Oligomers by Top-Down Electron Capture Dissociation Mass Spectromet

ARTICLE pubs.acs.org/ac

Conformer-Specific Hydrogen Exchange Analysis of Aβ(1
42) Oligomers by Top-Down Electron Capture Dissociation Mass Spectrometry Jingxi Pan,† Jun Han,‡ Christoph H. Borchers,‡ and Lars Konermann*,† † ‡

Department of Chemistry, The University of Western Ontario, London, Ontario, N6A 5B7 Canada University of Victoria-Genome BC Proteomics Centre, Victoria, British Columbia, V8Z 7X8 Canada ABSTRACT: Protein structural studies are particularly challenging under conditions in which several conformational species (e.g., monomers and aggregated forms) coexist in solution. Most spectroscopic techniques provide population-averaged data. Hence, it is usually not possible to obtain detailed structural information on individual protein species in heterogeneous samples. The current work employs an experimental strategy that addresses this issue. Solution-phase hydrogen exchange (HX) is used in combination with tandem mass spectrometry. Electrosprayed intact ions exhibiting specific HX mass shifts are selected in the gas phase, followed by electron capture dissociation. The resulting fragment ion deuteration pattern provides amide hydrogen bonding information in a conformerspecific and spatially resolved fashion. The feasibility of this approach is demonstrated by applying it to neurotoxic Aβ(1
42) oligomers that coexist with disordered monomers in solution. The findings of this study point to similarities between oligomers and mature amyloid fibrils with regard to the Aβ(1
42) backbone organization. Specifically, fibrils and oligomers appear to share a βloop-β secondary structure motif. The spatial resolution obtained with the “top-down” approach used here exceeds that of earlier proteolysis-based HX data on Aβ.

P

rotein structural studies are often conducted under solution conditions in which the native state is prevalent. However, it is also common to encounter situations in which proteins exhibit considerable structural heterogeneity. For example, under denaturing conditions, various partially unfolded species can coexist.1 Proteins may also form mixtures of different binding and aggregation states.2 Obtaining detailed structural insights for specific conformers in heterogeneous samples remains a key challenge for many protein studies.3 Commonly used spectroscopic techniques suffer from the shortcoming that they provide only populationaveraged information. Optical spectra are very broad, such that contributions associated with individual protein structures show extensive overlap. It is generally not possible to deconvolute these data into contributions arising from individual subpopulations. Modern NMR techniques offer a partial solution to this problem, although the analyses required for this approach are nontrivial.3 Mass spectrometry (MS) offers a number of relatively straightforward avenues for the detection of coexisting protein structures. For example, different conformers can give rise to unique ESI charge state distributions.4,5 In addition, mass measurements using “native” electrospray ionization (ESI) MS allow the detection of proteins that coexist in free and ligand-bound forms.6
8 Similarly, various oligomerization states of a given protein can be distinguished on the basis of their mass.9,10 Ion mobility spectrometry allows the detection of coexisting species through collision r 2011 American Chemical Society

cross section measurements.11
13 Although all of these approaches represent valuable experimental tools, they provide only lowresolution (global) structural information, while molecular details remain hidden. Amide hydrogen exchange (HX) MS represents another approach that is capable of detecting coexisting protein conformers. H atoms in stable hydrogen bonds exhibit significant protection. In contrast, rapid exchange can take place at amide groups that are not involved in hydrogen bonding and that are accessible to the solvent.14 Spatially resolved deuteration data can be obtained by subjecting proteins to acid quenching after HX, followed by proteolytic digestion and HPLC/MS analysis of the resulting peptides.15
17 Different conformers can be distinguished by HXMS in cases when they exhibit distinct HX mass shifts.18
22 Similar to the approaches mentioned above, however, this selectivity is largely restricted to intact protein measurements. Incorporation of a proteolysis step leads to digestion of all protein molecules in the sample, such that conformational selectivity is dramatically reduced.23 An even lower selectivity is obtained in HX-NMR experiments.1,18 Currently, there is considerable interest in the development of “top-down” techniques that can complement or replace traditional Received: April 8, 2011 Accepted: June 2, 2011 Published: June 02, 2011 5386

