Analytica Chimica Acta 926 (2016) 36e47

Contents lists available at ScienceDirect

Analytica Chimica Acta

journal homepage: www.elsevier.com/locate/aca

Amyloideb peptides time-dependent structural modifications: AFM

and voltammetric characterization
Teodor Adrian Enache, Ana-Maria Chiorcea-Paquim, Ana Maria Oliveira-Brett*
Department of Chemistry, Faculty of Sciences and Technology, University of Coimbra, 3004-535, Coimbra, Portugal

h i g h l i g h t s g r a p h i c a l a b s t r a c t

The Ab peptide fibrilization process

was followed by AFM and DP
voltammetry.

The human Ab1-40 and Ab1-42 pep-
tides undergo oxidation in two steps.

The Ab peptides oxidation peak po-
tentials corresponded to amino acid
residues, the first to tyrosine, and the
second to histidine and methionine.

The first time the second oxidation
reaction of Ab peptides, the His and
Met oxidation, was investigated.

Human Ab peptides structural
modifications were compared to in-
verse Ab40-1 and Ab42-1, mutants
Ab1-40Phe10 and Ab1-40Nle35, and rat
Ab1-40Rat.

a r t i c l e i n f o a b s t r a c t

Article history: The human amyloid beta (Ab) peptides, Ab1-40 and Ab1-42, structural modifications, from soluble
Received 15 December 2015 monomers to fully formed fibrils through intermediate structures, were investigated, and the results
Received in revised form were compared with those obtained for the inverse Ab40-1 and Ab42-1, mutant Ab1-40Phe10 and Ab1-
14 March 2016
40Nle , and rat Ab1-40Rat peptide sequences. The aggregation was followed at a slow rate, in chloride free
35
Accepted 8 April 2016
Available online 21 April 2016
media and room temperature, and revealed to be a sequence-structure process, dependent on the
physicochemical properties of each Ab peptide isoforms, and occurring at different rates and by different
pathways. The fibrilization process was investigated by atomic force microscopy (AFM), via changes in
Keywords:
Amyloid-b
the adsorption morphology from: (i) initially random coiled structures of ~0.6 nm height, corresponding
Fibrilization to the Ab peptide monomers in random coil or in a-helix conformations, to (ii) aggregates and proto-
Aggregation fibrils of 1.5e6.0 nm height and (iii) two types of fibrils, corresponding to the Ab peptide in a b-sheet
Adsorption configuration. The reactivity of the carbon electrode surface was considered. The hydrophobic surface
Oxidation AFM induced rapid changes of the Ab peptide conformations, and differences between the adsorbed fibrils,
Voltammetry formed at the carbon surface (beaded, thin, <2.0 nm height) or in solution (long, smooth, thick, >2.0 nm
height), were detected. Differential pulse voltammetry showed that, according to their primary structure,
the Ab peptides undergo oxidation in one or two steps, the first step corresponding to the tyrosine amino
acids oxidation, and the second one to the histidine and methionine amino acids oxidation. The

* Corresponding author.
E-mail address: brett@ci.uc.pt (A.M. Oliveira-Brett).

http://dx.doi.org/10.1016/j.aca.2016.04.015
0003-2670/© 2016 Elsevier B.V. All rights reserved.
 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47 37

fibrilization process was electrochemically detected via the decrease of the Ab peptide oxidation peak
currents that occurred in a time dependent manner.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction understanding the AD mechanisms of action, but the research is

Protein structure dysregulations have critical effects in cellular
function, structure and recognition, such as misfolding and aggre-
gation, and can lead to cellular damage [1e4]. Amyloid fibril for-
mation, or amyloidogenesis, is occurring through endogenous
native proteins self-assembling into highly ordered aggregates with
cross b-sheet rich structures [4], specific to different unrelated
diseases, such as Parkinson's disease, type II diabetes, prion dis-
eases, familial amyloidosis, and Alzheimer's disease (AD) [5e7].
AD is the most widespread form of dementia and was recently
estimated to affect 44.4 million people worldwide [8]. Since its first
report in 1906 [9] till now, no specific cause or cure was discovered
[10]. The main clinical manifestations of AD are loss of memory,
cognitive skills, speech, and normative behaviour; however, accu-
mulation of fibrillary amyloid beta (Ab) peptides in plaques can
occur and be observed decades before the AD symptoms [11,12].
Native human Ab peptides exist as short ~4 kDa isoforms, sol-
uble monomers, varying from 39 to 43 amino acids, that are
generated from the transmembrane amyloid precursor protein
(APP) by hydrolytic cleavage by the proteases b- and g-secretase
(Scheme 1) [12e14]. The predominant isoforms are Ab1-40
(85e90%) and the faster-aggregating Ab1-42 (10e15%) (Scheme 1)
[14].
In AD brain, dysregulation and overproduction of Ab peptides
occurs, leading to Ab peptides fibrillization, in a process that
involves numerous highly neurotoxic intermediate structures
(Scheme 2). These intermediate aggregates were associated with
AD neurodegeneration through the lipid peroxidation, destabili-
zation of neuronal calcium level, membrane damage and tau pro-
tein hyperphosphorylation (Scheme 2) [4,13e16]. Although they
have been identified and partially characterized, the correlations
between Ab peptide isoforms conformation and neurotoxicity
remain unclear [4,7,10,14]. Structural, bioanalytical and biochemical
studies of the Ab peptides aggregation are likely to be critical in
complicated due to the Ab peptide aggregates inhomogeneity and
instability [4,7,10,14,17,18].
The Ab peptides complete fibrilization in vitro, at physiological
conditions (pH ¼ 7.4, 150 mM chloride) in Tris or phosphate buffer
saline at 37  C, occurs in 12 h for Ab1-40 and 5 h for Ab1-42 [19].
Although in these conditions the aggregation process is fast
[17,20e22], the in vivo aggregation rate is very slow, and AD brain
neurodegeneration appears a long time after the plaques formation
[11,12]. The difference in aggregation time may be due to the
complexity of the biological processes [14,23], as well as to different
Ab peptide concentration of several nM in vivo vs. 10e100 mM
in vitro [24].
The question that arises is whether, in the same physiological
conditions, the Ab peptide aggregates formed slowly in vivo at low
concentration, are different and more toxic, than the Ab peptide
aggregates formed faster in vitro at higher Ab peptide concentra-
tion. For a better characterization of the intermediary structures
formed during the fibrilization process, the fast aggregation rep-
resents an inconvenient that can be surpassed by lowering the
temperature and removing the salt conditions [25e27]. This
approach may be helpful for a better understanding of the mech-
anisms and identification of toxic species involved in the AD
development, as well as of other amyloid diseases.
Atomic force microscopy (AFM) combined with voltammetric
methods have been successfully used for the detection of small
perturbations of the morphological structure of many biological
compounds [28e31] and they represent an effective approach to
study the Ab peptides fibrilization process [32]. The Ab peptides
aggregation was investigated by electrochemistry onto glassy car-
bon electrode (GCE), as well as by AFM and STM onto mica and
highly oriented pyrolytic graphite (HOPG) [19,20,32e35], but all
studies were performed in chloride media, to promote peptides
aggregation. Successive stages in Ab peptides aggregation were
followed electrochemically via oxidation of tyrosine Y residues

Scheme 1. The Ab isoform peptides: human Ab1-40 and Ab1-42, mutants Ab1-40Phe10 and Ab1-40Nle35, and rat Ab1-40Rat. Red dots mark the electroactive amino acid residues. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
38 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47

Scheme 2. The neurodegenerative processes in Alzheimer's disease.

