intensities of the simulated spectra at different MAS rotation frequencies, we assumed that the integral over the total spectral range for all spectra is equal, and used this to re-normalize the peak intensities at the slower MAS rotation frequencies. 6 ACS Paragon Plus Environment

Page 6 of 14

Page 7 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

To account for the T2' signal decay in the INEPT based experiments, we corrected the simulated intensities according to

INT jexp (i) = κ j INT jsim (i) e

−∆/T2 '

(Eq. 1)

where κj is an empirical fitting factor that has been determined for each experiment (recorded at the MAS rotation frequency j). ∆ refers to the total INEPT magnetization transfer delay (4*τ according to Figure S10A), and was set to 7.9 ms in all experiments. κj is obtained by analysis of the correlation diagrams in Figs. 3A-C. For each MAS frequency (70, 90 and 111 kHz), we obtain a linear fit yielding an excellent correlation (R2 > 0.7) between experiments and numerical simulations. Since the numerical simulations describe the proton spectra and their relative intensities for each MAS frequency rather well, we set out next to predict the MAS frequency that is necessary to completely average out the proton-proton dipolar interactions for these kinds of samples. Figure 3D, E show simulated (black) and experimental (red) proton peak intensities for a few representative residues in the SH3 domain sample. Simulated MAS dependent proton spectra for all methyl groups of the valine and leucine residues of the α-spectrin SH3 domain are represented in the Supporting Information (Fig. S7) in the MAS frequency range from 10 kHz to 10 MHz. Matched simulated and experimental intensities for all experimentally accessible valine and leucine methyl groups are shown in Fig. S8.

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 14

Figure 3. Comparison of simulated and experimental cross-peak intensities at 70 (A), 90 (B) and 111 kHz (C) MAS and at an external magnetic field of 11.7 T (500 MHz), respectively. The dashed line represents a linear fit. The correlation coefficient improves significantly if the outliers (L8δ2, L10δ1 and L12δ2) are excluded from the fit (in red). (D),(E) Simulated intensities (black), as well as experimental intensities (red) for the methyl groups L34δ1 and V53γ2 as a function of the MAS frequency. Inclusion of side chain dynamics (open symbols) yields higher proton intensities at a given MAS frequency, as the strength of the proton dipolar coupling network is reduced due to local structural fluctuations. The MAS dependent intensities for all residues are represented in the Supporting Information (Fig. S8). (F) ሺହ଴ሻ Characteristic MAS frequencies ߥெ஺ௌ (frequency that yields 50% of the maximum possible intensity) for each methyl group in α-spectrin SH3. The average characteristic MAS frequency in the absence (black) and in the presence of dynamics (red) amounts to (135.0 ± 88.0) kHz and (104.0 ± 68) kHz, respectively, and is indicated with a dashed line.

All simulated MAS dependent intensity curves show all a sigmoidal like shape in the semi-log plots. To simplify the quantitative description of the MAS dependent peak ሺହ଴ሻ

intensities, we introduce the characteristic MAS frequency ߥெ஺ௌ which is defined as the MAS frequency that is required to yield 50 % of the maximum possible intensity in the numerical simulations for a given residue (Fig. 3F). We find that the ሺହ଴ሻ

characteristic MAS frequency ߥெ஺ௌ covers a relatively broad range of frequencies. For L34δ2, a MAS frequency of 20 kHz is sufficient, whereas for V53γ1 a rotation frequency of 324 kHz is necessary to yield 50 % of the maximum possible signal intensity or amplitude.