dx.doi.org/10.1021/ac200906v | Anal. Chem. 2011, 83, 5386–5393

Analytical Chemistry protease-based MS mapping avenues in proteomics and structural biology. Top-down measurements involve the dissociation of intact protein ions following electrospray ionization (ESI), with subsequent fragment ion analysis. Different ion activation methods can be used for this purpose, including collisions with neutrals, electron-ion reactions, ion
ion processes, and combinations thereof.24
28 A prerequisite for meaningful top-down HX-MS studies is the (near) absence of intramolecular 1H/2H scrambling. Only under these conditions can the solution-phase HX pattern of polypeptide chains be retained in the gas phase.29
32 Scrambling is a common occurrence in experiments that involve collisional activation.33 In contrast, both electron capture dissociation (ECD) and electron transfer dissociation (ETD)26,28,34
37 allow the preservation of deuterium and hydrogen atom locations during fragmentation.30,38,39 Hence, top-down ECD and ETD are suitable for monitoring the spatial 1H/2H distribution in electrosprayed protein ions.40
44 Advantages of top-down HXMS over protease-based measurements include a higher spatial resolution for small to medium-sized proteins, as well as the possibility to suppress the extent of isotope back-exchange.40,43 Possibly the most attractive feature of top-down HX-MS is the promise of combining spatial resolution with conformational selectivity. Attaining this goal requires precursor ion selection to ensure that only proteins exhibiting characteristic HX properties are subjected to dissociation while others are rejected. In this way, it should become feasible to obtain detailed structural information on specific protein conformers in heterogeneous samples. Although this tandem MS strategy has been proposed in the literature,23,41 to our knowledge, its full implementation has not yet been demonstrated. Some earlier top-down HX-MS studies did employ precursor ion selection, but the experiments were conducted on structurally homogeneous samples.42,43 Kaltashov et al. recently employed precursor ion selection for dissociating specific protein subpopulations, but a detailed analysis of fragment HX levels was not performed.23 The current study closes this gap by reporting the selective HX-ECD MS characterization of an aggregated polypeptide species in a structurally heterogeneous sample. The accumulation of amyloid fibrils consisting of Aβ protomers in the human brain represents a hallmark of Alzheimer’s disease. Both Aβ(1
40) and Aβ(1
42) are formed as degradation products of the amyloid precursor protein.45 Although Aβ(1
40) and Aβ(1
42) are generated in a 10:1 ratio, the latter is more toxic due to its higher aggregation and fibrillation propensity.46 HX with NMR detection and mutagenesis has led to a structural model of mature Aβ(1
42) amyloid fibrils47 according to which residues 18
42 in each protomer are arranged in a β-turn-β motif. The orientation of both β-strands is roughly perpendicular to the fibril axis. Hydrogen bonding between the β-strands of individual protomers results in extended sheets on both sides of the fibril. Early studies on the disease mechanism of Alzheimer’s have largely focused on the role of amyloid fibrils. However, recent work has revealed that the direct neurotoxicity of these fibrils is quite low. Instead, soluble Aβ oligomers appear to be the actual culprit responsible for synaptic dysfunction.11,48,49 A similar “oligomer-centric” view seems to emerge for other neurodegenerative disorders, as well.48,50,51 The characterization of Aβ oligomers is complicated by their metastable and heterogeneous nature as well as their sensitivity to solvent conditions. Despite intensive research efforts,11,52
57 no consensus structural model

ARTICLE

of Aβ oligomers has yet emerged. Recent proteolysis-based HXMS measurements on oligomerized Aβ(1
40) provided relatively low spatial resolution58,59 while implying that the structures of those species are likely different from those of Aβ(1
42) aggregates. Here, we employ HX-MS with top-down ECD for exploring structural properties of Aβ(1
42) oligomers in vitro. Our data reveal that oligomeric species coexist with unfolded monomers in solution. Precursor ion selection is used to ensure that only HX isotopologues corresponding to oligomers are subjected to ECD. The resulting HX protection pattern suggests that under the conditions used here, oligomers share certain secondary structure features with fully developed Aβ(1
42) amyloid fibrils.

’ EXPERIMENTAL SECTION Materials. Synthetic human Aβ(1
42) DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVIA47 (MW 4514.1 Da) was obtained from AnaSpec (San Jose, CA) as 1 mg vials of lyophilized powder. Deuterium oxide and ammonium deuteroxide were from Cambridge Isotope Laboratories (Andover, MA). Hexafluoroisopropyl alcohol (HFIP), F12 medium,60 and all other chemicals were purchased from Sigma-Aldrich (St. Louis, MO). Sample Preparation. Aβ(1
42) oligomers were prepared following an established protocol55,61 with minor modifications. Briefly, Aβ(1
42) was dissolved in cold HFIP at 1 mM in a microcentrifuge tube and incubated at room temperature for 1 h to ensure a monomeric structure.62 After HFIP removal by evaporation, the resulting clear peptide film was dissolved in anhydrous dimethyl sulfoxide (DMSO) at 5 mM concentration. To prepare fully deuterated protein, Aβ(1
42) in DMSO was diluted to 2.5 mM in D2O at pH 11.0 (adjusted by ammonium deuteroxide), followed by incubation for 1 h at room temperature. The resulting deuteration level was determined to be >99% by direct infusion using ESI-MS (Q-TOF Ultima, Waters, Milford, MA) with bradykinin as internal control. Oligomer formation was induced by 1:25 dilution of fully deuterated Aβ(1
42) in D2Obased F12 medium at pH 7.8 with vortexing. F12 represents a complex buffer mixture containing high concentrations of salts, amino acids, and other bio-organic compounds.60 After 24 h of aging at 5 °C, the sample was centrifuged at 13000g for 15 min, thereby removing small amounts of insoluble species that might have formed. The supernatant containing soluble Aβ(1
42) oligomers was used for the experiments. Previous work has shown that samples prepared in this way remain fibril-free for days.55,61 All pH values are reported as read, without correction for isotope effects. Sample Characterization by Size Exclusion Chromatography (SEC). F12-aged Aβ(1
42) was fractionated using a Waters Acquity UPLC with a Superdex 75 10/300 GL column (GE Healthcare, Buckinghamshire, UK) and ice cold ammonium acetate (50 mM, pH 8.5) in H2O as mobile phase at 0.5 mL min
1. Both UV absorbance (280 nm) and Q-TOF ESI-MS were employed for detection. For the latter, a three-way splitter was inserted to decrease the flow going into the ion source to 50 μL min
1, and the column outflow was supplemented with 10% formic acid in acetonitrile (50 μL min
1). The SEC column was calibrated with bovine serum albumin (MW 67 kDa), β-lactoglobulin dimer (36.8 kDa), carbonic anhydase (29 kDa), and ubiquitin (8.6 kDa). The void volume was established with blue dextran (∼2 MDa). 5387

