Article pubs.acs.org/Langmuir
Infrared Probe Technique Reveals a Millipede-like Structure for Aβ(8−28) Amyloid Fibril Yachao Gao, Ye Zou, Yan Ma, Dan Wang, Ying Sun, and Gang Ma* Key Laboratory of Medicinal Chemistry and Molecular Diagnosis of Ministry of Education, Key Laboratory of Analytical Science and Technology of Hebei Province, College of Chemistry and Environmental Science, Hebei University, Baoding 071002, China S Supporting Information *
ABSTRACT: Amyloid fibrils are unique fibrous polypeptide aggregates. They have been associated with more than 20 serious human diseases including Alzheimer’s disease and Parkinson’s disease. Besides their pathological significance, amyloid fibrils are also gaining increasing attention as emerging nanomaterials with novel functions. Structural characterization of amyloid fibril is no doubt fundamentally important for the development of therapeutics for amyloid-related diseases and for the rational design of amyloid-based materials. In this study, we explored to use side-chain-based infrared (IR) probe to gain detailed structural insights into the amyloid fibril by a 21-residue model amyloidogenic peptide, Aβ(8−28). We first proposed an approach to incorporate thiocyanate (SCN) IR probe in a site-specific manner into amyloidogenic peptide using 1-cyano-4-dimethylaminopyridinium tetrafluoroborate as cyanylating agent. Using this approach, we obtained three Aβ(8−28) variants, labeled with SCN probe at three different positions. We then showed with thioflavin T fluorescence assay, Congo red assay, and atomic force microscopy that the three labeled Aβ(8−28) peptides can quickly form amyloid fibrils under high concentration and high salt conditions. Finally, we performed a detailed IR spectral analysis of the Aβ(8−28) fibril in both amide I and probe regions and proposed a millipede-like structure for the Aβ(8−28) fibril.
■
INTRODUCTION Amyloid fibrils are a unique form of polypeptide aggregates featuring cross-β supramolecular structure and fibrillar microscopic morphology.1 In vivo, the deposition of amyloid fibrils in human tissues and organs is a devastating event, and more than 20 serious human diseases are associated with such a phenomenon.2 These diseases include some well-known disorders such as Alzheimer’s disease and Parkinson’s disease. Besides their pathological significance, amyloid fibrils are also gaining increasing attention when people develop novel nanomaterials owing to their excellent material properties such as high Young’s modulus and high tensile strength.3−5 Potential uses of amyloid fibrils as biosensors, microattenuators, cell culture scaffolds, drug delivery vehicles, and enzyme immobilization substrates have been demonstrated recently.4,6−8 Structural characterization of amyloid fibril is no doubt fundamentally important for the development of therapeutics to treat and prevent amyloid-related diseases and for the rational design of amyloid-based materials. However, due to the semicrystalline nature of amyloid fibril, structural characterization of amyloid fibril has been a challenging task. In this study, we explore to use side-chain-based infrared (IR) probe to build a structural model for the amyloid fibril by a 21-residue amyloidogenic peptide, Aβ(8−28). This model peptide, with a sequence of SGYEVHHQKLVFFAEDVGSNK, is a segment © XXXX American Chemical Society
derived from amyloid-β (Aβ), a well-known peptide relevant to Alzheimer’s disease. Compared with the sequence of full length Aβ (e.g., Aβ(1−40) peptide), Aβ(8−28) lacks the first seven residues (i.e., residues 1−7) in the N-terminal region and the last 12 residues (i.e., residues 29−40) in the C-terminal region of Aβ(1−40). Side-chain-based IR probe is an emerging vibrational spectroscopic technique particularly useful to study protein structure and function. This type of IR probe is a small IRactive molecular moiety covalently attached to the side chain of an amino acid residue in a protein or a peptide and has attracted a lot of attention in recent years in the field of protein science.9−21 Two reasons make side-chain-based IR probe a valuable technique in amyloid research. First, the IR probe can be in small size, thus causing minimal perturbation to the native structure of amyloidogenic peptide in both monomeric and fibrillar states. Second, side-chain-based IR probe technique can tackle amyloid structure at both secondary and quaternary structural levels simultaneously through the detection of amide I and probe absorptions within the same IR spectrum. By doing so, the amide I vibrational mode can serve as a global secondary structural reporter for interstrand arrangement, while the IR Received: September 28, 2015 Revised: January 2, 2016
A
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
range of 1.8−2.5 mM. This salt-free peptide solution was subjected to FTIR characterization first. Then, a concentrated 4 M NaCl solution in the same D2O (90%)−DMSO (10%) (v/v) mixed solvent was introduced into the peptide solution in the glass vial to initiate fibrillation. The NaCl concentration used to induce fibrillation is 400 mM. The peptide solution after salt-induced fibrillation was subjected to ThT fluorescence assay, AFM characterization, CR assay, and FTIR characterization. All these experiments were performed at room temperature of 22 °C. One technical issue that we would like to point out here is why we used D2O rather than H2O in the mixed solvent system. This is due to the fact that monitoring both probe and amide I regions simultaneously in the same FTIR spectrum requires using D2O as the solvent. Technically, detecting a rather weak probe absorption requires the use of an IR cell with long optical path length (e.g., >20 μm), while long path length is not suitable for FTIR measurement in the amide