8 ACS Paragon Plus Environment

Page 9 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

In order to account for side chain dynamics and its consequences on scaling of the 1

H,1H dipolar couplings, we repeated the simulations, assuming the experimental

values for methyl side chain order parameters

12

. As only very few residues (V23,

L31 and V46) undergo significant side chain dynamics, inclusion of side chain dynamics has a negligible effect on the matching of experimental and simulated intensities. Surprisingly, we find that these dynamic residues have experimental proton intensities that are lower than expected from the simulations or are even only barely visible. We speculate that these side chains undergo ns-µs dynamics

26

. The

intensity of their proton methyl peaks is presumably affected by transverse relaxation effects. In the past, the root sum squared dipolar coupling (dRSS) has been introduced to describe the effects of the effective dipolar coupling in a dense proton network 27-28:

RSS i

d

µ = 0 γ H2 4π

1 ∑ r3  j  i, j 

2

(Eq. 2)

dRSS is a simple descriptor of the dipole-dipole coupling environment and does not include explicit chemical shifts, the spin topology and the dependence on the MAS frequency. In order to find out if dRSS can be employed to characterize the proton spin density, we have represented the characteristic MAS frequency νMAS as a function of dRSS. Even though dRSS drastically simplifies the proton spin system, we find that it describes the experimental results rather well (Fig. S9). We note that the proton spectrum resulting from a methyl group (13CH3) under MAS is found sensitive to chemical shift differences, relative orientation of the methyl groups and MAS frequency (Fig. S4-6).

CONCLUSION We have simulated proton spectral line shapes and intensities of the methyl groups in a selectively methyl protonated (CH3) microcrystalline sample of the chicken αspectrin SH3 domain. The simulated spectra were validated by solid-state NMR experiments recorded at MAS rotation frequency of 70, 90 and 110 kHz. We find that the characteristic MAS frequency, which is needed to yield 50 % of the maximum peak intensity, ranges from as low as 20 kHz up to 324 kHz with the average value of (135.0 ± 88.0) kHz at an external magnetic field strength of 11.7 T (500 MHz). Higher

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

rotation frequencies are presumably required to achieve dipolar decoupling for fully protonated protein samples.

MATERIALS AND METHODS The perdeuterated, selectively methyl protonated sample of the microcrystalline SH3 domain was prepared as described previously

11

. In brief, expression was carried out

in 100 % D2O M9 medium, supplemented with 13

15

N-ammonium chloride and u-[2H,

C]-D-glucose. α-ketoisovalerate (2-keto-3-(methyl-d3)-butyric acid-4-13C sodium

salt, Sigma-Aldrich) was added to the M9 medium 1 h prior to induction with 1 mM IPTG (at OD600 0.5-0.6). Subsequent to overnight expression, the SH3 domain was purified via anion exchange and size exclusion chromatography as described before. For crystallization, pure protein was lyophilized and dissolved in 100 % D2O (final concentration: 8-10 mg/ml). Ammonium sulfate (dissolved in 100 % D2O) was added to a final concentration of 100 mM and the pH was adjusted to 8.0 by adding NaOD. All NMR experiments were carried out using a 0.7 mm H/C/N triple resonance MAS probe operating at a static magnetic field of 11.7 T (500 MHz 1H Larmor frequency). RF field strengths of 156 kHz and 100 kHz on the 1H and 13C channel was applied for hard pulses, respectively. For decoupling, a 1H π pulse was applied in the middle of t1, and WALTZ-16 (ω13C/2π = 5 kHz) during acquisition. As the sample was recrystallized from 100 % D2O, no solvent suppression was employed. For all experiments, the effective sample temperature was adjusted to 15 °C, using DSS and the residual water signal for calibration 29. The acquisition times in the F1 and F2 dimensions (13C, 1H) were set to 70 ms. A recycle delay of 1.5 s was employed. 1H T2’ relaxation times were measured employing a CP based experiment, inserting a (τ-π-τ) refocussing element on the 1H channel prior to the 13C t1 evolution period. The pulse sequence that have been employed are represented in the Supporting Information (Fig. S10). The 1H T2’ data was acquired as an interleaved pseudo 3D experiment to minimize systematic errors in the measurement. The site-specific intensities were extracted using NMRGLUE

