Factors influencing compact-extended structure equilibrium in oligomers of Aβ1–40
peptide – an ion mobility mass spectrometry study.
Ewa Sitkiewicz1,*, Marcin Kłoniecki1,*, Jarosław Poznański1, Wojciech Bal1 & Michał
Dadlez1,2,&
Institute of Biochemistry and Biophysics, Pol. Acad. Sci. Pawińskiego 5A, 02-106 Warszawa,
1
Poland
2
Institute of Genetics and Biotechnology, Biology Department, Warsaw University, Miecznikowa 1,
02-185 Warszawa, Poland
1,*
Ewa Sitkiewicz: ewa@ibb.waw.pl
Marcin Kłoniecki: marcinkloniecki@wp.pl
1,*
Jarosław Poznański: jarek@ibb.waw.pl
1
1
Wojciech Bal: wbal@ibb.waw.pl
Michał Dadlez: michald@ibb.waw.pl
1, 2, &
*
These authors contributed equally to this work.
&
Corresponding author: Michał Dadlez, Pawinskiego 5A, 02-106 Warszawa,
michald@ibb.waw.pl, tel. +48 22/5923471, fax.+48 22/8237194
Running Title: Compact-extended structure of Aβ1–40 oligomers.
1
2016-02-16
Abstract
Oligomers formed by amyloid beta peptide (Aβ) are widely believed to be the main
neurotoxic agent in Alzheimer’s disease. Studies discovered a broad variety of oligomeric
forms, which display different levels of toxicity. Some of these forms may further assemble
into mature fibrils, while other might be off-pathway from conversion to fibrils and assemble
into alternative forms. To better understand a relationship between the structure and toxicity
of Aβ oligomers, systematic characterization and classification of all possible forms is
required, facilitating rational design of the beneficial modifiers of their activity. In previous
ion-mobility analysis of Aβ1–40 oligomers, we have detected the coexistence of two
alternative structural forms (compact and extended) in a pool of low-order Aβ1–40 oligomers.
These forms may represent two pathways of the oligomer evolution, either leading to fibrils or
to off-pathway oligomers, potential candidates for the neurotoxic species. Here, we have
analyzed the impact of incubation time, the presence of selected metal ions and the effect of a
series of point mutations on mutual population of alternative forms. We have shown that a salt
bridge D23K28 provides stabilization of the compact form whereas G25 is required for the
existence of the extended form. We have found that binding of metal ions also stabilizes the
compact form. These results improve our understanding of the possible molecular mechanism
of the bifurcation of structural evolution of non-monomeric Aβ species into an off-fibril
pathway, ultimately leading to the formation of potentially neurotoxic species.
Keywords: Alzheimer’s disease; Aβ1–40 peptide; oligomer structure; ion mobility separation
mass spectrometry; collisional cross-section
Abbreviations used:
Aβ, amyloid β peptide; AD, Alzheimer's disease; IMS, ion mobility separation; MS, mass
spectrometry; Ω, collisional cross-section; tD, drift time; MON, monomer; DIM, dimer; TRI,
trimer; TET, tetramer; PEN, pentamer
2
2016-02-16
Introduction
Causative link between the non-monomeric forms of Aβ peptide (a product of APP protein’s
proteolytic cleavage) and Alzheimer’s disease is supported by several lines of evidence,1
providing basis for amyloid cascade hypothesis.2 This theory has recently evolved, giving
more recognition to soluble, non-monomeric pool of Aβ as a causative factor in AD, while
previously more significance was attributed to the insoluble aggregated peptide forms.3 It was
found that the correlation between the level of soluble oligomers and decreased cognitive
performance is stronger than the one between aggregated Aβ burden and cognitive decline.4,5
Subsequent experiments performed in wide variety of setups, both with synthetic oligomers
and samples extracted from the tissue of AD patients6–8, confirmed the neuro- and
synaptotoxic properties of soluble Aβ forms. The “oligomer hypothesis” is a key theory in
current AD-related research and was frequently reviewed in recent years. 9–12 Aβ oligomeric
assemblies ranging from dimers to high molecular weight (HMW) aggregates9 have been
identified both endogenously and in in vitro preparations. No consensus has been reached yet
on the identity of the major neurotoxic species. Thus, a systematic definition, classification
and characterization of all species that can be formed by oligomeric Aβ is necessary.9,13
The problem becomes more complex if we take under consideration the interaction of
Aβ peptide with metal ions. The formation of Aβ1–40 aggregates is accelerated by the
presence of the Cu(II) or Zn(II) ions by ca. four orders of magnitude – from hours to
seconds.14,15 These aggregates, while also fibrillar, are morphologically different from those
formed in the absence of metal ions. Cu(II), Zn(II) and Fe(II) ions associated with Aβ are also
components of senile plaques. However, little is known about the mechanism(s) according to
which metal ions affect aggregation and whether they include oligomer formation. The
Cu(II)/Aβ interactions have been studied in detail for the monomeric peptide.16–20 The binding
residues are recruited from the N-terminal part of the peptide. In the physiologically relevant
range of pH (6.9-7.8) Aβ forms a mixture of complexes, which contain a Cu(II) ion bonded to
three nitrogen (N-terminus, His-6 and either His-13 or His-14) and one or two oxygen donors.
The latest dissociation constant obtained for this complex at pH 7.4 is 36 pM for the Aβ1–40
peptide.21 On the basis of EPR spectra it was concluded that the Cu(II) binding mode in the
aggregates is very similar to that in the monomeric peptide.22 There is also indirect evidence
for the formation of minute amounts of Cu(II) complex with two Aβ molecules in this pH
range, which may be considered as dimeric species. No structural data are available for this
putative complex.23 The pattern of Zn(II) coordination is thought to be similar to that of
Cu(II), although the Zn(II) binding is much weaker.16,24 Nevertheless, some structural
3
2016-02-16
differences between complexes of these two metal ions must be responsible for a much lower
solubility/faster aggregation, seen for Zn(II) with respect to Cu(II).
It is a general feature of peptide/protein amyloid formation process that the fibril
assembly undergoes a cycle of lag phase followed by nucleation and subsequent fast fibril
growth phase.25–29 Lag-phase is caused by the process of a self-assembly of a “nucleus” from
which fibrils can elongate fast. This leads to a characteristic sigmoidal curve of fibril growth
with no fibrils present during lag phase followed by a spontaneous fast increase in fibril
numbers during growth phase. This implies that an obligatory oligomeric nucleus is a
transient, high free energy, low probability form. Its population is thus expected to be low,
unlikely to be detected at all times of fibrilization process, similar to protein folding
intermediates. On the other hand a stunning variety of oligomeric forms was detected in Aβ
preparations by different methods (SEC, DLS, analytical ultracentrifugation, electron
microscopy, AFM, PICUP, etc.)30, also in the absence of fibrils. These forms may represent
pre-nucleation species which are on-pathway to fibrils, but also off-pathway oligomeric forms
which do not evolve into fibrils.31 Indeed, recent work indicates the presence of alternative
pathways of aggregation32–34 and suggests that neurotoxic oligomers can also be populated via
alternate pathways.35,36 Since the mature fibrils do not seem to represent the major neurotoxic
form, off-pathway oligomers gain special attention, as the potential candidates for the main
neurotoxic agent or agents.30,36 In the evolution pathway from monomeric units to high
molecular weight fibrillar assemblies several structurally different oligomeric forms coexist in
solution. Oligomers of both classes may retain a dynamic equilibrium but it is also likely that
with increasing structural complexity an interconversion of species between classes becomes
less probable, and in the end their final form is defined.
