Research Article Received: 18 April 2014

Revised: 12 June 2014

Accepted: 26 June 2014

Published online in Wiley Online Library: 17 July 2014

(wileyonlinelibrary.com) DOI 10.1002/psc.2677

Synthesis and disaggregation of asparagine repeat-containing peptides Xiaomeng Lua and Regina M. Murphyb* Of all amino acid repeats in eukaryotes, polyglutamine (polyQ) is the most frequent, followed by polyasparagine (polyN). Glutamine repeats are expanded in proteins associated with several neurodegenerative disorders. The expanded polyQ domain is known to induce aggregation, and it is hypothesized that aggregation is directly causative of pathology. Despite the widespread presence of asparagine repeats in invertebrate eukaryotes, polyN is curiously quite rare in vertebrates. Several investigators have characterized the conformational and aggregation properties of polyQ-containing peptides and proteins, and to a lesser extent, peptides containing mixed glutamine and asparagine, but to our knowledge, there is no detailed characterization of polyN-containing peptides. Such a comparison could elucidate reasons for the paucity of asparagine repeats in humans. In this study, we synthesized a peptide containing a 24-asparagine repeat (N24). For aggregation studies, it is critical to start with monomeric unaggregated peptide. A protocol involving dissolution in mixed trifluoroacetic acid and hexafluoroisopropanol (TFA + HFIP) solvents is widely used for disaggregation of polyQ peptides. We used the same protocol for N24 but discovered that there was both oxidative damage and insufficient disaggregation. Oxidation of tryptophan, used as a flanking residue, was common. Moreover, we found evidence of Förster resonance energy transfer between Trp and its oxidation product N-formylkynurenine, even in chemical denaturants. This suggested that N24 was insufficiently disaggregated, a conclusion that was further supported by gel electrophoresis analysis. Oxidation was reduced, but not eliminated, by addition of methionine to the buffer. Formic acid proved to be a better disaggregator and caused no oxidative damage. The glutamine repeat peptide Q24 also underwent some oxidation after extended incubation in TFA + HFIP, but there was no evidence of Förster resonance energy transfer, and samples appeared monomeric by gel electrophoresis. This result indicates that polyN-containing peptides self-associate more strongly than polyQcontaining peptides. Circular dichroism spectra reveal a greater propensity for β-turn formation in polyN than polyQ, providing an explanation for the increased stability of polyN aggregates relative to polyQ. Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd. Additional supporting information may be found in the online version of this article at the publisher’s web site. Keywords: polyglutamine; polyasparagine; aggregation; tryptophan; oxidation

Introduction

860

Homo-amino acid repeats are quite common in eukaryotes: One survey of the human genome identified 20% of genes as containing one or more repeat tracts [1]. Repeat-containing sequences are found frequently among the family of intrinsically disordered proteins [2], possibly because there are many energetically equivalent configurations [3]. Of all repeat-containing proteins (RCPs) in eukaryotes, those containing glutamine (Q) are the most common [4]. Asparagine (N) repeats are the second most common in eukaryotes: Curiously, these are abundant in primitive invertebrates but rare in vertebrates [4]. For example, the human genome contains 233 Q repeats but only 8 N repeats, all in a single protein. The opposite pattern occurs in Plasmodium falciparum, where N repeats are plentiful but Q repeats are nearly absent [4]. Glutamine repeats in proteins are polymorphic in length. Abnormal expansion is linked to at least nine progressive neurodegenerative disorders, including Huntington’s disease [5]. Proteins with expanded Q repeats are prone to aggregation, and it is hypothesized that the formation of aggregates is intimately associated with disease onset and progression. This has motivated several investigators to examine the conformational and aggregation properties of Q repeat-containing peptides (e.g. [6–14]). Briefly, polyglutamine (polyQ) peptide monomers shift from extended to compact as the length of the repeat increases, and the rate of aggregation increases rapidly with an increase in the number of

J. Pept. Sci. 2014; 20: 860–867

Qs. Monomers lack regular secondary structure, while mature aggregates are fibrillar in morphology with a β-sheet fold [6–14]. Mixed Q/N-rich domains appear in Sup35 and other prion proteins and play a critical role in aggregation and propagation of these prions [15–17]. Fibrillar aggregates are more common in N-rich prions, whereas nonfibrillar assemblies are more common in Q-rich proteins, a difference attributed to stereochemical considerations [18]. A comparison of the properties of long N versus long Q tracts might help to explain the striking difference in frequency of these repeats in invertebrates versus vertebrates. The purpose of the work reported here was to develop reliable methods for producing peptides containing long N tracts that would be suitable for detailed biophysical characterization of their conformational

* Correspondence to: Regina M. Murphy, Department of Chemical and Biological Engineering, University of Wisconsin-Madison, Madison, WI 53706, USA. E-mail: [email protected] a Biophysics Program, University of Wisconsin-Madison, Madison, WI 53706, USA b Department of Chemical and Biological Engineering, University of WisconsinMadison, Madison, WI 53706, USA Abbreviations: CD, circular dichroism; FRET, Förster resonance energy transfer; HFIP, hexafluoroisopropanol; NFK, N-formylkynurenine; RCP, repeat-containing protein; SEC, size-exclusion chromatography; TFA, trifluoroacetic acid.

Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.

ASPARAGINE REPEAT PEPTIDES and aggregation properties. To our knowledge, there are no reports of synthesis or biophysical characterization of peptides containing pure N repeats. Here, we report the synthesis of N24, a peptide with 24 contiguous Ns. We chose the sequence KKWN24KK, to parallel our previous work with polyQ. For biophysical characterization, it is critical that fully disaggregated (monomeric) peptide solutions are obtained. Previous researchers have documented the difficulties in preparing disaggregated Q repeat peptides, and they established a protocol for disaggregation [19] that has been widely used [9,20–24]. We applied the protocol for disaggregation of Q repeat peptides to our N repeat peptides and found, rather surprisingly, that the accepted protocol was ineffective at disaggregation and moreover caused oxidative damage. We show that an alternative protocol is simpler to perform, highly effective at disaggregating N repeat peptides, and avoids oxidative damage. Finally, we report that the greater stability of polyN aggregates vis-à-vis polyQ can be traced to greater propensity for β-turn formation.

Materials and Methods Peptide Synthesis and Purification All materials were from Fisher Scientific (Pittsburgh, PA, USA) except where indicated. Peptides K2WN24K2 (N24) and K2WQ24K2 (Q24) were synthesized using standard Fmoc solid-phase methods on a Protein Technology (Tucson, AZ, USA) Inc. Symphony synthesizer. Q and N with trityl side-chain protecting groups, and lysine and tryptophan with Boc side-chain protecting groups, were purchased from Novabiochem (Gibbstown, NJ, USA). The resin used was Fmoc-PAL-PEG-PS from Applied Biosystems (Foster City, CA, USA). Half the resin sites were blocked with lysine-Boc to reduce on-bead aggregation. Extended cycles and double coupling were used to improve yield: Specifically, the first five amino acids used extended coupling cycles, the next five used double coupling cycles, and this pattern was repeated until the last N or Q. Extended coupling cycles were also used for the N-terminal KKW. The peptide was C-terminal amidated and N-terminal acetylated. Peptides were cleaved from the resin using 95% trifluoroacetic acid (TFA), 2.5% ethanedithiol (Fluka, Buchs, Switzerland), and 2.5% H2O, precipitated into cold t-butylmethyl ether, and digested in aqueous solvent for 1 h before lyophilization. Crude lyophilized peptide was solubilized in a 1 : 1 solution of TFA and hexafluoroisopropanol (HFIP). This solution was evaporated under gentle N2 flow, and the peptide was resuspended in 55% TFA in water before purifying by reverse-phase highperformance liquid chromatography on a Vydac C18 column. Peptide was eluted from the column with a linear gradient of acetonitrile and water with 0.1% TFA, starting at 5% and ending at 30% acetonitrile. The major eluting fraction was collected and lyophilized. Peptide identity was confirmed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. TFA + HFIP Disaggregation Protocol

J. Pept. Sci. 2014; 20: 860–867

To determine peptide concentration by measuring Trp fluorescence, samples were diluted 5- to 20-fold into 6 M GuHCl. Trp fluorescence spectra were obtained with a QuantaMaster Series spectrofluorometer from PTI (Birmingham, NJ, USA) with excitation wavelength at 295 nm and emission wavelength measured from 310 to 540 nm. The emission intensity at 350 nm was compared with a standard curve of pure tryptophan in 6 M GuHCl to determine the concentration. The peptide concentration was also determined in triplicate with a bicinchoninic acid (Pierce, Rockford IL, USA) assay in the microplate format after tenfold dilution into phosphate buffered saline buffer. Color formation was measured using a BioTek (Winooski, VT, USA) EL 800 universal microplate reader and following the manufacturer’s protocol. Gel Electrophoresis Peptide samples (100–200 μM, pH 3) were typically diluted twofold by adding urea to a final 8 M urea concentration. Urea was used rather than GuHCl because the latter precipitates in sodium dodecylsulfate (SDS). Tricine SDS sample buffer (2×) was added to peptide samples at a 1 : 1 ratio. Samples were loaded on a Novex (Life Technologies, Carlsbad, CA, USA) 10–20% tricine gel, along with spectra multicolor low range protein ladder (Pierce), and electrophoresed using tricine SDS running buffer for 100 min at 125 V. Gels were silver stained (Pierce) following the manufacturer’s protocol. Size-exclusion Chromatography A BioSil (BioRad, Hercules, CA, USA) size-exclusion chromatography (SEC)-125™ column was connected to ÄKTA Explorer 100 FPLC system (GE Healthcare Life Sciences, Piscataway, NJ, USA). The column was equilibrated with 50 mM sodium phosphate buffer at pH 3, and 100 μl of sample (100–200 μM) was eluted using the same buffer at pH 3. Peptide was detected by absorbance at 280 and 330 nm. Circular Dichroism N24 and Q24 stock solutions were diluted into phosphate/sodium fluoride buffer (10 mM K2HPO4/KH2PO4 and 140 mM NaF, pH 7.4) to a peptide concentration of 30 μM. Samples were filtered through a 0.45 μm filter immediately before transfer to a 1-mm cell. Circular dichroism (CD) spectra were collected using an Aviv (Lakewood, NJ, USA) 202SF CD spectrometer at 37 °C. Solvent spectra were collected and subtracted.

