Directly Probing Intermolecular Structural Change ... - ACS Publications

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article 2

Directly Probing Intermolecular Structural Change of a Core Fragment of #Microglobulin Amyloid Fibrils with Low-Frequency Raman Spectroscopy Shinsuke Shigeto, Chun-Fu Chang, and Hirotsugu Hiramatsu J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.6b10779 • Publication Date (Web): 02 Jan 2017 Downloaded from http://pubs.acs.org on January 11, 2017

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 free 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 accessible to all readers and 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.

The Journal of Physical Chemistry B 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 29

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

Directly Probing Intermolecular Structural Change of a Core Fragment of

β2-Microglobulin Amyloid Fibrils with Low-Frequency Raman Spectroscopy

Shinsuke Shigeto,*,† Chun-Fu Chang,† and Hirotsugu Hiramatsu*,‡

†

Department of Applied Chemistry, National Chiao Tung University, Hsinchu 30010, Taiwan.

‡

Graduate School of Pharmaceutical Sciences, Tohoku University, Sendai 980-8578, Japan.

1

ACS Paragon Plus Environment


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 Amyloid fibrils, ordered aggregates of proteins or peptides, have attracted keen interest because their deposition causes severe human diseases. Despite many studies employing X-ray crystallography, solid-state NMR, and other methods, intermolecular interactions governing the fibril formation remain largely unclear. Here, we used low-frequency Raman (LFR) spectroscopy to investigate the intermolecular β-sheet structure of a core fragment of

β2-microglobulin amyloid fibrils, β2m21–29, in aqueous buffer solutions. The LFR spectra (approximately 10 to 200 cm−1) of β2m21–29 amyloid fibrils measured at different pH values (ranging from 6.8 to 8.0) revealed a broad spectral pattern with a maximum at ~80 cm−1 below pH 7.2 and at ~110 cm−1 above pH 7.4. This observation is attributed to a pH-dependent structural change from an antiparallel to a parallel intermolecular β-sheet structure. Multivariate curve resolution–alternating least-squares (MCR–ALS) analysis enabled us to decompose the apparently monotonous LFR spectra into three distinctly different contributions: intermolecular vibrations of the parallel and antiparallel β-sheets, and intramolecular vibrations of the peptide backbone. Peak positions of the obtained LFR bands not only exhibit a much more pronounced difference between the two β-sheets than the conventional amide I band, but they also suggest stronger intermolecular interaction, due presumably to the hydrophobic effect, in the parallel β-sheet than in the antiparallel β-sheet. The present results show that LFR spectroscopy in combination with the MCR–ALS analysis holds promise for real-time tracking of the intermolecular dynamics of amyloid fibril formation under physiological conditions.

2


Page 2 of 29

Page 3 of 29

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


Introduction Amyloid fibrils constitute an important class of well-ordered supramolecular aggregates of proteins or peptides that are associated with many serious human diseases including Alzheimer’s disease and prion diseases.1,2 They possess a common β-sheet structure in which

β-strands are aligned perpendicular to the fibril axis. Because nonpathogenic proteins and peptides are also shown to form similar fibrous structures, amyloid fibrils are deemed a fundamental structure that misfolded proteins and peptides can adopt through various intermolecular interactions. Intensive studies over the past decades using X-ray crystallography, solid-state NMR, electron microscopy, and other methods have focused on elucidation of the atomic structure of amyloid fibrils and their function–structure relationships,3-6 but a thorough understanding of the intermolecular interactions underlying the formation of insoluble fibrous aggregates is still lacking. The intermolecular interactions in β-sheets of amyloid fibrils have been studied by analyzing the intramolecular vibrational mode coupling of amide I (mainly C=O stretching) modes of the peptide main chain. The coupling gives rise to the inhomogeneous width of the amide I band. The coupling strength is a function of the orientation and distance of each pair of amide I oscillators,7,8 and these parameters are highly sensitive to the spatial distribution of the oscillators and hence protein secondary structure. The vibrational coupling between amide I modes in peptides and proteins has been intensively studied by linear and nonlinear IR spectroscopies. Two-dimensional (2D) IR spectroscopy allows for structural and kinetic characterization of peptides and proteins through analysis of cross-peaks observed in amide I 2D spectra.9-12 Conventional linear IR spectroscopy of the amide I band is a more facile method that can clearly distinguish between different peptide structures such as parallel and antiparallel β-sheets. Hiramatsu and co-workers analyzed the amide I band (at ~1630 cm−1) in the IR spectra of a core fragment of β2-microglobulin (β2m) amyloid fibrils measured at different pH conditions (pH 2.0–8.6).13,14 This peptide fragment comprises amino acid 3



