Copper(II), nickel(II) and zinc(II) complexes of the N-terminal nonapeptide fragment of amyloid-β and its derivatives

Article abstract of DOI:10.1016/j.jinorgbio.2014.06.001

Copper(II), nickel(II) and zinc(II) complexes of the nonapeptide fragment of amyloid-β Aβ(1-9) (NH₂-DAEFRHDSG-NH₂) and its two derivatives: NH₂-DAAAAHAAA-NH₂ and NH ₂-DAAAAAHAA-NH₂ have bee

Full text of DOI:10.1016/j.jinorgbio.2014.06.001

Copper(II), nickel(II) and zinc(II) complexes of the N-terminal nonapeptide
fragment of amyloid-β and its derivatives
´
Agnes Grena´cs, Imre So´va´go´
PII:
S0162-0134(14)00162-7
DOI:
Reference:
doi: 10.1016/j.jinorgbio.2014.06.001
JIB 9534
To appear in:
Journal of Inorganic Biochemistry
Received date:
Revised date:
Accepted date:
3 March 2014
2 June 2014
2 June 2014
´
Please cite this article as: Agnes Gren´acs, Imre So´va´g´o, Copper(II), nickel(II) and zinc(II)
complexes of the N-terminal nonapeptide fragment of amyloid-β and its derivatives,
Journal of Inorganic Biochemistry (2014), doi: 10.1016/j.jinorgbio.2014.06.001
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

ACCEPTED MANUSCRIPT
Copper(II), nickel(II) and zinc(II) complexes of the N-terminal nonapeptide
fragment of amyloid-β and its derivatives
Ágnes Grenács and Imre Sóvágó*
Department of Inorganic and Analytical Chemistry, University of Debrecen, H-4010
Debrecen, Hungary
* Corresponding author: Tel.: +36 52 512900/22303; Fax: +36 52 518660, E-mail:
sovago@science.unideb.hu
1

ACCEPTED MANUSCRIPT
Abstract
Copper(II), nickel(II) and zinc(II) complexes of the nonapeptide fragment of amyloid-β Aβ(1-
9) (NH2-DAEFRHDSG-NH2) and its two derivatives: NH2-DAAAAHAAA-NH2 and NH2-
DAAAAAHAA-NH2 have been studied by potentiometric, UV-visible and CD spectroscopic
methods. The results reveal the primary role of the amino terminus of peptides in copper(II)
and nickel(II) binding. The formation of dinuclear complexes was also possible in the
copper(II) containing systems but only the first six amino acids from the amino terminus were
involved in metal binding in the physiologically relevant pH range. The coordination
chemistry of the two alanine mutated peptides is almost the same as that of the native
nonapeptide, but the thermodynamic stability of the copper(II) complexes of the mutants is
significantly reduced. This difference probably comes from the secondary interactions of the
polar side chains of Asp, Glu, Ser and Arg residues present in the native peptide. Moreover,
this difference reveals that the amino acid sequence of the N-terminal domains of amyloid
peptides is especially well suited for the complexation with copper(II) ions.
Keywords:
copper(II), nickel(II), zinc(II), amyloid-β, stability constants, CD spectroscopy
2

ACCEPTED MANUSCRIPT
1. Introduction
The role of metal ions in the biological processes associated with Alzheimer’s disease gave a
big impetus to the studies on the coordination chemistry of amyloid-β peptide and its
fragments. Huge number of papers has been published in this field in the past few years
including several reviews [1-12]. Most of the previous studies were devoted to the
characterization of the copper(II) and zinc(II) complexes but the metal ion promoted redox
chemistry of the peptides is also widely investigated. In spite of the extremely high number of
related publications the metal binding sites and the data reported for the thermodynamic
stability of complexes remain contradictory. This dichotomy partly comes from the low
solubility of the 42-residue amyloid-β peptide and especially its metal complexes. At the same
time it is widely accepted that the N-terminal domain of amyloid-β is the primary metal
binding site which is rich in polar and coordinating side chains. It is also obvious from the
previous studies that the N-terminal hexadecapeptide fragment, amyloid-β(1-16), is the most
promising candidate for stable interactions with transition metal ions. Its sequence
corresponds to NH2-DAEFRHDSGYEVHHQK-NH2 containing a free amino terminus from
Asp(1), three imidazole-N donors from His(6), His(13) and His(14) and four carboxylates
from Asp(1), Glu(3), Asp(7) and Glu(11) residues. Moreover, the peptide contains a
phenolate-O and amino-N donors from Tyr(10) and Lys(16) residues, respectively, but they
are generally not considered as metal binding sites. The high number of protonation sites and
the solubility problems make even the acid-base characterization of the peptide difficult. As a
consequence, the description of metal binding under physiological conditions only was in the
focus of most publications. The data obtained under these conditions are relevant from a
biological point of view but do not provide a complete description of the metal-ligand
interaction.
3