[19,32]. surface. When compared with previous electrochemical studies

In the present paper, the structural modifications undergone by
the synthetic human Ab isoform peptides containing 40 and 42
amino acids, Ab1-40 and Ab1-42 (Scheme 1), were studied by AFM at
a HOPG surface and differential pulse (DP) voltammetry at a GCE.
Lower aggregation rates were promoted by incubating the Ab
peptide solutions at room temperature and in free chloride media,
within time windows from 0 h to 48 h. The time dependent
structural modifications undergone by the Ab1-40 and Ab1-42 pep-
tides, i.e. the intermediary structures which appear during fibrili-
zation, were investigated and compared with those obtained for
five control sequences: inverse Ab40-1 and Ab42-1 peptides, mutant
peptides Ab1-40Phe10 and Ab1-40Nle35, and rat Ab1-40Rat peptide
(Scheme 1). Different than previous AFM studies onto HOPG [32],
closer attention was paid to the carbon electrode surface reactivity,
which directly influenced the formation of fibrils at the electrode
based on the oxidation of only Y [19,32], the paper novelty con-
sisted on the investigation of the Ab peptides second oxidation
peak, at higher potential, corresponding to the histidine H and
methionine M residues.
2. Experimental
2.1. Materials and reagents
The synthetic amyloid b (Ab) isoform peptides of 40 and 42
amino acids, human Ab1-40 and Ab1-42, inverse Ab40-1 and Ab42-1,
mutants Ab1-40Phe10 (with phenylalanine F10 instead of tyrosine
Y10) and Ab1-40Nle35 (with norleucine Nle35 instead of methionine
M35), and rat Ab1-40Rat peptide (glycine G5 instead of arginine R5,
phenylalanine F10 instead of tyrosine Y10 and arginine R13 instead of
 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47
histidine H13), Scheme 1, were bought lyophilized from Sigma-
eAldrich and Bachem, and used without further purification. All
experiments were done at room temperature (25 ± 1  C) in 0.1 M
phosphate buffer pH 7.4 prepared with analytical grade reagents
and purified water from a Millipore Milli-Q system
(conductivity  0.1 mS cm1).
2.2. Amyloid-beta samples preparation
Prior to all experiments, the Ab1-40, Ab1-42, Ab40-1, Ab42-1, Ab1-
10
40Phe , Ab1-40Nle35, and Ab1-40Rat solutions were prepared
according to the following procedure. The lyophilized powder was
firstly dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) at a
concentration of 1 mg mL1 to monomerize pre-existing aggre-
gates. After that, the HFIP was evaporated under a vacuum pump,
leaving a film that was re-dissolved in 0.1 M phosphate buffer pH
7.4. The suspension was further monomerized by sonication for
1 min, followed by dilution to 100 mM Ab peptide in 0.1 M phos-
phate buffer pH 7.4 (stock solutions). The solutions prepared by this
procedure were transparent without evident aggregation.
The Ab peptide stock solutions were then kept at room tem-
perature up to 48 h incubation time, while fibrilization took place.
During the fibrilization process, after different aggregation states
(0 h e freshly prepared solution, after 24 h and 48 h incubation),
aliquots of stock solutions were collected, diluted and analysed by
AFM (dilution to 1.0, 5.0 and 10 mM Ab peptide) and DP voltam-
metry (dilution to 50 mM Ab peptide).
2.3. Atomic force microscopy
Atomic force microscopy (AFM) was performed in the acoustic
AC (AAC) mode, with a PicoScan controller and a CS AFM S scanner
with a scan range of 6 mm in x-y and 2 mm in z, from Agilent
Technologies, USA. AppNano type FORT of 225 mm length, 3.0 N m1
spring constants and 47e76 kHz resonant frequencies in air
(Applied NanoStructures, Inc., USA) were used. All AFM images
were topographical and were taken with 512 samples/line  512
lines and scan rates of 0.8e2.5 lines s1. When necessary, the AFM
images were processed by flattening in order to remove the back-
ground slope and the contrast and brightness were adjusted.
Highly oriented pyrolytic graphite (HOPG), grade ZYB of
15  15  2 mm3 dimensions, from Advanced Ceramics Co., USA,
and mica, from Agilent Technologies, USA, were used as substrates
in the AFM study, because they are atomically flat. The GCE used for
the voltammetric characterization is rough and therefore unsuit-
able for AFM surface characterization. Furthermore, the voltam-
metric experiments using HOPG and GCE showed similar
electrochemical behaviours. The HOPG and mica were freshly
cleaved with adhesive tape prior to each experiment and imaged by
AFM in order to establish its cleanliness.
Unless stated otherwise, for the AFM study, the four different Ab
peptide sequences, human Ab1-40 and Ab1-42 and mutant Ab1-
40Phe
10
and Ab1-40Nle35, were immobilized by spontaneous
adsorption on the surface of HOPG or mica, by depositing 50 mL of
the appropriate Ab solution in 0.1 M phosphate buffer pH 7.4 onto
the freshly cleaved substrates, over a period of 3 min. The excess of
solution was then gently cleaned with a jet of Millipore Milli-Q
water, and the HOPG or mica with adsorbed Ab peptides were
dried in a sterile atmosphere. The immobilized Ab peptides were
imaged by AAC Mode AFM in air.
2.4. Voltammetric parameters and electrochemical cells
Voltammetric experiments were carried out using an IVIUM
potentiostat in combination with IviumSoft program version 2.219
39
(Ivium Technologies, Eindhoven, The Netherlands). Measurements
were carried out using a glassy carbon working electrode (GCE,
d ¼ 1 mm), a Pt wire counter electrode, and an Ag/AgCl (3 M KCl)
reference electrode, in a one-compartment 2 mL electrochemical
cell. Differential pulse (DP) voltammetry experimental conditions
were: pulse amplitude 50 mV, pulse width 100 ms, and scan rate of
5 mV s1. All voltammograms were background-subtracted and
baseline-corrected using the IviumSoft software, to improve the
visualization and identification of peaks over the baseline without
introducing any artefact.
The GCE was polished using diamond spray (particle size 1 mm)
on a microcloth pad before each experiment. After polishing, the
GCE was rinsed thoroughly with Milli-Q water for 30 s; then it was
electrochemically pretreated, by recording various DP voltammo-
grams in buffer supporting electrolyte until a steady state baseline
voltammogram was obtained.
For the DP voltammetric study, solutions of 50 mM of different
Ab peptides, human Ab1-40 and Ab1-42, control inverse Ab40-1 and
Ab42-1, mutants Ab1-40Phe10 and Ab1-40Nle35, and rat Ab1-40Rat,
were incubated at room temperature during 0 he48 h time periods.
Prior to each DP voltammetric experiment, 5 mL of the appropriate
solution was dropped onto the clean GCE surface and allowed to
spontaneously adsorb during 10 min. Then the GCE with adsorbed
Ab peptide was rinsed with Millipore Milli-Q water, placed in the
electrochemical cell and DP voltammograms were recorded in
0.1 M phosphate buffer pH 7.4 supporting electrolyte.
3. Results
The structural modifications of the human Ab1-40 and Ab1-42
isoform peptides (Scheme 1) were investigated by AFM and DP
voltammetry. The results were compared with control experiments
performed with the inverse sequences Ab40-1 and Ab42-1 peptides,
the mutants Ab1-40Phe10 and Ab1-40Nle35 peptides, and the rat Ab1-
40Rat peptide.
Stock solutions of 100 mM Ab peptides were kept at room tem-
perature up to 48 h, while fibrilization took place. During the
fibrilization process, at different aggregation states (0 h e freshly
prepared solution, 24 h and 48 h incubation), aliquots of stock so-
lutions were collected, diluted and analysed.
3.1. Human amyloids
3.1.1. Atomic force microscopy characterization
Unless stated otherwise, the human Ab1-40 and Ab1-42 peptides
were immobilized by spontaneous adsorption during 3 min on the
HOPG surface, using 1.0 mM and 10 mM of Ab1-40 or Ab1-42 peptide
solutions after different aggregation states (0 h e freshly prepared
solution, after 24 h and 48 h incubation). Very small concentrations
of Ab peptides have been chosen when compared with previous
studies, in order to observe the main Ab peptides time-course
species and the influence of the electrode substrate on the fibrili-
zation process.
3.1.1.1. Human Ab1-40 peptide. A control AFM experiment of human
Ab1-40 peptides, at 0 h incubation e freshly prepared solution,
spontaneously adsorbed onto mica, from 1.0 mM Ab1-40 peptide,
was performed. The images showed an Ab1-40 peptide film with
globular morphology that completely covered the mica surface,
composed of small aggregates densely packed together (Fig. 1A).
AFM images under the same conditions of Ab1-40 peptides, at 0 h
incubation, spontaneously adsorbed onto HOPG, from 1.0 mM Ab1-40
peptide, showed almost no adsorption (Fig. 1B). Very rarely, a
number of very thin, fibril-like adsorbed structures, FA, of
0.65 ± 0.10 nm height and heterogeneous length, some over 1 mm,
40 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47