dx.doi.org/10.1021/ac200906v |Anal. Chem. 2011, 83, 5386–5393

Analytical Chemistry

ARTICLE

Hydrogen Exchange and MS Analyses. HX was carried out under exchange-out (D f H) conditions. Portions (1.5 μL) of F12-aged and fully deuterated Aβ(1
42) were mixed with 28.5 μL F12 in H2O (5 °C, pH 7.8) for a 95% H2O content in the labeling solvent. After 10 s of exchange-out, HX was acid-quenched by injection onto an equilibrated reversed-phase column (YMC C8, 23 mm  2 mm, Waters). Rapid HPLC elution was conducted using a water/acetonitrile gradient in the presence of 0.1% formic acid (pH ∼ 2.5) delivered by a Waters 1525μ pump at 100 μL min
1. In addition to HX quenching, the HPLC step also served to desalt the sample and to disassemble Aβ(1
42) oligomers into monomers. The column, accessories, injector (Rheodyne model 7125, loop volume 25 μL), and extensively coiled solvent delivery lines were embedded in a 0 °C ice bath to minimize exchange-out during analysis. Aβ(1
42) eluted as a single peak between 5 and 7 min postinjection, regardless of its initial aggregation state (monomer or oligomer) in bulk solution. During elution of Aβ(1
42), the HPLC flow rate was lowered to 10 μL min
1 to maximize the time available for MS data acquisition. Timedependent changes in deuteration level during the elution time window were negligible. Top-down ECD of electrosprayed Aβ(1
42) was conducted on a 12 T Apex-Qe hybrid quadrupole Fourier transform ion cyclotron resonance (ICR) mass spectrometer (Bruker Daltonics, Billerica, MA). The ion optics and parameter settings used have been described previously.40 Approximately 50 scans were accumulated over the m/z range 250
2500 for each ECD spectrum. All ECD data were acquired from front end quadrupole-isolated [M þ 5H]5þ ions, which represent the most abundant species in the spectrum of intact Aβ(1
42). The fwhm of the transmitted isotope distribution window corresponded to roughly 2 m/z units. Mass calibration was performed with intact ions and ECD fragments of bovine ubiquitin. Data Analysis. Neutral Aβ(1
42) contains 68 exchangeable hydrogens, 41 of which are on backbone amides (there is no proline), 24 are on side chains, and 3 are on the termini. N
CR backbone cleavage during ECD occurred at 28 out of 41 possible bonds, generating 28 c (N-terminal) and 16 z (C-terminal) fragment ions63 that were identified for nondeuterated Aβ(1
42) using ProteinProspector (http://prospector.ucsf.edu). Unfortunately, isotope envelope broadening after HX lowers the S/N ratio64 such that a reliable determination of mass shifts was possible only for a subset of ions that encompassed c3
c6, c9
c13, c19, c21, c23, c27, c34, c37, and intact Aβ(1
42). The centroid m/z value (R) for each partially deuterated ECD fragment after 10 s of HX followed by HPLC/MS was determined as

Figure 1. SEC analysis of F12-aged Aβ(1
42). (A) UV chromatogram, peaks are attributable to oligomers and monomers; (B) total ion chromatogram (TIC) obtained by Q-TOF ESI-MS analysis of the eluent; (C) as in panel B, but with postcolumn addition of a denaturing makeup solvent; (D) 4þ region of mass spectra, obtained by integration of monomer (red) and oligomer (blue) TIC segments. Different mass shifts in panel D reflect the HX behavior of the two species under exchange-out conditions. Note that under the conditions of panels C and D, the use of a highly denaturing makeup solvent causes oligomer dissociation into monomers immediately prior to mass analysis.

obtained from ProteinProspector. The number of protected amide sites Nprotected in a given fragment ion is equal to the corrected number of retained deuterium atoms. It can be calculated as 

k

R ¼

∑ ð R i  Ii Þ i¼1 k

∑ Ii

Nprotected ¼ Ntot  ð1Þ

i¼1

where Ri is the m/z value of a given peak in the isotope distribution, Ii is the corresponding peak intensity, and k is the total number of isotope peaks for that fragment ion. To account for deuterium exchange-out during quenching and desalting,15 fully deuterated control samples (corresponding to a zero labeling time prior to HPLC) were analyzed, as well. Centroid m/z values, Rdeuterated control, for these samples were also determined on the basis of eq 1. Centroids of nondeuterated controls (R0) were

R
R0

Rdeuterated control
R0

 ð2Þ

where Ntot is the total number of backbone amide groups in the fragment. Ion cn encompasses the backbone amide sites of residues 2 to n þ 1, keeping in mind that residue 1 carries the free amino terminus that undergoes exchange-out during HPLC. Transformation of these Nprotected data was performed as described40 to obtain the deuteration level D at individual amide linkages. Gaps in the ion progression were filled by inserting the appropriate average values at intervening sites. HX analyses were based on three independent experiments. Error bars correspond to (σ12 þ σ22)1/2, where σ1 and σ2 are the standard deviations of R and Rdeuterated control, respectively. 5388

dx.doi.org/10.1021/ac200906v |Anal. Chem. 2011, 83, 5386–5393

Analytical Chemistry

ARTICLE

Figure 3. Fragment ion spectrum, obtained by subjecting the high-mass portion of the [M þ 5H]5þ peak envelope (Figure 2B) to ECD. [M þ 5H]4þ 3 represents an undissociated charge-reduced ion.