The studies of such a complex set of structures require tools which allow to separate
signals of different oligomeric forms coexisting in solution. Recently, a limited spectrum of
methods available (gel filtration, light scattering, atomic force microscopy, analytical
ultracentrifugation) has been expanded by ion mobility separation (IMS) coupled with mass
spectrometry (IMS-MS). It allows to resolve species not only according to their molecular
mass but also according to their collisional cross section () (characteristic for each resolved
form). In such a way basic structural characterization of separate species can be obtained in
spite of the complexity of the starting mixture. This allows to avoid averaging of the structural
information over a set of interconverting or co-existing states as characteristic for other, more
classic techniques (DLS, fluorescence, CD, etc.). NMR also leads to averaging of the signal of
species interconverting during the time of experiment, which is in this case much longer than
4
2016-02-16
MS measurement, though recently 19F NMR allowed to detect six coexisting oligomeric
forms in Aβ preparations.37 IMS-MS, thought relatively new, became an established
analytical tool especially for peptide/protein aggregation studies.38 In our previous work39,
using IMS-MS we have detected structurally alternate forms of Aβ oligomers. For a given nmeric oligomer IMS drift time profiles revealed a distribution of signals of different drift
times (i.e. different collisional cross-section and thus different structure), indicating the
presence of more compact and more extended alternate forms. The signals were well
separated and relatively narrow, indicating that the detected forms have a well-defined three
dimensional structure and do not interconvert within the time of the experiment. We have also
constructed molecular models of compact and extended forms that best account for the
measured structural constraints.
Structural conversion between compact and extended oligomers may be critical for the
development of the disease directing the Aβ monomers either to relatively benign fibrils or to
more aggressive off-pathway species. Understanding the factors that influence the relative
population of structural forms of oligomers seems thus to be crucial for understanding the
disease and modifying its progression in a therapeutically beneficial way. Small molecules
selectively stabilizing the non-neurotoxic pathway of oligomer evolution (or destabilizing the
intermediates on the neurotoxic pathway) may be tested by measuring IMS profiles in their
absence and presence and provide the tool to select the most efficient drugs.38 Therefore, it is
of importance to elucidate factors influencing the mutual population of structural forms of
both classes, more compact and more extended. Here, we have studied the effects of
incubation, metal binding and point mutations on relative population of compact/extended
forms of Aβ oligomers.
Results
We have carried out the measurements of IMS-MS drift time profiles of Aβ1–40
oligomeric forms obtained from in vitro preparations and their changes upon a) incubation, b)
metal binding, c) mutation. Drift time profiles are extracted from two dimensional IMS-MS
spectra in which signals corresponding to different oligomeric forms observed in a typical MS
spectrum of Aβ1–40 (Figure 1) are additionally resolved in the domain of an ion-mobility
drift time. These profiles characterize the distribution of oligomeric forms of different
collisional cross-section in a given m/z range.
5
2016-02-16
First, the dependence of the drift time profiles on the incubation time was tested. Aβ1–
40 was incubated for different times prior to Ion Mobility MS experiment. The distribution of
various oligomeric forms in the domain of drift times was measured at different times of
incubation, as long as 24h. Some models of amyloid formation pathways40,41 assume that the
nucleation step requires a slow structural conversion of monomers or low-order oligomers.
Such structural change in oligomers would be expected to change their collisional crosssection and shift the position of peaks in the domain of the drift time which could be detected
during IMS-MS experiment. This is however not the case. The drift time profiles remain very
well reproducible irrelevant on the length of incubation. The only difference observed upon
prolonged incubation is the gradual decrease of the heights of all detected signals. Also, the
drift time profiles do not differ between batches of Aβ1–40 provided that all other experiment
parameters, including the spectrometer setup, remain unchanged. In conclusion, incubation
and reproducibility experiments showed that low-order oligomers do not undergo long-time
structural changes, any changes in the drift time profiles observed in subsequent experiments
thus cannot be attributed to incubation.
However, the drift time profiles of the Aβ1–40 peptide change when metal ions are
added to solution. Two metal ions were studied, namely Cu(II) and Zn(II), added in molar
ratio of 1:0.5 and 1:2, respectively. Figure 2 illustrates drift time profiles of selected
oligomeric forms and charge states. The assignment of compact and extended form to signals
in drift time domain was carried out in previous work.39 The shift towards compact state is
significant for metal-bound forms of all low order oligomers, except dimers for which the
changes are small. For DIM5+ signal (Figure 2a) a small decrease in the fraction of extended
form was observed, both for Cu(II) and Zn(II) ions for signal groups corresponding to the
form containing one metal ion per monomer (OMPM form), whereas in the signal group
containing one metal ion per oligomer (OMPO form) no significant difference with control
was observed. For TRI7+ (Figure 2b) a signal corresponding to the compact state, absent in
apo-Aβ1–40 control, appears, stronger for Cu(II) than for Zn(II). For TRI6+ signals in control,
both compact and extended forms (Figure 2c) are present in equal proportion (with drift times
of 9.5 ms and 12.4 ms, respectively), and the profile contains also signals for DIM4+ (10.9 ms)
and MON2+ (15.5 ms), as they have the same m/z for apo-Aβ1–40. Compact form becomes
dominating upon addition of metal ions in proportion 1:0.5 of peptide/metal in OMPO form
(middle panels). The effect seems to be stronger for Cu than for Zn. For OMPM form (lowest
panel) the extended form becomes barely detectable. The monomeric and dimeric signals
visible in these panels come from their sodium and potassium adducts. The pattern recurs for
6
2016-02-16
TET7+ (Figure 2d) and PEN8+ (Figure 2e) where the domination of extended forms in control
sample is replaced by the prevalence of compact form in the presence of metal ions. The
impact of Cu(II) is more pronounced than for Zn(II), especially for tetramers and pentamers.
For OMPM Cu/peptide complexes (lowest panels in Figure 2d, e) even for lower proportion
of metal added (peptide/metal 1:0.5) the extended form is absent, whereas for OMPO form
(panels 2 in Figure 2d, e) it is barely detectable. Zn/peptide OMPO complexes retain much of
the extended form (panels 3 in Figure 2d, e) even with higher proportion of metal 1:2 (panel
4 in Figure 2d), however a pentameric signal for this conditions could not be detected.
A shift towards the compact forms caused by metal binding is retained in the gas phase
even if metal ion dissociates from the oligomer (but not monomer) during ionization. Spectra
collected in the presence of metal ions contain signals of oligomeric species corresponding to
their apo forms, without metal bound. In Figure 3 drift time profiles at corresponding
fragments of the spectra are compared to profiles for Cu-bound species and profiles obtained
in the absence of metal ions. For monomer (Figure. 3a) apo form profile (panel 2) is identical
to the spectrum collected in the absence of metal (panel 1) and does not contain the signal of
the compact form, prominent in Cu-bound profile (panel 3). In the case of monomer,
dissociation of metal leads to rapid equilibration and decay of compact species. In contrast,
for trimers, tetramers and pentamers (Figure 3c-e) the profiles in apo form strictly retain the
distribution, detected for the Cu-bound species. For these oligomers the interconversion
between compact and extended species is relatively slow in the millisecond timescale of the
experiment. In conclusion, strong relative stabilization of compact forms of Aβ oligomers,
from trimers to pentamers, was shown based on the analysis of IMS-MS spectra collected in
the presence of Cu(II) or Zn(II).
