Solution structure of amyloid β-peptide (25-35) in different media

Article abstract of DOI:10.1021/jm040773o

The design of molecules able to interact with the amyloid peptides either as inhibitors of fibril formation or as inhibitors of amyloid membrane pore formation represents one of the most relevant approaches in the development of anti-Alzheimer therapies. Aβ-(25-35), sequence GSNKGAIIGLM, is a highly toxic synthetic derivative of amyloid β-peptides (Aβ-peptides), which forms fibrillary aggregates. Here, we report the NMR and CD investigation of Aβ-(25-35) in a membrane-mimicking environment and in isotropic mixtures of water and fluoroalcohols to scan its conformational properties as a function of the medium. The analysis of the 3D structures in the mentioned conditions indicates a propensity of the peptide to behave as a typical transmembrane helix in the lipidic environment. In media characterized by different polarity, it loses the structural regularity at specific points of the sequence as a function of the environment. Furthermore, a comparison with the solution structure of full-length amyloid peptides suggests a role for the 25-27 kink region, which appears to be a general feature of all peptides under the solution conditions explored.

Full text of DOI:10.1021/jm040773o

J . Med. Chem. 2004, 47, 4231-4238
4231
Solu tion Str u ctu r e of Am yloid â-P ep tid e (25-35) in Differ en t Med ia
Anna M. D’Ursi,§ Maria R. Armenante,§ Remo Guerrini,† Severo Salvadori,† Giuseppe Sorrentino,# and
Delia Picone*,‡
Dipartimento di Scienze Farmaceutiche, Universita` di Salerno, Via Ponte Don Melillo 11c, 84084 Fisciano, Italy,
Dipartimento di Scienze Farmaceutiche, Universita` di Ferrara, Via Fossato di Mortara 17-19, 44100 Ferrara, Italy,
Universita` degli Studi di Napoli “Parthenope”, Via Acton 38, 80133 Napoli, Italy, and Dipartimento di Chimica,
Universita` degli Studi di Napoli “Federico II”, Via Cintia 26, Complesso Universitario di Monte S. Angelo, 80126 Napoli, Italy
Received J anuary 16, 2004
The design of molecules able to interact with the amyloid peptides either as inhibitors of fibril
formation or as inhibitors of amyloid membrane pore formation represents one of the most
relevant approaches in the development of anti-Alzheimer therapies. Aâ-(25-35), sequence
GSNKGAIIGLM, is a highly toxic synthetic derivative of amyloid â-peptides (Aâ-peptides),
which forms fibrillary aggregates. Here, we report the NMR and CD investigation of Aâ-(25-
35) in a membrane-mimicking environment and in isotropic mixtures of water and fluoro-
alcohols to scan its conformational properties as a function of the medium. The analysis of the
3D structures in the mentioned conditions indicates a propensity of the peptide to behave as
a typical transmembrane helix in the lipidic environment. In media characterized by different
polarity, it loses the structural regularity at specific points of the sequence as a function of the
environment. Furthermore, a comparison with the solution structure of full-length amyloid
peptides suggests a role for the 25-27 kink region, which appears to be a general feature of
all peptides under the solution conditions explored.
In tr od u ction
still problematic, as argued by Terry,9 is the weak
correlation between fibrillary amyloid load and mea-
sures of neurological dysfunction. Recent studies on Aâ
support the possibility that nonfibrillary oligomeric
species are pathogenic.10,11 These data have led to the
hypothesis that the Aâ neurotoxicity is related to
different processes of membrane interaction and desta-
bilization culminating in membrane pore formation and
disruption. This hypothesis is supported also by the
surprising similarity of the structure of Aâ-(1-42)
recently solved in our lab12 to the fusion domain of
influenza hemagglutinin in an apolar environment.
According to this hypothesis, the R-helical peptide would
induce formation of membrane channels, allowing the
penetration of substances that can cause neuronal
death. The ability of Aâ to induce vesicle fusion and the
recent observation that it can enhance infection at the
stage of attachment or entry into the cell of several
viruses13 further support this hypothesis. Whatever the
pathogenic mechanism of the AD would be, the confor-
mational behavior of soluble forms of Aâ appears to play
a critical role. Although Aâ peptides were identified
several years ago, still relatively few structural studies
describe the conformation of these molecules in their
soluble or aggregated forms. Structural studies of these
peptides are indeed strongly hampered by the low
solubility in water and by their tendency to aggregate.