Figure 2. ESI-MS data for Aβ(1
42) in F12 medium after HX, analyzed by HPLC and Fourier transform ICR MS. (A) Entire envelope of the 5þ charge state, which represents the most intense peak in the ESI mass spectrum. (B) Envelope after high-mass precursor ion selection, corresponding to Aβ(1
42) molecules that existed as oligomers during solution-phase HX. Isotope labeling was conducted under exchange-out conditions.

’ RESULTS AND DISCUSSION Size-Exclusion Chromatography. The goal of the current work is to characterize the HX behavior of Aβ(1
42) oligomers by top-down MS, using a solvent environment (F12 medium) where these species can be preserved for extended periods without fibrillation.55,61 Prior to performing gas-phase fragmentation experiments, the properties of the samples used here were characterized by SEC. The purpose of this initial step was 2-fold: (i) It had to be verified that the sample behavior is consistent with earlier investigations.55,61 (ii) Correlating the oligomerization properties of Aβ(1
42) with specific isotope exchange patterns is a prerequisite for the interpretation of top-down MS data. The UV SEC chromatogram of Aβ(1
42) after aging in F12 medium reveals two peaks (Figure 1A). The first peak (14.9 min) elutes close to the void volume. This signal corresponds to Aβ(1
42) oligomers, which are the species of interest for the current work. Previous studies estimated the size of these aggregates to be between ∼2455 and 10061 protomers. The shape of these oligomers has been shown to be spherical, and their cytotoxicity has been demonstrated in biological assays.55,61 A second peak in the SEC elution profile reflects the presence of a less abundant species that migrates with an apparent MW of 11.5 kDa (25.4 min). Although this elution behavior appears to indicate a dimer or trimer, recent investigations point to a monomeric structure.61 The seemingly unusual SEC behavior of this monomeric species can be understood by noting that mass calibration of the column is based on folded globular proteins. Monomeric Aβ(1
42) is extensively disordered52,65 such that an elevated apparent MW under SEC conditions is not surprising. Additional support for the monomeric nature of the low-MW species comes from native ESI-MS, as well as its lack of HX protection (see below).

Direct analyses of the SEC effluent by ESI-MS reveals a monomer signal at 25.4 min, whereas no signal is observed for the oligomeric species at 14.9 min (Figure 1B). It is well-known that some protein analytes exhibit an exceedingly low ESI-MS response under certain conditions.66 In the present case, possible reasons for the lack of MS signal at 14.9 min include oligomer heterogeneity and poor desolvation behavior. This problem can be overcome by postcolumn supplementation of the SEC eluent with a highly denaturing and “ESI friendly” makeup solution (10% formic acid in acetonitrile). Under these conditions, both elution peaks are observable in the ESI-MS chromatogram (Figure 1C). The addition of a denaturing makeup solvent to the SEC outflow causes solution-phase dissociation of oligomers after separation. As a result, all Aβ(1
42) molecules are detected as monomers, regardless of their original oligomerization behavior in bulk solution and on the column (Figure 1D). As outlined in the Experimental Section, the Aβ(1
42) samples used for this work were initially prepared in the fully deuterated form. Hence, exchange-out takes place in the H2O-based SEC mobile phase. The chromatographic conditions of Figure 1 therefore provide a first opportunity to examine the HX properties of Aβ(1
42) monomers and oligomers. Complete deuterium loss occurs for the monomeric species, consistent with a disordered structure52,65 (Figure 1D, low mass envelope). In contrast, oligomers retain about nine deuterium atoms, attesting to significant protection (Figure 1D, high-mass envelope). The SEC elution profile of the samples used in this work (Figure 1A) is in agreement with earlier investigations.55,61 Our data reveal that Aβ(1
42) monomers and oligomers exhibit dramatic differences in their isotope exchange properties. Significant deuterium retention is observed only for the latter. The HX data of Figure 1A are indicative of a monomer
oligomer equilibrium, reminiscent of the behavior reported for mature Aβ amyloid fibrils.67 HPLC
MS Analysis of Aβ(1
42) After HX. ESI-MS signal intensities after SEC (Figure 1D) were quite low, thereby precluding meaningful top-down measurements. The experiments described in the subsequent sections therefore followed a different HX strategy, in which fully deuterated F12-aged Aβ(1
42) was exposed to H2O in bulk solution at pH 7.8. After 10 s of exchange-out, the samples were acid-quenched and desalted by HPLC. The 10 s labeling interval ensures complete deuterium 5389

dx.doi.org/10.1021/ac200906v |Anal. Chem. 2011, 83, 5386–5393

Analytical Chemistry loss at unprotected amides, whereas deuterium is retained at hydrogen-bonded sites.68 The denaturing HPLC mobile phase ensures oligomer dissociation, causing all Aβ(1
42) in the sample to elute in a single monomer peak after ∼6 min. Importantly, HX information on the structure and interactions in bulk solution is retained in the backbone deuteration pattern of the electrosprayed gas-phase ions. Mass analysis of intact Aβ(1
42) by Fourier transform ICR MS after HX and HPLC reveals a bimodal distribution (Figure 2A). On the basis of our SEC data (Figure 1D), an interpretation of the mass distribution in Figure 2A is straightforward. Accordingly, the low-mass component (around m/z 904.5) can be assigned to disordered monomers that have experienced complete deuterium loss during exchange-out. In contrast, the high-mass component (around m/z 907) originates from oligomers that exhibit significant deuterium retention. The HPLC strategy of Figure 2 leads to greatly improved ESI-MS signal intensities when compared with the SEC approach (Figure 1). In addition, the amide

Figure 4. Fragment ions detected after ECD of precursor ion-selected Aβ(1
42). Only fragments that are indicated in bold were used for HX data analysis.