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

residues 21–29 of β2m (21NFLNCYVSG29, hereafter denoted β2m21–29) and has a propensity to spontaneous fibril formation.15 Because the deposition of amyloid fibrils of β2m is known to be a typical symptom of dialysis-related amyloidosis,16,17 structural characterization of β2m21– 29

is highly important for developing new strategies for diagnosis and treatment of the disease.

Interestingly, the researchers found that β2m21–29 fibrils have an antiparallel β-sheet structure below pH 7.8, whereas they switch the stacking direction to form parallel β-sheets at more basic conditions.14 Although the amide I mode is often useful for distinguishing between the parallel and antiparallel β-sheet structures of amyloids, it remains to be fully elucidated what intermolecular interaction(s) dictate the intermolecular β-sheet structures of β2m21–29. An alternative approach is thus required to gain further insight into the nature of the intermolecular interactions that are responsible for the observed pH-dependent structural change of β2m21–29 fibrils. In this paper, we report that low-frequency Raman (LFR) spectra (below 200 cm−1 down to as low as 10 cm−1) can be used as a direct and sensitive probe for the intermolecular β-sheet structure of β2m21–29 amyloid fibrils in aqueous solutions. We have recently demonstrated the potential of fast, multichannel LFR spectroscopy for structural characterization by investigating crystal polymorphs of 1,1′-binaphthyl.18,19 Distinctly different LFR spectra of the polymorphs, which arise mainly from lattice vibrations, enabled us not only to track thermally induced polymorphic transformation of this compound,18 but also to visualize polycrystalline structures within microcrystals of the polymorphs.19 In addition to the lattice vibrations probed in our previous work, LFR spectra can provide molecular-specific information on collective intermolecular vibrations. Here, we exploit this capability of LFR spectroscopy to surmount the difficulty in the conventional experiments and to shed new light on the structural change that β2m21–29 amyloid fibrils exhibit. Unlike crystallography and electron microscopy, LFR spectroscopy does not require any sample pretreatment except for centrifugation so as to sediment the fibrils. We are therefore able to 4


Page 4 of 29

Page 5 of 29

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


measure amyloid fibrils as they are in phosphate buffer solution, which resembles the actual biological environment. In the present work, we measured the LFR spectra of β2m21–29 fibrils at different pH conditions (from pH 6.8 to 8.0 at 0.2 pH intervals) with high signal-to-noise ratio using a laboratory-built multichannel confocal Raman microspectrometer equipped with volume Bragg gratings as ultranarrow notch filters.19 With the help of multivariate curve resolution– alternating least-squares (MCR–ALS) analysis, we derived for the first time the intermolecular vibrational spectra of the parallel and antiparallel β-sheet structures of the

β2m21–29 fibril. Based on the MCR–ALS-derived spectra, we discuss possible intermolecular interactions that give rise to those structures.

Experimental Section Sample Preparation. The procedures for preparing β2m21–29 amyloid fibrils are basically the same as those in the literature.14 β2m21–29 peptide fragments were dissolved at 150 µM in 50 mM phosphate buffer (pH 6.8, 7.0, 7.2, 7.4, 7.6, 7.8, and 8.0) supplemented with 100 mM NaCl. The pH of the solution was adjusted by adding aliquots of concentrated NaOH or HCl. The peptide solution was incubated at 37 °C for 10 h, yielding amyloid fibrils spontaneously.15 The incubated fibril solution was centrifuged at 16100 × g for 5 min to sediment fibrils. A sedimented portion (10 µL) of the solution was transferred onto a concave microscope slide (depth of concave: 0.5 mm) and used for subsequent Raman spectroscopic measurements. Fibril formation was indirectly confirmed with an IR linear dichroism experiment (see Supporting Information, Figure S1). Raman Measurements. A laboratory-built confocal Raman microspectrometer, which has been described previously,19 was used to record LFR as well as high-frequency Raman (HFR) spectra of the β2m21–29 amyloid fibrils prepared at the seven pH values in wet condition. The second harmonic (532 nm) of a continuous-wave Nd:YVO4 laser (Verdi-V5, Coherent) 5