I region if H2O is the solvent (which requires 85% purity by high performance liquid chromatography (HPLC). Wild type Aβ(8− 28) (SGYEVHHQKLVFFAEDVGSNK) with >98% purity by HPLC and Aβ(13−23) cysteine mutant (HHQKLVFFCED) with >98% purity by HPLC were purchased from ChinaPeptides (Shanghai, China). The wild type Aβ(8−28) was obtained as trifluoroacetic acid (TFA)-free peptide. Deuterium oxide (D2O) with >99.8% purity and methyl thiocyanate (MeSCN) with 99% purity were obtained from J&K Chemical (Beijing, China). Dimethyl sulfoxide (DMSO) with 99% purity, sodium chloride (NaCl) with 99% purity, Congo red (CR) with 85% purity, and 1-cyano-4-dimethylaminopyridinium (CDAP) tetrafluoroborate with 97% purity were obtained from Sigma-Aldrich (Saint Louis, MO). Thioflavin T (ThT) was obtained from Acros (Geel, Belgium). Deionized water with a resistivity of 18.2 MΩ·cm was obtained from a Millipore system (Billerica, USA). SCN IR Probe Labeling Procedure. In a typical run, 200 mg of Aβ(8−28) cysteine mutant and 60 mg of CDAP tetrafluoroborate were dissolved in 100 mL of 0.1 M acetic acid (HOAc). The molar ratio between the peptide and CDAP is approximately 1:3. The mixture was incubated under room temperature for 1 h to allow reaction completion. To further purify the labeled peptide, the mixture was filtered first and then subjected to CXTH P3050 preparative HPLC (Beijing, China). The preparative HPLC procedure was performed on a C18 column. The HPLC conditions are as follows: eluent: A = H2O + 0.1% TFA, B = acetonitrile (ACN); gradient: 40 min linear gradient from 18% to 28% of B; flow rate: 30 mL/min; detection wavelength: 214 nm. After this step of separation, pure target peptides with TFA as counterion were obtained. The TFAcontaining peptides were then dissolved into H2O−ACN (4:1) mixed solvent. The pH of the mixture was adjusted to neutral with ammonium hydroxide. Then the solution was subjected to another round of preparative HPLC in order to remove TFA. The HPLC conditions are as follows: eluent: A = H2O + 0.3% HCl (pH = 1), B = ACN; gradient: 5% B in 80 min and then 80% B; detection wavelength: 214 nm. The eluent was then concentrated by rotary evaporation and further dried by lyophilization. A purity of 98% was achieved for the final peptides as determined by CXTH P3000 analytical HPLC (Beijing, China). The peptide mass was confirmed by Shimadzu LC-MS 2020 electrospray ionization mass spectrometry (ESI-MS) (Kyoto, Japan). The yield of the labeling was calculated using the ratio between the peptide mass after TFA removal with HPLC and the peptide mass before labeling and was determined to be ∼20%. Fibrillation Conditions. A concentrated Aβ(8−28) peptide solution in DMSO was prepared first. An aliquot of such solution was diluted into neat D2O to obtain a peptide solution in the D2O (90%)−DMSO (10%) (v/v) mixed solvent in a glass vial. The concentration is determined by UV−vis spectroscopy, and it is in the B
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir determined using the flat baseline in the 1850−1720 cm−1 window region as the subtraction criterion. Prior hydrogen/deuterium (H/D) exchange on Aβ(8−28) was not performed as the peptide backbone quickly became fully deuterated once dissolved into D2O due to its intrinsically disordered nature. This was evidenced by the disappearance of the amide II band at 1540 cm−1 in the absorption spectrum. Residual H2O in the overwhelming D2O forms HOD and has a negligible effect on the spectral analysis in the amide I′ region. This approach eases the work of sample preparation and has been adopted by us as well as by others.26−30
peptide that carries a free cysteine thiol (SH) resulting from site-specific mutagenesis. However, when dealing with amyloidogenic peptides like Aβ, the low solubility and aggregation-prone nature of such peptides can jeopardize the success of the DTNB approach. For example, oligomerization and aggregation can hinder the access of DTNB to the free SH; addition of cyanide salt may induce quick aggregation. In fact, we actually failed to label a short amyloidogenic peptide, Aβ(13−23) cysteine mutant (HHQKLVFFCED), using the DTNB approach simply due to the quick aggregation of this peptide under labeling condition. To overcome the obstacles caused by the aggregation of amyloidogenic peptide during labeling, we seek to find an alternative labeling approach completely different from the DTNB approach. Scheme 1 also illustrates the newly proposed approach using CDAP as the cyanylating reagent and 0.1 M HOAc as the solvent system. CDAP is an alternative cyanylating reagent that works well under acidic conditions and had been used in the selective cleavage of peptide bonds for peptide mapping in previous mass spectrometry studies.31−33 HOAc is the commonly chosen solvent to solubilize hydrophobic peptides and thus should be able to prevent peptide aggregation during labeling. In addition, we used HPLC to perform purification because it can effectively separate the target peptide from the side products due to additional side chain modifications by CDAP. Using this alternative approach, we successfully labeled three Aβ(8−28) cysteine mutants: G9C, Q15C, and S26C. The three labeling positions are located at the N-terminal, middle, and Cterminal regions, respectively. When choosing the labeling positions, we avoided choosing charged residues and we also avoided mutating the residues in the segment of KLVFF, which is believed to be crucial for Aβ fibrillation.34 The three SCNlabeled Aβ(8−28) are termed as G9C-SCN, Q15C-SCN, and S26C-SCN, respectively. Their ESI-MS and HPLC characterizations are shown in Figures S1−S6 of the Supporting Information. We now look at the amyloidogenic properties of the above three SCN-labeled Aβ(8−28) peptides. We discover that the three peptides can all quickly form amyloid fibrils under high concentration and high salt conditions. Figures S7A, S7B, and S7C in the Supporting Information show the ThT fluorescence assay for the fibrillation kinetics of G9C-SCN, Q15C-SCN, and S26C-SCN, respectively. Fibrillation was initiated by inducing NaCl into the Aβ(8−28) peptides solution in the D2O (90%)− DMSO (10%) (v/v) mixed solvent under high peptide concentration (∼2 mM) and high salt concentration (400 mM) conditions. As evidenced by the instant increase in ThT fluorescence intensity in Figure S7, the high peptide and high salt concentrations result in rapid fibrillation for all of the three