ACCEPTED MANUSCRIPT
The first pH-dependent solution equilibrium studies on the copper(II) complexes of
various fragments of amyloid-β were published more than ten years ago, but only the species
formed in equimolar samples were taken into account in these reports [13-14]. The results
obtained in our laboratories in the last few years [15-18], however, provided a more
comprehensive picture on the complex formation with these peptides. The synthesis of the
PEG (polyethyleneglycol)-conjugate of the peptide Aβ(1-16)PEG helped to overcome the
solubility problems, while the use of short fragments Aβ(1-6) and Ac-Aβ(8-16)Y10A made it
possible to suggest the major metal binding sites in solution. It was found that the
hexadecapeptide can bind as much as 4 copper(II) ions and the terminal-NH2, His(6), His(13)
and His(14) residues were identified as the major copper(II) binding sites [15]. Furthermore, a
high zinc(II) binding affinity of the peptide was reported with a preference for the internal
histidyl sites in metal binding [16]. In the case of nickel(II) the formation of only mono- and
di-nuclear species was suggested via the terminal amino and internal histidyl sites [17].
Moreover, it was reported that zinc(II) ions cannot replace but can alter the distributions of
copper(II) among the available binding sites [18]. More recent studies provided further
support for this observation [19,20].
The solution equilibria of the copper(II), nickel(II) and zinc(II) complexes of amyloid
fragments have been described in our previous publications [15-18] but some structural
details of these interactions remained unanswered, yet. For example, a high number of
different coordination isomers of the mononuclear species can coexist in these systems.
Among them the binding mode of His(6) residue is probably one of the most questionable,
because there are only five amide groups between Asp(1) and His(6) sites. Thus, the
saturation of the coordination spheres of two copper ions is not possible towards the N-
termini providing a chance for the similar processes at the C-terminal side of the peptide.
Systematic equilibrium and structural studies on the complexes of additional small fragments
4

ACCEPTED MANUSCRIPT
can help to answer this question. Now in this paper we report the synthesis of the nonapeptide
fragment Aβ(1-9) (=NH2-DAEFRHDSG-NH2) of the native peptide and its two derivatives:
NH2-DAAAAHAAA-NH2 and NH2-DAAAAAHAA-NH2. The comparison of the data
obtained for the complexes of the native and alanine mutated fragments helps to understand
the influence of weakly coordinating side chains in Aβ(1-9). The positions of histidines are
different (H6 and H7) in the two alanine-mutated fragments providing a chance for the
distinction of amide deprotonation and metal binding of amide nitrogens at the N-or C-
terminal side of the histidyl residues.
2. Experimental
2.1. Peptide synthesis and other materials
The N-terminally free peptides were obtained by solid phase peptide synthesis using the Fmoc
technique
with
the
sequences
of
NH2-AspAlaGluPheArgHisAspSerGly-NH2
(DAEFRHDSG), NH2-AspAlaAlaAlaAlaHisAlaAlaAla-NH2 (DAAAAHAAA) and NH2-
AspAlaAlaAlaAlaAlaHisAlaAla-NH2 (DAAAAAHAA). All chemicals and solvents used for
synthesis were obtained from commercial sources in the highest available purity and used
without further purification. Rink Amide AM resin, 2-(1-H-benzotriazole-1-yl)-1,1,3,3-
tetramethyluronium tetrafluoroborate (TBTU) and all N-fluorenylmethoxycarbonyl (Fmoc)-
protected amino acids (Fmoc-Asp(OtBu)-OH, Fmoc-Ala-OH, Fmoc-Glu(OtBu)-OH, Fmoc-
Phe-OH, Fmoc-Arg(Pbf)-OH, Fmoc-His(Trt)-OH, Fmoc-Ser(tBu)-OH and Fmoc-Gly-OH)
were from Novabiochem (Switzerland). N,N-diisopropyl-ethylamine (DIEA), trifluoroacetic
acid (TFA) and analytical grade N,N-dimethylformamide (DMF), were purchased from
Merck Kft. N-methyl-pyrrolidone (NMP), 1-hydroxybenzotriazole hydrate (HOBt∙H2O), 2,2’-
5

ACCEPTED MANUSCRIPT
(ethylenedioxy)diethanethiol, triisopropylsilane (TIS), 2-methyl-2-butanol and HPLC grade
trifluoroacetic acid were Sigma-Aldrich products. Dichloromethane (DCM), diethyl ether
(Et2O), acetic acid (AcOH) and piperidine were Molar solvents as well as acetic anhydride
and HPLC grade acetonitrile (ACN) were from VWR.
All peptides were synthesized by solid phase peptide synthesis using a microwave-assisted
Liberty 1 Peptide Synthesizer (CEM, Matthews, NC), introducing the amino acid derivatives
following the TBTU/HOBt/DIEA activation strategy on the Rink Amide AM resin. Removal
of the Fmoc was carried out by means of 20% piperidine/0.1 M HOBt∙H2O in DMF at 75 °C
with 35 Watts microwave power for 180 s. 0.5 M HOBt∙H2O /0.5 M TBTU in DMF and 2 M
DIEA in NMP were used for coupling at 75 °C with 25 Watts microwave power, for 300 s,
adding 4 times excess of amino acids. Finally, the N-terminal Fmoc group was removed as
described before. Cleaving of the peptides from their respective resins and the simultaneous
removal of the side chain protective groups were carried out by treatment with a mixture
containing TFA/TIS/H2O/2,2’-(ethylenedioxy)diethanethiol (94/2.5/2.5/1 v/v) at room
temperature for 1.5 h. After cleaving each solution the free peptide was separated from the
resin by filtration. Cold diethyl ether was used to precipitate the crude peptides from the
pertinent solution and to wash from the contaminants of the reagents of the synthesis and
cleaving agents. After separating from it, the products were dried under argon gas, redissolved
in water, and finally lyophilized. The purity of the peptides was checked by analytical rp-
HPLC analyses using a Jasco instrument, equipped with a Jasco MD-2010 plus
multiwavelength detector. The analyses were performed on a Vydac C18 chromatographic
column (250 x 4.6 mm, 300 Ǻ pore size, 5 m particle size) by eluting 2% of solvent A
(0.1% TFA in acetonitrile) and 98% of solvent B (0.1% TFA in water) at a flow rate of 1
mL/min monitoring the absorbance at 222 nm. Analytical rp-HPLC for all peptides and pH-
1
dependent H NMR spectra for the A(1-9) fragment was used to check the purity of the
6