In the search for the conditions that overcome these
limitations, the most interesting datum is that Aâ
peptides are crucially affected by the environmental
conditions. Consistently the study of the conformational
preferences of these peptides in different media appears
to be crucial for shedding light on their intrinsic
structural properties related to the Aâ fibril formation
or to membrane disruption activity.
Alzheimer’s disease (AD) is a member of a heteroge-
neous family of diseases referred to as amyloidosis,
characterized by deposits of amyloid in a variety of
organs.1,2 At present, almost 20 human diseases associ-
ated with the presence of insoluble aggregates of
normally soluble proteins resulting in the formation of
extracellular deposits have been identified.3 Through
self-assembly these proteins produce regular fibrillary
structures possessing a predominant â-sheet conforma-
tion.4 The amyloid deposits are the major constituent
of the extracellular senile, or neuritic, plaques charac-
teristic of brain tissue affected by AD. The primary
proteinaceous component of neuritic plaques is a peptide
composed of 39-42 amino acid residues5 and referred
to as amyloid â-peptide (Aâ). The Aâ molecules are
derived from a type I transmembrane protein containing
695-770 amino acids, known as Aâ precursor protein
(APP), by the action of the â- and γ-secretases.6 Aâ-(1-
42) is the most prone to aggregation and is produced in
larger quantities in familial forms of AD.
Although a large consensus exists about the central
role of Aâ in AD, the biophysical and biological mech-
anisms that underlie its toxicity are still controversial.
A great deal of evidence supports the hypothesis that
Aâ fibrillogenesis is a seminal pathogenetic event in
AD.7 It is natural to think that plaques adjacent to the
membrane can damage it or even that the mechanical
blockage by plaques of cellular exchanges with the outer
environment can lead to the cellular death.8 However,
* To whom correspondence should be addressed. Phone: +39-081-
674406. Fax: +39-081-674409. E-mail: picone@unina.it.
§ Universita` di Salerno.
† Universita` di Ferrara.
#
Universita` degli Studi di Napoli “Parthenope”.
‡ Universita` degli Studi di Napoli “Federico II”.
10.1021/jm040773o CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/21/2004

4232 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 17
D’Ursi et al.
It has been proposed that Aâ-(25-35) represents the
biologically active region of Aâ because it represents the
shortest fragment that exhibits large â-sheet aggregated
structures and retains the toxicity of the full-length
peptide.14 Several reports have also indicated that Aâ-
(25-35), in a manner similar to that of Aâ-(1-42),
undergoes a conformational transition from a soluble,
random-coil form to aggregated fibrillary â-sheet struc-
35) was achieved according to the Wu¨thrich procedure20
via the usual systematic application of DQF-COSY,
TOCSY, and NOESY experiments21-23 with the support
of the SPARKY software package.24 The analysis of the
NOE pattern revealed the presence of the typical helical
(i, i + 2) and (i, i + 3) medium range connectivities
encompassing the C-terminal residues, whereas less
regularity was detected in the N-terminal segment,
characterized by the prevalence of more disordered
structures. In particular RN(i, i + 2) involves the
residues Gly 29-Ile 31, Ile 31-Gly 33, and Ile 32-Leu 34,
whereas RN(i, i + 3) connectivity involves the residues
Ile 32-Met 35.
tures, depending on the environmental conditions.15
A
more detailed conformational study performed by 2D
NMR in LiDS micelles, pH 4.0, has indicated the
presence of a well-ordered R-helical stretch encompass-
ing residues 28-34.16 Furthermore, CD and Fourier
transform infrared spectroscopy have shown that upon
addition of lipid vesicles containing negatively charged
lipids the equilibrium shifts almost completely toward
â-sheet structure formation.15,17 Because of these prop-
erties Aâ-(25-35) has been often chosen as a model for
full-length Aâ in structural and functional studies.
We describe here the conformational analysis by CD
and NMR spectroscopies of Aâ-(25-35) in different
experimental conditions to evaluate the structural
properties of the Aâ segment in a membrane-mimicking
environment and in isotropic media with different
polarity. Micellar solutions of SDS are widely used to
test the structural behavior of biomolecules known to
interact with cellular membrane. The NMR study of Aâ-
(25-35) in SDS in the presence of spin labels allowed
us to investigate the structural behavior of the peptide
in a lipid environment and to establish the orientation
of the peptide with respect to the membrane. In addi-
tion, we have performed a structural investigation by
CD and 2D NMR spectroscopy in water/fluorinated
alcohol mixtures. These mixtures have been reported
as helical stabilizing agents18 and are also able to
dissolve fibrillar aggregates of Aâ-(1-40).19 To scan the
conformational characteristics of this peptide in differ-
ent polarity conditions, we collected data moving from
a basically aqueous medium of water/HFIP, 80/20 v/v,
to a basically apolar water/HFIP, 20/80 v/v. Although
this shorter fragment does not give rise to aggregates,
a conformational transition leading to a drastic reduc-
tion of the folded ordered structure is observable.