Results Peptide synthesis and purification Building on our prior work with Q repeat peptides [9], we synthesized N repeat peptides. The full sequence was K2WN24K2 (N24) where tryptophan serves for concentration determination by fluorescence and lysines facilitate solubilization. The crude peptide was cleaved from the resin, lyophilized, and then resuspended in 55% TFA in water. The resuspended N24 peptide was cloudy. The solution was filtered prior to purification by reversephase high-performance liquid chromatography, and purified peptide was lyophilized. By mass spectrometry, the molar mass for purified N24 was determined (3495.5 measured, 3496.5 expected) along with some N deletion peptides (delta = 114 per missing N, Figure S1). For comparison, Q24 was synthesized and purified using a similar protocol. Unlike N24, solutions of Q24 in

Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/jpepsci

861

Lyophilized purified peptide was incubated in a 1 : 1 solution of TFA and HFIP for several hours or overnight. Solvent was evaporated under a gentle flow of N2, and the peptides were redissolved in water adjusted to pH 3 with TFA to a final stock solution concentration of 100–200 μM. Peptide stock solutions were filtered through a 0.45 μm filter, aliquoted, snap frozen in dry ice and ethanol, and stored at 80 °C. Prior to each experiment, a vial was thawed, filtered through a 0.22 μm filter, and used immediately.

Concentration Determination

LU AND MURPHY 55% TFA were clear. The correct molar mass (3832.9 measured, 3833.1 expected) was obtained, along with some Q deletion peptides (delta = 128 per missing Q, Figure S2). Evaluation of TFA + HFIP disaggregation protocol for N repeat peptides In any study of aggregating proteins or peptides, rendering the initial material aggregate-free and monomeric poses a significant hurdle. A sequential TFA/HFIP protocol has previously been validated for disaggregation of aggregation-prone peptides such as β-amyloid, wherein the peptide is dissolved in TFA, dried, solubilized in HFIP, and then dried again before dilution into water or buffer [25]. However, this sequential TFA/HFIP treatment was reported to be ineffective at disaggregating Q repeat-containing peptides, especially longer ones [19]. Chen and Wetzel demonstrated that mixed TFA + HFIP was more effective at disrupting extensive intramolecular and/or intermolecular hydrogen bonding. Their protocol called for incubation of lyophilized Q repeat peptide in a 1 : 1 mixture of TFA + HFIP for 0.5 to 4 h, or longer if needed to dissolve all aggregates, followed by solvent evaporation, dissolution in pH 3 water, snap-freezing, and storage [19,26]. Similar protocols have been used by several research groups to prepare Q repeat and other aggregation-prone peptides [9,20–24]. We therefore used the mixed TFA + HFIP protocol, previously developed for Q repeats, to prepare N24. Fluorescence spectra of purified N24 before and after TFA + HFIP treatment were collected. Prior to treatment, the Trp emission peak at 350 nm was observed as expected. However, after treatment, the Trp peak disappeared, and a new peak centered at 423 nm appeared (Figure 1a). We varied the length of time that the sample was incubated in TFA + HFIP from 2 to 48 h and observed that the

350 nm peak decreased, and the 423 nm peak increased, as incubation time increased (Figure 1b). This result demonstrates that the fluorescence changes are directly related to the TFA + HFIP treatment. We also collected absorbance spectra on N24 before and after TFA + HFIP treatment. Before treatment, N24 exhibited the expected Trp peak near 280 nm (Figure 1c). After TFA + HFIP treatment (Figure 1d), maximum absorbance shifted to lower wavelengths (~250 nm), and there was a long tail of absorbance greater than 300 nm that was absent in the purified sample prior to TFA + HFIP disaggregation. We hypothesized that chemical changes to N24 during the TFA + HFIP treatment were responsible for the changes in absorbance and fluorescence. To test this, we collected mass spectra of N24 after disaggregation. Besides the peak for unmodified N24 at m/z 3496.6, we observed the appearance of a new peak at 3512.5 (+16) and a smaller broad peak from 3524.6 to 3529.5 (+28 to +33) (Figure 2a). The molecular weight of these new peaks is consistent with partial oxidation of the tryptophan residue to oxindolylalanine, 5-hydroxytryptophan, and/or N-formylkynurenine (NFK) [27–30]. This interpretation is consistent with the observed changes in fluorescence and absorbance spectra (Table 1). Other minor oxidation byproducts (Table 1) may also exist in the sample after disaggregation. To check whether the TFA + HFIP protocol was effective at removing aggregates, we analyzed N24 by gel electrophoresis. As shown in Figure 2b, the treated N24 sample contained two bands, corresponding to protein calibration standards at 10 and 14 kDa. Disordered peptides, because they typically contain large numbers of hydrophilic residues, tend to bind less SDS, and it is a common phenomenon for the apparent molar mass to be significantly larger than the actual [31]. Therefore, the band at ~10 kDa is likely monomer. Assignment of this band as a monomer was further supported by similar migration on gel electrophoresis of a Q repeat peptide of similar molecular weight containing interrupting prolines (KKWQ5PQ5PPQ5PQ5KK, data not shown), which is not aggregation-prone [32,33]. We attribute the 14 kDa band to N24 dimers, because the difference in apparent molar mass between the two bands corresponds closely to N24’s molecular weight. We observed different amounts of these dimers in various samples, but they were always present in N24. In a few samples, a third faint band was detected, which was likely a trimer (data not shown). The persistence of N24 dimers (and possibly higher molecular weight aggregates) after disaggregation, dilution into chemical denaturant, and addition of SDS speaks to the remarkable self-associating nature of N24. Evidence of Förster resonance energy transfer between Trp and its oxidative byproducts