The intra- and intermolecular interactions between the peptide residues are also of
great interest, since they may direct the mutual population of compact and extended forms. In
the present work we have studied the effect of selected point mutations in Aβ sequence on the
ratio of compact/extended forms. Previously, we have constructed molecular models of
compact and extended state (Figure 4 in39). In these models a fibril-like sandwich structure of
the compact form is converted to extended form due to breaking of the sandwich interactions
and rotation of the two β-sheet parts in the hinge region composed of turn residues 23-28. To
verify this model we have introduced mutations in residues of hinge region with the
expectation that these mutations will change the mutual population of compact and extended
forms. Since, according to the model, the turn in compact form is stabilized by a salt bridge
D23-K28 we have introduced mutation K28A. Glycine at position 25 may ensure flexibility in
7
2016-02-16
the hinge region so in the second mutant we have introduced a mutation G25P. We have
measured IMS-MS spectra of these two mutated peptides and compared them with spectra
obtained for wild type Aβ1–40 obtained in the same experimental conditions. Drift time
profiles (Figure 4) show significant change upon mutation. The effect is similar for all
oligomeric forms compared and for all signal groups analyzed, namely signals corresponding
to DIM5+, TRI6+, TET7+, PEN8+ (Figure 4a-d, respectively). Attribution of signals to two
alternate structural forms of given oligomer, carried out in previous work39 was in each case
confirmed here by the analysis of isotopic envelopes. In G25P, the coexistence of compact
and extended forms observed for WT is changed to strong preference to the compact form and
in K28A to the predominance of the extended form. In case of DIM5+ (Figure 4a) in K28A
(lowest panel) also a new form of intermediate  value (of drift time 9.9 ms) appears,
whereas in G25P (middle panel) the compact form dominates and extended form is nearly
absent. Of the two TRI6+ forms (Figure 4b), equally populated in WT, again the extended
form is nearly absent in G25P whereas the compact form is prominent. In K28A the compact
form signal is weak whereas the visible extended form signal partially co-migrated in the drift
chamber with the strong DIM4+ peak. However, its characteristic charge 6+ isotopic envelope
can be detected on 2D profiles, co-migrating at nearly the same drift time as a stronger charge
+4 envelope (Supporting Material Figure S1). Similarly, TET7+ and PEN8+ (Figure 4c, d,
respectively) are characterized by the predominance of compact form in G25P and extended
form in K28A, replacing a mixture of the two forms in WT. Amino acid K28 provides a
partner for salt bridge-mediated stabilization of the compact state, removal of which leads to
the extended form. On the other hand, a flexible Gly25 residues is necessary to allow for
opening the compact state, rigidity of chain caused by Pro at this position leads to the strong
predominance of the compact form. This experiment shows that turn residues are important
determinants, shaping the equilibrium of forms in Aβ1–40 oligomers. This supports the
molecular model in which the extended conformation arises from opening of the sandwich
structure of the fibril-like oligomer.
We have also tested the requirement for β-sheet preference within the two stretches of
hydrophobic residues, namely Leu17-Ala21 and Ala31-Leu34, by introducing up to 5 βbreaking proline residues at selected positions. Figure S2 (Supporting Material) shows drift
time profiles for mutants 3xPro (L17P, V18P, F19P) and 5xPro in which I32P, I34P
mutations were added to 3xPro mutations. For these two mutants the equilibrium shifts
strongly towards the compact form, in case of five prolines present leading to complete
absence of the extended form for all oligomer forms DIM5+, TRI6+, TET7+ and PEN8+ (Figure
8
2016-02-16
S2a-d, respectively). Moreover, the drift time of the compact state signal decreases with the
number of prolines introduced indicating the increased compactness of oligomeric forms,
which could accommodate numerous prolines substituting hydrophobic core residues of the
Aβ oligomer. These apparently counterintuitive results can be explained by molecular
modeling of 5xPro compact oligomer (Supporting Material Figure S3) which shows that a
parallel alignment of monomers allows to accommodate prolines into an oligomer with
appropriate bending in the sheet structure decreasing the overall collisional cross section of
the oligomer. Also, the hydrophobic nature of proline residue provides sufficient stability of
the hydrophobic core of the oligomer. The result of this experiment is in agreement with
parallel alignment of monomers in oligomer.
Discussion.
In our previous work39 we have shown in an Ion Mobility - Mass Spectrometry (IMS-MS)
analysis the coexistence of the two major structural forms of Aβ1–40 oligomers – compact
and extended. Here, we have studied several factors modifying the mutual population of
compact and extended structural forms. We have shown that the binding of metal ions
stabilizes the compact state and that single point mutations at several sites can drive the
equilibrium either towards compact or to extended form. Similar coexistence of
compact/extended forms was observed previously in the study of the differences between
human IAPP amyloidogenic peptide and its non-amyloidogenic rat counterpart42, and the
time-dependent evolution from more extended species to more compact forms has been noted
for PrP 20-aa fragment 106-126.43 Coexistence of alternative structural variants may lead to
the divergent, competing pathways leading to non-interconverting final structures of different
morphology, as described for β2-microglobulin.44–47 For Aβ peptide the bifurcation into onpathway and off-pathway evolution is proposed in numerous schemes.48,49 However, no
molecular details on the participating structural variants were presented in these schemes.
The analysis of IMS-MS data in our previous work39 allowed us to propose a
molecular model of compact and extended states that accounts best for the structural
constraints in the form of collisional cross section values obtained for a large set of Aβ1–40
oligomers. According to the model, a compact state represents on-pathway protofibril-like βsheet sandwich with a turn at positions 23-28, the placement of which was confirmed in a
solid state NMR-based fibril model50 and EPR study.51 Moreover, in an extended form the
sandwich structure becomes open, leading to oligomers of globulomer type, as described in48.
The coexistence of the two alternative forms leads to the molecular scheme of the two
9
2016-02-16
competing pathways of the Aβ1–40 oligomer evolution (Figure 5). The compact forms
populate an on-pathway branch leading to protofibrils and fibrils. Low-order oligomers might
undergo structural conversion to the extended form with subsequent exposure of patches of
hydrophobic residues, for instance providing opportunity for the oligomer to interact with the
non-polar milieu of a membrane. In the extended form the remaining C-terminal β-sheet
acquires additional flexibility, leading to the possibility for higher order oligomeric species to
form channel-like membrane structures which can account for some of the observed
neurotoxic events52, so modeled compact-extended structural transition may represent a
bifurcation point leading to the off-fibril pathway of oligomer evolution and formation of
potentially neurotoxic species.
IMS-MS, though relatively new, already became an established method of
characterization of amyloidogenic peptides/proteins.38 Aβ was also studied before using IMSMS.39,53–56 However, in the work of Bernstein and colleagues53 and the following papers from
the same group54–56 the two alternative structural forms were not identified and split drift time
profiles were interpreted as indicating the coexistence of oligomers of different order (like
dimers/tetramers/hexamers) and not of different compact-extended forms of an oligomer of
the same order (like compact tetramer/extended tetramer). For instance a bimodal drift time
distribution of the signal, centered at m/z 1731 (denoted z/n = -5/2 in54) was observed for
Aβ1–40 in negative mode, and the split was interpreted as coexistence of dimers and
tetramers. Measuring well resolved isotopic envelopes, we were able to show (Figure 1B
in39) that in positive mode both forms of the split drift time profile for the signal at m/z 1733
correspond to a dimer, one of the smaller (compact) and the other of a larger (extended)
collisional cross section, and not to the mixture of dimers and tetramers. To resolve this
inconsistency, in the present work we have carried out an additional IMS-MS analysis in
negative mode. Two regions of spectra obtained for Aβ1–40 in negative mode are shown in
Supporting Material Figure S4. One, corresponding to the above mentioned isotopic
envelope, centered at m/z 1731 and the second of the region at m/z 2164 (corresponding to
region denoted z/n = -2 in54). Similarly to the results reported in54 drift time profile at m/z
1731 in the negative mode is also split into two signals (Figure S4a), however both signals
are clearly characterized by the same spacing in the isotopic envelope, indicating the presence
of the two structural variants of -5 charged dimer (DIM5-) and absence of tetrameric signals in
this region of spectrum. Ionization mode polarity does not seem to affect the drift time
distributions in case of Aβ, as it was observed before for synuclein.57 Absence of tetrameric
signal in this region of spectrum does not mean the absence of tetramers in the sample.