ACCEPTED MANUSCRIPT
peptides. Potentiometric titrations also confirmed the purity and the identity of the peptides.
The purity was greater than 95% for all three peptides.
CuCl2, NiCl2 and ZnCl2 containing metal ion stock solutions were prepared from analytical
grade reagents and their concentrations were checked by gravimetry via the precipitation of
oxinates.
2.2. Potentiometric measurements
The pH-potentiometric titrations were performed with the use of carbonate-free stock solution
(0.2 M) of potassium hydroxide in 3 mL samples at 2 mM ligand concentration. The metal ion
to ligand ratios were selected as 1:1 and 2:1 for binary systems. While the equilibration of the
nickel(II)-peptide systems required a few minutes for each titration point in the pH range 8.0-
9.0 the copper(II) and zinc(II) complexes were formed in fast reactions. This is reflected in
the increased standard deviations of stability constants of the nickel(II) complexes comparing
with those of the other two metal ions. To ensure the absence of oxygen and carbon dioxide,
argon was bubbled through the samples and they were stirred by a VELP Scientific magnetic
stirrer. All pH-potentiometric measurements were carried out at a constant ionic strength of
0.2 M KCl at 298 K. A MOLSPIN pH-meter equipped with a 6.0234.100 combination glass
electrode (Metrohm) and a MOL-ACS microburette controlled by a computer was used to
carry out pH measurements. The recorded pH values were converted to hydrogen ion
concentration. Protonation constants of the ligands and overall stability constants (log βpqr) of
the metal complexes were calculated by means of the general computational programs,
(PSEQUAD and SUPERQUAD) as described in our previous publications [17,21]. The
equilibrium constants were defined by equations (1) and (2):
7

ACCEPTED MANUSCRIPT
pM +qH + rL = MpHqLr
(1)
(2)
[MpHqLr ]
pqr

[M]p [H]q [L]r
2.3. Spectroscopic measurements
A Perkin Elmer Lambda 25 scanning spectrophotometer was used to record the UV-visible
(UV-Vis) spectra of the copper(II) and nickel(II) complexes in the wavelength range of 250 to
800 nm. The ligand and metal ion concentrations of the samples were the same as reported for
the pH-potentiometric measurements.
CD spectra of the same complexes were recorded on a JASCO J-810 spectropolarimeter in the
220-800 nm wavelength range using 1 and/or 10 mm cells at the same concentration as used
for pH-potentiometry.
ESI-TOF-MS analysis was carried out with a Bruker micrOTOF-Q 9 ESI-TOF instrument in
the negative mode. The measurements were performed in water at cL=3.5×10-4 M at Cu(II): L
= 1.8:1 ratio at pH=7 by adding 0.2 M KOH solution. Temperature of drying gas (N2) was
180 °C and the pressure of the nebulizing gas (N2) was 0.3 bar. The capillary voltage applied
was 4000 V and the spectra were accumulated and recorded by a digitalizer at a sampling rate
of 2 GHz.
8

ACCEPTED MANUSCRIPT
3. Results and Discussion
Potentiometric titrations were used to determine the protonation constants of the ligands and
these values are included in Table 1. The data reported for the hexapeptide Aβ(1-6) are also
included for comparison [15]. It is clear form Table 1 that the number of protonation sites of
the three ligands is different. The terminal Asp and His(6) or His(7) residues are present in all
molecules and their amino- and imidazole-N donors represent the most basic sites. A slight
overlap can be in their protonation reactions but the highest pK values describe well the
deprotonation of the terminal ammonium group. Moreover, these values are in a good
agreement with the pK values of other terminally free histidine containing peptides [22]. The
other pK values belong to the carboxylate functions of aspartyl and glutamyl residues, the -
carboxylic group of glutamic acid being generally the less acidic site among them [23].
Although the metal binding properties of the three peptides are quite similar, the speciation
for the three metal ions is different and the results are discussed separately for each metal ion.
Table 1
3.1. Copper(II) complexes
The stability constants of the copper(II) complexes were determined by pH-metric titrations in
both equimolar samples and in the presence of excess metal ion or ligand. The data are listed
in Table 2 and the stepwise pK values of various protonated complexes are also shown. The
computer evaluation of the experimental data unambiguously ruled out the formation of
bis(ligand) complexes in measurable concentration but dinuclear species were formed in all
systems. Table 2 reveals that the speciation of the four systems is rather similar and it is
depicted only for the complexes of the native nonapeptide in Figure 1.
9