Fig. 1. AFM images of Ab1-40 peptides after different incubation times, immobilised by spontaneous adsorption onto (A) mica and (B, C) HOPG. Each scale bar represented 100 nm.

were observed. influence on the self-assembling properties of human Ab1-40 pep-

The Ab1-40 peptide different adsorption onto HOPG and mica tide in a b-sheet configuration. The HOPG cleavage process reveals
suggested that immediately after the solution preparation, the Ab1- many different atomically flat terraces on the basal plane of
40 peptide was in a soluble, monomeric, random coil or a-helix graphite; on these terraces, the individual crystalline domains of
conformation, conferring to the molecule a very hydrophilic char- the polycrystalline HOPG are rotated against each other. The AFM
acter, with the hydrophobic residues hidden, which favoured the images showed that both FA fibril-like adsorbed species and P
adsorption onto mica. Although the adsorption of the hydrophilic protofibrils adsorbed on the HOPG atomically flat terraces were
Ab1-40 peptide on the hydrophobic HOPG is not favoured, the HOPG oriented along three directions, at 60 and 120 to each other,
induced a rapid change of the Ab1-40 peptide secondary structure imposed by the threefold symmetry of the graphite substrate
into a b-sheet configuration, leading to the spontaneous adsorption (Fig. 2D-ii and 2D-iii). These FA fibril-like species and P protofibrils
of an Ab1-40 peptide monolayer assembled as FA fibril-like were not oriented at a fixed angle relative to the direction of the
structures. step edges of the HOPG, indicating that their orientation was not
AFM images of Ab1-40 peptides after 24 h (Fig. 1C) and 48 h in- induced by particular surface defects on the HOPG step edges, but
cubation, spontaneously adsorbed onto HOPG from 1.0 mM Ab1-40, induced by the Ab1-40 peptide epitaxial growth on the HOPG
also showed only a small number of FA fibril-like adsorbed struc- terraces.
tures, with a few embedded A aggregates into their structure.
AFM images of Ab1-40 peptides at 0 h incubation, spontaneously
3.1.1.2. Human Ab1-42 peptide. AFM images of Ab1-42 peptides, at
adsorbed onto HOPG from more concentrated 10 mM Ab1-40 pep-
0 h incubation e freshly prepared solution, spontaneously adsor-
tide, showed the formation of a 1.97 ± 0.38 nm height film that
bed onto HOPG from 1.0 mM Ab1-42 peptide, showed a very week
covered the HOPG almost completely (Fig. 2A), corresponding to
adsorption onto HOPG (Fig. 3A). The majority of Ab1-42 peptides
the adsorption of the Ab1-40 peptide in a random coil or an a-helix
adsorbed on the plane HOPG terraces as RC random coiled struc-
conformation. Embedded into the film structure, a number of A
tures, of 0.45 ± 0.16 nm height (Fig. 3A-ii), associated with the
spherical aggregates and short P protofibrils of 2.5e6.0 nm height
adsorption of Ab1-42 peptide monomers in random coil and/or a-
were also observed, formed by the Ab1-40 peptide in a b-sheet
helix conformations. Occasionally the RC structures get oriented
configuration. AFM images of Ab1-40 peptides after 24 h incubation
along the HOPG step edges (Fig. 3A-iii). Some RC structures started
showed a decrease of the adsorption, due to a decrease in the
to aggregate, forming 1.0e2.5 nm height A aggregates (Fig. 3A-i, red
number of free Ab1-40 peptide monomers, and an increase in the
arrows).
number of A spherical aggregates of heterogeneous, 2.0e7.0 nm,
In addition, similarly to the Ab1-40 peptide, the HOPG hydro-
heights (Fig. 2B). Increasing the incubation time to 48 h (Fig. 2C),
phobic surface played a crucial role in amyloid fibril formation,
the number of aggregates decreased drastically, and a number of
inducing rapid changes of the Ab1-42 peptide into a b-sheet
long, smooth and branched F fibrils formed in solution could be
configuration that favoured the formation of very thin FA fibril-like
observed, thicker than FA fibril like adsorbed species that self-
structures onto HOPG (Fig. 3A-i). The FA fibril-like adsorbed struc-
assembled directly onto HOPG. The HOPG surface coverage by
tures were 0.37 ± 0.09 nm height, slightly smaller than the RC
Ab1-40 peptides was very low.
height, but with lengths up to 430 nm. Many of the A aggregates
In a different experiment, 50 mL of 10 mM Ab1-40 peptides after
were accumulating on the FA fibril-like structures, which behaved
48 h incubation, were placed onto HOPG, and the solution was
as nucleation points for the formation of larger and thicker Ab1-42
allowed to evaporate completely. Afterwards the HOPG modified
peptide fibrils.
surface was rinsed with Milli-Q water and dried in a sterile atmo-
AFM images of Ab1-42 peptides incubated for 24 h (Fig. 3B-i and
sphere. As expected, large-scale AFM images showed the formation
amplification 3B-ii) and 48 h (Fig. 3C), showed a decrease and
of thick, long and smooth F fibrils of 2.0e7.0 nm height formed in
disappearance of the RC structures, an increase of A aggregates, and
solution (Fig. 2D-i). Higher-amplification AFM images showed the
the occurrence of two types of fibrils, FA fibril-like adsorbed
adsorption of thin, up to 2.3 nm height FA fibril-like adsorbed
structures and the F fibrils. The FA fibril-like adsorbed structures,
species, as well as A aggregates and smooth P protofibrils, of
self-assembled directly on the HOPG surface, were all thin (gener-
1.6e5.0 nm height, formed directly on the HOPG surface (Fig. 2D-ii
ally less than 2.0 nm height) and with beaded morphology due to
and larger amplification 2D-iii).
embedded A aggregates (Fig. 3C-ii), whereas the F fibrils, self-
The hydrophobic structure of the HOPG surface had a strong
assembled in solution, and were thick (larger than 2.0 nm height)
 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47 41