862

Figure 1. Fluorescence and absorbance changes after disaggregation. Purified N24 was disaggregated using the TFA + HFIP protocol. (a) Fluorescence emission spectra for N24 in 6 M GuHCl with excitation at 295 nm. Trp standards (–   –) and (), N24 at 5-fold (—), 10-fold (–  –), and 20-fold dilution (– –). (b) Fluorescence emission spectra for N24 after disaggregation in TFA + HFIP for 2 h (—), 4 h (), 18 h (– –), and 48 h (–  –). Absorbance spectra for N24 (c) before and (d) after TFA + HFIP disaggregation treatment. N24 concentration was 224 μM in (c) and 54 μM in (d), as determined by BCA assay.

wileyonlinelibrary.com/journal/jpepsci

Fluorescence excitation and emission spectra for NFK are redshifted compared with Trp (Table 1). Thus, Trp emission overlaps with NFK excitation, and there is the possibility of energy transfer [Förster resonance energy transfer (FRET)] between Trp and NFK [34], as long as Trp and NFK are separated by only a few nm. N24 suffered Trp oxidation damage post-disaggregation, but the mass spectral analysis still showed a significant amount of unmodified Trp (Figure 2a). To observe whether FRET occurred in our samples, we treated N24 with the TFA + HFIP disaggregation protocol overnight, then diluted the stock solution into 6 M GuHCl and measured emission spectra with excitation at 280 nm. We chose 280 nm instead of 295 nm to selectively excite Trp with no excitation of NFK [34]. We also collected emission spectra with excitation at 330 nm, to selectively excite NFK but not Trp. In both

Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.

J. Pept. Sci. 2014; 20: 860–867

ASPARAGINE REPEAT PEPTIDES

Figure 2. Evaluation of effect of HFIP + TFA disaggregation protocol on N24. (a) Mass spectra of N24 after disaggregation with the TFA + HFIP protocol. + Peptides are singly protonated, so peaks are M + H . (b) Gel electrophoresis of N24 and Q24 after disaggregation. (c) Fluorescence emission spectra for N24 disaggregated by TFA + HFIP. Samples were excited at 280 nm (– –) and 330 nm (—).

Table 1. Properties of Trp oxidation productsa Compound

Δm

Tryptophan

0

Kynurenine

+4

5-Hydroxytryptophan

+16

Oxindolylalanine 2,4-BisTrp-6,7-dione 6,7-Dione N-Formylkynurenine

+16 +28 +30 +32

Dioxindolylalanine

+32

Absorbance Max λ at 280 nm No absorbance past ~300 nm Max λ at 258 and 360 nm Min λ at 280 nm Max λ at 277 nm Tail to about 325 nm Max λ at 250 nm, shoulder from 280 to 320 nm N.A. N.A. Max λ at 260 and 322 nm Min λ at 280 nm N.A.

Fluorescence λex max at 280 nm λem max at 353 nm λex max at 365 nm λem max at 480 nm λem max at 339 nm N.A. N.A. N.A. λex max at 325 nm λem max at 430 nm N.A.

Reference 44 34 44 45

34, 46,47 [30]

a

Possible Trp oxidation products were identified by analyzing mass spectra using the Delta Mass Database at http://www.abrf.org. N.A., not known or data not available.

cases, we observed emission spectra with maxima ~423 nm (Figure 2c). This result indicates that selective excitation of Trp produces emission from NFK and is evidence of energy transfer from Trp to NFK. Because each N24 monomer contains only Trp or NFK, and not both, observation of FRET is further confirmation of the existence of dimers or higher-order oligomers, even after disaggregation and in the presence of a strong chemical denaturant. An improved disaggregation protocol

J. Pept. Sci. 2014; 20: 860–867

Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/jpepsci

863

Purging the TFA and HFIP and strict avoidance of exposure to light were not sufficient to prevent oxidation (data not shown).

Solvent evaporation with argon rather than nitrogen is unlikely to prevent oxidation, because the amount of oxidation increased with the dissolution time in TFA + HFIP (Figure 1) and thus cannot occur solely during the solvent evaporation phase. To see if oxidation could be prevented, we added methionine (Met) to the TFA + HFIP solvent during disaggregation, because Met is more readily oxidized than Trp and could act as a scavenger of trace amounts of oxidant. Adding Met suppressed oxidation, as measured by mass spectrometry (Figure 3a), absorbance (not shown), and fluorescence (Figure 3c), in a dose-dependent manner. However, aggregates were still present, as ascertained by gel electrophoresis (Figure 3b). We conclude that addition of Met during the

LU AND MURPHY

Figure 3. Effect of methionine use during disaggregation. N24 was disaggregated using the TFA + HFIP protocol, except methionine was added to the solvent. (a) Mass spectra of N24 disaggregated using the TFA + HFIP protocol with methionine added to the solvent. (b) Gel electrophoresis of N24 after disaggregation in presence of 17 or 100 mM methionine. (c) Fluorescence emission spectra (excited at 295 nm) for N24 with 17 mM (–  –) and 100 mM () methionine added during disaggregation.