The structure calculation by means of DYANA soft-
ware25 and based on the NOE-derived interprotonic
distances as restraints confirmed the previous struc-
tural results in LiDS.16 In the SDS micelles environ-
ment, Aâ-(25-35) is characterized by a well-ordered
R-helical structure involving the residues 28-34. The
rmsd for the backbone between residues 28-34 and 25-
27 are respectively 0.56 and 1.27 Å.
1.2. Sp in -La bel Stu d ies. The NMR analysis in SDS
was the starting point to investigate the positioning of
the peptide with respect to the surface and interior of
the micelle, using 5-doxylstearic acid and 16-doxyl-
stearic acid as paramagnetic probes. Both compounds
contain a doxyl headgroup, a cyclic nitroxide with
unpaired electron, that is bound to the aliphatic chain
carbon in position 5 or 16, respectively. Unpaired
electrons lead to dramatically accelerated longitudinal
and transverse relaxation rates of protons in spatial
proximity via highly efficient spin-electron relaxation.
It has been shown that these paramagnetic probes are
able to induce broadening of the NMR signals or a
decrease of resonance intensities for residues close to
the surface (5-doxyl) or deeply buried in the micelle (16-
doxyl), respectively.26,27 TOCSY spectra of Aâ-(25-35)
either in the presence or in the absence of the spin
labels, keeping constant all other conditions, were
compared. All signal intensities remained constant upon
addition of 5-doxylstearic acid, thereby proving that no
residue of Aâ-(25-35) interacts with the micelle surface
(data not shown). In Figure 1 a comparison of the HR-
NH region of TOCSY spectra of Aâ-(25-35) acquired
in the presence (A) and in absence (B) of 16-doxylstearic
acid is reported. Ala 30, Ile 31, and Gly 33 are drasti-
cally affected by a spin label effect with a significant
decrease of the HR-NH signals. Figure 2 shows the best
NMR structure obtained in SDS, selected on the basis
of the target function value; the side chains of the
residues affected by the 16-doxylstearic acid are dis-
played. This result provides evidence that the C-
terminal portion of Aâ-(25-35), arranged in an R-helical
structure, is deeply embedded in the SDS micelle.
The study of the Aâ-(25-35) in the mentioned condi-
tions could open the way to the design and the synthesis
of molecules able to antagonize the membrane pore
formation activity of Aâ peptides. Alternatively the
detailed description of the conformational transition is
a key step for designing aggregation inhibitors. Last but
not least it seems very exciting the possibility that
peptides designed from Aâ can be used to enhance the
viral cell permeation13 and may be useful for setting up
a gene delivery therapy based on retroviral vectors.
Resu lts
2. HF IP An a lysis. 2.1. CD Sp ectr oscop y. The
conformational preferences of Aâ-(25-35) were deter-
mined from the CD spectra acquired in mixtures of
water/HFIP with different ratios (v/v). The spectra
corresponding to 20%, 30%, and 40% of HFIP in water
have the typical shape of random coil structures, while
a significant change in the spectrum intensity is evident
starting from 50/50 HFIP/water composition. The CD
curves corresponding to mixtures 50/50, 60/40, and 70/
30 HFIP/water are typical of turn-helical structures.
1. SDS An a lysis. 1.1. NMR Sp ectr om etr y. A whole
set of 1D and 2D proton spectra were recorded in
aqueous solution of 100 mM SDS, pH 4.15. To check
the absence of aggregation, spectra were acquired in the
concentration range of 0.5-15 mM. No significant
changes were observed in the distribution and in the
shape of the H resonances, indicating that no aggrega-
tion phenomena occurred in this concentration range.
Complete assignment of the proton spectra of Aâ-(25-
1

Amyloid â-Peptide 25-35
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 17 4233
of the mixtures tested is shown in Figure 3. In the
selected mixtures, the CD spectra are essentially un-
changed in the temperature range 10-45 °C (data not
shown), suggesting a high conformational stability of
Aâ-(25-35) in these solvent media. Moreover, the Aâ-
(25-35) solutions in aqueous HFIP were very stable
because there was no evidence of aggregation or pre-
cipitation and the NMR spectra did not change over
several weeks.