Fig. 2. AFM images of Ab1-40 peptides after different incubation times, immobilized onto HOPG: (AeC) spontaneous adsorption and (D) evaporation.

Fig. 3. AFM images of Ab1-42 peptides after different incubation times, immobilized by spontaneous adsorption onto HOPG. Each scale bar represented 100 nm.
42 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47
and very smooth (Fig. 3C-iii). The F fibrils diameter increased with
increasing the incubation time, and only their head terminals
showed the presence of A globular aggregates (Fig. 3C-iii, red ar-
rows), meaning that their growing was in progress.
AFM images of Ab1-42 peptides, at 0 h incubation e freshly
prepared solution, spontaneously adsorbed onto HOPG from more
concentrated 10 mM Ab1-42 peptide, showed a 1.58 ± 0.32 nm height
film, with embedded A spherical aggregates of 2.0e5.5 nm height
(Fig. 4A and 4A inset). FA fibril-like structures with beaded structure
and heterogeneous diameter size were rarely observed. As the A
aggregates and FA fibril-like structures were adsorbed on top the
Ab1-42 peptide film, their height was overestimated. As previously
observed for the adsorption from 1.0 mM Ab1-42 peptide (Fig. 3A),
the thin fibrils formed at the HOPG surface served as nucleation
points for A aggregates (Fig. 4A), an intermediary step in the for-
mation of thicker fibrils. Increasing the incubation time to 24 h
(Fig. 4B) and 48 h, similar patterns of adsorption were observed, but
the number of A aggregates increased.
For all incubation times, the Ab1-42 peptide showed an increased
adsorption onto HOPG (Figs. 3, 4A and B) when compared with the
Ab1-40 peptide immobilized under the same conditions (Figs. 1B, C
and 2AeC), due to the presence of the hydrophobic amino acid
Fig. 4. AFM images of Ab1-42 peptides after different incubation times, immobilized by spontaneous adsorption onto HOPG. Each scale bar represented 100 nm.
residues isoleucine I and alanine A at the -C terminus, which
resulted in an increase of the total hydrophobicity from 42.50% for
the Ab1-40 peptide to 45.24% for the Ab1-42 peptide.
Due to its faster fibrilization, the human Ab1-42 peptide has been
chosen as a model to investigate the morphology after a very long
incubation time. AFM images of Ab1-42 peptides after 2 weeks in-
cubation, spontaneously adsorbed from 10 mM Ab1-42, showed
packs of thin, 1.60 ± 0.23 nm height, FA fibril-like adsorbed species
densely packed together (Fig. 4C-i and 4C-ii). Except very few A
aggregates and RC random coiled structures (Fig. 4C-ii and 4C-ii
inset), no thick F fibrils were observed, meaning that the F fibrils
self-assembled in solution were extremely compact, and conse-
quently were not spontaneously adsorbing onto the HOPG. The
packs self-assembled mainly along three directions on the HOPG
flat terraces, and rotated at 60 and 120 relative to each other, as
imposed by the graphite symmetry (Fig. 4C-i). This epitaxial growth
was common to both human Ab1-40 and Ab1-42 peptide sequences.
3.1.2. Voltammetric characterization
The protein electrochemical oxidation is due either to exposed
electroactive amino acids residues or to the redox prosthetic
groups. Among twenty amino acids, only five are electroactive:
 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47
tyrosine (Y), tryptophan (W), histidine (H), cysteine (C) and
methionine (M) [36]. At GCE at physiological pH, the free amino
acids oxidation potentials are: Ep ¼ þ0.55 V for cysteine [37],
Ep ¼ þ0.65 V for tyrosine [31,38] and tryptophan [39] and
Ep ~ þ1.10 V for histidine [36] and methionine [37]. The protein
electroactive amino acid residues can be easier oxidized when they
are present on the protein exterior, in contact with the electrode
surface.
The DP voltammograms at the Ab1-40 peptide and Ab1-42 peptide
modified GCE, 0 h incubation e freshly prepared solution, showed,
for each peptide, two consecutive oxidation peaks, Fig. 5A. The first
anodic charge transfer reaction corresponded to the tyrosine Y10
amino acid residue oxidation, at Ep ¼ þ0.65 V, and the second more
positive peak corresponded to the three histidines, H6, H13 and H14,
and the methionine M35 amino acid residues oxidation, at
Ep ¼ þ1.00 V (Scheme 1).
The Ab1-40 and Ab1-42 peptides after 24 h incubation showed a
decrease of the anodic peak current of both oxidation peaks, Fig. 5B.
This indicated a decrease of the available Ab peptide monomers,
adsorbed at the GCE surface, due to the formation of amyloid
aggregated structures in solution, with the electroactive amino acid
residues inside the molecule [19]. The oxidation peak currents
decrease continued until 48 h incubation, Fig. 5C, when the faradic
signal of the DP voltammograms tended to merge with the back-
ground current, meaning that the Ab1-40 and Ab1-42 peptides
monomers have been completely converted into highly ordered
aggregates and/or fibrils with the electroactive amino acid residues
buried inside and not able to reach the electrode surface.
The inverse Ab40-1 and Ab42-1 peptides did not aggregate.
Although the human and inverse peptide structures presented
exactly the same amino acids residues and hydrophobic/hydro-
philic pattern, the secondary structures were very different. Since
they contain the same residues as the corresponding native se-
quences, they were used as a control in DP voltammetry, in order to
confirm that the decrease of the oxidation peaks is due to aggre-
gation. For the inverse Ab40-1 and Ab42-1 peptides, the same
experimental procedure as for human Ab1-40 and Ab1-42 peptides
was applied.
The DP voltammograms recorded at the inverse Ab40-1 and Ab42-
Fig. 5. DP voltammograms baseline corrected of human Ab1-40 and Ab1-42 peptides and inverse Ab40-1 and Ab42-1 peptides, after different incubation times.
43
1 peptides modified GCE, at 0 h incubation e freshly prepared so-
lution, showed two oxidation peaks, Fig. 5A, at the same potentials
as those for human Ab1-40 and Ab1-42 peptides, corresponding to