■
RESULTS AND DISCUSSION Before we perform amyloid structural analysis using the sidechain-based IR probe technique, we first describe a new method to label Aβ(8−28) with SCN IR group through a postsynthetic approach. The SCN moiety covalently attached to the side chain of a cysteine residue on protein or peptide is an excellent IR probe due to its high sensitivity to environmental perturbation and relatively large extinction coefficient.20,21 Though SCN probes have been incorporated into a variety of proteins and peptides in previous studies of protein structure and function, the incorporation of SCN probe into amyloidogenic peptides has never been reported. This is largely due to the fact that the conventional method of labeling SCN probe onto protein or peptide is not very suitable for amyloidogenic peptides. This method, as depicted in Scheme 1, Scheme 1. DTNB Approach versus CDAP Approach
was introduced by Fafarman et al. in 2006 using a two-step cyanylating process involving 5,5-dithiobis(2-nitrobenzoic acid) (DTNB, also termed as Ellman’s reagent).10 This DTNB approach can be easily implemented to any soluble protein or
Figure 1. AFM characterizations of the amyloid fibrils by the three labeled Aβ(8−28) peptides: (A) G9C-SCN, (B) Q15C-SCN, and (C) S26CSCN. C
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir labeled peptides, and the fibrillation process reaches completion in less than 1 h. One unique feature of the fibrillation kinetics shown in Figure S7 is that the lag phase that is typical in amyloid fibrillation kinetics was not observed under our fibrillation conditions. Such fibrillation behavior is very similar to that of the 14-residue prion (109−122) peptide in the mixed solvent of Hepes buffer and acetonitrile as observed by Petty et al.35 In Figures 1A, 1B, and 1C, we show the AFM images of the fibrils by G9C-SCN, Q15C-SCN, and S26C-SCN, respectively. These images were taken 1 h after the initiation of fibrillation to ensure the fibrils were sampled at the fibrillation end point. The CR assay was also performed to further confirm the amyloid nature of the observed fibrils in Figure 1, and the results are shown in Figures 2A, 2B, and 2C. The UV−vis absorption spectra of CR after binding to the fibrils by G9C-SCN, Q15C-SCN, and S26C-SCN all show a shoulder peak at ∼540 nm, which is an indicator of the presence of amyloid structure.36 The AFM characterizations of the amyloid fibrils by the three labeled Aβ(8−28) peptides lead to the following two observations. First, the amyloid fibrils by G9C-SCN, Q15CSCN, and S26C-SCN show identical morphologies, supporting the argument that the perturbation by the small size SCN group on the amyloidogenic property of Aβ(8−28) is minimal. Second, height analysis reveals that the average height of these amyloid fibrils is ∼1.5 nm, suggesting a relatively simple quaternary structure for these amyloid fibrils. The argument is as follows. Previous cryo-electron microscopy (cryo-EM) study of the amyloid fibril by Aβ(1−40) showed that the protofilament containing two β-sheets has a thickness of less than 2 nm;37 previous AFM study of Aβ(25−35) supports that one βsheet has a thickness of ∼0.8 nm;38 previous molecular dynamic simulation showed that a four-sheet amyloid fibril had a thickness of 2.8 nm, implying one β-sheet has a thickness of 0.7 nm.39 On the basis of these previous studies, we can conclude that the amyloid fibril by Aβ(8−28) under our fibrillation condition contains only two β-sheets. This two-βsheet architecture will serve as one constraint when we build the structural model for Aβ(8−28) amyloid fibril. In Figures 3A, 3B, and 3C, we show the FTIR spectra of G9C-SCN, Q15C-SCN, and S26C-SCN in the 1700−1600 cm−1 amide I′ region before and after salt-induced fibrillation in the D2O−DMSO mixed solvent. The FTIR spectra of G9CSCN, Q15C-SCN, and S26C-SCN before salt-induced fibrillation all feature a broad absorption band with the absorption maximum at ∼1645 cm−1. According to the empirical correlation between amide I (and amide I′) frequency and protein secondary structure,40 this observed 1645 cm−1 is indicative of the disordered nature of the three labeled Aβ(8− 28) peptides in the mixed D2O−DMSO solvent before fibrillation. After salt-induced fibrillation, the FTIR spectra of G9C-SCN, Q15C-SCN, and S26C-SCN change their shapes significantly. These FTIR spectra are all dominated with two relatively sharp peaks: one in the low-frequency region at 1616 cm−1 and the other in the high-frequency region at 1686 cm−1; in addition, the central part of the spectra is basically featureless with flat absorptions. In our two recent works, we have discussed in great detail whether IR spectroscopy can be used to distinguish antiparallel and parallel β-sheet structures in protein aggregates through some characteristic spectral features in the amide I region.26,27 On the basis of evidence from previous experimental and theoretical studies,41−48 we have proposed the following IR assignment criteria for antiparallel
Figure 2. CR assay for the amyloid fibrils by the three labeled Aβ(8− 28) peptides: (A) G9C-SCN, (B) Q15C-SCN, and (C) S26C-SCN.