TFA + HFIP disaggregation process reduces oxidation, but oxidative damage does not cause aggregation, and prevention of oxidation does not lead to effective disaggregation. Given that the TFA + HFIP protocol caused oxidative damage and did not completely eliminate aggregates even if oxidation was controlled by addition of Met, we tested several other disaggregation protocols. We dissolved N24 in TFA alone and found that here too there was oxidative modification (not shown). Neither HFIP alone nor DMSO was able to dissolve N24. We reasoned that formic acid might be a preferable solvent: It is a polar protic solvent that readily participates in hydrogen bonding as both donor and acceptor, and, as a common reducing agent, it is unlikely to cause oxidative damage. Lyophilized N24 was dissolved in 20 μl of neat formic acid, vortexed and centrifuged briefly but repeatedly, then after 1 min or less, resuspended in pH 3 water. Samples were filtered, snap frozen, and stored at 80 °C. Prior to use, samples were thawed and filtered through a 0.22 μm filter. By mass spectrometry (Figure 4a), fluorescence (Figure 4c), and absorbance (not shown), there was no evidence of oxidation or other chemical modifications. By gel electrophoresis (Figure 4b), we observed that formic acid-disaggregated N24 was monomeric. We conclude that formic acid treatment is superior for disaggregation of N24 compared with the widely used TFA + HFIP protocol. SEC analysis of oxidation

864

We discovered that SEC provides a useful method for quantitatively assessing the extent of oxidation during disaggregation. Figure 5a is a chromatogram of N24 disaggregated using the mixed TFA + HFIP protocol for 4 h. There are two major peaks, at 19.5 and 21 min. Both major peaks elute earlier than expected for monomers, on the basis of protein calibration standards, consistent with the hypothesis that N24 is relatively disordered and

wileyonlinelibrary.com/journal/jpepsci

expanded compared with compact folded proteins. Absorbance traces at both 280 and 330 nm are shown; because the 330/280 ratio is higher for the 21 min peak, we conclude that the 21 min peak contains peptide where the Trp is mostly oxidized, while the 19.5 min peak contains peptide with primarily unoxidized Trp. To confirm, we analyzed an aged and highly oxidized N24 sample; there was only one predominant peak, at 21 min (Figure 5b). If N24 was disaggregated in the presence of Met (Figure 5c) or using the formic acid protocol (Figure 5d), the only major peak was at 19.5 min. These data suggest that oxidized and nonoxidized N24 peptides are conformationally distinct: Oxidation leads to an apparently smaller size (longer elution time) on SEC, because of greater compaction, more nonspecific interaction with the column, or both. Comparison of N24 and Q24 TFA + HFIP treatment of Q24 caused some oxidation damage, as detected by mass spectrometry (data not shown). If Met was added, oxidation was inhibited, and a single 19-min elution peak was observed by SEC (data not shown). In contrast to N24, for Q24, we observed only a single monomer band by gel electrophoresis (Figure 2b). In addition, FRET was not detected with Q24 (data not shown). Taken together, these data show that both Q24 and N24 are susceptible to oxidation when the TFA + HFIP protocol is used. Q24 differs from N24, though, in that the former is more readily disaggregated to monomers. The greater difficulty of fully disaggregating N24 compared with Q24 indicates that N24 aggregates are more thermodynamically stable and suggests that there may be structural differences between these two peptides. To explore this question, we collected CD spectra on solutions of N24 and Q24 (Figure 6). Relative to Q24, N24 spectra showed a blueshift in the position of the minima (from 200 to 197 nm), a decrease in (absolute) ellipticity at

Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.

J. Pept. Sci. 2014; 20: 860–867

ASPARAGINE REPEAT PEPTIDES

Figure 4. Formic acid treatment of N24. (a) Mass spectra of N24 disaggregated using formic acid protocol. (b) Gel electrophoresis of N24 disaggregated by either TFA + HFIP or formic acid. (c) Fluorescence emission spectra for N24 disaggregated by formic acid. Samples were excited at 280 nm (—), 295 nm (– –), and 330 nm ().

Figure 6. Circular dichroic spectra of N24 (✕) and Q24 (○).

Figure 5. SEC analysis of peptides. Molecular weight markers are ubiquitin (8.5 kDa) and ribonuclease A (13.7 kDa). (a) Partially oxidized N24 disaggregated in TFA + HFIP, detected at 280 nm (—) and 330 nm (– –). (b) Highly oxidized N24 disaggregated in TFA + HFIP, detected at 280 nm. (c) N24 disaggregated in TFA + HFIP with added methionine, detected at 280 nm. (d) N24 disaggregated in formic acid, detected at 280 nm.