ARTICLE

deuteration level of Aβ(1
42) oligomers in Figure 2A is increased by a factor of 2. Conformer-Specific Top-Down MS. ECD measurements were conducted by employing quadrupole selection for transmitting only the highly deuterated (oligomer) portion of the [M þ 5H]5þ peak envelope into the ICR cell (Figure 2B). An overview of the complete fragment ion spectrum obtained under these conditions is depicted in Figure 3. ECD resulted in 28 c and 16 z fragment ions, corresponding to a coverage that is comparable to earlier ECD investigation on Aβ.69 Only a subset of these fragment ions, however, was suitable for HX analyses, as noted in the Experimental Section. Nonetheless, the spatial resolution of our top-down HX data (Figure 4) remains significantly higher than in earlier HX studies that employed proteolytic peptide mapping.58,59 We reiterate that the fragment ion deuteration levels in Figure 3 exclusively reflect the HX properties of Aβ(1
42) oligomers, and contributions from monomers that coexisted in solution are eliminated by precursor ion selection. Figure 5 provides a close-up view of selected top-down fragment ion signals. Also included in Figure 5 are data for a nondeuterated control and for a sample that had undergone complete deuteration prior to HPLC. The number of deuterium atoms retained in each protein segment, Nprotected, can be determined from analyses of fragment ion mass shifts (Figure 6). The strategy used for this purpose has to take into account deuterium exchange-out after the actual HX step, a factor that is unavoidable when employing HPLC desalting (eq 2).15 Transformation40 of the Nprotected progression

Figure 5. Close-up views of selected ion signals, obtained by subjecting the 5þ charge state of Aβ(1
42) to ECD after HPLC desalting. First row, nondeuterated control; second row, after precursor ion selection of the high mass envelope following 10 s of HX (oligomers, Figure 2B); third row, fully deuterated control. Data are shown for three different fragment ions (columns 1
3) and for intact Aβ(1
42) (last column). 5390

dx.doi.org/10.1021/ac200906v |Anal. Chem. 2011, 83, 5386–5393

Analytical Chemistry

Figure 6. Total number of deuterium atoms retained in ECD fragments of Aβ(1
42) oligomers after 10 s of HX and HPLC desalting. The dotted line with a slope of unity represents the hypothetical case that every backbone amide completely retains its deuterium. In the experimental data, regions with a zero slope represent segments that are unprotected. Partial protection is reflected by a slope between zero and unity. The last data point, denoted as “c4100 , represents Nprotected for intact Aβ(1
42).

yields the deuteration level D at amide sites along the backbone (Figure 7). D = 1 corresponds to sites that are completely protected, such that all deuterium is retained. Fractional D values represent partial protection, whereas amide groups that exhibit a D value of zero do not offer any protection. The latter case applies to the low-mass (monomer) component of Figure 2A, such that an ECD analysis of that species is not meaningful. Structural Implications for Aβ(1
42) Oligomers. All of the backbone amides in oligomeric Aβ(1
42) show partial deuterium loss after the 10 s HX period (Figure 7). When comparing this behavior with data previously obtained for natively folded proteins,40 it can be concluded that the hydrogen bonding network in these oligomers must be quite dynamic, heterogeneous, or both. The N-terminal residues 2
14 exhibit a protection level that is quite low, with D values around 0.3. This is followed by a segment that shows substantially higher deuterium retention (residues 15
24). The next four residues are completely unprotected (25
28, D ≈ 0), whereas the C-terminal segment 29
42 again shows relatively high deuteration. It seems likely that much of the HX protection seen in Figure 7 stems from intermolecular hydrogen bonds that mediate cohesive interactions within Aβ(1
42) oligomers. Earlier studies suggested that the backbone organization in Aβ(1
42) oligomers is β-sheet-rich, sharing certain similarities with fully developed amyloid fibrils.52,53,56,57 It is noted, however, that those earlier investigations employed conditions that were different from those used here. Mature Aβ(1
42) fibrils have residues 18
26 and 31
42 arranged in β-strands. Intermolecular hydrogen bonding leads to extended sheets. The two β-strands in each fibril protomer are linked by a flexible turn (residues 27
30).47 This fibril arrangement largely mirrors the protection pattern seen here for oligomers (Figure 7). In particular, we observe two partially protected regions (15
24 and 29
42)

ARTICLE

Figure 7. Deuteration level D of backbone amides in Aβ(1
42) oligomers after 10 s of HX under exchange-out conditions. Shown along the bottom is the polypeptide sequence. Oligomer segments that form relatively stable hydrogen bonds, as indicated by their D values, are highlighted in solid red. The N-terminal region is weakly hydrogenbonded (red/white hatch pattern). Also shown are the locations hydrogen bonded segments in mature amyloid fibrils, based on work by others.47