the same residues, i.e., the first oxidation peak to tyrosine Y10 amino
acid residue oxidation, and the second more positive peak corre-
sponded to the three histidines, H6, H13 and H14, and the methio-
nine M35 amino acid residues oxidation. Increasing the incubation
time to 24 h, Fig. 5B, and to 48 h, Fig. 5C, no changes in the oxidation
potentials and peak currents were observed.
3.2. Mutant human and rat amyloids
In order to determine how specific amino acids residues influ-
ence the human Ab peptides fibrilization, adsorption and redox
behaviour, the mutant Ab1-40Phe10 and Ab1-40Nle35 peptides and rat
Ab1-40Rat peptide were also studied.
In the mutant Ab1-40Phe10 peptide, the tyrosine Y10 amino acid
residue is substituted by phenylalanine F10, and in the mutant Ab1-
35
peptide, the methionine M35 is substituted by norleucine
40Nle
Nle35 (Scheme 1). In the Ab1-40Rat peptide, three amino acids are
substituted: argine R5 by glycine G5, tyrosine Y10 by phenylalanine
F10, and histidine H13 by arginine R13 (Scheme 1).
3.2.1. Atomic force microscopy characterization
The mutant Ab1-40Phe10 and Ab1-40Nle35 peptides, after different
aggregation states (0 h e freshly prepared solution, after 24 h and
48 h incubation), were immobilized by spontaneous adsorption
during 3 min on the HOPG surface, using solutions of 10 mM Ab1-
40Phe
10
peptide and 1.0 and 5.0 mM Ab1-40Nle35 peptide.
AFM images of Ab1-40Phe10 peptides, 0 h incubation e freshly
prepared solution, spontaneously adsorbed onto HOPG from
10 mM Ab1-40Phe10, showed a 1.65 ± 0.22 nm height network film,
with several 2.0e3.0 nm height A spherical aggregates embedded
into its structure (Fig. 6A). The Ab1-40Phe10 peptides were more
loosely packed and no protofibriles or fibrils were observed,
comparing to the human Ab1-40 peptide for the same solution
concentration (Fig. 2A). The Ab1-40Phe10 peptides were very stable
over time, and AFM images of Ab1-40Phe10 peptide after 24 h and
48 h incubation showed similar results.
44 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47
Fig. 6. AFM images of mutant Ab peptides, Ab1-40Phe10 and Ab1-40Nle35, at 0 h incubation e freshly prepared solution, immobilized by spontaneous adsorption onto HOPG. Each
scale bar represented 100 nm.
AFM images of Ab1-40Nle35 peptides 0 h incubation e freshly
prepared solution, spontaneously adsorbed onto HOPG from
1.0 mM Ab1-40Nle35, showed the adsorption of only RC random
coiled structures (Fig. 6B), but the adsorption increased when
compared with the Ab1-40 peptide (Fig. 1B). AFM images of Ab1-
35
40Nle at 0 h incubation, spontaneously adsorbed onto HOPG from
a more concentrated solution of 5.0 mM Ab1-40Nle35, showed a
HOPG completely covered surface by a multilayer film with very
small pores (Fig. 6C). The increased adsorption of Ab1-40Nle35
peptide compared with human Ab1-40 peptide is due to the higher
hydrophobicity of Nle35 when compared with Met35, which
enhanced the adsorption of the -C terminus hydrophobic tail of the
mutant Ab1-40Nle35 peptide. Increasing the incubation time to 24 h
and 48 h no protofibriles or fibrils were observed.
3.2.2. Voltammetric characterization
The DP voltammograms, in 0.1 M phosphate buffer pH 7.4, of the
mutant Ab1-40Phe10 and Ab1-40Nle35 peptides and rat Ab1-40Rat
peptide modified GCE, showed different aggregation and redox
behaviour (Fig. 7), when compared with human Ab1-40 peptide
(Fig. 5 ).
The DP voltammograms recorded at the Ab1-40Phe10 peptides
modified GCE, at 0 h incubatione freshly prepared solution,
showed only the second oxidation peak, corresponding to the H6,
H13, H14 and M35 amino acid residues oxidation, at Ep ¼ þ 1.00 V
(Fig. 7A ). The disappearance of the first oxidation peak is due to
the lack of Y10 in the Ab1-40Phe10 peptide, which was substituted by
the non-electroactive F10. Increasing the incubation time to 24 h
(Fig. 7B ) and 48 h (Fig. 7C ), the oxidation potential and peak
current remained unchanged. This was confirmed integrated area
of the peaks: 2.28  1010 V A at 0 h, 2.41  1010 V A at 24 h, and
2.37  1010 V A at 48 h.
The DP voltammograms recorded at the Ab1-40Nle35 peptide
modified GCE, at 0 h incubation e freshly prepared solution,
showed two oxidation peaks (Fig. 7A ), similar to what was ob-
tained for human Ab1-40 peptide (Fig. 5A ). Although M35 was
substituted by the non-electroactive Nle35, no significant second
oxidation peak current decrease was observed, when compared
with human Ab1-40 peptide (Fig. 5 ). Increasing the incubation
time, a decrease of both Ab1-40Nle35 peptide oxidation peak cur-
rents was observed up to 24 h incubation (Fig. 7B ) , then the peak
currents remained constant (Fig. 7C , 48 h incubation).
Comparing with human Ab1-40 peptide, in the rat Ab1-40Rat
peptide the arginine R5 is substituted by glycine G5, tyrosine Y10 by
phenylalanine F10, and histidine H13 by arginine R13. The electro-
chemical oxidation of adsorbed Ab1-40Rat peptide showed the
disappearance of the first oxidation peak (Fig. 7 ), due to the lack
of Y10, similar to what was obtained for Ab1-40Phe10 peptide
(Fig. 7 ). In addition, the Ab1-40Rat peptide second oxidation peak
presented lower currents when compared with human Ab1-40
peptide and mutant Ab1-40Phe10 peptide, due to the lack of H13 and
the substitution of R5 by G5, which may also result in less avail-
ability to oxidation of the neighbour electroactive H6 residue
(Scheme 1).
4. Discussion
The Ab peptides undergo morphological changes, from soluble,
monomeric random coil or a-helix conformations into aggregated
b-sheet structures, a fibrillization process that involved numerous
intermediates, including the formation of low-molecular weight
structures (e.g. dimers, trimers, and tetramers), globular oligomers
and protofibrils [4,13e16]. Although, these oligomers and proto-
fibrils are considered to be primarily responsible to the neuro-
degeneration in AD, they are difficult to be investigated due to the