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

was used as the excitation light. A small portion ( 0 ) or anti-Stokes (ν~ < 0 ) region. The difference as well as similarity in the low-frequency spectral pattern is now more evident in the reduced spectra: the fibril and buffer spectra both show prominent peaks at approximately 40 and 160 cm−1, which can be assigned, respectively, to H-bond bending and stretching of water,25 but the reduced Raman intensity below ~150 cm−1 is greater for the β2m21–29 fibril solution than for the buffer. The pure LFR spectrum of the

β2m21–29 amyloid fibril can finally be obtained by appropriately subtracting the reduced Raman spectrum of the buffer (Figure 1b, black trace) from that of the fibril solution (Figure 1b, red trace). The resulting difference spectrum (Figure 1c) shows a very broad pattern that resembles neither the fibril nor buffer spectrum in Figure 1b. This broad feature is due to intermolecular vibrations and possibly low-frequency intramolecular vibrations of the

β2m21–29 fibril. The difference reduced LFR spectra of β2m21–29 fibrils prepared at pH 6.8, 7.0, 7.2, 7.4, 7.6, 7.8, and 8.0 are shown in Figure 2. These spectra have been subjected to a normalization to the intensities at around 10 and 170 cm−1. The normalization resulted in nearly the same intensity of an intramolecular Raman band at 620 cm−1, ensuring its validity. The LFR spectra below pH 7.2 look almost identical having a maximum at around 80 cm−1. In contrast, those above pH 7.4, all of which are again similar to each other, appear to be blueshifted with a 8


Page 9 of 29

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


peak located around 110 cm−1. This observation is in good agreement with the previous result on the amide I IR band of the β2m21–29 fibril, where a pH-dependent change in the stacking direction of β-strands in the β2m21–29 fibril was demonstrated by the analyses of IR band shape, IR linear dichroism, and

13

C isotope shift.14 It has been established that below pH 7.2, the

β2m21–29 fibril forms an antiparallel intermolecular β-sheet structure in which the β-strands stack in an antiparallel manner, whereas above pH 7.4, it forms a parallel intermolecular

β-sheet structure in which the β-strands align in the same direction (see the schematic illustration in Figure 2). We also measured the HFR spectra of the same β2m21–29 fibrils. Figure 3 shows the 1550– 1750 cm−1 region of the observed HFR spectra, which include the amide I band at ~1670 cm−1. Consistent with the LFR spectra, the amide I band exhibits a change in bandwidth and a slight peak shift (1670 → 1675 cm−1) above pH 7.4. This finding is assignable to the pH-dependent change in the coupling of the amide I oscillators and thus provides support for the structural change of our fibers as a function of pH. We note, however, that the observed change in the amide I band is subtle compared with that in the LFR spectrum (see Figure 2).

9



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

Figure 1. (a) Low-frequency Raman spectra of the β2m21–29 amyloid fibril in pH 6.8 buffer solution (red trace) and of the corresponding buffer (black trace). The buffer spectrum has been scaled by a factor of 0.79. (b) Reduced Raman spectra of the fibril solution (red trace) and the buffer (black trace) calculated using eq 1. (c) Difference reduced Raman spectrum of

10


Page 10 of 29

Page 11 of 29

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 β2m21–29 fibril prepared at pH 6.8.

Figure 2. Difference reduced low-frequency Raman spectra of β2m21–29 amyloid fibrils at pH 6.8, 7.0, 7.2, 7.4, 7.6, 7.8, and 8.0 (from top to bottom) obtained in the same way as in Figure 1c. Also shown on the right is a schematic representation of the pH-dependent change in the stacking direction in the β2m21–29 fibril from the antiparallel to parallel direction.