30

based python scripts and fitted with a

1

mono-exponential function to obtain the H T2’ values. In order to estimate the MAS frequency that is necessary to average residual methyl proton dipolar interactions for a specific methyl group in the α-SH3 domain, we simulated first numerically the proton line shapes considering the explicit geometry of the atoms in the structure, PDB-ID: 2NUZ

23

. For the calculation, we assumed that

the protein is perdeuterated except for the methyl groups of valine and leucine 10 ACS Paragon Plus Environment

Page 10 of 14

Page 11 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

residues. The amide groups were assumed to be fully deuterated, as the sample was recrystallized from 100 % D2O. Protein biosynthesis yielded labelling of

13

CH3 in only

one of the two valine / leucine methyl groups, while the other methyl group is NMR silent 12CD3. 20. The numerical simulations were carried out using a 9-proton spin system, thus accounting for two neighbouring methyl containing side chains. Since the incorporation of

13

CH3 and

12

CD3 into the pro-R and pro-S position occurs at random,

selecting the two closest neighbouring methyl groups for a given site overestimates the involved dipole-dipole couplings. Using the program SIMPSON

21-22

, we have

therefore calculated the methyl proton spectra for all permutations to reflect the actual isotope labelling of the sample. Subsequently, the average spectrum has been calculated.

ACKNOWLEDGEMENT We acknowledge support from the Helmholtz-Gemeinschaft and the Deutsche Forschungsgemeinschaft (Grants Re1435 and SFB-1035, project B07). In addition, we are grateful to the Center for Integrated Protein Science Munich (CIPS-M) for financial support. ZT acknowledges funding from the European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 657682, and access to computing and storage facilities owned by parties and projects contributing to the National Grid Infrastructure MetaCentrum provided under the programme "Projects of Large Research, Development, and Innovations Infrastructures" (CESNET LM2015042). We acknowledge further NMR measurement time at the Bavarian NMR Center (BNMRZ).

Supporting Information Figure S1. Simulated 1H spectra of an isolated methyl groupE.through SI 10

REFERENCES 1. Castellani, F.; van Rossum, B.-J.; Diehl, A.; Schubert, M.; Rehbein, K.; Oschkinat, H., Structure of a protein determined by solid-state magic-angle spinning NMR. Nature 2002, 420, 98-102.