ACCEPTED MANUSCRIPT
Table 2
Figure 1
The equilibrium data and structural characterization of the complexes of the C-terminally
shortened hexapeptide have already been published in our previous paper [15] but the data are
also included in Table 2 for comparison. It can be seen form Table 2 that the complex
formation processes of the native hexa- and nona-peptides are quite similar suggesting the
existence of the same coordination modes for all species with the same stoichiometry. The
pH-dependent concentration distribution curves in Figure 1.a. reveal the consecutive
deprotonation of the various protonated complexes. It is a common feature of the copper(II)
complexes of simple oligopeptides when the terminal amino group is the primary metal
binding site [24-26]. The results of UV-Vis and CD spectroscopic measurements provided
further evidence for the coexistence of these coordination modes in slightly acidic samples.
The spectroscopic parameters calculated for the mononuclear copper(II) species are included
in Table 3.
Table 3.
It can be seen from the comparison of Table 3 and Figure 1 that the absorption maximum of
the species [CuL] is recorded at 696 nm but measurable CD activity cannot be obtained below
pH 5.5. This observation rules out the binding of amide nitrogens in this species. On the other
hand, the stability constants of [CuL] are much higher than those of the (amino, carbonyl) or
(amino, carboxylate)-coordinated common peptide complexes [22,26]. These data suggest the
binding of amino and carboxylate functions of the terminal aspartyl residue in the formation
of a 6-membered chelate, which is supported by a macrochelate from the distant histidyl
residue (see Scheme 1.a). Similar stability enhancement has already been reported for
peptides containing distant histidyl residues [21,22]. The protonated complexes ([CuH2L] and
[CuHL]) are present in low concentration and formed in overlapping processes. Their
10

ACCEPTED MANUSCRIPT
absorption maxima cannot be accurately determined but occurs above or around 700 nm
suggesting the same coordination mode of the amino terminus with protonated histidyl and
glutamyl side chains.
Subsequent deprotonation reactions or the formation of species [CuH-nL] (n=1-3) is
accompanied with a significant blue shift of the absorption maxima and the appearance of
new CD extrema by increasing intensity (see data in Table 3). Similar observations have
already been reported for the corresponding hexapeptide fragment of amyloid-β and were
explained by the copper(II) coordination of subsequent amide nitrogens starting from the
amino terminus [15]. These data strongly support that terminal amino and subsequent amide
nitrogens are the primary metal binding sites up to pH 9.0. However, a new CD extrema can
be recorded in more alkaline samples with a low intensity positive Cotton effect around 660
nm. The species [CuH-3L] predominates in these alkaline samples and its CD spectrum is
plotted with those of the same species of Aβ(1-4) and Ac-Aβ(1-6) [15] and the dinuclear
complex [Cu2H-6L] of Aβ(1-9) in Figure 2. The comparison of the CD spectra of the
mononuclear complex of Aβ(1-9) with those of Aβ(1-4) and Ac-Aβ(1-6) reveals the co-
existence of coordination isomers in the species [CuH-3L]. Both isomers are 4N-coordinated
complexes but the primary coordinating donors start from the amino terminus or from the
internal histidyl sites (see Scheme 1.b and c). The ratio of the (NH2,N-,N-,N-) and (N-,N-,N-
,Nim,) isomers can be estimated from the intensity of CD spectra and the value 3:1 was
obtained reflecting the predominance for binding to the amino group.
Scheme 1
Figure 2
Formation of dinuclear complexes with the general stoichiometry of [Cu2H-nL] (n=3-5 or 6)
was also recorded with all peptides. It can be seen from Figure 1.b that the formation of these
species starts above pH 6 in parallel with the coordination of amide nitrogens. ESI-MS
11

ACCEPTED MANUSCRIPT
measurements provided a direct evidence for the existence of dinuclear species. Figure 3 is
used to demonstrate the mass spectra of the 2:1 copper(II)-Aβ(1-9) samples in the
physiological pH range. In accordance with the speciation curves (see Figure 1.b.) the
existence of two major species the mononuclear [CuH-1L]2- and the dinuclear [Cu2H-3L]2- can
be unambiguously seen in the mass spectra. The agreement of the isotope distribution of the
calculated and experimental spectra justifies the coexistence of the mono- and di-nuclear
complexes under these conditions.
Figure 3
In the case of the hexapeptide Aβ(1-6) the binding of the second copper(II) ion can
occur only at the N-terminal side of the histidyl residue and only two amide nitrogens are
available for copper(II) binding in agreement with the [Cu2H-5L] stoichiometry of the
dinuclear complex at high pH. For the nonapeptide, the existence of the species [Cu2H-6L],
can also be calculated. The corresponding pK value is, however, very high (pK =10.66). The
high pK value cannot rule out the formation of a hydroxo complex but from the CD spectra
the existence of a new coordination mode can also be suggested. In the case of the
nonapeptide Aβ(1-9) there are three additional amino acids on C-terminal side of histidine
providing a chance for the (Nim,N-,N-,N-) coordination mode towards the C-termini. The
chelate ring sizes are (7,5,5) for this binding mode (Scheme 1.d.) while it is (6,5,5) towards
the N-termini (Scheme 1.c.). It is evident that the binding mode with 7-membered chelate has
reduced stability but the coexistence of the two coordination modes is possible at high pH and
this is in accordance with the relatively low intensity of the positive Cotton effect in the CD
spectra of the dinuclear samples at high pH values (see Figure 2.d).
It is clear from Table 2 that the stoichiometry of major species formed with the alanine
mutated nonapeptides is very similar to that of Aβ(1-9). The stability constants especially for
the dinuclear species are, however, significantly different. This can be demonstrated by Figure
12