On the basis of these results, we performed an NMR
investigation of Aâ-(25-35) in the presence of a low
(HFIP/water 20/80) and a high (HFIP/water 80/20)
amount of HFIP. The two mixtures are representative
of a mainly aqueous medium and of a mainly apolar
environment, respectively. Furthermore, the NMR data
of Aâ-(25-35) in HFIP/water 80/20 allows a comparison
with the NMR structure of Aâ-(1-42), which has been
recently determined by us in the same solvent mix-
ture.12
2.2. NMR Sp ectr om etr y. A whole set of 1D and 2D
protonic spectra were recorded in 20/80 and 80/20 HFIP/
water mixtures. Also, under these experimental condi-
tions, no significant changes were observed in the
concentration range of 0.5-15 mM, showing the absence
of aggregated forms. The assignment of the proton
resonances of Aâ-(25-35) was achieved by following the
same procedure reported above.
In Figure 4 all short- and medium-range NOE effects
involving the backbone protons of Aâ-(25-35) in both
the HFIP/water mixtures are summarized.
The inspection of the bar diagram corresponding to
the 20/80 HFIP/water mixture shows a low number of
diagnostic (i, i + 2) and (i, i + 3) effects. Moreover, all
of them are located in the N-terminal segment 25-29,
while the C-terminal region of the peptide is character-
ized by disordered structures. In contrast, a more
regular pattern was found in the bar diagram corre-
sponding to the 80/20 HFIP/water mixture, where
sequential NHi-NHi+1 NOEs and diagnostically critical
(i, i + 2) and (i, i + 3) effects encompassing residues
28-34 are present. The data are consistent with the
presence of a well-defined R-helical structure in the
C-terminal region. A less regular pattern consistent
with the presence of partially folded structures was
observed for the N-terminal segment.
F igu r e 1. Fingerprint region of the 600 MHz TOCSY spectra
of Aâ-(25-35) recorded in 100 mM SDS water solution at 300
K (A) and upon addition of 5-doxylstearic acid at a concentra-
tion of one spin label per micelle (B).
2.3. Str u ctu r e Ca lcu la tion . The same calculation
procedure followed in SDS experiments was adopted.
Among 50 calculated structures the resulting best 20
were selected according to the lowest values of their
target function. They were subjected to further proce-
dures of minimization with the SANDER module of the
AMBER 5.0 software29,30 using the DYANA derived
restraints.
The model of Aâ-(25-35) calculated on the basis of
NOE data in the 20/80 HFIP/water mixture shows the
prevalence of disordered structures particularly in the
C-terminal region, whereas a significant presence of
ordered conformations is detectable at level of the
N-terminal residues. The conformers of the N-terminal
segment of Aâ-(25-35) in 20/80 HFIP/water are well
fitted (the rmsd for the backbone between residues 25
and 28 is 0.46 Å) (Figure 5). The PDB files subjected to
the PROMOTIF software31 for the validation of the
structural results show the presence of a type I â-turn
F igu r e 2. NMR structure of Aâ-(25-35) in 100 mM SDS
aqueous solution. The structure was derived from DYANA
calculations and was chosen according to the lowest value of
the target function. The residues affected by paramagnetic
spins are in red and highlighted with ball-and-stick rendering.
The helix content, as estimated by standard linear
combination fits of the spectra,28 increases almost
linearly from 17% in neat water to about 80% in HFIP/
water 80/20 (v/v), i.e., reaching a value very similar to
that of Aâ-(1-42) under the same experimental condi-
tions.12 A comparison of the CD spectra recorded in some

4234 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 17
D’Ursi et al.
F igu r e 3. Circular dichroism curves of Aâ-(25-35) at different HFIP/water compositions. The spectra were recorded at room
temperature at 80/20 (a), 70/30 (b), 30/70(c), and 20/80 (d) (v/v) HFIP/water composition. Total molar ellipticities are reported vs
wavelength in nm.
F igu r e 5. The best 20 calculated structures of Aâ-(25-35) in
HFIP/water 20/80 as derived from DYANA calculations and
energy-minimized using the SANDER module of the AMBER
5.0 software. The structures are fitted on the heavy backbone
atoms of the 25-28 segment.