10
2016-02-16
Oligomers are often stabilized by a network of ionic interactions, so higher order oligomers
may carry less proportional charge than lower order oligomers.39,54 Thus, it may not be
justified to expect that the dimeric -5 signals should be accompanied by tetrameric -10 signal.
Instead, tetrameric signals bearing lower charge 5 to 8, including +8 charged tetramer were
unequivocally identified in positive ion mode spectra at m/z 2164, accompanied by two
trimeric, a dimeric and monomeric signal (Figure 2 in39). The analysis of the same region in
negative ion mode (denoted z/n = -2, Figure 1b54) lead in the cited work to the conclusion that
only dimeric and monomeric forms are represented. Moreover, lack of signals corresponding
to trimers and higher order oligomer signals at m/z 2164 was interpreted as their lack in the
sample. Contrary to that, negative mode spectra carried out in present work (Figure S4b), due
to their higher resolution and sensitivity, clearly show the presence of trimeric signals
accompanying the dimeric and monomeric signals. Thus, IMS-MS data, irrelevant of the
ionization mode polarity used, do not support the hypothesis of the absence of trimers and
higher order oligomers in Aβ1–40 preparations and invalidates the sequence of events in the
oligomerization pathway of Aβ1–40 described in Figure 5 of ref.54. In conclusion, the
assignment of a particular signal to the oligomeric form should be verified by the analysis of
the isotopic envelope, and results not validated by such analysis should be treated with
caution. The coexistence of compact-extended oligomeric forms overlooked in the previous
work due to inferior resolution, is in our opinion an important feature of the system under
study. The structural switch within the pool of Aβ peptide oligomers may be decisive for
either fibril formation or off-pathway evolution of Aβ towards oligomeric forms of increased
neurotoxicity.
IMS allows to measure collisional cross-section of the molecules in the gas phase.
This leads to the question of the correspondence of these results to in-solution situation and
the issue is discussed.58 In case of proteins it has been shown that ESI can retain solution
phase structures at least for some time, as ESI-MS analyses reproduce several known protein
structural
aspects
and
non-covalent
complexes
stoichiometries59,60,
although
the
counterexamples exist.61,62 However, in the presented work we have noted that, in contrast to
monomers, A oligomers stripped of metal during ESI do not equilibrate to apo-state
distribution, retaining the metal-bound drift time profile in their metal-free form. This showed
that the “structural memory” of A oligomeric structures during ESI can last longer than the
time of IMS experiment.
To test the oligomer structural model we have studied selected point mutants at
residues which, according to the model, are expected to shape the equilibrium between the
11
2016-02-16
compact and extended form, especially in the region 23-28 of the turn providing a crucial
hinge that enables compact extended form transition. In K28A mutant we have observed a
strong preference for the extended form whereas in G25P to the compact form. In K28A
mutant the salt bridge D23K28, stabilizing the compact form according to the model, is not
possible and in G25P mutant the flexibility of the peptide chain necessary to open the sandwich structure is strongly restricted. Mutations of K28 are also known to change the
morphology of A aggregates and K28D mutation in Aβ1–42 leads to spherical aggregates
instead of fibrils.63 Also, a lactam bridge introduced at positions D23K28 leads to 1000 times
faster fibril formation.64 Thus, these results are in full agreement with the model assuming
stabilization of the compact form by D23K28 salt bridge on the pathway to fibrils.
Interestingly G25P mutation strongly destabilizes fibril formation65, which indicates different
requirements for turn conformation in rigid fibril structure and in more flexible oligomer
structure.
We have also studied the effects of mutations of selected hydrophobic residues. Since
proline residues are known to be disfavored in β-sheets and easily accommodated in the
disordered regions66 proline substitutions were probed at sheet positions in Aβ oligomers.
However, we have found that even up to 5 proline substitutions strongly stabilize compact
forms of oligomers. Proline substitution at β-sheet positions V18, F19, F20 were found to
strongly destabilize fibrils.65 However, L17P mutation had only moderate effect, which
indicates that proline mutation at some β-sheet positions can be accommodated even in rigid
fibril structure. A polyglutamine sequence was shown to be capable of strong aggregation
despite containing multiple Pro replacements at regular intervals.67 For oligomers molecular
modeling (Figure S3) shows that indeed prolines at β-sheet positions can be accommodated
in the model of a parallel β-sheet A oligomer. It remains unclear though why the extended
form becomes relatively destabilized in proline mutants.
We have also shown that binding of metal ions leads to relative stabilization of the
compact state, and Cu(II) is more effective in that respect than Zn(II). These features can be
tentatively explained on the basis of known similarities and differences in the interactions of
these ions with Aβ monomers and oligomers. In monomers, at neutral pH range, the Cu(II)
ion is coordinated as a 3N complex (i.e. with coordination to three nitrogen atoms). The major
structure (component I) contains Cu(II) bound to the N-terminal amine, His 6 imidazole, and
His 13 or His14 imidazole, but three imidazole donors are largely exchangeable. The minor
structure (component II) involves the N-terminal amine, the neighboring peptide nitrogen, and
12
2016-02-16
one of three His imidazoles. His13/His14 residues are alternatively engaged in their binding
in the monomeric state, when they have a significant conformational freedom19. Although
Cu(II) ions are thought to bind to oligomeric Aβ in a fashion qualitatively similar to that
studied exhaustively for the Aβ monomer22, the binding is known to be stronger in oligomeric
structures than in monomer.68 This observation suggests that individual Cu(II) ions may be
bound in the oligomers to nitrogen donors coming from different Aβ monomers. The
coordination mode can thus be different in details also because the conformational freedom of
His13/14 is lost in oligomers, as these residues are located at the beginning of the β structure,
which enforces location of His13 and His14 on the opposite sides of the β-sheet. Such fixing
will not have a negative effect on the efficiency of metal binding because either of these
histidines will remain available for simultaneous binding with unordered N-terminus,
depending on its location on either side of the β-sheet. In oligomers histidines from monomers
adjacent in the sheet are kept in orientation that enables both of them to contribute
simultaneously to metal binding (either Hi13+His13 or His14+His14).
Molecular modeling of the possible binding modes of the metal ion to oligomer
structures obtained before39 showed that the canonical 3N character of the complex can be
retained in oligomers (Figure 6a-c), although in oligomeric structures the metal-binding
imidazole of His-13 points upwards, and that of His-14 downwards, with respect to the
location of the C-terminus in the compact structure. The structures obtained can also explain
the observed stabilization of the compact state. Three different Cu(II) 3N structures consistent
with both the experimental data presented above and the previous literature are shown in
Figure 6. Structure A retains the N-terminal amine and His 6 imidazole pattern of component
I, while the second imidazole is provided by His-6 of the neighboring Aβ molecule. Metal
binding stabilizes the oligomer by keeping N-termini of Aβ monomers together. Structure B
in which the complex engages two His 14 imidazoles, stabilizes β-sheet region in the
oligomer. Structure C is analogous to the previous one, but involves two His 13 imidazoles.
This enables participation of the C-terminal carboxylates in the equatorial and/or axial Cu(II)
binding, which provides the mechanistic support to the observation that Cu(II) ions stabilize
the closed structure conformers. The presence of exchangeable carboxylate(s) in the
coordination spheres of Cu(II) and Zn(II) is known for the monomer. In these terms, the Cterminal carboxylate simply joins this pool in oligomers. According to this line of reasoning,
Cu(II) is more effective than Zn(II) in stabilizing the compact structure, due to its higher
affinity to imidazole nitrogens, and thus the higher population of His-13 in the complex.