J. Pept. Sci. 2014; 20: 860–867

Discussion We report the successful synthesis and purification of a long N repeat-containing peptide, N24. Such peptides are of interest in studies of the conformational and aggregation properties of RCPs. For such studies, a disaggregation protocol is needed. We

Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/jpepsci

865

the minima, and a flattening of the shoulder from 212 to 225 nm. Nearly identical changes were reported previously when CD spectra of a Q repeat peptide was compared with that of a Q-rich peptide

containing a central DPro-Gly motif, a β-turn template [33]. These data show that N24 has greater propensity for forming β-turns than Q24, a result that has been predicted in simulations [18] but not previously shown experimentally.

LU AND MURPHY

866

initially adopted a TFA + HFIP disaggregation protocol from the literature, a protocol that has been used extensively to disaggregate Q repeat-containing peptides. Given the close similarity between Q and N, it is reasonable to imagine the protocol would be easily adapted to N repeats. However, we discovered that the protocol was of limited utility for two reasons: First, N24 was oxidatively damaged, and second, the peptide was not fully disaggregated. We believe that oxidation arises because of contact with TFA. This conclusion is supported by studies showing that hydrogen peroxide, a strong oxidant, may be generated in TFA [35]. Trp oxidation caused changes in the effective size of N24 monomers, as detected by SEC, but oxidation does not cause aggregation, because prevention of oxidation by addition of Met to TFA + HFIP did not eliminate aggregation. Further examination of solutions of TFA + HFIP-treated N24 led to the detection of FRET between Trp and NFK. For our system, FRET could occur only if a fraction of the peptide was oxidized and if disaggregation was incomplete; FRET was useful therefore to confirm the presence of multimers. We suspect that the selfassociation of N24 occurs during solid-phase synthesis, because we observed NFK and other oxidation products, but less FRET, if the synthesis yield was low. Our observation of FRET between Trp and NFK could be used to advantage in other systems. Synthetic routes have been worked out to deliberately and quantitatively incorporate NFK or other Trp oxidation products during solid-phase synthesis [36]. Deliberate incorporation of NFK as a FRET acceptor, with Trp as donor, could be advantageous over other acceptors (such as dansyl-lysine), because of NFK’s relatively smaller size and hydrophilicity. Switching to the formic acid protocol produced the desired results: no oxidation and complete disaggregation. Crick et al. used formic acid to disaggregate Q-containing peptides [37]. Other researchers working with Q RCPs or other aggregation-prone peptides may wish to consider using the formic acid disaggregation protocol, especially when working with peptides that contain Met, Cys, or Trp. Examples of these include the N-terminal domain of htt exon1 [38,39], and β-amyloid, where partial oxidation of Met was reported after 1-day incubation in 0.5% TFA [40]. Inadvertent oxidative damage during disaggregation could influence the conformational and/or aggregation properties, as suggested by our SEC analysis. N24 dimers persisted even after TFA + HFIP treatment and dissolution in strong chemical denaturants. The greater difficulty we had in disaggregating N24 compared with Q24 indicates that N24 dimers are more thermodynamically stable. This is consistent with the notion that polyN peptides form tighter β-turns, which we confirmed by CD. The greater β-turn propensity and greater stability of N24 aggregates compared with Q24 should produce a faster aggregation rate and more structured fibrillar aggregates in N24. Indeed, deliberate incorporation of a β-turn template into a Q repeat peptide shifts the CD spectra, markedly increases the aggregation rate, and alters the morphological appearance of aggregates [33]. Increasing the N content of Q/N-rich prion proteins led to more stable fibrillar aggregates and less toxicity [18], a phenomena that is consistent with the hypothesis that soluble partially disordered aggregates are more toxic than fibrils. Given our successful establishment of methods to synthesize, purify, and disaggregate N24, we plan now to more fully characterize the conformational and aggregational properties of long N repeat-containing peptides and compare with Q repeats of identical repeat length and flanking residues. Whether these biophysical differences between Q and N repeats can explain the differences in frequency of repeats in vertebrates

wileyonlinelibrary.com/journal/jpepsci

compared with invertebrates remains an unanswered question. Q repeats occur frequently in flexible disordered domains of human proteins and are important in transcription regulation in eukaryotes and the assembly of large macromolecular complexes [4,41]. Q repeat domains can act as flexible spacers between structured domains, helpful in tuning orientation in protein-protein interactions [1]. It could be that the greater disorder of Q repeats leads to greater opportunity for promiscuous interactions, or coupled folding and binding, and thereby greater regulation of transcription or assembly of macromolecular complexes and that these functions are more crucial to vertebrate biology. Perhaps lower organisms take advantage of the greater propensity of N repeats to self-assemble into structured aggregates for other functions, such as regulating cell phenotype or responding to environmental stress [42,43]. Acknowledgements The authors would like to acknowledge the National Science Foundation (CBET-1262729) for their sponsorship.