Figure 6 reports the overlap of the best 20 NMR
structures of Aâ-(25-35) calculated from NMR data of
HFIP/water 80/20. The models are fitted at the level of
the residues 28-34, where Aâ-(25-35) is characterized
by the presence of a helical structure (rmsd for the
backbone is 0.68 Å). As expected from the NOE data, a
significant decrease of structural regularity is observ-
able in the N-terminal segment. Notwithstanding, a
careful glance toward the dihedral angles relative to the
26-28 residues shows that they are generally compat-
ible with the presence of a turn structure. The dihedral
angles relative to the structures calculated in the two
HFIP/water mixtures are reported in Table 1.
F igu r e 4. Sequential and medium-range NOEs for Aâ-(25-
35). Data were obtained from a 600 MHz NOESY experiment
with a mixing time of 150 ms and collected in HFIP/water 20/
80 v/v (A) and HFIP/water 80/20 (B) at 300 K.
centered on residues 26 and 27. Unconstrained mini-
mization of the structures did not produce any major
rearrangement in this region, which instead would be
expected if the observed dihedrals were imposed by the
influence of artifactual NMR restraints.

Amyloid â-Peptide 25-35
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 17 4235
F igu r e 6. The best 20 calculated structures of Aâ-(25-35) in
HFIP/water 80/20 as derived from DYANA calculations and
energy-minimized using the SANDER module of the AMBER
5.0 software. The structures are fitted on the heavy backbone
atoms of the 28-34 segment.
Ta ble 1. Values of the φ and ψ Dihedral Angles of Aâ-(25-35)
Structures Derived from NMR Data in HFIP/Water (20/80 and
80/20 v/v) at 600 MHz and 300 K
HFIP/water 20/80
φ (deg) ψ (deg)
-99.0 ( 33
-88.8 ( 49 -35.9 ( 38 -74.5 ( 26 -151.6 ( 34
HFIP/water 80/20
φ (deg) ψ (deg)
-169.3 ( 35
residue
Gly25
Ser26
Asn27 -107.8 ( 30 -15.4 ( 25 -60.9 ( 15 -15.7 ( 22
Lys28 -92 ( 34
-42.1 ( 16 -5.3 ( 14
-40.3 ( 15
Gly29
Ala30
Ile31
53.1 ( 27
-18.9 ( 14 -55.8 ( 13 -17.4 ( 16
-76.2 ( 19 -120.5 ( 13 -103.4 ( 12 -26.4 ( 14
-37.8 ( 16 -49.5 ( 12 -47.7 ( 12 -24.8 ( 20
-85.8 ( 12 -135.1 ( 15 -51.1 ( 13 -23.8 ( 16
Ile32
F igu r e 8. Structural comparison of the surfaces relative to
the NMR structure of Aâ-(25-35) in HFIP/water 80/20 (blue)
and Aâ-(1-42) (cyan) in HFIP/water 80/20. The surfaces are
colored according to the hydrophobicity scale where the
hydrophobic residues are in red and the hydrophilic residues
are in blue.
Gly33
111.0 ( 13
-50.2 ( 14 -67.5 ( 15 -4( 13
Leu34 46.0 ( 18
Met35 -75.2 ( 22
-79.5 ( 23 170.7 ( 19
-41.6 ( 19
48.2 ( 12
2.4. Aâ-(25-35) a n d Aâ-(1-42) Str u ctu r e Com -
p a r ison . Recently we have determined the NMR struc-
ture of Aâ-(1-42) in aqueous HFIP.12 In this environ-
ment the amyloid peptide shows two helical stretches
connected by a kink region centered on a regular type
I â-turn involving the residues 25-28. The same struc-
tural motif has been found in Aâ-(25-35) in the explored
fluorinated mixtures. Moreover, depending on the ex-
perimental conditions, a helical stretch spanning resi-
dues 28-34 has been found in full-length Aâ as well as
in Aâ-(25-35). Finally, Figure 8 shows that the molec-
ular surfaces of Aâ-(1-42) and Aâ-(25-35) in HIFP 80/
20 are very similar, showing a common orientation of
the hydrophobic side chains in the more significantly
ordered region.
Discu ssion
The fight against the Alzheimer disease is an impor-
tant challenge in pharmaceutical research because of
the increasing incidence of this pathology correlated to
the life elongation.
The design of molecules able to interact with the
amyloid peptides as inhibitors of fibril formation or as
inhibitors of amyloid membrane pore formation consti-
tutes a basic approach in the development of anti-
Alzheimer therapies. Despite the large amount of
experimental data, it is not clear whether the fibrils of
amyloid peptides are the cause of Alzheimer pathology
or if alternatively they have a protective function, while
F igu r e 7. Overlapping of the best NMR structures of Aâ-(25-
35) in HFIP/water 80/20 (blue) and HFIP/water 20/80 (green)
mixtures.