fast aggregation rate of the Ab peptides.
The conformational change of the Ab peptide structure, within
the aggregation process, was also influenced by a number of factors,
such as peptide length, solvent hydrophobicity, ionic strength, pH,
peptide concentration, initial aggregation state, buffer, peptide
counterions, etc. [14,15,18,20,21,40,41]. The effectiveness of a given
salt to reduce protein's solubility is predicted to some degree by the
Hofmeister series and the protein charge. The Hofmeister series
ranks the ability of different anions and cations to stabilise (salt-
out) or destabilise (salt-in) proteins.
In this context, the aim of this study was to investigate the Ab1-
 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47
Fig. 7. DP voltammograms baseline corrected of human mutant Ab1-40Phe10 and Ab1-40Nle35 peptides and of rat Ab40-1Rat peptide, after different incubation times.
40 and Ab1-42 structural modifications, from monomers to fully-
formed fibrils, i.e. the intermediary structures that appeared dur-
ing fibrilization. Lower aggregation rates were promoted by incu-
bating the Ab peptide solutions at room temperature and in free
chloride media. The use of only phosphate buffer, without chloride,
limited the salt out effect induced by ionic species (Hofmeister
series).
The Ab1-40 and Ab1-42 peptides fibrilization was followed by DP
voltammetry, via the decrease and disappearance of the oxidation
peak currents of the electroactive residues (Fig. 5), and compared at
each particular time with the morphological structures observed by
AFM (Figs. 1e4).
The research novelty consisted on the investigation of the Ab
peptides redox behaviour, taking into consideration two electrode
process reactions, the first one corresponding to the Y oxidation,
and the second one corresponding to the H and M oxidation. It was
the first time that the second oxidation reaction of Ab peptides, the
H and M oxidation, was investigated.
The AFM and voltammetric results were consistent with the
in vitro fibrilization mechanism, observed for different Ab peptide
sequences [42e48]. The Ab peptide monomers followed a nucle-
ation process, leading to the formation of small nuclei, as observed
in Fig. 3A (red arrows). Once the nucleation process started, Ab free
monomers in solution were added to the nuclei and their di-
mensions grown forming various aggregate types, such as small
oligomers and long fibrils with different morphologies. This poly-
merization reaction followed first-order kinetics, and is thermo-
dynamically favoured by the hydrophobic regions of Ab peptides, as
well as their b-sheet secondary structure [14,48]. As fibrilization
proceeds, the electroactive residues become more protected, and
the transfer of electrons became more difficult.
The AFM studies showed differences between fibrils formed in
solution or onto the carbon surface. The HOPG played an important
role in amyloid fibril formation and adsorption, inducing rapid
changes of the Ab peptide monomeric random coil or a-helix
conformations into a b-sheet configuration. The fibrils formed onto
HOPG were thin, very reactive and presented a beaded morphology
due to the adsorption of Ab peptide aggregates onto their structure
(Fig. 3C-ii). These FA fibril-like adsorbed species presented epitaxial
growth onto HOPG (Fig. 2D-ii, 2D-iii, 4C-i, and 4C-ii). The fibrils
formed in solution grown attaching Ab peptide monomer units to
the end of the fibril and were long, continuous and twisted together
45
into helices (Fig. 3C-iii).
Using control sequences that: (i) do not aggregate (inverse Ab40-
1 and Ab42-1 peptides, Fig. 5, and rat Ab1-40Rat peptide, Fig. 7), and
(ii) Ab peptides specially designed to lack specific electroactive
amino acids (mutant Ab1-40Phe10 and Ab1-40Nle35 peptides, Figs. 6
and 7), it was demonstrated that the decrease of the Ab peptide
oxidation peaks current was effectively due to aggregation and not
due to another process. For all these control Ab peptide sequences,
no protofibriles or fibrils were observed, and the oxidation peaks
current did not changed increasing the incubation time from 24 h
to 48 h. To the best of our knowledge, this represents the first study
where control voltammetric experiments have been performed, in
order to fully understand how the electron transfer reaction was
influenced by both the specific amino acid side-chain position, and
the fibrilization process.
Generally the Ab peptide secondary structure is related with the
specific arrangement pattern of the residues i, iþ2 (compatible with
the formation of b-sheets) or i, iþ4 (compatible with the formation
of a helices). However, it is not fully proved in proteomics if the
order N to C terminus or C to N terminus dictates this specific
arrangement pattern of amino acid residues to form the secondary
and tertiary structures [49]. The specific arrangement pattern of the
residues of the native and inverse peptides was different; therefore
the Ab conformations for native and inverse sequences are
different. Using the inverse sequences of human Ab peptides (i.e.
Ab40-1 and Ab42-1 peptides), the human mutants Ab1-40Phe10 and
Ab1-40Nle35 peptides and the rat Ab1-40Rat peptide, it was observed
that the Ab peptides aggregation was a process dependent on the
sequence (i.e. primary structure) and on the specific arrangement
pattern of the amino acid residues, which involved a substantial
structural recognition, therefore affecting their adsorption
morphology and redox behaviour.
5. Conclusions
The time-dependent structural modifications of the human Ab
isoform peptides, containing 40 and 42 amino acids, at the single-
molecule level, that take place at room temperature and in chloride
free media, were investigated. An excellent correlation was
observed between the Ab peptides structural changes and their
redox behaviour. The fibrilization process was detected by AFM via
the adsorption of Ab peptide monomers as random coiled