References 1 Karlin S, Brocchieri L, Bergman A, Mrazek J, Gentles AJ. Amino acid runs in eukaryotic proteomes and disease associations. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 333–338. 2 Dyson HJ. Expanding the proteome: disordered and alternatively folded proteins. Q. Rev. Biophys. 2011; 44: 467–518. 3 Tran HT, Mao A, Pappu RV. Role of backbone – solvent interactions in determining conformational equilibria of intrinsically disordered proteins. J. Am. Chem. Soc. 2008; 130: 7380–7392. 4 Faux NG, Bottomley SP, Lesk AM, Irving JA, Morrison JR, de la Banda MC, Whisstock JC. Functional insights from the distribution and role of homopeptide repeat-containing proteins. Genome Res. 2005; 15: 537–551. 5 Gatchel JR, Zoghbi HY. Diseases of unstable repeat expansion: mechanisms and common principles. Nat. Rev. Genet. 2005; 6: 743–755. 6 Chen SM, Ferrone FA, Wetzel R. Huntington’s disease age-of-onset linked to polyglutamine aggregation nucleation. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 11884–11889. 7 Chen SM, Berthelier V, Hamilton JB, O’Nuallain B, Wetzel R. Amyloidlike features of polyglutamine aggregates and their assembly kinetics. Biochem. 2002; 41: 7391–7399. 8 Slepko N, Bhattacharyya AM, Jackson GR, Steffan JS, Marsh JL, Thompson LM, Wetzel R. Normal-repeat-length polyglutamine peptides accelerate aggregation nucleation and cytotoxicity of expanded polyglutamine proteins. Proc. Natl. Acad. Sci. U. S. A. 2006; 103: 14367–14372. 9 Walters RH, Murphy RM. Examining polyglutamine peptide length: a connection between collapsed conformations and increased aggregation. J. Mol. Biol. 2009; 393: 978–992. 10 Kar K, Jayaraman M, Sahoo B, Kodali R, Wetzel R. Critical nucleus size for disease-related polyglutamine aggregation is repeat-length dependent. Nat. Struct. Mol. Biol. 2011; 18: 328–336. 11 Vitalis A, Wang XL, Pappu RV. Atomistic simulations of the effects of polyglutamine chain length and solvent quality on conformational equilibria and spontaneous homodimerization. J. Mol. Biol. 2008; 384: 279–297. 12 Singh VR, Lapidus LJ. The intrinsic stiffness of polyglutamine peptides. J. Phys. Chem. B 2008; 112: 13172–13176. 13 Crick SL, Jayaraman M, Frieden C, Wetzel R, Pappu RV. Fluorescence correlation spectroscopy shows that monomeric polyglutamine molecules form collapsed structures in aqueous solutions. Proc. Natl. Acad. Sci. U. S. A. 2006; 103: 16764–16769. 14 Sharma D, Shinchuk LM, Inouye H, Wetzel R, Kirschner DA. Polyglutamine homopolymers having 8–45 residues form slablike beta-crystallite assemblies. Proteins 2005; 61: 398–411. 15 Michelitsch MD, Weissman JS. A census of glutamine/asparagine-rich regions: implications for their conserved function and the prediction of novel prions. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 11910–11915. 16 Sawaya MR, Sambashivan S, Nelson R, Ivanova MI, Sievers SA, Apostol MI, Thompson MJ, Balbirnie M, Wiltzius JJW, McFarlane HT, Madsen AO, Riekel C, Eisenberg D. Atomic structures of amyloid cross-beta spines reveal varied steric zippers. Nature 2007; 447: 453–457.

Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.

J. Pept. Sci. 2014; 20: 860–867

ASPARAGINE REPEAT PEPTIDES 17 DePace AH, Santoso A, Hillner P, Weissman JS. A critical role for amino-terminal glutamine/asparagine repeats in the formation and propagation of a yeast prion. Cell 1998; 93: 1241–1252. 18 Halfmann R, Alberti S, Krishnan R, Lyle N, O’Donnell CW, King OD, Berger B, Pappu RV, Lindquist S. Opposing effects of glutamine and asparagine govern prion formation by intrinsically disordered proteins. Mol. Cell 2011; 43: 72–84. 19 Chen SM, Wetzel R. Solubilization and disaggregation of polyglutamine peptides. Protein Sci. 2001; 10: 887–891. 20 Sen S, Dash D, Pasha S, Brahmachari SK. Role of histidine interruption in mitigating the pathological effects of long polyglutamine stretches in SCA1: a molecular approach. Protein Sci. 2003; 12: 953–962. 21 Darnell G, Orgel JPRO, Pahl R, Meredith SC. Flanking polyproline sequences inhibit beta-sheet structure in polyglutamine segments by inducing PPII-like helix structure. J. Mol. Biol. 2007; 374: 688–704. 22 Bieschke J, Cohen E, Murray A, Dillin A, Kelly JW. A kinetic assessment of the C-elegans amyloid disaggregation activity enables uncoupling of disassembly and proteolysis. Protein Sci. 2009; 18: 2231–2241. 23 Digambaranath JL, Campbell TV, Chung A, McPhail MJ, Stevenson KE, Zohdy MA, Finke JM. An accurate model of polyglutamine. Proteins 2011; 79: 1427–1440. 24 Michalek M, Salnikov ES, Werten S, Bechinger B. Membrane interactions of the amphipathic amino terminus of Huntingtin. Biochem. 2013; 52: 847–858. 25 Zagorski MG, Yang J, Shao HY, Ma K, Zeng H, Hong A. Methodological and chemical factors affecting amyloid beta peptide amyloidogenicity. Meth. Enzymol. 1999; 309: 189–204. 26 O’Nuallain B, Thakur AK, Williams AD, Bhattacharyya AM, Chen SM, Thiagarajan G, Wetzel R. Kinetics and thermodynamics of amyloid assembly using a high-performance liquid chromatography-based sedimentation assay. Amyloid, Prions, Protein Aggregates, Pt C 2006; 413: 34–74. 27 Simat TJ, Steinhart H. Oxidation of free tryptophan and tryptophan residues in peptides and proteins. J. Agr. Food Chem. 1998; 46: 490–498. 28 Finley EL, Dillon J, Crouch RK, Schey KL. Identification of tryptophan oxidation products in bovine alpha-crystallin. Protein Sci. 1998; 7: 2391–2397. 29 Itakura K, Uchida K, Kawakishi S. Selective formation of oxindole-type and formylkynurenine-type products from tryptophan and its peptides treated with a superoxide-generating system in the presence of Iron(iii) EDTA – a possible involvement with iron oxygen complex. Chem. Res. Toxicol. 1994; 7: 185–190. 30 van de Weert M, Lagerwerf FM, Haverkamp J, Heerma W. Mass spectrometric analysis of oxidized tryptophan. J. Mass Spectrom. 1998; 33: 884–891. 31 Receveur-Brechot V, Bourhis JM, Uversky VN, Canard B, Longhi S. Assessing protein disorder and induced folding. Proteins 2006; 62: 24–45. 32 Thakur AK, Wetzel R. Mutational analysis of the structural organization of polyglutamine aggregates. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 17014–17019. 33 Walters RH, Murphy RM. Aggregation kinetics of interrupted polyglutamine peptides. J. Mol. Biol. 2011; 412: 505–519.