and parallel β-sheets: the antiparallel β-sheet features both a low-frequency component below 1640 cm−1 and a highfrequency component above 1680 cm−1 in the amide I (or amide I′) region, whereas parallel β-sheet lacks the highfrequency component above 1680 cm−1. Meanwhile, in a recent review, Sarroukh et al. basically proposed the same IR assignment criteria for antiparallel and parallel β-sheets, and they further point out that for β-sheet structure in amyloid fibril the low-frequency component usually shows up between 1630 and 1611 cm−1.49According to these arguments, the presence of the two peaks at 1616 and 1686 cm−1 in the spectra in Figure 3 is indicative of antiparallel β-sheet configuration within the amyloid fibril by Aβ(8−28). This is consistent with previous D
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
β-sheet and disordered structures and shows no turn or helix structures as supported by the good match between FTIR experimental observation and theoretical calculation.39 These spectral observations thus support that amyloid fibril by Aβ(8− 28) contains only two types of secondary structures: antiparallel β-sheet structure and disordered structure. The structural insights from Figure 3 will serve as the constraints at the secondary structural level when we build the structural model for Aβ(8−28) amyloid fibril. From Figure S7 and Figures 1−3, we can see that the three SCN-labeled peptides, G9C-SCN, Q15C-SCN, and S26CSCN, show similar amyloidogenic properties. This observation supports the argument that SCN probe has minimal impact on the fibrillation of Aβ(8−28). In addition, we have performed ThT assay, AFM characterization, CR assay, and FTIR characterization on the fibrillation of wild type Aβ(8−28). These results are shown in Figures S8−S11. These additional control experiments further strengthen the argument that SCN probe impacts the fibrillation of Aβ(8−28) minimally. Yet, we think whether an IR probe has an impact on the aggregation behavior of an amyloidogenic peptide should always be evaluated in a case-by-case manner. In Figures 4A, 4B, and 4C, we show the FTIR secondderivative spectra of G9C-SCN, Q15C-SCN, and S26C-SCN in the 2170−2145 cm−1 SCN probe region before and after saltinduced fibrillation in the D2O−DMSO mixed solvent. The FTIR second-derivative spectra of D2O−DMSO mixed solvent are also shown for comparison. Unlike in previous studies where the actual frequency of IR probe was obtained by a delicate solvent subtraction approach,15 here we propose to use the second-derivative technique as a convenient and sensitive way to reveal the rather weak absorption of SCN probe which is overwhelmed by solvent D2O absorption in the original FTIR absorption spectrum. The success of this alternative approach is well demonstrated through the comparison between the peptide second-derivative spectra and the solvent secondderivative spectra in Figure 4. Furthermore, in Figure 5, we present the second-derivative spectra of the model compound of MeSCN in different D2O−DMSO solvent environments to demonstrate the sensitivity of second-derivative technique to detect the frequency shift of SCN probe caused by environmental perturbations. Certainly, we should keep in mind that though this second-derivative approach is a convenient approach in determining the actual frequency of the SCN probe, it is not suitable for extracting the line shape information about the SCN probe like the solvent subtraction approach does. Using SCN as a sensitive environmental probe, we are able to reveal the solvent exposure status of the SCN probes at the three different positions (i.e., 9, 15, and 26) within Aβ(8− 28) fibril. Such information will provide additional constraints when we build the structural model for the Aβ(8−28) amyloid fibril. We now look at the SCN frequencies before and after the fibrillation of G9C-SCN, Q15C-SCN, and S26C-SCN. As we can see in Figures 4A, 4B, and 4C, before fibrillation, the SCN frequencies of the three peptides are all located at the same position, 2162 cm−1; after fibrillation, the SCN frequencies of the three peptides are all red-shifted to new positions: the SCN frequency of G9C-SCN is now at 2160 cm−1, while the SCN frequencies of Q15C-SCN and S26C-SCN are both shifted to 2156 cm−1. To understand the structural implications of the observed SCN frequencies, we first address the relationship between the IR frequency and the solvent exposure status of
Figure 3. FTIR absorption spectra of the three labeled Aβ(8−28) peptides in the amide I′ region: (A) G9C-SCN, (B) Q15C-SCN, and (C) S26C-SCN. Dashed line: before fibrillation; solid line: after fibrillation.
observations that short segments of full length amyloid-forming protein or peptide such as Aβ(1−40) tends to form amyloid fibrils with antiparallel β-sheet configuration, while full length amyloid-forming protein or peptide tends to form amyloid fibrils with parallel β-sheet configuration.50,51 Furthermore, in Figure 3 we do not observe additional spectral growth (e.g., 1670 cm−1 for turn structure40) except the two peaks corresponding to antiparallel β-sheet structures. In fact, the FTIR spectral feature of the 21-residue Aβ(8−28) amyloid fibril observed here is very similar to that of the amyloid fibril by the 14-residue prion(109−122) peptide which only contains E
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 5. Second-derivative spectra of MeSCN in different solvent environments in the SCN probe region: (A) 1% DMSO in D2O (v/v); (B) 15% DMSO in D2O (v/v); (C) 30% DMSO in D2O (v/v).
fully solvent exposed to the aqueous environment, the SCN frequency is at 2163 cm−1. They further found that when the peptides are in micelle solution but the SCN probe is placed at solvent exposed sites, the SCN frequency is also located at 2163 cm−1. Bischak et al. studied the frequency of SCN labeled on the C-terminal domain of the measles virus nucleoprotein (NTAIL). NTAIL is an intrinsically disordered protein; when it is in buffer solution, the SCN probe on NTAIL remains fully exposed to the aqueous environment. They again found that the SCN frequency is located at 2163 cm−1.15 These results establish 2163 cm−1 as a reference frequency for SCN probe in fully solvent-exposed status. Fafarman et al. introduced SCN probe at residue 105 of ketosteroid isomerase (KSI) D40N31 from P. putida and measured the SCN frequency.53 X-ray diffraction showed that residue 105 is at a location where the SCN probe is buried deep within the protein interior and not within hydrogen-bonding distance of any polar groups. The measured frequency of SCN probe at this location in KSI was found to be ∼2154 cm−1. This result thus establishes 2154 cm−1 as a reference frequency for SCN probe in completely solvent-protected (or buried) status. People also investigated the frequencies of SCN probe in partially solvent-exposed status. Alfieri et al. used SCN probe to map the intermolecular contacts between the antimicrobial peptide CM15 and lipid membrane and found that the SCN frequency can go down to 2159 cm−1 while SCN still remains partially water exposed.54 Stafford et al. used SCN probe to study protein−protein, and their work showed that SCN probe at partially water-exposed position can exhibit frequency within the range of 2161−2159 cm−1.55 These results suggest that 2159 cm−1 can be viewed as a “borderline” reference frequency for SCN probe. Namely, above this frequency, SCN probe is considered as solvent exposed; below this frequency, SCN probe is considered as solvent protected. On the basis of these previous results, we include a “frequency ruler” as the figure inset in Figure 4 to visually illustrate the relationship between SCN frequency and solvent exposure status of SCN probe. As we can see, this ruler covers the frequency range from 2163 to 2154 cm−1, with the SCN frequency above 2159 cm−1 corresponding to SCN probe in solvent exposed state (blue-bar region) and the SCN frequency below 2159 cm−1 corresponding to SCN probe in solvent protected state (red-bar region). This ruler can be used
Figure 4. FTIR second-derivative spectra of the three labeled Aβ (8− 28) peptides in the SCN probe region: (A) G9C-SCN, (B) Q15CSCN, and (C) S26C-SCN. Solid line: before fibrillation; dashed line: after fibrillation; dotted line: solvent; blue bar: solvent exposed; red bar: solvent protected.