In Figure 7 a comparison of the best NMR structures
of Aâ-(25-35) in the two solvent mixtures is displayed.
It is evident that there is a significant degree of
similarity in the N-terminal residues, where the pres-
ence of a turn conformation is conserved.

4236 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 17
D’Ursi et al.
the pathological cause is the amyloid membrane disrup-
tion activity. In both the cases the study of the confor-
mational properties of Aâ peptides is the basis for the
design of molecules with “anti-amyloid” activity.
peptide folding. To follow the conformational modifica-
tions derived from increasing medium polarity, we have
selected two extreme conditions: a 20/80 HFIP/water
mixture, which may resemble the aqueous interface
between the membrane surface and the extracellular
medium, and a 80/20 HFIP/water mixture with a higher
apolar character to simulate the lipid environment.
Interestingly, the structure of Aâ-(25-35) in the 80/20
HFIP/water mixture is characterized by a high degree
of ordered, folded conformation particularly in the
C-terminal portion that is almost identical to the one
previously discussed in SDS. In comparison, the N
terminus appears to be less ordered with a slight
presence of conformers characterized by a â-turn in the
25-27 region. The structural similarity between Aâ-(1-
42) and Aâ-(25-35) in HFIP/water 80/20 suggests a
common neurotoxic mechanism possibly based on the
formation of ion channels into the membrane.35 Inter-
estingly the conformational peculiarities of Aâ-(25-35)
in the HFIP/water 80/20 apolar environment further
support the recent hypothesis that amyloid peptides can
enhance the association of liposomes with cells, produc-
ing an enhancement of viral infection.13 This result gives
further relevance to the previously highlighted analogies
between the fusion domain of the influenza virus
haemagglutinin,36 several membrane interacting pep-
tides,37 and Aâ-peptides.
The solution structure of Aâ-(25-35) in an inverted
HFIP/water ratio (20/80 v/v) is characterized by a
decrease of regular folding in the C-terminal region,
while a significant regularity in the structure is con-
served only in the 25-28 segment, characterized by a
â-turn structure. The structural peculiarity of this
region emphasizes the role of the kink conformation
encompassing residues 25-27 that appears to be a
general feature of amyloid peptides in all the solution
conditions explored as well as in the crystal state. This
region in the full-length Aâ corresponds to a kink
connecting two helical stretches12,32,33 and has a similar
arrangement also in the model of fibrillar Aâ-(1-40)
reported very recently.38
The difficulty of manipulating the natural proteins
to test their structural properties has prompted the
scientific community to devise suitable models sharing
with natural proteins the in vitro amyloidogenic prop-
erty. Similar to Aâ-(1-42), Aâ-(25-35) may assume
different conformations in vitro, depending on the
experimental conditions. However, a reduced size and
a higher solubility allow for a better biophysical char-
acterization that might help to dissect the contribution
of single interactions as hydrophobic, electrostatic, or
hydrogen bonds in the different conformational states.
In the present study we have determined the high-
resolution solution structure of the Aâ-(25-35) under
different experimental conditions by 2D NMR spectrom-
etry. The comparison among the structures obtained
might help unveil the regions that are mainly affected
by the environmental conditions.
The SDS micellar solution defines a nonisotropic
apolar environment where the membrane interaction
activities of the biomolecules can be tested. The solution
structure of Aâ-(25-35) in SDS confirmed the presence
of a helical stretch encompassing residues 28-34,
already reported in LiDS micelles;16 the region 25-27
characterized by a low number and poor quality of NOE
data appears less ordered in this medium. A comparison
of our SDS structure with the corresponding region of
the full length Aâ-(1-40)32 and Aâ-(1-42)33 in SDS
micelles shows that Aâ-(25-35) retains the structural
features of its parent peptide Aâ-(1-42) in similar
environments. The NMR analysis in the presence of spin
labels allows the definition of the orientation of Aâ-(25-
35) with respect to the micelles as a model of the
orientation with respect to the cell membrane. Indeed,
the perturbation of chemical shifts and intensities of the
signals in the presence of 16-doxylstearic acid shows
that only the C-terminal part of Aâ-(25-35) is inserted
into SDS micelles. These results suggest a natural
propensity of the C-terminal region, which shows the
typical amino acid composition of a transmembrane
helix,34 to interact with an apolar environment. More-
over, the resemblance to the solution structure of Aâ-
(1-40)32 and Aâ-(1-42)33 in SDS supports the hypoth-
esis of a similar orientation of the full-length Aâ with
respect to the cell membrane.