ACCEPTED MANUSCRIPT
4 where the distribution of copper(II) among the three peptide ligands is plotted as a function
of pH. It can be seen from this Figure that the alanine mutated peptides are less effective
metal binders than the native peptide fragments. The side chain carboxylate functions of
Glu(3) and Asp(7) are missing from the mutants and this results in a big change in the overall
charge of the peptides. The fully deprotonated form of Aβ(1-9) is L3-, while it is only L- for
NH2-DAAAAHAAA-NH2 and NH2-DAAAAAHAA-NH2. This change in the charge of the
ligands will influence the overall charge of the corresponding complexes, too. Namely, the
most common dinuclear species [Cu2H-5L] is a two-, three- or four-negatively charged species
for the mutants, and native hexa- and nona-peptides, respectively. However, it is also clear
from Table 2 that the different charges of the complexes cannot give the sufficient
explanation for the differences in their stability constants. The stability order of the peptide
complexes is: Aβ(1-6) > Aβ(1-9) >> mutants. This trend suggests some weak axial
interactions form the side chain carboxylate residues. Moreover, the presence of polar side
chains in the native fragments may also have some contribution to the overall stability of the
copper(II) complexes of amyloid fragments. Similar observations have been reported recently
by Alies et al. for the copper(II) complexes of N-terminally modified peptides [27]. In
general, the stabilizing role of weakly coordinating or polar side chains have already been
demonstrated in reviews, by several authors [24-26,28,29]. These weak interactions, however,
do not change the major structural properties of the various complexes as it can be seen from
the comparison of spectroscopic parameters in Table 3. The binding modes in the
mononuclear copper(II) complexes of alanine mutants are the same as those of Aβ(1-9) and
even their stability is not much reduced. The high decrease of stability is obtained for the
dinuclear species of the mutant peptides and it is reflected in the absence of [Cu2H-6L] species
of NH2-DAAAAHAAA-NH2 containing only five amide groups between the anchoring
amino and histidyl sites. The pK value for the formation of this deprotonated species is,
13

ACCEPTED MANUSCRIPT
however, very high even for the native fragment (pK = 10.66) and it is further increased for
the other mutant with six amides between the anchoring sites (pK = 11.76).
Figure 4
Another small structural difference can be seen in the high pH CD spectra of the
peptides. The development of the positive Cotton effect at 660 nm in the [CuH-3L] species of
Aβ(1-6) and Aβ(1-9) was interpreted by the coexistence of coordination isomers with a
preference for the binding to the amino terminus. The intensity of these CD extrema is
especially low for the mutants reflecting a reduced tendency for binding at the internal
histidyl sites in accordance with the reduced affinity for the formation of dinuclear species.
3.2. Nickel(II) complexes
Nickel(II) is not considered as an essential element for humans and it is not involved in the
development of neurodegeneration. At the same time, nickel(II) ions play an important role in
the biochemical processes of various plants and microorganisms. His and/or Cys residues are
the most common nickel(II) binding sites in the corresponding proteins justifying the studies
on nickel(II) peptide complexes. Stability constants of the nickel(II) complexes are included
in Table 4, while the spectroscopic data for the major nickel(II) species were listed in Table 3.
The absence of dinuclear complexes in the nickel(II) containing systems represent the major
difference in the complex formation processes of copper(II) and nickel(II) ions. The titration
of 2:1 = nickel(II):peptide samples resulted in precipitation above pH 8.0 in all cases and it
was not dissolved completely even by pH 12.0. This suggests that dinuclear species can exist
only under extremely alkaline conditions without any biological significance. The
stoichiometry of the mononuclear complexes is, however, very similar to that of copper(II).
The formation of 3 or 4 species was obtained from the computer calculations but only 2 of
14

ACCEPTED MANUSCRIPT
them ([NiL] and [NiH-3L]) are present in measurable concentration. The species [NiL] exists
between pH 6 to 8 but without measurable CD activity. The weak absorption bands clearly
indicate that nickel(II) ion has an octahedral coordination geometry in this species and the
same binding mode as reported for the corresponding copper(II) species can be suggested (see
Scheme 1.a.). The characteristic yellow colour of the samples above pH 8.0 indicates the
formation of square planar complexes in this pH range. Only the intensity of both UV-visible
and CD spectra is changing by increasing pH supporting the cooperative deprotonation of
three amide nitrogens and the formation of [NiH-3L] in alkaline samples. In principle, the
coordination isomers (Scheme 1.b and 1.c) can be present in the nickel(II) containing systems,
too. The comparison of the CD spectra of the [NiH-3L] species of Aβ(1-9) with those of Aβ(1-
4) and Ac-Aβ(1-6), however, unambiguously supports the exclusive binding of nickel(II) to
the amino terminus of the nonapeptides (Figure 5). Moreover, the comparison of the stability
constants of the three nonapeptide complexes rules out any stabilizing or destabilizing effect
from the non-coordinated side chain residues of Aβ(1-9). This observation is not surprising if
one takes into account the strictly square planar geometry of these nickel(II) complexes.
Furthermore, the involvement of polar side chains were suggested to be more effective in the
dinuclear copper(II) complexes but this type of species is not formed with nickel(II).
Table 4
Figure 5
3.3. Zinc(II) complexes
Stability constants of zinc(II) complexes were also determined by potentiometric titrations
and these data are included in Table 5. Before the interpretation of the results it is important to
note that the amino terminus of Aβ(1-6) was suggested as the primary copper(II) and
15