the SCN probe on peptide or protein in aqueous environment. Previous studies on the IR frequencies of SCN probe on SCNlabeled peptides and proteins under different solvent exposure status in aqueous environment have laid a very good foundation for us to establish such relationship. McMahon et al. labeled two model peptides with SCN probea membrane-binding sequence of the human myelin basic protein (MBP81−95) and the antimicrobial peptide CM15and they investigated the frequencies of the SCN probe on the two peptides under different solvent exposure status.52 They found that when these peptides are in buffer solution where the SCN probe remains F
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir as a qualitative measure for the solvent exposure status of SCN probe in amyloid structure. According to this ruler, the observed 2160 cm−1 frequency in Figure 4A suggests that the SCN probe at position 9 still remains solvent exposed after fibrillation, though it is not as well solvated when compared with the SCN probe at position 9 before fibrillation. This further implies that the residue at position 9 is in relatively disordered state within Aβ(8−28) fibril. The observed 2156 cm−1 frequency in Figures 4B and 4C appears to support that the SCN probes at positions 15 and 26 are exposed to similar hydrophobic environment after fibrillation, yet the exact interpretation for this observation is actually not this simple. The argument is as follows. If the SCN probe with 2156 cm−1 frequency is simply assigned to be the probe buried within the hydrophobic dry interface between the two β-sheets of Aβ(8− 28) fibril, it would be difficult to understand why the two SCNs (at positions 15 and 26) pointing to opposite directions with respect to the β-sheet plane are exposed to similar hydrophobic environments; furthermore, as we show in Figure 6, the frequencies of the SCN probes in dried G9C-SCN, S26C-SCN, and Q15C-SCN are all located at 2154 cm−1 (identical to the frequency of the SCN probe buried deep within the interior of protein KSI as observed by Fafarman et al. as we mentioned above), implying that it is not appropriate to simply assign the SCN probe with 2156 cm−1 frequency to be the SCN buried within the hydrophobic dry interface between the two β-sheets of Aβ(8−28) fibril. The assignment for the SCN probe with 2156 cm−1 frequency will be addressed in the following section with the proposal of the structural model for Aβ (8−28) fibril. In Scheme 2, we propose a structural model for Aβ (8−28) fibril. The construction of the model is based on the following two criteria. First, the model needs to satisfy the constraints set in Figures 1, 3, and 4; second, we will allow the key segment of KLVFFA (i.e., Aβ(16−21) to be involved in the steric zipper structure. The steric zipper motif is widely accepted to be the basic requirement for amyloid structure. Previous theoretical prediction shows that Aβ(16−21) is one of the most amyloidogenic segments for Aβ (along with other segments including Aβ(27−32), Aβ(29−34), Aβ(30−35), Aβ(35−40), Aβ(35−42), and Aβ(37−42)), and it can form steric zipper type microcrystal with class 7 symmetry according to the definition by Sawaya et al.1,56 As shown in Scheme 2, the proposed model consists of two in-register β-sheets with the strands being aligned in antiparallel configuration within each β-sheet; at each end of the strand, there are two residues with disordered conformation; other residues within the strands are in β-sheet configuration. The symmetry of this architecture is class 7. Class 7 is so far the only experimental observed symmetry for the amyloid structure by KLVFFA segment in microcrystal X-ray diffraction.56 With this model, β-sheets and disordered conformations are the two secondary structure in Aβ(8−28) amyloid fibril, consistent with the observation in amide I region in Figure 3. This model can also allow us to interpret the observation in the probe region in Figure 4. With this model, the SCN probes at position 9 are in disordered state; it thus shows an IR frequency at 2160 cm−1. For the SCN probes at positions 15 and 26, there are equal numbers of solvent exposed probes and solvent protected probes. The SCN probes in solvent protected states (i.e., in the dry interface between the two β-sheets) would show an IR frequency at 2154 cm−1, while the SCN probes in solvent exposed states would show an IR frequency between 2163 and 2159 cm−1. The combined contributions from these two types of SCN probes
Figure 6. FTIR characterization of the three labeled Aβ(8−28) peptides in dry states: (A) G9C-SCN, (B) Q15C-SCN, and (C) S26CSCN. Solid line: absorption; dashed line: second derivative.
could show an IR frequency at 2156 cm−1. This structural model also explains why the frequencies of the two probes at positions 15 and 26 are identical. Overall, if we do not consider the possible twist of the structure model shown in Scheme 2, the model can then be intuitively described as a millipede-like structure, with the compact β-sheet central region being the main “body” and the disordered peripheral regions being the short “legs”. Certainly, we should point out that the interstrand registry shown in the proposed model is hypothetical. In other words, this millipede-like structure is still a coarse model, and we believe a combination of side-chain-based IR probe and backbone-based IR probe (i.e., isotope-edited IR spectroscopy57) techniques will lead to a more delicate model. In the well-known amyloid model of the full length Aβ(1− 40), each Aβ(1−40) monomer contributes two β-strands with G
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
peptides by ThT assay; Figures S8−S11 for ThT assay, AFM characterization, CR assay, and FTIR characterization on wild type Aβ(8−28) fibrillation (PDF)
Scheme 2. Millipede-like Structural Model for Aβ(8−28) Amyloid Fibrila
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected] (G.M.). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We gratefully acknowledge the financial support from the National Natural Science Foundation of China (No.21075027), the Natural Science Foundation of Hebei Province (No. B2011201082) and JUREN PLAN, the Key Project of Chinese Ministry of Education (No. 211014), and Research Fund for the Doctoral Program of Higher Education of China (20121301110003). We thank Ms. Yilin Wu for her assistances on HPLC and ESI-MS.