13
2016-02-16
In the presented study we have identified several factors shaping the equilibrium
between structurally different Aβ oligomeric forms, thus improving our understanding of
chain of events leading to the formation of neurotoxic species in Alzheimer’s disease. This
knowledge may be helpful in the engineering of a successful therapeutic strategy, since
rational drug design requires sound knowledge of the target structures.
Materials and Methods
Aβ variants expression
Aβ1–40 peptide and its variants were obtained by expression in E. coli and purified using
HPLC as described previously.69 Peptide variants were obtained by Site-Directed
Mutatagenesis (Stratagene Kit). Aβ peptide mutants have changed at position: a) K28A; b)
G25P; c) L17P, V18P, F19P (3xPro), d) L17P, V18P, F19P, I32P, I34P (5xPro). The identity
and purity of each molecule was verified using a Q-ToF Premier ESI-MS instrument
(Waters). Typically, the concentration of the peptide in the stock solution was within the
range of 30 - 80 μM and pH was adjusted to 7.4 with ammonia. The stocks were stored at
4 °C and used within 48 h.
Experiments with metal
Copper (II) acetate and zinc acetate dihydrate were purchased from Sigma-Aldrich
Chemical Co. Metal solutions were prepared from weighted amount using analysis grade
water (Baker) to attain concentrations of 9.4 mM (Cu) and 15.6 mM (Zn). Additionally, the
concentration of Cu(CH3COO)2 was verified using the Lambert-Beer law (A = εlc), with a
molar extinction coefficient value for Cu2+aq of 12.0 M-1 cm-1 at λ = 810 nm.70
Ion mobility separation – mass spectrometry (IMS-MS)
Experiments were performed using a Synapt G2 HDMS instrument (Waters). Aβ1–40
peptide or its variants in the presence or absence of metal ions at 80 μM concentration in 10
mM CH3COONH4, pH 7.4 was infused directly (at 5l/min) to the ion source of a mass
spectrometer, with a glass Hamilton syringe, through a stainless steel capillary. The mass
signals were measured in the range of 400–4100 m/z at the rate of 1 s per scan. The instrument
was working in electrospray positive ion mode with a capillary voltage of 2.8 kV and sample
cone voltage of 38 V. The mobility T-wave cell was operated at a pressure of 3.2 mbar of
nitrogen, with a wave velocity of 650 m/s and amplitude of 39 V. Data acquisition and
processing were carried out with MassLynx (V4.1) and DriftScope (V2.1) software supplied
14
2016-02-16
with the instrument. Each analysis of drift time profiles in the presence of metal or in mutant
was directly preceded by the experiment on WT Aβ1–40 carried out in the same experimental
conditions. This excludes the possibility of changes in drift time profiles caused by
instrumental irreproducibility. Since the profiles obtained for WT Aβ1–40 were highly
reproducible on a batch-to-batch and day-to-day basis, in the presented figures only one
control profile is shown for each experiment.
Modeling of Aβ1–40 oligomers.
All the molecular modeling and molecular dynamics simulations were performed
using Yasara Structure package. The original Yasara 271 forcefield was extended with QMderived bond lengths and ESP charges, which were calculated as follows. Various topologies
of possible Cu2+ coordinating centers were initially preoptimized using the semi-empirical
PM3 method with methylimidazole, methylamine, acetamide and acetic acid as models for Cu
coordination by Histidine, N-terminal amino group, backbone carbonyl, and C-terminal
carboxyl, respectively. These systems were then optimized using the DFT B3LYP/6-31G(d,p)
method implemented in Firefly 8.0 package,72 which is partially based on the GAMESS
(US)73 source code. Corrections for solute-solvent interactions were also introduced by the
polarizable continuum model (PCM).74
Initial protofibril-like structure of a non-covalently stabilized hexadecamer of Aβ1–40
was prepared previously39 using constraints deduced from accessible fibril structural
information: intermolecular D23–K28 salt bridge; residues L17–V24 and A30–V39 kept in
the extended conformation to build a β-sheet core; the twist of the backbone in the turn
region; the intrachain distance constraints (2.9–8.5 Å between Cα atoms) set, in agreement
with mutational data for F19–G38 and A21–V36 residue pairs to simulate their spatial
proximity. For the purpose of the present work a central hexamer (peptide molecules 6-11)
was extracted, placed in water box and subjected for 50 ns molecular dynamics in NTP
ensemble (T = 298 K and p = 1 atm). The weights of all constraints were iteratively
decreased, the last 20 ps MD was performed without any constraints.
Acknowledgements
Financial support was received from the Foundation for Polish Science TEAM program
(TEAM/2011-7/1), CEPT (POIG.02.02.00-14-024/08-00), and NanoFun (POIGT.02.02.0000-025/09-00). Many thanks to Damian Matak, Hubert Czepik and Dorota Wasilak for their
15
2016-02-16
help with overexpression of peptide variants and to Agnieszka Jabłonowska for Aβ
overexpression protocol.
Reference
1. Walsh, D. M. & Selkoe, D. J. (2007). Aβ Oligomers – a decade of discovery. J
Neurochem. 101, 1172–1184.
2. Hardy, J. A. & Higgins, G. A. (1992). Alzheimer’s disease: the amyloid cascade
hypothesis. Science 256, 184–185.
3. Terry, R. D. (1996). The pathogenesis of Alzheimer disease: an alternative to the amyloid
hypothesis. J. Neuropathol. Exp. Neurol. 55, 1023–1025.
4. McLean, C. A., Cherny, R. A., Fraser, F. W., Fuller, S. J., Smith, M. J., Beyreuther, K.,
Bush, A. I. & Masters, C. L. (1999). Soluble pool of Abeta amyloid as a determinant of
severity of neurodegeneration in Alzheimer’s disease. Ann. Neurol. 46, 860–866.
5. Lue, L. F., Kuo, Y. M., Roher, A. E., Brachova, L., Shen, Y., Sue, L., Beach, T., Kurth, J.
H., Rydel, R. E. & Rogers, J. (1999). Soluble amyloid beta peptide concentration as a
predictor of synaptic change in Alzheimer’s disease. Am. J. Pathol. 155, 853–862.
6. 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. (2008). Amyloid-beta protein dimers isolated directly from
Alzheimer’s brains impair synaptic plasticity and memory. Nat. Med. 14, 837–842.
7. Walsh, D. M., Klyubin, I., Fadeeva, J. V., Cullen, W. K., Anwyl, R., Wolfe, M. S.,
Rowan, M. J. & Selkoe, D. J. (2002). Naturally secreted oligomers of amyloid beta
protein potently inhibit hippocampal long-term potentiation in vivo. Nature 416, 535–539.
8. Cleary, J. P., Walsh, D. M., Hofmeister, J. J., Shankar, G. M., Kuskowski, M. A., Selkoe,
D. J. & Ashe, K. H. (2005). Natural oligomers of the amyloid-beta protein specifically
disrupt cognitive function. Nat. Neurosci. 8, 79–84.
9. Benilova, I., Karran, E. & De Strooper, B. (2012). The toxic Aβ oligomer and
Alzheimer’s disease: an emperor in need of clothes. Nat. Neurosci. 15, 349–357.
10. Klyubin, I., Cullen, W. K., Hu, N.-W. & Rowan, M. J. (2012). Alzheimer’s disease Aβ
assemblies mediating rapid disruption of synaptic plasticity and memory. Mol Brain 5, 110.