ACCEPTED MANUSCRIPT
nickel(II) binding site, while the internal histidines (especially His13 and His14) were
suggested for zinc(II) [12]. On the basis of these results, relatively low zinc(II) binding
affinity can be expected for the nonapeptides. The experimental data are in agreement with
this expectation. The speciation of the three nonapeptides is very similar and precipitation of
zinc(II)-hydroxide prevent the measurements in alkaline samples. [ZnL] is the major species
in all cases and its coordination geometry support the formation of β-alanine like 6-membered
chelate supported by the macrochelation from the side chain imidazole. This binding mode
enhances the stability constants as compared to simple oligoglycines but does not prevent
hydrolysis above pH 8. The species [ZnH-1L] is probably a mixed hydroxo complex and the
lowest value is obtained for the native nonapeptide. In other words it means that the
uncoordinated polar side chains of Aβ(1-9) suppress the hydrolytic reactions and in an
indirect way they contribute to the enhanced zinc(II) binding of amyloid peptides.
Table 5
4. Conclusions
Previous studies on the copper(II), nickel(II) and zinc(II) complexes of amyloid peptides [15-
17] have already demonstrated the outstanding copper(II) and nickel(II) binding affinity of the
amino terminal domain of the peptide, some structural details of the interactions, however,
remained unanswered. The results of this study provide a significant contribution to the better
understanding of these questions. The combined application of potentiometric and UV-
visible, CD and mass spectrometric measurements on the complexes of the N-terminal
nonapeptide fragment, Aβ(1-9), undoubtedly prove the primary role of the amino terminus in
copper(II) and nickel(II) binding. The macrochelate created by the histidine nitrogens,
16

ACCEPTED MANUSCRIPT
however, enhances the thermodynamic stability of the species existing in the physiological pH
range ([CuL] and [CuH-1L]). In the case of copper(II) the formation of dinuclear complexes
was also recorded with the involvement of His(6) residue in copper(II) binding. The
comparison of the present data with those reported earlier for Aβ(1-6) made it possible to
clarify the binding modes around the histidyl site of the peptide. It was found that only the
first six amino acids from the amino terminus are involved in metal binding in the
physiological pH range. Among them, the amino and the subsequent amide nitrogens are the
primary ligating sites but the coexistence of two different 4N complexes (NH2,N-,N-,N-) and
(N-,N-,N-,Nim) is also possible in alkaline samples with a high preference for the former
isomer. Moreover, the deprotonation of amide nitrogens towards the C-terminus cannot be
excluded but only in the dinuclear species [Cu2H-6L] formed above pH 11.0.
Another important conclusion can be obtained from the comparison of the data
reported for Aβ(1-9) and its alanine mutated counterparts. The major metal binding sites of
the two mutants are the same as those of the native nonapeptide, but the thermodynamic
stability of the copper(II) complexes of the mutants is significantly reduced. This was
demonstrated by Figure 4 where the distribution of copper(II) ion among the three ligands is
calculated for a model system containing the metal ion and the three peptides in equimolar
concentrations. Figure 4 reveals that the majority of copper(II) ions is bonded to the native
peptide fragment at any pH values. This difference probably comes from the secondary
interactions of the polar side chains of Asp, Glu, Ser and Arg residues. Moreover, this
difference reveals that the amino acid sequence of the N-terminal domains of amyloid
peptides is well suited for the complexation with copper(II) ions.
In addition to copper(II) the amyloid-β N-terminus can also effectively bind nickel(II)
ions but only mononuclear species can be formed with the exclusive binding from the amino
terminus. On the other hand, there is no indication for extra stabilization from the polar
17

ACCEPTED MANUSCRIPT
residues and the thermodynamic stability of the nickel(II) complexes corresponds well to
those of the common oligopeptides. Zinc(II) ions are not able to promote deprotonation and
metal ion coordination of amide nitrogens resulting in a moderate stability of the zinc(II)
complexes. This is in accordance with previous findings that the amino terminus is not the
preferred zinc(II) binding site of amyloid-β. On the other hand, the comparison of the stability
constants of zinc(II) complexes of oligoglycines with that of the native nonapeptide reveals a
stability enhancement for the latter one. This enhanced stability is accompanied with a
reduced tendency of hydrolysis of zinc(II)-Aβ(1-9) complexes, which can be explained by the
simultaneous coordination of terminal aspartyl and histidyl sites in a macrochelate and by the
weak secondary interactions of the polar side chains of the other amino acids.
5. Abbreviations
CD – circular dichroism
MS – mass spectra
Fmoc – N-fluorenylmethoxycarbonyl
DMF – dimethyl formamide
TFA – trifluoracetic acid
TIS – triisopropylsilane
Trt – trytil
Pbf – (2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl)
18

ACCEPTED MANUSCRIPT
Acknowledgements
The research was supported by EU and co-financed by the European Social Fund under the
project ENVIKUT (TAMOP-4.2.2.A-11/1/KONV-2012-0043).
19