11



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

Figure 3. High-frequency Raman spectra of the same β2m21–29 amyloid fibrils as in Figure 2, at pH 6.8, 7.0, 7.2, 7.4, 7.6, 7.8, and 8.0 (from top to bottom). The buffer spectrum has been subtracted off.

MCR–ALS Analysis of the LFR Spectra of β2m21–29 Fibrils. Compared with the intramolecular amide I band at ~1670 cm−1, our LFR spectra probe intermolecular interactions more directly and sensitively. They are anticipated to reveal otherwise unobtainable information on the change in the intermolecular β-sheet structure. The broad spectral pattern indicates that intermolecular vibrations of different origins contribute together with low-frequency intramolecular vibrations, so it is essential to resolve those different spectral components. A commonly employed approach is to decompose a spectrum into several Gaussian and/or Lorentzian bands by fitting the raw spectrum and its second derivative simultaneously, but it is often beset by large uncertainties particularly when the target spectrum is very broad and less structured. Instead, we here use MCR–ALS to analyze our 12


Page 12 of 29

Page 13 of 29

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


LFR spectra of β2m21–29 fibrils at different pH conditions. Previously we have applied MCR– ALS analysis to Raman imaging data of living cells and demonstrated its ability to resolve the pure component spectra and spatial distributions of several chemical species that we assumed to make dominant contributions to the original data.26,27 In the present work, we performed MCR–ALS of the pH-dependence data of the LFR spectra (10–180 cm−1) of β2m21–29 amyloid fibrils assuming three components. The number of components, k, was estimated from a SVD analysis of the matrix A (Figure 4). Inspection of a plot of singular values (Figure 4a) and the corresponding component spectra (Figure 4b) shows that there are at least two components to be taken into account, so k = 2 was first assumed. However, the results were unsatisfactory in the sense that the pH-dependent intensity profiles of the resolved components had a constant offset in common. k was then set to be 3, which turned out to yield chemically sound results in terms of both their spectral and pH-dependence profiles, as shown and discussed below. The MCR–ALS results obtained with k = 3 are presented in Figure 5. Figure 5a displays the pure LFR spectra of the three components (denoted 1–3), and Figure 5b shows the pH dependence of the intensity of each component. The vector representing each of the pure LFR spectra is normalized so that the sum of its components is equal to unity. It is worth mentioning that, owing to non-negativity constraints imposed in our MCR–ALS analysis, physically and/or chemically interpretable results have been derived, on the basis of which we are able to discuss the assignments of the three components.

13



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

Figure 4. Results of singular value decomposition of the LFR spectral data matrix A. (a) Plot of singular values. (b) Component spectra. Each spectrum is offset by 0.5 for clarity of display.

14


Page 14 of 29

Page 15 of 29

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


Figure 5. Results of the MCR–ALS analysis of the difference reduced Raman spectra (10– 180 cm−1) of β2m21–29 fibrils prepared at the seven pH values. Pure LFR spectra (a) and pH-dependent intensity profiles (b) of the three components (1–3) assumed in our MCR–ALS analysis.

The spectrum of component 1 (Figure 5a, black trace) is as broad as those in Figure 3,

15



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

with a maximum at ~90 cm−1 and a tiny but recognizable hump at ~27 cm−1, and its intensity (Figure 5b, black circles) does not depend markedly on pH. Previous studies also observed similar peaks in the low-frequency vibrational spectra of various proteins with different structures.28-31 According to the assignments in those studies, component 1 is most likely to originate from low-frequency motions of the peptide backbone (including acoustic, accordion-like motions). This interpretation is consistent with the nearly constant pH dependence of the component (Figure 5b, black circles), because such intramolecular modes would remain virtually unchanged even though the intermolecular alignment of β-strands varies. The remaining two components reveal an interesting behavior in their pH dependence (Figure 5b, blue triangles and red squares). Component 2, having a maximum at ~120 cm−1 in its LFR spectrum (Figure 5a, blue trace), shows an intensity jump at pH 7.4 (Figure 5b, blue triangles), and component 3, whose spectrum (Figure 5a, red trace) has maxima at around 60 and 170 cm−1, shows a concomitant decrease in intensity (Figure 5b, red squares). This behavior corresponds well to the intermolecular structural change from the antiparallel to parallel β-sheet with increasing pH value. It is therefore reasonable to attribute component 2 to the parallel β-sheet structure of the β2m21–29 fibril and component 3 to the antiparallel