11. Mucke, L. & Selkoe, D.J. (2012). Neurotoxicity of Amyloid β-Protein: Synaptic and
Network Dysfunction. Cold Spring Harb Perspect Med 2, 1-17.
12. Larson, M. E. & Lesné, S. E. (2012). Soluble Aβ oligomer production and toxicity. J.
Neurochem. 120, 125–139.
13. State of aggregation. (2011). Nat Neurosci 14, 399.
14. Huang, X., Atwood, C., Moir, R., Hartshorn, M., Tanzi, R. & Bush, A. (2004). Trace
metal contamination initiates the apparent auto-aggregation, amyloidosis, and
oligomerization of Alzheimer’s Aβ peptides. J Biol Inorg Chem 9, 954–960.
15. Tõugu, V. & Palumaa, P. (2012). Coordination of zinc ions to the key proteins of
neurodegenerative diseases: Aβ, APP, α-synuclein and PrP. Coordination Chemistry
Reviews 256, 2219–2224.
16. Drew, S. C., Noble, C. J., Masters, C. L., Hanson, G. R. & Barnham, K. J. (2009).
Pleomorphic Copper Coordination by Alzheimer’s Disease Amyloid-β Peptide. J. Am.
Chem. Soc. 131, 1195–1207.
17. Faller, P., Hureau, C., La Penna G. & Furlan S. (2013). Coordination of Metal Ions to Aβ
Peptide: Impact on Alzheimers disease. Alzheimer’s Disease: Insights into Low Molecular
Weight and Cytotoxic Aggregates from In Vitro and Computer Experiments. In The
16
2016-02-16
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
Molecular Medicine and Medicinal Chemistry 7 (Derreumaux, P.), pp. 127-155, London:
Imperial College Press.
Alies, B., Bijani, C., Sayen, S., Guillon, E., Faller, P. & Hureau, C. (2012). Copper
Coordination to Native N-Terminally Modified versus Full-Length Amyloid-β: SecondSphere Effects Determine the Species Present at Physiological pH. Inorg. Chem. 51,
12988–13000.
Hureau, C. & Dorlet, P. (2012). Coordination of redox active metal ions to the amyloid
precursor protein and to amyloid-β peptides involved in Alzheimer disease. Part 2:
Dependence of Cu(II) binding sites with Aβ sequences. Coordination Chemistry Reviews
256, 2175–2187.
Zawisza, I., Rózga, M. & Bal, W. (2012). Affinity of copper and zinc ions to proteins and
peptides related to neurodegenerative conditions (Aβ, APP, α-synuclein, PrP).
Coordination Chemistry Reviews 256, 2297–2307.
Alies, B., Renaglia, E., Rózga, M., Bal, W., Faller, P. & Hureau, C. (2012). Cu(II)
Affinity for the Alzheimer’s Peptide: Tyrosine Fluorescence Studies Revisited. Anal.
Chem. 85, 1501–1508.
Gunderson, W. A., Hernández-Guzmán, J., Karr, J. W., Sun, L., Szalai, V. A. & Warncke,
K. (2012). Local Structure and Global Patterning of Cu2+ Binding in Fibrillar Amyloid-β
[Aβ(1–40)] Protein. J. Am. Chem. Soc. 134, 18330–18337.
Pedersen, J. T., Teilum, K., Heegaard, N. H. H., Østergaard, J., Adolph, H.-W. &
Hemmingsen, L. (2011). Rapid Formation of a Preoligomeric Peptide–Metal–Peptide
Complex Following Copper(II) Binding to Amyloid β Peptides. Angewandte Chemie
International Edition 50, 2532–2535.
Faller, P. (2009). Copper and Zinc Binding to Amyloid-β: Coordination, Dynamics,
Aggregation, Reactivity and Metal-Ion Transfer. ChemBioChem 10, 2837–2845.
Xue, W.-F., Homans, S. W. & Radford, S. E. (2008). Systematic analysis of nucleationdependent polymerization reveals new insights into the mechanism of amyloid selfassembly. Proc Natl Acad Sci USA 105, 8926–8931.
Lomakin, A., Teplow, D. B., Kirschner, D. A. & Benedek, G. B. (1997). Kinetic theory of
fibrillogenesis of amyloid β-protein. Proc Natl Acad Sci USA 94, 7942–7947.
Harper, J. D. & Lansbury, P. T. (1997). Models of amyloid seeding in Alzheimer’s
disease and scrapie: mechanistic truths and physiological consequences of the timedependent solubility of amyloid proteins. Annu Rev Biochemistry 66, 385–407.
Roychaudhuri, R., Yang, M., Hoshi, M. M. & Teplow, D. B. (2009). Amyloid β-Protein
Assembly and Alzheimer Disease. J Biol Chem. 284, 4749–4753.
Wetzel, R. (2006). Kinetics and Thermodynamics of Amyloid Fibril Assembly. Acc.
Chem. Res. 39, 671–679.
Kumar, A., Paslay, L. C., Lyons, D., Morgan, S. E., Correia, J. J. & Rangachari, V.
(2012). Specific soluble oligomers of amyloid-β peptide undergo replication and form
non-fibrillar aggregates in interfacial environments. J. Biol. Chem. 287, 21253–21264.
Necula, M., Rakez, K., Kayed, R., Milton, S. & Glabe, G. Ch. (2007). Small Molecule
Inhibitors of Aggregation Indicate That Amyloid β Oligomerization and Fibrillization
Pathways Are Independent and Distinct. J Biol Chem 282, 1031-1024.
Jahn, T. R. & Radford, S. E. (2008). Folding versus aggregation: Polypeptide
conformations on competing pathways. Arch Biochemistry Biophys. 469, 100–117.
Goldsbury, C., Frey, P., Olivieri, V., Aebi, U. & Müller, S. A. (2005). Multiple Assembly
Pathways Underlie Amyloid-β Fibril Polymorphisms. J Mol Biol. 352, 282–298.
Gorman, P. M., Yip, C. M., Fraser, P. E. & Chakrabartty, A. (2003). Alternate
Aggregation Pathways of the Alzheimer β-Amyloid Peptide: Aβ Association Kinetics at
Endosomal pH. J Mol Biol. 325, 743–757.
17
2016-02-16
35. Huang, T. H. J., Yang, D. S., Fraser, P. E. & Chakrabartty, A. (2000) Alternate
Aggregation Pathways of the Alzheimer β-Amyloid Peptide. J Biol Chem 275, 3643636440.
36. Lambert, M. P., Barlow, A. K., Chromy, B. A., Edwards, C., Freed, R., Liosatos, M.,
Morgan, T. E., Rozovsky, I., Trommer, B., Viola, K. L., Wals, P., Zhang, C., Finch, C. E.,
Krafft, G. A. & Klein, W. L. (1998). Diffusible, nonfibrillar ligands derived from Aβ1–42
are potent central nervous system neurotoxins. Proc Natl Acad Sci USA 95, 6448–6453
37. Suzuki, Y., Brender, J. R., Soper, M. T., Krishnamoorthy, J., Zhou, Y., Ruotolo, B. T.,
Kotov, N. A., Ramamoorthy, A. & Marsh, E. N. G. (2013). Resolution of Oligomeric
Species during the Aggregation of Aβ1–40 Using 19F NMR. Biochemistry 52, 1903–
1912.
38. Woods, L. A., Radford, S. E. & Ashcroft, A. E. (2013). Advances in ion mobility
spectrometry–mass spectrometry reveal key insights into amyloid assembly. Biochim
Biophysica Acta (BBA) - Proteins and Proteomics 1834, 1257–1268.