■
ABBREVIATIONS FTIR, Fourier transform infrared; IR, infrared; AFM, atomic force microscopy; cryo-EM, cryo-electron microscopy; HPLC, high performance liquid chromatography; ESI-MS, electrospray ionization mass spectrometry; Aβ, amyloid β; SCN, thiocyanate; ThT, thioflavin T; CR, Congo red; D2O, deuterium oxide; CDAP, 1-cyano-4-dimethylaminopyridinium; DTNB, 5,5dithiobis(2-nitrobenzoic acid); DMSO, dimethyl sulfoxide; MeSCN, methyl thiocyanate; TFA, trifluoroacetic acid; ACN, acetonitrile; HOAc, acetic acid.
Arrow: peptide strand; shaded area in the arrow: ordered β-sheet region; blank area at the two ends of the arrow: disordered region; octagon with Arabic number: amino acid residue at a specific position; shaded octagon: residue with side chain pointing away from the viewer; blank octagon: residue with side chain pointing toward the viewer; the front and back β-sheets are identical, so only front β-sheet is shown with a clear view. a
residues 10−22 and residues 30−40; in addition, residues 23− 29 form a bend or loop and residues 1−9 are disordered.50,51 Therefore, the whole Aβ(1−40) molecule assumes a U-shaped conformation in amyloid fibril. Different from Aβ(1−40) amyloid model, in the Aβ(8−28) amyloid model proposed above, each Aβ(8−28) monomer only contributes one single βstrand. This is due to the fact that Aβ(8−28) is a significantly truncated version of Aβ(1−40), lacking residues 30−40 that could form another β-strand if present.
■
■
SUMMARY To the best of our knowledge, this work is the first application of side-chain-based IR probe technique in building a structural model for amyloid fibril. We believe the side-chain-based IR probe technique with the capability of tackling the structural details of amyloid aggregates at both secondary and quaternary structural levels can be a valuable complementary tool in future amyloid research. It can be used to provide more constraints that can allow us to fine-tune the amyloid structural models made by other techniques such as solid-state NMR, cryo-EM, and isotope-edited FTIR spectroscopy; it can also be potentially used to map the binding sites for amyloid inhibitors on amyloid fibrils and oligomers.
■
REFERENCES
(1) 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.; Madsen, A. O.; Riekel, C.; Eisenberg, D. Atomic Structures of Amyloid Cross-β Spines Reveal Varied Steric Zippers. Nature 2007, 447, 453−457. (2) Chiti, F.; Dobson, C. M. Protein Misfolding, Functional Amyloid, and Human Disease. Annu. Rev. Biochem. 2006, 75, 333−366. (3) Hauser, C. A. E.; Maurer-Stroh, S.; Martins, I. C. Amyloid-Based Nanosensors and Nanodevices. Chem. Soc. Rev. 2014, 43, 5326−5345. (4) Shimanovich, U.; Efimov, I.; Mason, T. O.; Flagmeier, P.; Buell, A. K.; Gedanken, A.; Linse, S.; Akerfeldt, K. S.; Dobson, C. M.; Weitz, D. A.; Knowles, T. P. J. Protein Microgels from Amyloid Fibril Networks. ACS Nano 2015, 9, 43−51. (5) Wang, X. Y.; Li, Y. F.; Zhong, C. Amyloid-Directed Assembly of Nanostructures and Functional Devices for Bionanoelectronics. J. Mater. Chem. B 2015, 3, 4953−4958. (6) Li, C.; Bolisetty, S.; Mezzenga, R. Hybrid Nanocomposites of Gold Single-Crystal Platelets and Amyloid Fibrils with Tunable Fluorescence, Conductivity, and Sensing Properties. Adv. Mater. 2013, 25, 3694−3700. (7) Reynolds, N. P.; Charnley, M.; Mezzenga, R.; Hartley, P. G. Engineered Lysozyme Amyloid Fibril Networks Support Cellular Growth and Spreading. Biomacromolecules 2014, 15, 599−608. (8) Zhou, X.-M.; Shimanovich, U.; Herling, T. W.; Wu, S.; Dobson, C. M.; Knowles, T. P. J.; Perrett, S. Enzymatically Active Microgels from Self-Assembling Protein Nanofibrils for Microflow Chemistry. ACS Nano 2015, 9, 5772−5781. (9) Getahun, Z.; Huang, C. Y.; Wang, T.; De Leon, B.; DeGrado, W. F.; Gai, F. Using Nitrile-Derivatized Amino Acids as Infrared Probes of Local Environment. J. Am. Chem. Soc. 2003, 125, 405−411. (10) Fafarman, A. T.; Webb, L. J.; Chuang, J. I.; Boxer, S. G. SiteSpecific Conversion of Cysteine Thiols into Thiocyanate Creates an