β-sheet structure. Their LFR bands, which evidence quite different peak frequencies (~120 cm−1 for the parallel β-sheet and ~60 cm−1 for the antiparallel β-sheet), represent intermolecular vibrations of the β-sheets. Because an apparent peak at ~170 cm−1 in the LFR spectrum of the antiparallel β-sheet form (Figure 5a, red trace) is affected to a great extent by a relatively large intensity fluctuation around 170 cm−1, this peak will not be interpreted in detail.

Discussion Having determined the intermolecular vibrational spectra of the parallel and antiparallel 16


Page 16 of 29

Page 17 of 29

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


β-sheets, we discuss three principal intermolecular interactions that might differ between those structures and hence affect the peak frequencies of the LFR bands: (i) the H-bonding interaction, (ii) the hydrophobic interaction between side chains, and (iii) the electrostatic interaction between terminal charges. First, we consider H-bond lengths in the β-sheet structures of β2m21–29 amyloid fibrils. Asakura and co-workers investigated crystalline polyalanine using solid-state NMR and found that the H-bond in a parallel β-sheet form of polyalanine is longer than that in an antiparallel

β-sheet form.32,33 Because amyloid fibrils are ordered aggregates of an infinitely large number of β-sheets just as crystalline polyalanine, we can safely assume that the above relationship between H-bond length and β-sheet structure also holds for our amyloid fibrils. It is then expected that the H-bond intermolecular interaction is weaker and the corresponding vibrational frequency is lower in the parallel β-sheet structure of the β2m21–29 fibril than in the antiparallel counterpart. Second, we consider the effect of side-chain hydrophobic interactions in β2m21–29 fibrils. Side chains of the same kind tend to align in the parallel β-sheet. As a result, the hydrophobic effect (an attractive force) would be stronger in the parallel β-sheet than in the antiparallel

β-sheet. In line with this idea, a recent vacuum-UV circular dichroism study exemplified the presence of the hydrophobic interactions between the side chains of Phe22 and Tyr26 in the parallel β-sheets of β2m21–29.34 Third, the electrostatic interaction between the charges at the N- and C-termini is repulsive in the parallel β-sheet structure and attractive in the antiparallel β-sheet structure. The electrostatic repulsion in the parallel β-sheet of the β2m21–29 fibril should be suppressed on the grounds that the salt concentration in the buffer used here is high (100 mM NaCl) and that the N-terminus is probably deprotonated in the parallel β-sheet of β2m21–29.14 All of these interactions can, in principle, contribute to the observed difference in the LFR bands, but given that the peak frequency of the parallel β-sheet structure of the β2m21–29 17



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

fibril is considerably higher than that of the antiparallel form (see Figure 5a), the hydrophobic effect might play a key role. Amyloid polymorphism is a subject of intensive studies,35-38 and a recent paper by Chatani and co-workers reports that the hydrophobic interaction does affect the polymorphism of β2m amyloid fibrils.39 It will be of great interest to correlate the LFR spectra of the parallel and antiparallel structures of the β2m21–29 fibril with their morphological characteristics revealed by electron microscope and atomic force microscope techniques. In conclusion, using LFR spectroscopy in combination with MCR–ALS analysis, we have observed separately intermolecular vibrations of the parallel and antiparallel β-sheet structures of the β2m21–29 amyloid fibril that depend on pH, and intramolecular backbone motions of the fibrils that are independent of pH. These low-frequency spectral signatures of the intermolecular interactions in amyloid fibrils will allow for a real-time tracking of the intermolecular dynamics of the amyloid fibril formation (e.g., to examine whether the antiparallel β-sheet structure of the fibril prepared at pH 6.8 will evolve into the parallel

β-sheet structure upon increasing the pH above 7.4). Furthermore, they could provide a reliable test for rapidly growing computer simulations of protein aggregates.