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2. Wasmer, C.; Lange, A.; Van Melckebeke, H.; Siemer, A. B.; Riek, R.; Meier, B. H., Amyloid fibrils of the HET-s(218-289) prion form a beta solenoid with a triangular hydrophobic core. Science 2008, 319 (5869), 1523-1526. 3. Loquet, A.; Sgourakis, N. G.; Gupta, R.; Giller, K.; Riedel, D.; Goosmann, C.; Griesinger, C.; Kolbe, M.; Baker, D.; Becker, S.; Lange, A., Atomic model of the type III secretion system needle. Nature 2012, 486 (7402), 276-9. 4. Xiao, Y. L.; Ma, B. Y.; McElheny, D.; Parthasarathy, S.; Long, F.; Hoshi, M.; Nussinov, R.; Ishii, Y., A beta(1-42) fibril structure illuminates self-recognition and replication of amyloid in Alzheimer's disease. Nature Struct. Mol. Biol. 2015, 22 (6), 499-505. 5. Tuttle, M. D.; Comellas, G.; Nieuwkoop, A. J.; Covell, D. J.; Berthold, D. A.; Kloepper, K. D.; Courtney, J. M.; Kim, J. K.; Barclay, A. M.; Kendall, A.; Wan, W.; Stubbs, G.; Schwieters, C. D.; Lee, V. M. Y.; George, J. M.; Rienstra, C. M., Solidstate NMR structure of a pathogenic fibril of full-length human alpha-synuclein. Nature Struct. Mol. Biol. 2016, 23 (5), 409-415. 6. Colvin, M. T.; Silvers, R.; Ni, Q. Z.; Can, T. V.; Sergeyev, I.; Rosay, M.; Donovan, K. J.; Michael, B.; Wall, J.; Linse, S.; Griffin, R. G., Atomic Resolution Structure of Monomorphic A beta(42) Amyloid Fibrils. J. Am. Chem. Soc. 2016, 138 (30), 96639674. 7. Chevelkov, V.; van Rossum, B. J.; Castellani, F.; Rehbein, K.; Diehl, A.; Hohwy, M.; Steuernagel, S.; Engelke, F.; Oschkinat, H.; Reif, B., 1H detection in MAS solid state NMR spectroscopy employing pulsed field gradients for residual solvent suppression. J. Am. Chem. Soc. 2003, 125 (26), 7788-7789. 8. Paulson, E. K.; Morcombe, C. R.; Gaponenko, V.; Dancheck, B.; Byrd, R. A.; Zilm, K. W., Sensitive High Resolution Inverse Detection NMR Spectroscopy of Proteins in the Solid State. J. Am. Chem. Soc. 2003, 125 (51), 15831-15836. 9. Agarwal, V.; Diehl, A.; Skrynnikov, N.; Reif, B., High Resolution 1H Detected 1H,13C Correlation Spectra in MAS Solid-State NMR using Deuterated Proteins with Selective 1H,2H Isotopic Labeling of Methyl Groups. J. Am. Chem. Soc. 2006, 128 (39), 12620-12621. 10. Chevelkov, V.; Rehbein, K.; Diehl, A.; Reif, B., Ultra-high resolution in proton solid-state NMR at high levels of deuteration. Angew. Chem. Int. Ed. 2006, 45 (23), 3878-3881. 11. Agarwal, V.; Xue, Y.; Reif, B.; Skrynnikov, N. R., Protein side-chain dynamics as observed by solution- and solid-state NMR: a similarity revealed. J. Am. Chem. Soc. 2008, 130, 16611–16621. 12. Asami, S.; Reif, B., Proton-detected solid-state NMR at aliphatic sites: Applications to Crystalline Systems. Acc. Chem. Res. 2013, 46 (9), 2089-2097. 13. Sinnige, T.; Daniels, M.; Baldus, M.; Weingarth, M., Proton Clouds to Measure Long-Range Contacts between Nonexchangeable Side Chain Protons in Solid-State NMR. J. Am. Chem. Soc. 2014, 136 (12), 4452-4455. 14. Agarwal, V.; Penzel, S.; Szekely, K.; Cadalbert, R.; Testori, E.; Oss, A.; Past, J.; Samoson, A.; Ernst, M.; Bockmann, A.; Meier, B. H., De Novo 3D Structure 12 ACS Paragon Plus Environment