46 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47
structures, and Ab peptide in b-sheet configurations as aggregates,
protofibrils, and fibrils, and by DP voltammetry via the decrease of
the two oxidation peak currents, corresponding to the oxidation of
tyrosine (the first peak) and of histidine and methionine (the sec-
ond peak). The combination of these two techniques, AFM and DP
voltammetry, revealed to be very effective to study the interme-
diary structures formed both in solution and at the surface of hy-
drophobic carbon electrodes, and will trigger their use for a whole
range of studies concerning amyloid fibrilization involved in the
development of AD, as well as of other amyloid diseases, such as
Parkinson's and type II diabetes.
Acknowledgment
~o para a Cie
Financial support from Fundaça ^ncia e Tecnologia
(FCT), Grants SFRH/BPD/92726/2013 (A.-M. Chiorcea-Paquim),
SFRH/BPD/80195/2011 (T. A. Enache) projects PTDC/SAU-BMA/
118531/2010, PTDC/QEQ-MED/0586/2012, UID/EMS/00285/2013
CEMUC-R (Research Unit 285), (co-financed by the European
Community Fund FEDER), FEDER funds through the program
COMPETE e Programa Operacional Factores de Competitividade,
are gratefully acknowledged.
References
[1] R. Grünberg, L. Serrano, Strategies for protein synthetic biology, Nucleic Acids
Res. 38 (2010) 2663e2675, http://dx.doi.org/10.1093/nar/gkq139.
[2] B. Alberts, A. Johnson, J. Lewis, M. Raff, K. Roberts, P. Walter, The Shape and
Structure of Proteins, 2002. http://www.ncbi.nlm.nih.gov/books/NBK26830/.
[3] I. Moreno-Gonzalez, C. Soto, Misfolded protein aggregates: mechanisms,
structures and potential for disease transmission, Semin. Cell {&} Dev. Biol. 22
(2011) 482e487, http://dx.doi.org/10.1016/j.semcdb.2011.04.002.
[4] P. Salahuddin, Protein Folding, Misfolding, Aggregation And Amyloid Forma-
tion: Mechanisms of Ab Oligomer Mediated Toxicities, J. Biochem. Mol. Biol.
Res. 1 (2015) 36e45. http://www.ghrnet.org/index.php/jbmbr/article/view/
1027/1397.
[5] L. Bertram, R.E. Tanzi, The genetic epidemiology of neurodegenerative disease,
J. Clin. Investig. 115 (2005) 1449e1457, http://dx.doi.org/10.1172/JCI24761.
[6] J. Williamson, J. Goldman, K.S. Marder, Genetic aspects of Alzheimer disease,
Neurologist 15 (2009) 80e86, http://dx.doi.org/10.1097/
NRL.0b013e318187e76b.
[7] L. Bertram, R.E. Tanzi, Thirty years of Alzheimer's disease genetics: the im-
plications of systematic meta-analyses, Nat. Rev. Neurosci 9 (2008) 768e778,
http://dx.doi.org/10.1038/nrn2494.
[8] Dementia statistics j Alzheimer’s Disease International, (n.d.). http://www.alz.
co.uk/research/statistics (accessed August 18, 2015).
~
[9] A. Alzheimer, Aœber eine eigenartige Erkrankung der Hirnrinde, Allg. Z. Psy-
chiatr. Psych. Gerichtl. Med. 64 (1907) 146e148.
[10] B.D. James, S.E. Leurgans, L.E. Hebert, P.A. Scherr, K. Yaffe, D.A. Bennett,
Contribution of Alzheimer disease to mortality in the United States, Neurology
82 (2014) 1045e1050, http://dx.doi.org/10.1212/WNL.0000000000000240.
[11] C.R. Jack, D.S. Knopman, W.J. Jagust, L.M. Shaw, P.S. Aisen, M.W. Weiner, et al.,
Hypothetical model of dynamic biomarkers of the Alzheimer's pathological
cascade, Lancet Neurol 9 (2010) 119e128, http://dx.doi.org/10.1016/S1474-
4422(09)70299-6.
[12] D.J. Selkoe, Alzheimer's Disease: Genes, Proteins, and Therapy, Physiol Rev 81
(2001) 741e766. http://physrev.physiology.org/content/81/2/741.short.
[13] D.M. Walsh, D.J. Selkoe, A beta oligomers - a decade of discovery,
J. Neurochem. 101 (2007) 1172e1184, http://dx.doi.org/10.1111/j.1471-
4159.2006.04426.x.
[14] I.W. Hamley, The amyloid beta peptide: a chemist's perspective. Role in Alz-
heimer's and fibrillization, Chem. Rev. 112 (2012) 5147e5192, http://
dx.doi.org/10.1021/cr3000994.
[15] M. Amaro, K. Kubiak-Ossowska, D.J.S. Birch, O.J. Rolinski, Initial stages of beta-
amyloid Ab 140 and Ab 142 oligomerization observed using fluorescence
decay and molecular dynamics analyses of tyrosine, Methods Appl. Fluoresc 1
(2013) 15006, http://dx.doi.org/10.1088/2050-6120/1/1/015006.
[16] C.S. Atwood, M.E. Obrenovich, T. Liu, H. Chan, G. Perry, M.A. Smith, et al.,
Amyloid-beta: a chameleon walking in two worlds: a review of the trophic
and toxic properties of amyloid-beta, Brain Res. Brain Res. Rev. 43 (2003)
1e16. http://www.ncbi.nlm.nih.gov/pubmed/14499458.
[17] J.E. Straub, D. Thirumalai, Principles governing oligomer formation in amy-
loidogenic peptides, Curr. Opin. Struct. Biol. 20 (2010) 187e195. http://www.
sciencedirect.com/science/article/pii/S0959440X10000035.
[18] S. Tomaselli, V. Esposito, P. Vangone, N.A.J. van Nuland, A.M.J.J. Bonvin,
R. Guerrini, et al., The alpha-to-beta conformational transition of Alzheimer's
Abeta-(1-42) peptide in aqueous media is reversible: a step by step
conformational analysis suggests the location of beta conformation seeding,
Chembiochem 7 (2006) 257e267, http://dx.doi.org/10.1002/cbic.200500223.
[19] M. Vestergaard, K. Kerman, M. Saito, N. Nagatani, Y. Takamura, E. Tamiya,
A rapid label-free electrochemical detection and kinetic study of Alzheimer's
amyloid beta aggregation, J. Am. Chem. Soc. 127 (2005) 11892e11893, http://
dx.doi.org/10.1021/ja052522q.
[20] I.H. Cheng, K. Scearce-Levie, J. Legleiter, Accelerating amyloid-b fibrillization
reduces oligomer levels and functional deficits in Alzheimer disease mouse
models, J. Biol. Chem. 282 (2007) 23818e23828. http://www.jbc.org/content/
282/33/23818.abstract.
[21] M. Pannuzzo, A. Raudino, D. Milardi, C. La Rosa, M. Karttunen, a-helical
structures drive early stages of self-assembly of amyloidogenic amyloid
polypeptide aggregate formation in membranes, Sci. Rep 3 (2013) 2781,
http://dx.doi.org/10.1038/srep02781.
[22] C.L. Masters, D.J. Selkoe, Biochemistry of amyloid b-protein and amyloid de-
posits in Alzheimer disease, Cold Spring Harb. Perspect. Med 2 (2012)
a006262, http://dx.doi.org/10.1101/cshperspect.a006262.
[23] D.J. Selkoe, The molecular pathology of Alzheimer's disease, Neuron 6 (1991)