39. Kłoniecki, M., Jabłonowska, A., Poznański, J., Langridge, J., Hughes, C., Campuzano, I.,
Giles, K. & Dadlez, M. (2011). Ion Mobility Separation Coupled with MS Detects Two
Structural States of Alzheimer’s Disease Aβ1–40 Peptide Oligomers. J Mol Biol 407,
110–124.
40. Kelly, J. W. (2000). Mechanisms of amyloidogenesis. Nat Struct Mol Biol 7, 824–826.
41. Murphy, R. M. (2007). Kinetics of amyloid formation and membrane interaction with
amyloidogenic proteins. Biochim Biophysica Acta (BBA) - Biomembranes 1768, 1923–
1934.
42. Dupuis, N.F., Wu, C., Shea, J.-E. & Bowers, M.T. (2011). The Amyloid Formation
Mechanism in Human IAPP: Dimers Have β-Strand Monomer−Monomer Interfaces. J.
Am. Chem. Soc. 133, 7240–7243
43. Grabenauer, M., Wyttenbach, T., Sanghera, N., Slade, S.E., Pinheiro, T.J.T., Scrivens,
J.H. & Bowers, M.T. (2010). Conformational Stability of Syrian Hamster Prion Protein
PrP(90−231). J. Am. Chem. Soc. 132, 8816–8818
44. Smith, D.P., Woods, L.A., Radford, S.E. & Ashcroft, A.E. (2011). Structure and
Dynamics of Oligomeric Intermediates in β2-Microglobulin Self-Assembly. Biophysical
Journal 101, 1238–1247
45. Gosal, W., Morten, I., Hewitt, E., Smith, D., Thomson, N. & Radford, S. (2005).
Competing pathways determine fibril morphology in the self-assembly of beta2microglobulin into amyloid. J Mol Biol 351, 850–864
46. Baskakov, I.V., Legname, G., Baldwin, M.A., Prusiner, S.B. & Cohen, F.E. (2002).
Pathway Complexity of Prion Protein Assembly into Amyloid. Journal of Biological
Chemistry 277, 21140–21148
47. Bleiholder, C., Dupuis, N.F., Wyttenbach, T. & Bowers, M.T. (2011). Ion mobility–mass
spectrometry reveals a conformational conversion from random assembly to β-sheet in
amyloid fibril formation. Nat Chem 3, 172–177
48. 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. (2009). Structural
Characterization of a Soluble Amyloid β-Peptide Oligomer. Biochemistry 48, 1870–1877
49. Gellermann, G.P., Byrnes, H., Striebinger, A., Ullrich, K., Mueller, R., Hillen, H. &
Barghorn, S. (2008). Aβ-globulomers are formed independently of the fibril pathway.
Neurobiology of Disease 30, 212–220
50. Petkova, A.T., Ishii, Y., Balbach, J.J., Antzutkin, O.N., Leapman, R.D., Delaglio, F. &
Tycko, R. (2002). A structural model for Alzheimer’s β-amyloid fibrils based on
18
2016-02-16
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
experimental constraints from solid state NMR. Proceedings of the National Academy of
Sciences 99, 16742–16747
Török, M., Milton, S., Kayed, R., Wu, P., McIntire, T., Glabe, C.G. & Langen, R. (2002).
Structural and Dynamic Features of Alzheimer’s Aβ Peptide in Amyloid Fibrils Studied
by Site-directed Spin Labeling. Journal of Biological Chemistry 277, 40810–40815
Kawahara, M., Ohtsuka, I., Yokoyama, S., Kato-Negishi, M. & Sadakane, Y. (2011).
Membrane Incorporation, Channel Formation, and Disruption of Calcium Homeostasis by
Alzheimer’s β-Amyloid Protein. International Journal of Alzheimer’s Disease,Article ID
304583
Bernstein, S.L., Wyttenbach, T., Baumketner, A., Shea, J.-E., Bitan, G., Teplow, D.B. &
Bowers, M.T. (2005). Amyloid β-Protein: Monomer Structure and Early Aggregation
States of Aβ42 and Its Pro19 Alloform. J. Am. Chem. Soc. 127, 2075–2084
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. (2009).
Amyloid-β protein oligomerization and the importance of tetramers and dodecamers in
the aetiology of Alzheimer’s disease. Nat. Chem. 1, 326Gessel, M.M., Wu, C., Li, H., Bitan, G., Shea, J.-E. & Bowers, M.T. (2011). Aβ(39–42)
Modulates Aβ Oligomerization but Not Fibril Formation. Biochemistry 51, 108–117
Gessel, M.M., Bernstein, S., Kemper, M., Teplow, D.B. & Bowers, M.T. (2012). Familial
Alzheimer’s Disease Mutations Differentially Alter Amyloid β-Protein Oligomerization.
ACS Chem. Neurosci. 3, 909–918
Frimpong, A.K., Abzalimov, R.R., Uversky, V.N. & Kaltashov, I.A. (2010).
Characterization of intrinsically disordered proteins with electrospray ionization mass
spectrometry: Conformational heterogeneity of α-synuclein. Proteins: Structure,
Function, and Bioinformatics 78, 714–722
Ashcroft, A. (2010). Mass spectrometry and the amyloid problem—How far can we go in
the gas phase? J Am Soc Mass Spectrom 21, 1087–1096
Ruotolo, B.T. & Robinson, C.V. (2006). Aspects of native proteins are retained in
vacuum. Current Opinion in Chemical Biology 10, 402–408
Kaltashov, I.A. & Eyles, S.J. (2002). Studies of biomolecular conformations and
conformational dynamics by mass spectrometry. Mass Spectrometry Reviews 21, 37–71
Breuker, K. & McLafferty, F.W. (2008). Stepwise evolution of protein native structure
with electrospray into the gas phase, 10−12 to 102 s. Proceedings of the National
Academy of Sciences 105, 18145–18152
Ly, T. & Julian, R.R. (2010). Elucidating the Tertiary Structure of Protein Ions in Vacuo
with Site Specific Photoinitiated Radical Reactions. J. Am. Chem. Soc. 132, 8602–8609
Lührs, T., Ritter, C., Adrian, M., Riek-Loher, D., Bohrmann, B., Döbeli, H., Schubert, D.
& Riek, R. (2005). 3D structure of Alzheimer’s amyloid-β(1–42) fibrils. Proceedings of
the National Academy of Sciences of the United States of America 102, 17342–17347
Sciarretta, K.L., Gordon, D.J., Petkova, A.T., Tycko, R. & Meredith, S.C. (2005). Aβ40Lactam(D23/K28) Models a Conformation Highly Favorable for Nucleation of Amyloid†.
Biochemistry 44, 6003–6014
Williams, A.D., Portelius, E., Kheterpal, I., Guo, J., Cook, K.D., Xu, Y. & Wetzel, R.
(2004). Mapping Aβ Amyloid Fibril Secondary Structure Using Scanning Proline
Mutagenesis. Journal of Molecular Biology 335, 833–842
Chou, P.Y. & Fasman, G.D. (1978). Empirical Predictions of Protein Conformation.