that are separated by an unprotected segment (25
28). This HX pattern strongly suggests that residues 15
42 form a β-loop-β motif in our Aβ(1
42) oligomers, resembling the secondary structure of mature fibrils (red and blue elements in Figure 7).47 Hence, the observations of this work are consistent with the view52,53,56,57 that some aspects of the backbone arrangement are similar in Aβ(1
42) oligomers and amyloid fibrils. Not all features of the oligomer protection pattern, however, are consistent with a fibril-like architecture. The N-terminal 17 residues are extensively disordered in mature fibrils.47 In contrast, our oligomer data reveal nonzero protection for residues 2
14, which indicates a certain degree of hydrogen bonding in this region (D ≈ 0.3, red/white hatch pattern in Figure 7). This latter finding may suggest a role of the N-terminal segment for intermolecular contacts in Aβ(1
42) oligomers, a possibility that has received little attention in the previous literature. A notable feature of the oligomer deuteration data (Figure 7) is the strong protection of the C-terminal residues 41 and 42. In mature Aβ(1
42) fibrils, these two amino acids form intermolecular hydrogen bonds.47 Although HX-MS cannot distinguish between inter- and intramolecular bonds, our observations suggest that also in the oligomeric state 41 and 42 form interprotomer contacts. Aβ(1
40) is missing these two residues, such that the maximum number of intermolecular bonds is lower. The capability to form two additional hydrogen bond linkages with other protomers provides a simple explanation for the greater aggregation propensity of Aβ(1
42) relative to Aβ(1
40).70

’ CONCLUSIONS This work has demonstrated the implementation of a combined solution-phase/gas-phase approach for interrogating the hydrogen bonding characteristics of specific polypeptide conformers in structurally heterogeneous samples. In contrast to complementary MS-based approaches such as ion mobility spectrometry,11,12 the strategy used here does not rely on the preservation of protein structural features upon transfer from bulk solution into the gas phase. On the contrary, the current work employs acid quenching after isotope labeling, thereby ensuring oligomer dissociation prior to mass analysis. Nonetheless, information regarding the hydrogen bonding behavior is 5391

dx.doi.org/10.1021/ac200906v |Anal. Chem. 2011, 83, 5386–5393

Analytical Chemistry retained in the amide deuteration pattern along the polypeptide backbone. The detection of coexisting conformers by HX-MS, based on intact-protein signals with distinct mass shifts, is already well established.18
22 The novel aspect introduced here is the use of tandem MS for fragmenting only precursor ions that correspond to one selected type of conformer. As a result, structural data can be obtained with conformational specificity. The system studied in this work represents a relatively straightforward case in which the oligomeric species of interest coexists with only one other type of conformer. In addition, the small size of Aβ(1
42) makes this species an easy target for top-down MS. Electron-based topdown HX studies on larger proteins have already been demonstrated,40,42,43 albeit without conformer-specific precursor ion selection. With currently available mass analyzers, the practical size limit of top-down HX-MS is likely in the range of 20 kDa. Our data strongly suggest that the Aβ(1
42) backbone in F12-aged oligomers adopts a β-turn-β secondary structure, resembling the arrangement seen in mature fibrils (Figure 7).47 However, it remains to be established whether the aggregates studied here are representative of cytotoxic species formed in the human brain.55,61 Recent work indicates that assemblies much smaller than those considered here can also exhibit considerable cytotoxicity.11,71 In the future, it will be interesting to extend the approach employed here to different types of Aβ assemblies for gaining a better understanding of the factors that determine their aggregation behavior. In addition, it should be possible to apply precursor ion selection to more complex protein systems that are larger and that exhibit multimodal HX distributions.18,19 Applications can be envisioned for both equilibrium studies and kinetic experiments.

’ AUTHOR INFORMATION Corresponding Author

*Phone: (519) 661-2111, ext. 86313. Fax: (519) 661-3022. E-mail: [email protected]. Web: http://publish.uwo.ca/∼konerman.

’ ACKNOWLEDGMENT This work was supported by the Natural Sciences and Engineering Council of Canada, the Canada Foundation for Innovation, Genome Canada, Genome BC, and the Canada Research Chairs Program. ’ REFERENCES (1) Englander, S. W.; Mayne, L.; Krishna, M. M. G. Q. Rev. Biophys. 2008, 40, 287–326. (2) Dobson, C. M. Nature 2003, 426, 884–890. (3) Baldwin, A. J.; Kay, L. E. Nat. Chem. Biol. 2009, 5, 808–814. (4) Grandori, R. Protein Sci. 2002, 11, 453–458. (5) Kaltashov, I. A.; Eyles, S. J. Mass Spectrom. Rev. 2002, 21, 37–71. (6) Wang, W.; Kitova, E. N.; Klassen, J. S. Anal. Chem. 2003, 75, 4945–4955. (7) Daniel, J. M.; Friess, S. D.; Rajagopalan, S.; Wendt, S.; Zenobi, R. Int. J. Mass Spectrom. 2002, 216, 1–27. (8) Heck, A. J. R. Nat. Methods 2008, 5, 927–933. (9) Wang, G.; Abzalimov, R. R.; Kaltashov, I. A. Anal. Chem. 2011, 83, 2870–2876. (10) Smith, A. M.; Jahn, T. R.; Ashcroft, A. E.; Radford, S. E. J. Mol. Biol. 2006, 364, 9–19.