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b03616. Figures S1−S6 for HPLC and ESI-MS characterizations of the three labeled Aβ(8−28) peptides; Figure S7 for the fibrillation kinetics of the three labeled Aβ(8−28) H
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir IR Probe for Electric Fields in Proteins. J. Am. Chem. Soc. 2006, 128, 13356−13357. (11) Glasscock, J. M.; Zhu, Y.; Chowdhury, P.; Tang, J.; Gai, F. Using an Amino Acid Fluorescence Resonance Energy Transfer Pair to Probe Protein Unfolding: Application to the Villin Headpiece Subdomain and the Lysm Domain. Biochemistry 2008, 47, 11070−11076. (12) Ye, S.; Huber, T.; Vogel, R.; Sakmar, T. P. FTIR Analysis of GPCR Activation Using Azido Probes. Nat. Chem. Biol. 2009, 5, 397− 399. (13) Lindquist, B. A.; Furse, K. E.; Corcelli, S. A. Nitrile Groups as Vibrational Probes of Biomolecular Structure and Dynamics: An Overview. Phys. Chem. Chem. Phys. 2009, 11, 8119−8132. (14) Ye, S.; Zaitseva, E.; Caltabiano, G.; Schertler, G. F.; Sakmar, T. P.; Deupi, X.; Vogel, R. Tracking G-protein-Coupled Receptor Activation Using Genetically Encoded Infrared Probes. Nature 2010, 464, 1386−1389. (15) Bischak, C. G.; Longhi, S.; Snead, D. M.; Costanzo, S.; Terrer, E.; Londergan, C. H. Probing Structural Transitions in the Intrinsically Disordered C-Terminal Domain of the Measles Virus Nucleoprotein by Vibrational Spectroscopy of Cyanylated Cysteines. Biophys. J. 2010, 99, 1676−1683. (16) Taskent-Sezgin, H.; Chung, J.; Banerjee, P. S.; Nagarajan, S.; Dyer, R. B.; Carrico, I.; Raleigh, D. P. Azidohomoalanine: A Conformationally Sensitive IR Probe of Protein Folding, Protein Structure, and Electrostatics. Angew. Chem., Int. Ed. 2010, 49, 7473− 7475. (17) Jo, H.; Culik, R. M.; Korendovych, I. V.; DeGrado, W. F.; Gai, F. Selective Incorporation of Nitrile-Based Infrared Probes into Proteins via Cysteine Alkylation. Biochemistry 2010, 49, 10354−10356. (18) Smith, E. E.; Linderman, B. Y.; Luskin, A. C.; Brewer, S. H. Probing Local Environments with the Infrared Probe: L-4-Nitrophenylalanine. J. Phys. Chem. B 2011, 115, 2380−2385. (19) Waegele, M. M.; Culik, R. M.; Gai, F. Site-Specific Spectroscopic Reporters of the Local Electric Field, Hydration, Structure, and Dynamics of Biomolecules. J. Phys. Chem. Lett. 2011, 2, 2598−2609. (20) Kim, H.; Cho, M. Infrared Probes for Studying the Structure and Dynamics of Biomolecules. Chem. Rev. 2013, 113, 5817−5847. (21) Ma, J.; Pazos, I. M.; Zhang, W.; Culik, R. M.; Gai, F. SiteSpecific Infrared Probes of Proteins. Annu. Rev. Phys. Chem. 2015, 66, 357−377. (22) Oh, K.-I.; Lee, J.-H.; Joo, C.; Han, H.; Cho, M. β-Azidoalanine as an IR Probe: Application to Amyloid Aβ(16−22) Aggregation. J. Phys. Chem. B 2008, 112, 10352−10357. (23) Marek, P.; Mukherjee, S.; Zanni, M. T.; Raleigh, D. P. ResidueSpecific, Real-Time Characterization of Lag-Phase Species and Fibril Growth during Amyloid Formation: A Combined Fluorescence and IR Study of p-Cyanophenylalanine Analogs of Islet Amyloid Polypeptide. J. Mol. Biol. 2010, 400, 878−888. (24) Liu, H.; Lantz, R.; Cosme, P.; Rivera, N.; Andino, C.; Gonzalez, W. G.; Terentis, A. C.; Wojcikiewicz, E. P.; Oyola, R.; Miksovska, J. Site-Specific Dynamics of Amyloid Formation and Fibrillar Configuration of Aβ 1−23 Using an Unnatural Amino Acid. Chem. Commun. 2015, 51, 7000−7003. (25) Lin, D.; Luo, Y.; Wu, S.; Ma, Q.; Wei, G.; Yang, X. Investigation of the Aggregation Process of Amyloid-β-(16−22) Peptides and the Dissolution of Intermediate Aggregates. Langmuir 2014, 30, 3170− 3175. (26) Zou, Y.; Li, Y.; Hao, W.; Hu, X.; Ma, G. Parallel β-Sheet Fibril and Antiparallel β-Sheet Oligomer: New Insights into Amyloid Formation of Hen Egg White Lysozyme under Heat and Acidic Condition from FTIR Spectroscopy. J. Phys. Chem. B 2013, 117, 4003−4013. (27) Zou, Y.; Hao, W.; Li, H.; Gao, Y.; Sun, Y.; Ma, G. New Insight into Amyloid Fibril Formation of Hen Egg White Lysozyme Using a Two-Step Temperature-Dependent FTIR Approach. J. Phys. Chem. B 2014, 118, 9834−9843. (28) Dzwolak, W.; Smirnovas, V.; Jansen, R.; Winter, R. Insulin Forms Amyloid in a Strain-Dependent Manner: An FT-IR Spectroscopic Study. Protein Sci. 2004, 13, 1927−1932.