Author Information Corresponding Authors *E-mail: [email protected] (S.S.). *E-mail: [email protected] (H.H.). Present Address S.S.: Department of Chemistry, School of Science and Technology, Kwansei Gakuin University, Sanda 669-1337, Japan. C.-F.C.: Department of Chemistry, School of Science, The University of Tokyo, Tokyo 113-0033, Japan. H.H.: Department of Applied Chemistry, National Chiao Tung University, Hsinchu 30010, Taiwan. Notes 18


Page 18 of 29

Page 19 of 29

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 authors declare no competing financial interest.

Supporting Information IR linear dichroism spectra of the β2m21–29 amyloid fibrils (Figure S1). This information is available free of charge via the Internet at http://pubs.acs.org

Acknowledgments This work was financially supported by Ministry of Education, Taiwan; National Chiao Tung University (“Aiming for the Top University” program); and Ministry of Science and Technology,

Taiwan

(Grants

MOST103-2745-M-009-001-ASP

and

MOST103-2113-M-009-008-MY2).

References (1)

Dobson, C. M. Protein Folding and Misfolding. Nature 2003, 426, 884-890.

(2)

Chiti, F.; Dobson, C. M. Protein Misfolding, Functional Amyloid, and Human Disease.

Annu. Rev. Biochem. 2006, 75, 333-366. (3)

Nelson, R.; Sawaya, M. R.; Balbirnie, M.; Madsen, A. Ø.; Riekel, C.; Grothe, R.; Eisenberg, D. Structure of the Cross-β Spine of Amyloid-Like Fibrils. Nature 2005, 435, 773-778.

(4)

Sawaya, M. R.; Sambashivan, S.; Nelson, R.; Ivanova, M. I.; Sievers, S. A.; Apostol, M. I.; Thompson, M. J.; Balbirnie, M.; Wiltzius, J. J. W.; McFarlane, H. T.; et al. Atomic Structures of Amyloid Cross-β Spines Reveal Varied Steric Zippers. Nature 2007, 447, 453-457.

(5)

Vilar, M.; Chou, H.-T.; Lührs, T.; Maji, S. K.; Riek-Loher, D.; Verel, R.; Manning, G.; Stahlberg, H.; Riek, R. The Fold of α-Synuclein Fibrils. Proc. Natl. Acad. Sci. USA 2008, 105, 8637-8642. 19



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

(6)

Tycko, R. Molecular Structure of Amyloid Fibrils: Insights from Solid-State NMR. Q.

Rev. Biophys. 2006, 39, 1-55. (7)

Krimm, S.; Abe, Y. Intermolecular Interaction Effects in the Amide I Vibrations of β Polypeptides. Proc. Natl. Acad. Sci. USA 1972, 69, 2788-2792.

(8)

Moore, W. H.; Krimm, S. Transition Dipole Coupling in Amide I Modes of β Polypeptides. Proc. Natl. Acad. Sci. USA 1975, 72, 4933-4935.

(9)

Demirdöven, N.; Cheatum, C. M.; Chung, H. S.; Khalil, M.; Knoester, J.; Tokmakoff, A. Two-Dimensional Infrared Spectroscopy of Antiparallel β-Sheet Secondary Structure. J.

Am. Chem. Soc. 2004, 126, 7981-7990. (10) Ganim, Z.; Chung, H. S.; Smith, A. W.; DeFlores, L. P.; Jones, K. C.; Tokmakoff, A. Amide I Two-Dimensional Spectroscopy of Proteins. Acc. Chem. Res. 2008, 41, 432-441. (11) Kim, Y. S.; Hochstrasser, R. M. Applications of 2D IR Spectroscopy to Peptides, Proteins, and Hydrogen-Bond System. J. Phys. Chem. B 2009, 113, 8231-8251. (12) Shim, S.-H.; Gupta, R.; Ling, Y. L.; Strasfeld, D. B.; Raleigh, D. P.; Zanni, M. T. Two-Dimensional IR Spectroscopy and Isotope Labeling Defines the Pathway of Amyloid Formation with Residue-Specific Resolution. Proc. Natl. Acad. Sci. USA 2009,

106, 6614-6619. (13) Hiramatsu, H.; Goto, Y.; Naiki, H.; Kitagawa, T. Structural Model of the Amyloid Fibril Formed by β2-Microglobulin #21-31 Fragment Based on Vibrational Spectroscopy. J.