34 Fukunaga Y, Katsuragi Y, Izumi T, Sakiyama F. Fluorescence characteristics of kynurenine and N’-formylkynurenine – their use as reporters of the environment of tryptophan-62 in hen egg-white lysozyme. J. Biochem.-Tokyo 1982; 92: 129–141. 35 Vishnetskaya MV, Ivanova MS, Solkan VN, Zhidomirov GM, Melnikov MY. Activation of molecular oxygen in trifluoroacetic acid. Russ. J. Phys. Chem A. 2012; 86: 889–891. 36 Todorovski T, Fedorova M, Hennig L, Hoffmann R. Synthesis of peptides containing 5-hydroxytryptophan, oxindolylalanine, N-formylkynurenine and kynurenine. J. Pept. Sci. 2011; 17: 256–262. 37 Crick SL, Ruff KM, Garai K, Frieden C, Pappu RV. Unmasking the roles of N- and C-terminal flanking sequences from exon 1 of huntingtin as modulators of polyglutamine aggregation. Proc. Natl. Acad. Sci. U. S. A. 2013; 110: 20075–20080. 38 Thakur AK, Jayaraman M, Mishra R, Thakur M, Chellgren VM, Byeon IJL, Anjum DH, Kodali R, Creamer TP, Conway JF, Gronenborn AM, Wetzel R. Polyglutamine disruption of the huntingtin exon 1 N terminus triggers a complex aggregation mechanism. Nat. Struct. Mol. Biol. 2009; 16: 380–389. 39 Burke KA, Kauffman KJ, Umbaugh CS, Frey SL, Legleiter J. The interaction of polyglutamine peptides with lipid membranes is regulated by flanking sequences associated with huntingtin. J. Biol. Chem. 2013; 288: 14993–15005. 40 Shen CL, Fitzgerald MC, Murphy RM. Effect of acid predissolution on fibril size and fibril flexibility of synthetic beta-amyloid peptide. Biophys. J. 1994; 67: 1238–1246. 41 Atanesyan L, Gunther V, Dichtl B, Georgiev O, Schaffner W. Polyglutamine tracts as modulators of transcriptional activation from yeast to mammals. Biol. Chem. 2012; 393: 63–70. 42 Halfmann R, Lindquist S. Epigenetics in the extreme: prions and the inheritance of environmentally acquired traits. Science 2010; 330: 629–632. 43 Tuite MF, Serio TR. The prion hypothesis: from biological anomaly to basic regulatory mechanism. Nat. Rev. Mol. Cell Bio. 2010; 11: 823–833. 44 Twine SM, Szabo AG. Fluorescent amino acid analogs. Biophotonics Pt A 2003; 360: 104–127. 45 Savige WE, Fontana A. Oxidation of tryptophan to oxindolylalanine by dimethyl-sulfoxide hydrochloric-acid – selective modification of tryptophan containing peptides. Int. J. Pept. Prot. Res. 1980; 15: 285–297. 46 Balasubramanian D, Du X, Zigler JS. The reaction of singlet oxygen with proteins, with special reference to crystallins. Photochem. Photobiol. 1990; 52: 761–768. 47 Dreaden TM, Chen J, Rexroth S, Barry BA. N-Formylkynurenine as a marker of high light stress in photosynthesis. J. Biol. Chem. 2011; 286: 22632–22641.

Supporting Information Additional supporting information may be found in the online version of this article at the publisher’s web site.

867

J. Pept. Sci. 2014; 20: 860–867

Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/jpepsci

Synthesis and disaggregation of asparagine repeat-containing peptides.

Of all amino acid repeats in eukaryotes, polyglutamine (polyQ) is the most frequent, followed by polyasparagine (polyN). Glutamine repeats are expande...
810KB Sizes 2 Downloads 3 Views