ARTICLE

(11) Bernstein, S. L.; Dupuis, N. F.; Lazo, N. D.; Wyttenbach, T.; Condron, M. M.; Bitan, G.; Teplow, D. B.; Shea, J.-E.; Ruotolo, B. T.; Robinson, C. V.; Bowers, M. T. Nat. Chem. 2009, 1, 326–331. (12) Ashcroft, A. E. J. Am. Soc. Mass Spectrom. 2010, 21, 1087–1096. (13) Koeniger, S. L.; Clemmer, D. E. J. Am. Soc. Mass Spectrom. 2007, 18, 322–331. (14) Chetty, P. S.; Mayne, L.; Lund-Katz, S.; Stranz, D. D.; Englander, S. W.; Phillips, M. C. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 19005–19010. (15) Wales, T. E.; Engen, J. R. Mass Spectrom. Rev. 2006, 25, 158–170. (16) Chalmers, M. J.; Busby, S. A.; Pascal, B. D.; He, Y.; Hendrickson, C. L.; Marshall, A. G.; Griffin, P. R. Anal. Chem. 2006, 78, 1005–1014. (17) Del Mar, C.; Greenbaum, E. A.; Mayne, L.; Englander, S. W.; Woods, V. L. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 15477–15482. (18) Miranker, A.; Robinson, C. V.; Radford, S. E.; Aplin, R.; Dobson, C. M. Science 1993, 262, 896–900. (19) Heidary, D. K.; Gross, L. A.; Roy, M.; Jennings, P. A. Nat. Struct. Biol. 1997, 4, 725–731. (20) Tsui, V.; Garcia, C.; Cavagnero, S.; Siuzdak, G.; Dyson, H. J.; Wright, P. E. Protein Sci. 1999, 8, 45–49. (21) Xiao, H.; Hoerner, J. K.; Eyles, S. J.; Dobo, A.; Voigtman, E.; Melcuk, A. I.; Kaltashov, I. A. Protein Sci. 2005, 14, 543–557. (22) Hoerner, J. K.; Xiao, H.; Kaltashov, I. A. Biochemistry 2005, 44, 11286–11294. (23) Kaltashov, I. A.; Bobst, C. E.; Abzalimov, R. R. Anal. Chem. 2009, 81, 7892–7899. (24) Siuti, N.; Kelleher, N. L. Nat. Methods 2007, 4, 817–821. (25) Han, X.; Jin, M.; Breuker, K.; McLafferty, F. W. Science 2006, 314, 109–112. (26) Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem. Soc. 1998, 120, 3265–3266. (27) Liu, J.; Huang, T.-Y.; McLuckey, S. A. Anal. Chem. 2009, 81, 1433–1441. (28) Coon, J. J. Anal. Chem. 2009, 81, 3208–3215. (29) Rand, K. D.; Zehl, M.; Jensen, O. N.; Jørgensen, T. J. Anal. Chem. 2010, 82, 9755–9762. (30) Rand, K. D.; Zehl, M.; Jensen, O. N.; Jørgensen, T. J. D. Anal. Chem. 2009, 81, 5577–5584. (31) Hoerner, J. K.; Xiao, H.; Dobo, A.; Kaltashov, I. A. J. Am. Chem. Soc. 2004, 126, 7709–7717. (32) Demmers, J. A. A.; Rijkers, D. T. S.; Haverkamp, J.; Killian, J. A.; Heck, A. J. R. J. Am. Chem. Soc. 2002, 124, 11191–11198.  (33) Jørgensen, T. J. D.; Gardsvoll, H.; Ploug, M.; Roepstorff, P. J. Am. Chem. Soc. 2005, 127, 2785–2793. (34) Zubarev, R. A.; Zubarev, A. R.; Savitski, M. M. J. Am. Soc. Mass Spectrom. 2008, 19, 753–761. (35) Syka, J. E. P.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.; Hunt, D. F. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9528–9533. (36) Sohn, C. H.; Chung, C. K.; Yin, S.; Ramachandran, P.; Loo, J. A.; Beauchamp, J. L. J. Am. Chem. Soc. 2009, 131, 5444–5459. (37) Syrstad, E. A.; Turecek, F. J. Am. Soc. Mass Spectrom. 2005, 16, 208–224. (38) Zehl, M.; Rand, K. D.; Jensen, O. N.; Jørgensen, T. J. D. J. Am. Chem. Soc. 2008, 130, 17453–17459. (39) Rand, K. D.; Adams, C. M.; Zubarev, R. A.; Jørgensen, T. J. D. J. Am. Chem. Soc. 2008, 130, 1341–1349. (40) Pan, J.; Han, J.; Borchers, C. H.; Konermann, L. J. Am. Chem. Soc. 2009, 131, 12801–12808. (41) Pan, J.; Han, J.; Borchers, C. H.; Konermann, L. Anal. Chem. 2010, 82, 8591–8597. (42) Abzalimov, R. R.; Kaplan, D. A.; Easterling, M. L.; Kaltashov, I. A. J. Am. Soc. Mass Spectrom. 2009, 20, 1514–1517. (43) Sterling, H. J.; Williams, E. R. Anal. Chem. 2010, 82, 9050–9057. (44) Stefanowicz, P.; Petry-Podgorska, I.; Kowalewska, K.; Jaremko, L.; Jaremko, M.; Szewczuk, Z. Biosci. Rep. 2009, 30, 91–99. (45) Aguzzi, A.; O’Connor, T. Nat. Rev. Drug Discovery 2010, 9, 237–248. 5392