ACCEPTED MANUSCRIPT
References
[1] J.A. Duce, A.I. Bush, Progress in Neurobiology 92 (2010) 1-18.
[2] A.I. Bush, Advances in Alzheimer’s Disease (ed. G. Perry) 3 (2013) 277-281.
[3] C. Hureau, Coord. Chem. Rev. 256 (2012) 2164-2174.
[4] C. Hureau, P. Dorlet, Coord. Chem. Rev. 256 (2012) 2175-2187.
[5] C. Migliorini, E. Porciatti, M. Luczkowski, D. Valensin, Coord. Chem. Rev. 256
(2012) 352-368.
[6] H. Kozlowski, M. Luczkowski, M. Remelli, D. Valensin, Coord. Chem. Rev. 256
(2012) 2129-2141.
[7] V. Tougu, P. Palumaa, Coord. Chem. Rev. 256 (2012) 2219-2224.
[8] A. Travaglia, A. Pietropaolo, D. La Mendola, V. G. Nicoletti, E. Rizzarelli, J. Inorg.
Biochem. 111 (2012) 130-137.
[9] J.H. Viles, Coord. Chem. Rev. 256 (2012) 2271-2284.
[10] C. Hureau, P. Faller, Biochimie 91 (2009) 1212-1217
[11] S. Chassaing, F. Collin, P. Dorlet, J. Gout, C. Hureau, P. Faller, Curr. Top. Med.
Chem. 12, (2012) 2573-2595.
[12] G. Arena, G. Pappalardo, I. Sóvágó, E. Rizzarelli, Coord. Chem. Rev. 256 (2012) 3-
12.
[13] T. Kowalik-Jankowska, M. Ruta-Dolejsz, K. Wisniewska, L. Lankiewicz, H.
Kozlowski, J. Chem. Soc., Dalton Trans. (2000) 4511-4519.
[14] T. Kowalik-Jankowska, M. Ruta, K. Wisniewska, L. Lankiewicz, J. Inorg. Biochem.
95 (2003) 270-282.
20

ACCEPTED MANUSCRIPT
[15] C.A. Damante, K. Ősz, Z. Nagy, G. Pappalardo, G. Grasso, G. Impellizzeri, E.
Rizzarelli, I. Sóvágó, Inorg. Chem. 47 (2008) 9669-9683.
[16] C.A. Damante, K. Ősz, Z. Nagy, G. Pappalardo, G. Grasso, G. Impellizzeri, E.
Rizzarelli, I. Sóvágó, Inorg. Chem. 48 (2009) 10405-10415.
[17] É. Józsa, K. Ősz, C. Kállay, P. de Bona, C.A. Damante, G. Pappalardo, E. Rizzarelli, I.
Sóvágó, Dalton Trans. 39 (2010) 7046-7053.
[18] C.A. Damante, K. Ősz, Z. Nagy, G. Grasso, G. Pappalardo, E. Rizzarelli, I. Sóvágó,
Inorg. Chem. 50 (2011) 5342-5350.
[19] B. Alies, I. Sasaki, O. Proux, S. Sayen, E. Guillon, P. Faller, C. Hureau, Chem.
Comm. 49, (2013) 1214-1216.
[20] K.I. Silva, S. Saxena, J. Phys. Chem. 117 (2013) 9386-9394.
[21] V. Jószai, I. Turi, C. Kállay, G. Pappalardo, G. Di Natale, E. Rizzarelli, I. Sóvágó, J.
Inorg. Biochem. 112 (2012) 17-24.
[22] Á. Grenács, A. Kaluha, C. Kállay, V. Jószai, D. Sanna, I. Sóvágó, J. Inorg. Biochem.
128 (2013) 17-25.
[23] C. Kállay, K. Várnagy, G. Micera, D. Sanna, I. Sóvágó, J. Inorg. Biochem. 99 (2005)
1514-1525.
[24] H. Sigel, R.B. Martin, Chem. Rev. 82 (1982) 385-426
[25] I. Sóvágó, K. Ősz, Dalton Trans. (2006) 3841-3854.
[26] I. Sóvágó, C. Kállay, K. Várnagy, Coord. Chem. Rev. 256 (2012) 2225-2233.
[27] B. Alies, C. Bijani, S. Sayen, E. Guillon, P. Faller, C. Hureau, Inorg. Chem. 51 (2012)
12988-13000.
[28] H. Kozlowski, W. Bal, M. Dyba, T. Kowalik-Jankowska, Coord. Chem. Rev. 184
(1999) 319-346.
[29] O. Yamauchi, A. Odani, M. Takani, J. Chem. Soc., Dalton Trans. (2002) 3411-3421.
21

ACCEPTED MANUSCRIPT
Table 1
Protonation constants of the nonapeptides.
T = 298 K, I = 0.2 M
Species
Aβ(1-6)
NH2-DAEFRH-
NH2
Aβ(1-9)
NH2-
DAEFRHDSG-
NH2
NH2-
NH2-
DAAAAHAAA- DAAAAAHAA-
NH2
NH2
[13]
[HL]
7.59
7.73(1)
7.69(1)
14.01(2)
16.63(2)
-
7.65(1)
14.01(1)
16.83(1)
-
[H2L]
[H3L]
[H4L]
[H5L]
13.86
18.04
20.85
-
14.06(1)
18.36(2)
21.83(3)
24.45(8)
-
-
pK(-NH2)
pK(Im)
7.59
6.27
7.73
6.33
4.30
3.47
2.62
7.69
6.32
2.62
7.65
6.36
2.82
pK(-COOH) 4.18
pK(-COOH) 2.81
pK(-COOH)
-
22