A possible biological role of this region is suggested
by the observation that the mutant of Aâ-(25-35) in
which the Asn 27 has been replaced with Ala loses the
neurotoxicity.16 The solution structure of this peptide
is very similar to that of the parent Aâ-(25-35) but for
the mutated residue, which is also included in the
N-terminal R-helix. The different structures of Aâ-(25-
35) in HFIP/water mixtures give experimental evidence
of the propensity of the peptide to lose the structural
regularity at a particular point of the sequence as a
function of the environment, behaving as a partially
unfolded intermediate in the aggregation process.
In the search for experimental conditions that may
favor the monitoring of different conformational states
using different polarity conditions, we resorted to HFIP/
water mixtures with different composition. These have
been proved, in the 80/20 HFIP/water composition, to
produce a completely stable, ordered solution of Aâ-(1-
40)19 and Aâ-(1-42).12 In contrast, the HFIP/water
mixtures with a content of water higher than 20%
produce insoluble aggregates. Thus, the observation of
the conformational properties of Aâ-(25-35) in mixtures
with different HFIP/water composition was performed
to monitor the conformational behavior of Aâ-(25-35)
in media with different polarity. The CD data recorded
in HFIP/water mixtures from 80/20 v/v HFIP/water to
20/80 v/v HFIP/water show that an amount of water
greater than 50% produces a drastic modification of the
Exp er im en ta l Section
Solid -P h a se P ep tid e Syn th esis a n d P u r ifica tion . Aâ-
(25-35) was synthesized according to published methods using
standard solid-phase synthesis techniques39 with a Milligen
9050 synthesizer. Protected amino acids and chemicals were
purchased from Bachem, Novabiochem, or Fluka (Switzer-
land). The resin loaded with methionine on the poly(ethylene
glycol)/polystyrene support (Fmoc-Met-PEG-PS) was from
Millipore (Waltham, MA). NR-Fmoc derivatives of amino acids
were used in the coupling reactions, and all lateral amino acid
protections were trifluoroacetic acid labile.

Amyloid â-Peptide 25-35
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 17 4237
Ta ble 2. Analytical Properties of Aâ-(25-35) Peptide
a
tR (min)
MH+ b
calculated
1061.3
sequence
purity (%)
I
II
found
H-Gly-Ser-Asn-Lys-Gly-Ala-Ile-Ile-Gly-Leu-Met-OH
>98
13.37
10.93
1061.2
a
b
tR is the retention time determined by analytical HPLC. The mass ion (MH+) was obtained from MALDI-TOF mass spectrometry.
Fmoc-Met-PEG-PS resin (0.21 mequiv/g, 0.5 g) was treated
with piperidine (20%) in DMF, and the NR-Fmoc amino acid
derivatives (4-fold excess) were sequentially coupled to the
growing peptide chain by using O-(7-azabenzotriazol-1-yl)-
1,1,3,3-tetramethyluronium hexafluorophosphate (HATU)40 (4-
fold excess) in DMF, and the coupling reaction time was 1 h.
Double coupling was required in the acylation step of Leu
and Ile. Piperidine (20%) in DMF was used to remove the Fmoc
group at all steps. After deprotection of the last NR-Fmoc
group, the peptide resin was washed with methanol and dried
in vacuo to yield the protected peptide-PEG-PS-resin. The
protected peptide was cleaved from the resin by treatment with
TFA/H2O/phenol/ethanedithiol/thioanisole (reagent K) (82.5:
5:5:2.5:5 v/v) 10 mL/0.5 g of resin at room temperature for 1
h.41 After filtration of the exhausted resin, the solvent was
concentrated in vacuo and the residue was triturated with
ether.
The crude peptide was purified by preparative reversed-
phase HPLC using a Water Delta Prep 4000 system with a
Waters PrepLC 40 mm assembly column C18 (30 cm × 4 cm,
300 Å, 15 mm spherical particle size column). The column was
perfused at a flow rate of 40 mL/min with a mobile phase
containing solvent A (water in 0.1% TFA), and a linear
gradient from 5% to 30% of solvent B (acetonitrile in 0.1% TFA)
in 25 min was adopted for the elution of the peptide. The pure
fraction was collected to yield a white powder after lyophiliza-
tion.