Annual Review of Biochemistry 47, 251–276
Thakur, A.K. & Wetzel, R. (2002). Mutational analysis of the structural organization of
polyglutamine aggregates. Proceedings of the National Academy of Sciences 99, 17014–
17019
19
2016-02-16
68. Jiang, D., Zhang, L., Grant, G.P.G., Dudzik, C.G., Chen, S., Patel, S., Hao, Y.,
Millhauser, G.L. & Zhou, F. (2012). The Elevated Copper Binding Strength of Amyloid-β
Aggregates Allows the Sequestration of Copper from Albumin: A Pathway to
Accumulation of Copper in Senile Plaques. Biochemistry 52, 547–556
69. Rózga, M., Kłoniecki, M., Jabłonowska, A., Dadlez, M. & Bal, W. (2007). The binding
constant for amyloid Aβ40 peptide interaction with human serum albumin. Biochemical
and Biophysical Research Communications 364, 714–718
70. Prenesti, E. & Berto, S. (2002). Interaction of copper(II) with imidazole pyridine
nitrogen-containing ligands in aqueous medium: a spectroscopic study. Journal of
Inorganic Biochemistry 88, 37–43
71. Krieger, E., Joo, K., Lee, J., Lee, J., Raman, S., Thompson, J., Tyka, M., Baker, D. &
Karplus, K. (2009). Improving physical realism, stereochemistry, and side-chain accuracy
in homology modeling: Four approaches that performed well in CASP8. Proteins:
Structure, Function, and Bioinformatics 77, 114–122
72. Granovsky, A.A. (2013) Firefly version 8.0;
http://classic.chem.msu.su/gran/firefly/index.html
73. Schmidt, M.W., Baldridge, K.K., Boatz, J.A., Elbert, S.T. & Gordon, M.S. (1993) General
atomic and molecular electronic-structure system. Journal of Computational Chemistry
14: 1347-1363.
74. Mennucci, B., Cammi, R. & Tomasi, J. (1999). Medium effects on the properties of
chemical systems: Electric and magnetic response of donor–acceptor systems within the
polarizable continuum model. International Journal of Quantum Chemistry 75, 767–781
Figure legends
Figure 1. Typical ESI-MS spectrum of Aβ1-40, described in detail previously in ref.39. In the
spectrum signals expected for monomeric species are accompanied by numerous signals
corresponding to oligomeric species, as exemplified by the isotopic envelopes of the selected
signals A-C (upper insets). For each selected m/z region a profile showing molecular species
present at different drift times in the ion mobility chamber can be obtained in the IMS-MS
experiment as shown for signals A-C in lower insets. Some profiles contain a single peak,
indicating the presence of single species of uniform collisional cross-section, whereas other
signals are split in the domain of the drift time into multiple species (see text).
Figure 2. IMS-MS drift time profiles of selected signals corresponding to Aβ peptide
oligomers collected in the presence of metal ions, either Cu(II) or Zn(II) and compared to the
control spectra collected in the absence of metals (upper panels). (a) dimer (DIM5+) for which
only a small decrease in the fraction of extended form (at 10.9-11.4 ms) as compared to
compact form (at 8.1 ms) is observed in one metal per monomer (OMPM) signal group (lower
panels) as contrasted to lack of change for one metal per oligomer (OMPO) spectra (panels
20
2016-02-16
2,3). Both signals, denoted “extended” and “compact”, correspond to two structurally
different forms of the dimer because both are characterized by a well resolved isotopic
envelope showing the 0.2 Da spacing, thus corresponding to a +5 charged species and the
molecular mass expected for the dimer (Supporting material Fig. S2 and Fig. S3 in39 (b)
trimer (TRI7+) present only in the extended form (at 10.5 ms) in control spectra (upper panel)
in the presence of metals shows the mixture of extended and compact (at 7.7 ms) forms in
OMPO spectra. OMPM signals could not be detected for this species. (c) trimer (TRI6+)
signals for apo-Aβ1–40 have the same m/z values as TET8+, DIM4+ and MON2+ signals and
are separated into a characteristic group of five well resolved signals only in IMS spectra (see
also Fig. 2 in39). The identity of each oligomeric form, as shown in the upper panel, was
assigned to IMS signals based on the analysis of well resolved isotopic envelopes as described
in the cited work, with clear coexistence of the two structurally different forms of the trimer
+6 (compact at 9.5 ms and extended at 12.4 ms). It is the ratio of the heights of these two
signals which is under study in the present work. In control spectra both extended and
compact forms are equally populated, in OMPO spectra (panels 2 and 3) compact form
dominates even with low (1:0.5) peptide/metal molar ratio. In these spectra signals at 11.2 ms
and 15.8 ms originate from potassium and sodium adducts of DIM4+ and MON2+ oligomers,
respectively. In OMPM form (panel 4) the extended form becomes negligible. (d) tetramer
(TET7+) in control coexists as a mixture of compact (at 10.1 ms) form dominated by a more
abundant extended form (at 13.2 ms). The presence of metals leads to the predominance of
the compact form, more pronounced in the case of Cu than for Zn (compare panels 2 and 3).
Also, increased number of metal ions associated stabilized compact state (compare Cu/peptide
OMPM – panel 5 and OMPO – panel 2 complex and Zn/peptide OMPO complexes in lower
fraction of 1:0.5 Zn – panel 3 and higher 1:2 Zn fraction – panel 4). (e) pentamer (PEN8+)
shows strong predominance of extended form in control (upper panel) exchanged into strong
predominance of the compact form in the presence of metals, however with much stronger
effect of Cu than Zn (compare panels 2 and 3) and absence of extended form in OMPM
profile (panel 4).
Figure 3. IMS-MS drift time profiles of Aβ peptide species observed in the presence of
Cu(II), corresponding to apo form (panels 2) or Cu-bound form (panels 3), compared to the
control spectra collected in the absence of metals (panels 1). Data for monomeric (a), dimeric
(b), trimeric (c) tetrameric (d) and pentameric (e) forms are shown. Metal-induced compact
21
2016-02-16
state is retained in the timescale of the experiment after dissociation of metal in oligomers but
not in monomer.
Figure 4. The effect of turn region mutants G25P and K28A on relative population of
compact and extended form of Aβ1–40 oligomeric species. For the dimeric (DIM5+) species
(a) the compact form, observed at 8.3 ms in control, becomes weaker in the mutant K28A,
instead the signal of the extended form at 11.4 ms is accompanied by a new signal at 9.9 ms,
for which an isotopic envelope indicates alternative form of DIM5+. On the other hand, for
G25P mutant the DIM5+ species profile becomes dominated by the compact form at 7.4 ms,
whereas the extended form becomes much less populated and characterized by a shorter drift
time of 10.9 ms (and thus a smaller collisional cross section) as compared to extended species
in control at 11.2 ms. Of the two trimeric (TRI6+) forms (b) observed with equal signal height
in control the compact form becomes barely detectable in K28A mutant and strongly
dominates in G25P. Similarly, for TET7+ (c) and PEN8+ (d) a bimodal distribution of compact
and extended species observed in control, upon mutation becomes unimodal, with extended
form present in K28A and compact form in G25P. All analyzed drift time profiles show a
shift towards extended species in K28A and to compact species in G25P.
Figure 5. Oligomer evolution scheme resulting from detected coexistence of the compact and
extended oligomeric forms. Compact forms represent the on-pathway protofibrillar species
evolving towards fibrils (lower part of the scheme). These forms coexist with extended
structures in which only the C-terminal β-sheet is retained (upper part of the scheme) opening
the possibility for interaction with biological membranes and the formation of dodecameric
channel-like structures. The possibility of equilibration between the compact and extended
species symbolized by double arrows most probably decreases with the order of oligomer at
some point leading to the bifurcation (red cross) of the on-pathway and off-pathway
evolution.
Figure 6. Molecular modeling of Aβ1–40 oligomer complexed with a metal ion. In structure
(a) N-terminal amine and His 6 imidazole re engaged and the second imidazole is provided by
His-6 of the neighboring Aβ molecule. In structure (b) the complex engages two His 14
imidazoles. Structure (c) is analogous to the previous one, but involves two His 13
imidazoles. This enables participation of the C-terminal carboxylates in the equatorial and/or
22
2016-02-16
axial Cu(II) binding, explaining the observation that Cu(II) ions stabilize the closed structure
conformers.
23
2016-02-16