Am. Chem. Soc. 2005, 127, 7988-7989. (14) Hiramatsu, H.; Lu, M.; Goto, Y.; Kitagawa, T. The β-Sheet Structure pH Dependence of the Core Fragments of β2-Microglobulin Amyloid Fibrils. Bull. Chem. Soc. Jpn. 2010,

83, 495-504. (15) Hasegawa, K.; Ohhashi, Y.; Yamaguchi, I.; Takahashi, N.; Tsutsumi, S.; Goto, Y.; Gejyo, F.; Naiki, H. Amyloidogenic Synthetic Peptides of β2-Microglobulin—a Role of the 20


Page 20 of 29

Page 21 of 29

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


Disulfide Bond. Biochem. Biophys. Res. Commun. 2003, 304, 101-106. (16) Gejyo, F.; Homma, N.; Suzuki, Y.; Arakawa, M. Serum Levels of β2-Microglobulin as a New Form of Amyloid Protein in Patients Undergoing Long-Term Hemodialysis. N.

Engl. J. Med. 1986, 314, 585-586. (17) Campistol, J. M.; Bernard, D.; Parastoitsis, G.; Solé, M.; Kasirsky, J.; Skinner, M. Polymerization of Normal and Intact β2-Microglobulin as the Amyloidogenic Protein in Dialysis-Amyloidosis. Kidney Int. 1996, 50, 1262-1267. (18) Chang, C.-F.; Wang, S.-C.; Shigeto, S. In Situ Ultralow-Frequency Raman Tracking of the Polymorphic Transformation of Crystalline 1,1'-Binaphthyl. J. Phys. Chem. C 2014,

118, 2702-2709. (19) Chang, C.-F.; Okajima, H.; Hamaguchi, H.; Shigeto, S. Imaging Molecular Crystal Polymorphs and Their Polycrystalline Microstructures in Situ by Ultralow-Frequency Raman Spectroscopy. Chem. Commun. 2014, 50, 12973-12976. (20) Paatero, P.; Tapper, U. Positive Matrix Factorization: A Non-Negative Factor Model with Optimal Utilization of Error Estimates of Data Values. Environmetrics 1994, 5, 111-126. (21) Lee, D. D.; Seung, H. S. Learning the Parts of Objects by Non-Negative Matrix Factorization. Nature 1999, 401, 788-791. (22) Boutsidis, C.; Gallopoulos, E. SVD Based Initialization: A Head Start for Nonnegative Matrix Factorization. Pattern Recognition 2008, 41, 1350-1362. (23) Shuker, R.; Gammon, R. W. Raman-Scattering Selection-Rule Breaking and the Density of States in Amorphous Materials. Phys. Rev. Lett. 1970, 25, 222-225. (24) Iwata, K.; Okajima, H.; Saha, S.; Hamaguchi, H. Local Structure Formation in Alkyl-Imidazolium-Based Ionic Liquids as Revealed by Linear and Nonlinear Raman Spectroscopy. Acc. Chem. Res. 2007, 40, 1174-1181. (25) Walrafen, G. E. Raman Spectral Studies of Water Structure. J. Chem. Phys. 1964, 40, 21



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

3249-3256. (26) Huang, C.-K.; Ando, M.; Hamaguchi, H.; Shigeto, S. Disentangling Dynamic Changes of Multiple Cellular Components during the Yeast Cell Cycle by in Vivo Multivariate Raman Imaging. Anal. Chem. 2012, 84, 5661-5668. (27) Hsu, J.-F.; Hsieh, P.-Y.; Hsu, H.-Y.; Shigeto, S. When Cells Divide: Label-Free Multimodal Spectral Imaging for Exploratory Molecular Investigation of Living Cells during Cytokinesis. Sci. Rep. 2015, 5, 17541. (28) Brown, K. G.; Erfurth, S. C.; Small, E. W.; Peticolas, W. L. Conformationally Dependent Low-Frequency Motions of Proteins by Laser Raman Spectroscopy. Proc.