(29) Sassi, P.; Giugliarelli, A.; Paolantoni, M.; Morresi, A.; Onori, G. Unfolding and Aggregation of Lysozyme: A Thermodynamic and Kinetic Study by FTIR Spectroscopy. Biophys. Chem. 2011, 158, 46− 53. (30) Giugliarelli, A.; Sassi, P.; Paolantoni, M.; Morresi, A.; Dukor, R.; Nafie, L. Vibrational Circular Dichroism Spectra of Lysozyme Solutions: Solvent Effects on Thermal Denaturation Processes. J. Phys. Chem. B 2013, 117, 2645−2652. (31) Wakselman, M.; Guibe-Jampel, E. 1-Cyano-4-DimethylaminoPyridinium Salts: New Water-Soluble Reagents for the Cyanylation of Protein Sulfhydryl Groups. J. Chem. Soc., Chem. Commun. 1976, 21− 22. (32) Wu, J.; Watson, J. T. A Novel Methodology for Assignment of Disulfide Bond Pairings in Proteins. Protein Sci. 1997, 6, 391−398. (33) Pipes, G. D.; Kosky, A. A.; Abel, J.; Zhang, Y.; Treuheit, M. J.; Kleemann, G. R. Optimization and Applications of CDAP Labeling for the Assignment of Cysteines. Pharm. Res. 2005, 22, 1059−1068. (34) Tjernberg, L. O.; Callaway, D. J.; Tjernberg, A.; Hahne, S.; Lilliehöök, C.; Terenius, L.; Thyberg, J.; Nordstedt, C. A Molecular Model of Alzheimer Amyloid β-Peptide Fibril Formation. J. Biol. Chem. 1999, 274, 12619−12625. (35) Petty, S. A.; Decatur, S. M. Intersheet Rearrangement of Polypeptides During Nucleation of β-Sheet Aggregates. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 14272−14277. (36) Klunk, W. E.; Jacob, R. F.; Mason, R. P. Quantifying Amyloid by Congo Red Spectral Shift Assay. Methods Enzymol. 1999, 309, 285− 305. (37) Sachse, C.; Faendrich, M.; Grigorieff, N. Paired β-Sheet Structure of an Aβ(1−40) Amyloid Fibril Revealed by Electron Microscopy. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 7462−7466. (38) Karsai, A.; Grama, L.; Murvai, U.; Soos, K.; Penke, B.; Kellermayer, M. S. Z. Potassium-Dependent Oriented Growth of Amyloid β 25−35 Fibrils on Mica. Nanotechnology 2007, 18, 345102. (39) Polzi, L. Z.; Amadei, A.; Aschi, M.; Daidone, I. New Insight into the IR-Spectra/Structure Relationship in Amyloid Fibrils: A Theoretical Study on a Prion Peptide. J. Am. Chem. Soc. 2011, 133, 11414−11417. (40) Barth, A.; Zscherp, C. What Vibrations Tell Us About Proteins. Q. Rev. Biophys. 2002, 35, 369−430. (41) Kubelka, J.; Keiderling, T. A. Differentiation of β-Sheet-Forming Structures: Ab Initio-Based Simulations of IR Absorption and Vibrational CD for Model Peptide and Protein β-Sheets. J. Am. Chem. Soc. 2001, 123, 12048−12058. (42) Decatur, S. M. Elucidation of Residue-Level Structure and Dynamics of Polypeptides via Isotopp-Edited Infrared Spectroscopy. Acc. Chem. Res. 2006, 39, 169−175. (43) Ling, Y. L.; Strasfeld, D. B.; Shim, S. H.; Raleigh, D. P.; Zanni, M. T. Two-Dimensional Infrared Spectroscopy Provides Evidence of an Intermediate in the Membrane-Catalyzed Assembly of Diabetic Amyloid. J. Phys. Chem. B 2009, 113, 2498−2505. (44) Fu, L.; Ma, G.; Yan, E. C. Y. In Situ Misfolding of Human Islet Amyloid Polypeptide at Interfaces Probed by Vibrational Sum Frequency Generation. J. Am. Chem. Soc. 2010, 132, 5405−5412. (45) Guo, Y.; Wang, J. Spectroscopic Evidence for Polymorphic Aggregates Formed by Amyloid-β Fragments. ChemPhysChem 2012, 13, 3901−3908. (46) Buchanan, L. E.; Dunkelberger, E. B.; Tran, H. Q.; Cheng, P.N.; Chiu, C.-C.; Cao, P.; Raleigh, D. P.; de Pablo, J. J.; Nowick, J. S.; Zanni, M. T. Mechanism of IAPP Amyloid Fibril Formation Involves an Intermediate with a Transient β-Sheet. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 19285−19290. (47) Shanmugam, G.; Polavarapu, P. L. Site-Specific Structure of Aβ (25−35) Peptide: Isotope-Assisted Vibrational Circular Dichroism Study. Biochim. Biophys. Acta, Proteins Proteomics 2013, 1834, 308− 316. (48) Welch, W. R.; Kubelka, J.; Keiderling, T. A. Infrared, Vibrational Circular Dichroism, and Raman Spectral Simulations for β-Sheet Structures with Various Isotopic Labels, Interstrand, and Stacking I
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir Arrangements Using Density Functional Theory. J. Phys. Chem. B 2013, 117, 10343−10358. (49) Sarroukh, R.; Goormaghtigh, E.; Ruysschaert, J.-M.; Raussens, V. ATR-FTIR: A “Rejuvenated” Tool to Investigate Amyloid Proteins. Biochim. Biophys. Acta, Biomembr. 2013, 1828, 2328−2338. (50) Tycko, R. Solid State Nmr Studies of Amyloid Fibril Structure. Annu. Rev. Phys. Chem. 2011, 62, 279. (51) Tycko, R.; Wickner, R. B. Molecular Structures of Amyloid and Prion Fibrils: Consensus versus Controversy. Acc. Chem. Res. 2013, 46, 1487−1496. (52) McMahon, H. A.; Alfieri, K. N.; Clark, C. A. A.; Londergan, C. H. Cyanylated Cysteine: A Covalently Attached Vibrational Probe of Protein-Lipid Contacts. J. Phys. Chem. Lett. 2010, 1, 850−855. (53) Fafarman, A. T.; Sigala, P. A.; Herschlag, D.; Boxer, S. G. Decomposition of Vibrational Shifts of Nitriles into Electrostatic and Hydrogen-Bonding Effects. J. Am. Chem. Soc. 2010, 132, 12811− 12813. (54) Alfieri, K. N.; Vienneau, A. R.; Londergan, C. H. Using Infrared Spectroscopy of Cyanylated Cysteine to Map the Membrane Binding Structure and Orientation of the Hybrid Antimicrobial Peptide CM15. Biochemistry 2011, 50, 11097−11108. (55) Stafford, A. J.; Ensign, D. L.; Webb, L. J. Vibrational Stark Effect Spectroscopy at the Interface of Ras and Rap1A Bound to the Ras Binding Domain of RalGDS Reveals an Electrostatic Mechanism for Protein-Protein Interaction. J. Phys. Chem. B 2010, 114, 15331−15344. (56) Colletier, J.-P.; Laganowsky, A.; Landau, M.; Zhao, M.; Soriaga, A. B.; Goldschmidt, L.; Flot, D.; Cascio, D.; Sawaya, M. R.; Eisenberg, D. Molecular Basis for Amyloid-β Polymorphism. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 16938−16943. (57) Moran, S. D.; Zanni, M. T. How to Get Insight into Amyloid Structure and Formation from Infrared Spectroscopy. J. Phys. Chem. Lett. 2014, 5, 1984−1993.
J
DOI: 10.1021/acs.langmuir.5b03616 Langmuir XXXX, XXX, XXX−XXX