Page 12 of 14

Page 13 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Determination from Sub-milligram Protein Samples by Solid-State 100 kHz MAS NMR Spectroscopy. Angewandt. Chem. Int. Edt. 2014, 53 (45), 12253-12256. 15. Lamley, J. M.; Iuga, D.; Oster, C.; Sass, H. J.; Rogowski, M.; Oss, A.; Past, J.; Reinhold, A.; Grzesiek, S.; Samoson, A.; Lewandowski, J. R., Solid-State NMR of a Protein in a Precipitated Complex with a FullLength Antibody. J. Am. Chem. Soc. 2014, 136 (48), 16800-16806. 16. Andreas, L. B.; Jaudzems, K.; Stanek, J.; Lalli, D.; Bertarello, A.; Marchand, T. L.; Cala-De Paepe, D.; Kotelovica, S.; Akopjana, I.; Knott, B.; Wegner, S.; Engelke, F.; Lesage, A.; Emsley, L.; Tars, K.; Herrmann, T.; Pintacuda, G., Structure of fully protonated proteins by proton-detected magic-angle spinning NMR. Proc. Natl Acad. Sci. U.S.A. 2016, 113 (33), 9187-9192. 17. Zhou, D. H.; Shah, G.; Cormos, M.; Mullen, C.; Sandoz, D.; Rienstra, C. M., Proton-detected solid-state NMR Spectroscopy of fully protonated proteins at 40 kHz magic-angle spinning. J. Am. Chem. Soc. 2007, 129 (38), 11791-11801. 18. Xue, K.; Sarkar, R.; Motz, C.; Asami, S.; Rodriguez Camargo, D. C.; Decker, V.; Wegner, S.; Tosner, Z.; Reif, B., Limits of Resolution and Sensitivity of Proton Detected MAS Solid-State NMR Experiments at 111 kHz in Deuterated and Protonated Proteins. Sci. Rep. 2017, 7, e7444. 19. Sternberg, U.; Witter, R.; Kuprov, I.; Lamley, J. M.; Oss, A.; Lewandowski, J. R.; Samoson, A., (1)H line width dependence on MAS speed in solid state NMR Comparison of experiment and simulation. J Magn Reson 2018, 291, 32-39. 20. Goto, N. K.; Gardner, K. H.; Mueller, G. A.; Willis, R. C.; Kay, L. E., A robust and cost-effective method for the production of Val, Leu, Ile (d1) methyl-protonated 15N-, 13 C-, 2H-labeled proteins. J. Biomol. NMR 1999, 13, 369-374. 21. Bak, M.; Rasmussen, J. T.; Nielsen, N. C., SIMPSON: A General Simulation Program for Solid-State NMR Spectroscopy. J. Magn. Reson. 2000, 147 (2), 296330. 22. Tosner, Z.; Andersen, R.; Stevenss, B.; Eden, M.; Nielsen, N. C.; Vosegaard, T., Computer-intensive simulation of solid-state NMR experiments using SIMPSON. J. Magn. Reson. 2014, 246, 79-93. 23. Chevelkov, V.; Faelber, K.; Schrey, A.; Rehbein, K.; Diehl, A.; Reif, B., Differential Line Broadening in MAS solid-state NMR due to Dynamic Interference. J. Am. Chem. Soc. 2007, 129 (51), 10195-10200. 24. Hologne, M.; Faelber, K.; Diehl, A.; Reif, B., Characterization of Dynamics of Perdeuterated Proteins by MAS Solid-State NMR. J. Am. Chem. Soc. 2005, 127 (33), 11208-11209. 25. Schanda, P.; Meier, B. H.; Ernst, M., Accurate measurement of one-bond H-X heteronuclear dipolar couplings in MAS solid-state NMR. J. Magn. Reson. 2011, 210 (2), 246-259. 26. Linser, R.; Fink, U.; Reif, B., Probing Surface Accessibility of Proteins Using Paramagnetic Relaxation in Solid-State NMR Spectroscopy. J. Am. Chem. Soc. 2009, 131, 13703–13708.

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

27. Zorin, V. E.; Brown, S. P.; Hodgkinson, P., Origins of linewidth in 1H magic-angle spinning NMR. J. Chem. Phys. 2006, 125 (14), 144508. 28. Zorin, V. E.; Brown, S. P.; Hodgkinson, P., Quantification of homonuclear dipolar coupling networks from magic-angle spinning H-1 NMR Mol. Phys. 2006, 104 (2), 293-304. 29. Linser, R.; Chevelkov, V.; Diehl, A.; Reif, B., Sensitivity Enhancement Using Paramagnetic Relaxation in MAS Solid State NMR of Perdeuterated Proteins. J. Magn. Reson. 2007, 189 (2), 209-216. 30. Helmus, J. J.; Jaroniec, C. P., Nmrglue: an open source Python package for the analysis of multidimensional NMR data. J. Biomol. NMR 2013, 55 (4), 355-367.

TOC GRAPHIC

14 ACS Paragon Plus Environment

Page 14 of 14