Natl. Acad. Sci. USA 1972, 69, 1467-1469. (29) Chou, K.-C. Low-Frequency Motions in Protein Molecules. β-Sheet and β-Barrel.

Biophys. J. 1985, 48, 289-297. (30) Giraud, G.; Karolin, J.; Wynne, K. Low-Frequency Modes of Peptides and Globular Proteins in Solution Observed by Ultrafast OHD-RIKES Spectroscopy. Biophys. J. 2003,

85, 1903-1913. (31) Hédoux, A.; Krenzlin, S.; Paccou, L.; Guinet, Y.; Flament, M.-P.; Siepmann, J. Influence of Urea and Guanidine Hydrochloride on Lysozyme Stability and Thermal Denaturation; a Correlation between Activity, Protein Dynamics and Conformational Changes. Phys. Chem. Chem. Phys. 2010, 12, 13189-13196. (32) Yazawa, K.; Suzuki, F.; Nishiyama, Y.; Ohata, T.; Aoki, A.; Nishimura, K.; Kaji, H.; Shimizu, T.; Asakura, T. Determination of Accurate 1H Positions of an Alanine Tripeptide with Anti-Parallel and Parallel β-Sheet Structures by High Resolution 1H Solid State NMR and GIPAW Chemical Shift Calculation. Chem. Commun. 2012, 48, 11199-11201. (33) Asakura, T.; Okonogi, M.; Nakazawa, Y.; Yamauchi, K. Structural Analysis of Alanine Tripeptide with Antiparallel and Parallel β-Sheet Structures in Relation to the Analysis 22


Page 22 of 29

Page 23 of 29

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


of Mixed β-Sheet Structures in Samia Cynthia Ricini Silk Protein Fiber Using Solid-State NMR Spectroscopy. J. Am. Chem. Soc. 2006, 128, 6231-6238. (34) Matsuo, K.; Hiramatsu, H.; Gekko, K.; Namatame, H.; Taniguchi, M.; Woody, R. W. Characterization of Intermolecular Structure of β2-Microglobulin Core Fragments in Amyloid Fibrils by Vacuum-Ultraviolet Circular Dichroism Spectroscopy and Circular Dichroism Theory. J. Phys. Chem. B 2014, 118, 2785-2795. (35) Petkova, A. T.; Leapman, R. D.; Guo, Z.; Yau, W.-M.; Mattson, M. P.; Tycko, R. Self-Propagating, Molecular-Level Polymorphism in Alzheimer's β-Amyloid Fibrils.

Science 2005, 307, 262-265. (36) Gosal, W. S.; Morten, I. J.; Hewitt, E. W.; Smith, D. A.; Thomson, N. H.; Radford, S. E. Competing Pathways Determine Fibril Morphology in the Self-Assembly of

β2-Microglobulin into Amyloid. J. Mol. Biol. 2005, 351, 850-864. (37) Kodali, R.; Wetzel, R. Polymorphism in the Intermediates and Products of Amyloid Assembly. Curr. Opin. Struct. Biol. 2007, 17, 48-57. (38) Tycko, R. Physical and Structural Basis for Polymorphism in Amyloid Fibrils. Protein

Sci. 2014, 23, 1528-1539. (39) Chatani, E.; Yagi, H.; Naiki, H.; Goto, Y. Polymorphism of B2-Microglobulin Amyloid Fibrils Manifested by Ultrasonication-Enhanced Fibril Formation in Trifluoroethanol. J.

Biol. Chem. 2012, 287, 22827-22837.

23



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

TOC Graphic

24


Page 24 of 29

Page 25 of 29

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


Figure 1 98x202mm (300 x 300 DPI)



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

Figure 2 110x101mm (300 x 300 DPI)


Page 26 of 29

Page 27 of 29

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


Figure 3 85x114mm (300 x 300 DPI)



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

Figure 4 94x177mm (300 x 300 DPI)


Page 28 of 29

Page 29 of 29

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


Figure 5 97x181mm (300 x 300 DPI)


Directly Probing Intermolecular Structural Change ... - ACS Publications

Recommend Documents