dx.doi.org/10.1021/ac200906v |Anal. Chem. 2011, 83, 5386–5393

Analytical Chemistry

ARTICLE

(46) Teplow, D. B.; Lazo, N. D.; Bitan, G.; Bernstein, S.; Wyttenbach, T.; Bowers, M. T.; Baumketner, A.; Shea, J. E.; Urbanc, B.; Cruz, L.; Borreguero, J.; Stanley, H. E. Acc. Chem. Res. 2006, 39, 635–645. (47) L€uhrs, T.; Ritter, C.; Adrian, M.; Riek-Loher, D.; Bohrmann, B.; Dobeli, H.; Schubert, D.; Riek, R. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 17342–17347. (48) Haass, C.; Selkoe, D. J. Nat. Rev. Mol. Cell Biol. 2007, 8, 101–112. (49) Koffie, R. M.; Meyer-Luehmann, M.; Hashimoto, T.; Adams, K. W.; Mielke, M. L.; Garcia-Alloza, M.; Micheva, K. D.; Smith, S. J.; Kim, M. L.; Lee, V. M.; Hyman, B. T.; Spires-Jones, T. L. Proc. Natl. Acad. Sci. 2009, 106, 4012–4017. (50) Campioni, S.; Mannini, B.; Zampagni, M.; Pensalfini, A.; Parrini, C.; Evangelisti, E.; Relini, A.; Stefani, M.; Dobson, C. M.; Cecchi, C.; Chiti, F. Nat. Chem. Biol. 2010, 6, 140–147. (51) Collinge, J.; Clarke, A. R. Science 2007, 318, 930–936. (52) Ono, K.; Condron, M. M.; Teplow, D. B. Proc. Nat. Acad. Sci. U.S.A. 2009, 106, 14745–14750. (53) Chimon, S.; Shaibat, M. A.; Jones, C. R.; Calero, D. C.; Aizezi, B.; Ishii, Y. Nat. Struct. Mol. Biol. 2007, 14, 1157–1164. (54) Cerf, E.; Sarroukh, R.; Tamamizu-Kato, S.; Breydo, L.; Derclaye, S.; Dufr^ene, Y. F.; Narayanaswami, V.; Goormaghtigh, E.; Ruysschaert, J. M.; Raussens, V. Biochem. J. 2009, 421, 415–423. (55) Chromy, B. A.; Nowak, R. J.; Lambert, M. P.; Viola, K. L.; Chang, L.; Velasco, P. T.; Jones, B. W.; Fernandez, S. J.; Lacor, P. N.; Horowitz, P.; Finch, C. E.; Krafft, G. A.; Klein, W. L. Biochemistry 2003, 42, 12749–12760. (56) Yu, L.; Edalji, R.; Harlan, J. E.; Holzman, T. F.; Lopez, A. P.; Labkovsky, B.; Hillen, H.; Barghorn, S.; Ebert, U.; Richardson, P. L.; Miesbauer, L.; Solomon, L.; Bartley, D.; Walter, K.; Johnson, R. W.; Hajduk, P. J.; Olejniczak, E. T. Biochemistry 2009, 48, 1870–1877. (57) Shafrir, Y.; Durell, S. R.; Anishkin, A.; Guy, H. R. Proteins 2010, 78, 3458–3472. (58) Zhang, A.; Qi, W.; Good, T. A.; Fermandez, E. J. Biophys. J. 2009, 96, 1091–1104. (59) Kheterpal, I.; Chen, M.; Cook, K. D.; Wetzel, R. J. Mol. Biol. 2006, 361, 785–795. (60) Ham, R. G. Proc. Natl. Acad. Sci. U. S. A. 1965, 53, 288–293. (61) Lauren, J.; Gimbel, D. A.; Nygaard, H. B.; Gilbert, J. W.; Strittmatter, S. M. Nature 2009, 457, 1128–1132. (62) Stine, W. B., Jr.; Dahlgren, K. N.; Krafft, G. A.; LaDu, M. J. J. Biol. Chem. 2003, 278, 11612–11622. (63) Kruger, N. A.; Zubarev, R. A.; Horn, D. M.; McLafferty, F. W. Int. J. Mass Spectrom. 1999, 185
187, 787–793. (64) Slysz, G. W.; Percy, A. J.; Schriemer, D. C. Anal. Chem. 2008, 80, 7004–7011. (65) Baumketner, A.; Bernstein, S. L.; Wyttenbach, T.; Bitan, G.; Teplow, D. B.; Bowers, M. T.; Shea, J. E. Protein Sci. 2006, 15, 420–428. (66) Peschke, M.; Verkerk, U. H.; Kebarle, P. J. Am. Soc. Mass Spectrom. 2004, 15, 1424–1434. (67) Sanchez, L.; Madurga, S.; Pukala, T.; Vilaseca, M.; LopezIglesias, C.; Robinson, C. V.; Giralt, E.; Carulla, N. J. Am. Chem. Soc. 2011, 133, 6505–6508. (68) Bai, Y.; Milne, J. S.; Mayne, L.; Englander, S. W. Proteins: Struct. Funct. Genet. 1993, 17, 75–86. (69) Sargaeva, N. P.; Lin, C.; O’Connor, P. B. Anal. Chem. 2009, 81, 9778–9786. (70) Kim, W.; Hecht, M. H. J. Biol. Chem. 2005, 280, 35069–35076. (71) Shankar, G. M.; Li, S.; Mehta, T. H.; Garcia-Munoz, A.; Shepardson, N. E.; Smith, I.; Brett, F. M.; Farrell, M. A.; Rowan, M. J.; Lemere, C. A.; Regan, C. M.; Walsh, D. M.; Sabatini, B. L.; Selkoe, D. J. Nat. Med. 2008, 14.

5393

dx.doi.org/10.1021/ac200906v |Anal. Chem. 2011, 83, 5386–5393