487e498, http://dx.doi.org/10.1016/0896-6273(91)90052-2.
[24] P.D. Mehta, T. Pirttila€, S.P. Mehta, E.A. Sersen, P.S. Aisen, H.M. Wisniewski,
Plasma and cerebrospinal fluid levels of amyloid b proteins 1-40 and 1-42 in
Alzheimer disease, Arch. Neurol 57 (2000) 100, http://dx.doi.org/10.1001/
archneur.57.1.100.
[25] S. Vivekanandan, J.R. Brender, S.Y. Lee, A. Ramamoorthy, A partially folded
structure of amyloid-beta(1-40) in an aqueous environment, Biochem. Bio-
phys. Res. Commun 411 (2011) 312e316, http://dx.doi.org/10.1016/
j.bbrc.2011.06.133.
[26] S. Narayanan, B. Reif, Characterization of chemical exchange between soluble
and aggregated states of beta-amyloid by solution-state NMR upon variation
of salt conditions, Biochemistry 44 (2005) 1444e1452, http://dx.doi.org/
10.1021/bi048264b.
[27] K. Klement, K. Wieligmann, J. Meinhardt, P. Hortschansky, W. Richter,
M. Fa€ndrich, Effect of different salt ions on the propensity of aggregation and
on the structure of Alzheimer's abeta(1-40) amyloid fibrils, J. Mol. Biol. 373
(2007) 1321e1333, http://dx.doi.org/10.1016/j.jmb.2007.08.068.
[28] A.M. Oliveira Brett, A.-M. Chiorcea Paquim, V.C. Diculescu, T.S. Oretskaya,
Synthetic oligonucleotides: AFM characterisation and electroanalytical
studies, Bioelectrochemistry 67 (2005) 181e190, http://dx.doi.org/10.1016/
j.bioelechem.2004.06.008.
[29] O. Corduneanu, A.-M. Chiorcea-Paquim, V. Diculescu, S.M. Fiuza,
M.P.M. Marques, A.M. Oliveira-Brett, DNA interaction with palladium chelates
of biogenic polyamines using atomic force microscopy and voltammetric
characterization, Anal. Chem. 82 (2010) 1245e1252, http://dx.doi.org/
10.1021/ac902127d.
[30] V.C. Diculescu, A.-M. Chiorcea-Paquim, R. Eritja, A.M. Oliveira-Brett, Evalua-
tion of the structureeactivity relationship of thrombin with thrombin binding
aptamers by voltammetry and atomic force microscopy, J. Electroanal. Chem.
656 (2011) 159e166, http://dx.doi.org/10.1016/j.jelechem.2010.11.037.
[31] V.C. Diculescu, T.A. Enache, Electrochemical evaluation of Abelson tyrosine-
protein kinase 1 activity and inhibition by imatinib mesylate and dan-
usertib, Anal. Chim. Acta 845 (2014) 23e29, http://dx.doi.org/10.1016/
j.aca.2014.06.025.
[32] P. Lopes, M. Xu, M. Zhang, T. Zhou, Y. Yang, C. Wang, et al., Direct electro-
chemical and AFM detection of amyloid-b peptide aggregation on basal plane
HOPG, Nanoscale 6 (2014) 7853e7857, http://dx.doi.org/10.1039/c4nr02413c.
[33] Z. Wang, C. Zhou, C. Wang, L. Wan, X. Fang, C. Bai, AFM and STM study of beta-
amyloid aggregation on graphite., Ultramicroscopy. 97 73e79. doi:10.1016/
S0304-3991(03)00031-7.
[34] H. Yang, M. Pritzker, S.Y. Fung, Y. Sheng, W. Wang, P. Chen, Anion effect on the
nanostructure of a metal ion binding self-assembling peptide, Langmuir 22
(2006) 8553e8562, http://dx.doi.org/10.1021/la061238p.
[35] X. Mao, X. Ma, L. Liu, L. Niu, Y. Yang, C. Wang, Structural characteristics of the
beta-sheet-like human and rat islet amyloid polypeptides as determined by
scanning tunneling microscopy, J. Struct. Biol. 167 (2009) 209e215, http://
dx.doi.org/10.1016/j.jsb.2009.05.009.
[36] T.A. Enache, A.M. Oliveira-Brett, Peptide methionine sulfoxide reductase A
(MsrA): direct electrochemical oxidation on carbon electrodes, Bio-
electrochemistry 89 (2013) 11e18, http://dx.doi.org/10.1016/
j.bioelechem.2012.08.004.
[37] T.A. Enache, A.M. Oliveira-Brett, Boron doped diamond and glassy carbon
electrodes comparative study of the oxidation behaviour of cysteine and
methionine, Bioelectrochemistry 81 (2011) 46e52, http://dx.doi.org/10.1016/
j.bioelechem.2011.02.001.
[38] T.A. Enache, A.M. Oliveira-Brett, Phenol and para-substituted phenols elec-
trochemical oxidation pathways, J. Electroanal. Chem. 655 (2011) 9e16,
http://dx.doi.org/10.1016/j.jelechem.2011.02.022.
[39] T.A. Enache, A.M. Oliveira-Brett, Pathways of electrochemical oxidation of
indolic compounds, Electroanalysis 23 (2011) 1337e1344, http://dx.doi.org/
10.1002/elan.201000671.
[40] W.B. Stine, L. Jungbauer, C. Yu, M.J. LaDu, Preparing synthetic Ab in different
aggregation states, Methods Mol. Biol. 670 (2011) 13e32, http://dx.doi.org/
10.1007/978-1-60761-744-0{_}2.
[41] S.K. Pachahara, N. Chaudhary, C. Subbalakshmi, R. Nagaraj, Hexa-
fluoroisopropanol induces self-assembly of b-amyloid peptides into highly
ordered nanostructures, J. Pept. Sci. 18 (2012) 233e241, http://dx.doi.org/
 T.A. Enache et al. / Analytica Chimica Acta 926 (2016) 36e47
10.1002/psc.2391.
[42] A. Trovato, F. Chiti, A. Maritan, F. Seno, Insight into the structure of amyloid
fibrils from the analysis of globular proteins, PLoS Comput. Biol. 2 (2006) e170,
http://dx.doi.org/10.1371/journal.pcbi.0020170.
[43] L.C. Serpell, Alzheimer's amyloid fibrils: structure and assembly, Biochim.
Biophys. Acta 1502 (2000) 16e30. http://www.ncbi.nlm.nih.gov/pubmed/
10899428.
[44] J.C. Rochet, P.T. Lansbury, Amyloid fibrillogenesis: themes and variations, Curr.
Opin. Struct. Biol. 10 (2000) 60e68. http://www.ncbi.nlm.nih.gov/pubmed/
10679462.
[45] L. Yu, R. Edalji, J.E. Harlan, T.F. Holzman, A.P. Lopez, B. Labkovsky, et al.,
Structural characterization of a soluble amyloid beta-peptide oligomer,
Biochemistry 48 (2009) 1870e1877, http://dx.doi.org/10.1021/bi802046n.
47
[46] O.S. Makin, L.C. Serpell, Structures for amyloid fibrils, FEBS J 272 (2005)
5950e5961, http://dx.doi.org/10.1111/j.1742-4658.2005.05025.x.
[47] A. Dubnovitsky, A. Sandberg, M.M. Rahman, I. Benilova, C. Lendel, T. H€ ard,
Amyloid-b protofibrils: size, morphology and synaptotoxicity of an engi-
neered mimic, PLoS One 8 (2013) e66101, http://dx.doi.org/10.1371/
journal.pone.0066101.
[48] B. Moores, E. Drolle, S.J. Attwood, J. Simons, Z. Leonenko, Effect of surfaces on
amyloid fibril formation, PLoS One 6 (2011) e25954, http://dx.doi.org/
10.1371/journal.pone.0025954.
[49] E. Lacroix, A.R. Viguera, L. Serrano, Reading protein sequences backwards,
Fold. {&} Des 3 (1998) 79e85, http://dx.doi.org/10.1016/S1359-0278(98)
00013-3.