ACCEPTED MANUSCRIPT
Table 2
Stability constants (log βpqr) of the copper(II) complexes of the peptides.
T = 298 K, I = 0.2 M
Species
Aβ(1-6)
Aβ(1-9)
NH2-
NH2-
NH2-DAEFRH- NH2-
DAAAAHAAA- DAAAAAHAA-
DAEFRHDSG- NH2 NH2
NH2
NH2
[13]
[CuH2L]
[CuHL]
17.36
16.94(3)
-
-
13.16
8.76
12.99(2)
8.12(2)
12.24(3)
7.66(2)
1.60(3)
-6.73(4)
-15.76(4)
-9.29(6)
-17.86(5)
-28.04(6)
-
12.23(3)
7.45(2)
1.65(2)
-6.83(2)
-15.94(2)
-9.79(4)
-18.57(4)
-28.00(5)
-39.76(3)
[CuL]
[CuH-1L]
[CuH-2L]
[CuH-3L]
[Cu2H-3L]
[Cu2H-4L]
[Cu2H-5L]
[Cu2H-6L]
3.02
1.97(2)
-4.86
-13.50
-7.18
-14.08
-21.55
-
-5.85(2)
-14.85(2)
-7.94(2)
-15.38(2)
-22.75(4)
-33.41(3)
pK1
pK2
pK3
pK4
pK5
4.20
4.40
5.74
7.88
8.64
3.95
4.87
6.15
7.82
9.00
-
-
4.58
6.06
8.33
9.03
4.78
5.80
8.48
9.11
23

ACCEPTED MANUSCRIPT
Table 3
UV-visible and CD spectroscopic parameters of the major mononuclear complexes formed in
the copper(II)- and nickel(II)-nonapeptide systems
Species
Aβ(1-9)
NH2-DAEFRHDSG-
NH2
NH2-DAAAAHAAA-
NH2
NH2-DAAAAAHAA-
NH2
(λmax/ε,
nm/
(λ/Δε,
nm/
(λmax/ε,
nm/
(λ/Δε,
nm/
(λmax/ε,
nm/
(λ/Δε,
nm/
M-1cm-1)
696/41
M-1cm-1)
-
M-1cm-1)
685/44
M-1cm-1)
-
M-1cm-1)
682/40
M-1cm-1)
-
[CuL]
[CuH-1L]
[CuH-2L]
619/94
591/-0.36 618/104
583/-0.38 566/109
496/-0.41
625/-0.35 623/97
560/-0.35 557/107
621/-0.38
554/-0.29
560/101
[CuH-3L]
514/162
418/211
664/0.25
521/-1.57
313/0.72
517/168
699/0.066 515/167
523/-0.90
673/0.057
529/-0.88
309/0.35
307/0.57
[NiH-3L]
469/-3.28 414/247
413/-
463/-4.11 414/173
466/-2.54
2.11(sh)
24

ACCEPTED MANUSCRIPT
Table 4
Stability constants (log βpqr) of the nickel(II) complexes of the peptides.
T = 298 K, I = 0.2 M
Species
Aβ(1-6)
Aβ(1-9)
NH2-
NH2-
NH2-DAEFRH- NH2-
DAAAAHAAA- DAAAAAHAA-
DAEFRHDSG- NH2 NH2
NH2
NH2
[15]
[NiHL]
[NiL]
10.60
11.31(8)
11.24(11)
5.29(5)
-3.05(7)
-
11.19(6)
5.63
-3.05
-
5.53(4)
5.31(5)
[NiH-1L]
[NiH-2L]
[NiH-3L]
-
-
-
-
-20.00
-20.63(4)
-20.38(6)
-20.74(5)
25

ACCEPTED MANUSCRIPT
Table 5
Stability constants (log βpqr) of the zinc(II) complexes of the peptides.
T = 298 K, I = 0.2 M
Species
Aβ(1-9)
NH2-
NH2-
NH2-
DAAAAHAAA- DAAAAAHAA-
DAEFRHDSG- NH2
NH2
NH2
[ZnHL]
[ZnL]
10.09(5)
4.25(1)
-4.50(3)
10.40(4)
10.28(4)
4.11(1)
-3.38(5)
4.37(1)
[ZnH-1L]
-3.70(1)
26

ACCEPTED MANUSCRIPT
Legends to Figures
Figure 1
Species distribution of the complexes formed in the copper(II)-Aβ(1-9) system at 1:1 (a) and
2:1 (b) metal to ligand ratios (cL = 2 mM).
Figure 2
Circular dichroism spectra of the [CuH-3L] species of the ligands Aβ(1-9) (a), Aβ(1-4) (b),
Ac-Aβ(1-6) (c) and [Cu2H-6L] of Aβ(1-9) (d).
Figure 3
ESI-MS spectra of the 2:1 copper(II)-Aβ(1-9) sample showing the isotopic distribution of the
measured (upper) and calculated (lower) spectra for the mononuclear [CuH-1L]2- and
dinuclear [Cu2H-3L]2- species.
Figure 4
Distribution of the copper(II) ion among the three ligands in a model sample containing the
metal ion and the three peptides in equimolar concentration (c = 2 mM).
Figure 5
Circular dichroism spectra of the NiH-3L complexes formed with ligands A(1-9) (a), A(1-4)
(b) and Ac-A(1-6) (c).
27

ACCEPTED MANUSCRIPT
Figure 1
28

ACCEPTED MANUSCRIPT
Figure 2
29