HPLC analyses were performed on a Beckman 125 liquid
chromatography fitted with an Alltech column C18 (4.6 mm
× 150 mm, 5 mm particle size) and equipped with a Beckman
168 diode array detector, using the above solvent system
(solvents A and B) programmed at a flow rate of 1 mL/min
with a linear gradient (I) from 0% to 25% or (II) from 0% to
40% B in 25 min. The molecular weight of the compound was
determined by a MALDI-TOF (matrix-assisted laser desorption
ionization time-of-flight) analysis using a Hewlett-Packard
G2025A LD-TOF mass spectrometer and R-cyano-4-hydroxy-
cinnamic acid as a matrix. The analytical properties of Aâ-
(25-35) peptide are reported in Table 2.
Cir cu la r Dich r oism . All CD spectra were recorded using
a J ASCO J -810 spectropolarimeter with a cell of 1.0 mm path
length. CD measurements were performed at 298 K, using a
range from 260 to 190 nm, 1 nm bandwidth, 4 accumulations,
and 10 nm/min of scanning speed. The SDS sample was
obtained by dissolving the peptide in an aqueous solution of
100 mM SDS at pH 4.15. Water/HFIP samples were prepared
by dissolution of Aâ-(25-35) in water/HFIP mixtures (compo-
sition 20/80, 30/70, 40/60, 50/50, 60/40, 70/30 by volume). The
concentration of the peptide in these solutions was 0.1 mM.
For estimation of secondary structure content, CD spectra were
analyzed by a linear combination fit using the reference data
of Greenfield and Fasman.42
signal was suppressed by low-power selective irradiation in
the homogated mode. DQF-COSY, TOCSY, and NOESY21-23
experiments were run in the phase-sensitive mode using
quadrature detection in ω1 by time-proportional phase increase
of the initial pulse. Data block sizes were 2048 addresses in t2
and 512 equidistant t1 values. Before Fourier transformation,
the time domain data matrices were multiplied by shifted sin2
functions in both dimensions. A mixing time of 70 ms was used
for the TOCSY experiments. NOESY experiments were run
at 300 K with mixing times in the range of 100-250 ms. The
qualitative and quantitative analysis of DQF-COSY, TOCSY,
and NOESY spectra was achieved using the SPARKY soft-
ware.24
Molecu la r Mod elin g. Peak volumes were translated into
upper distance bounds with the routine CALIBA of the
DYANA software.25
The necessary pseudoatom corrections were applied for
nonstereospecifically assigned protons at prochiral centers and
for the methyl group of aliphatic side chains. After the
redundant and duplicated constraints were discarded, the final
list for the structure of Aâ(25-35) in HFIP/water included 78
intraresidue constraints and 33 interresidue constraints, which
were used to generate an ensemble of 100 structures by the
standard protocol of simulated annealing in torsion angle space
implemented in DYANA (using 6000 steps). No dihedral angle
restraints and no hydrogen bond restraints were applied. The
best 20 structures, which had low values of the target functions
(0.83-1.19) were refined by in vacuo minimization in the
AMBER 1991 force field, using the program SANDER of the
AMBER 5.0 suite.30
To mimic the effect of solvent screening, all net charges were
reduced to 20% of their real value, and moreover, a distance-
dependent dielectric constant (ꢀ ) r) was used. The cutoff for
nonbonded interactions was 12 Å. The NMR-derived upper
bounds were imposed as semiparabolic penalty functions, with
force constants of 16 kcal mol-1 Å-2; the function was shifted
to linearity when the violation exceeded 0.5 Å. The best 10
structures after minimization had AMBER energies ranging
from -441.4 to -391.1 kcal/mol. The final structures were
analyzed using the Insight 98.0 program.44 Computations were
performed on SGI Indigo II computers.
Coor d in a tes. Coordinates have been deposited in the
Protein Data Bank. The access codes are 1QXC and 1QWP
for Aâ-(25-35) in HFIP/water 20/80 and HFIP/water 80/20
mixtures, respectively. The access code is 1QYT for Aâ-(25-
35) in 100 mM SDS aqueous solution.
Ack n ow led gm en t. We are very grateful to Profes-
sor P. A. Temussi for useful and stimulating discussions.
We thank Professor Kanfer for critical reading of the
manuscript. This work was supported by grants from
Regione Campania (legge regionale 41/94), Italy and
Progetto Strategico Alzheimer, Ministero della Salute-
Regione Lazio, Italy.
NMR Sp ectr om etr y. The samples for NMR spectroscopy
were prepared by dissolving the appropriate amount of Aâ-
(25-35) in 0.5 mL of solution to obtain a concentration 1.0
mM of peptide. For SDS measurements, the concentration of
deuterated SDS-d25 was 100 mM. Assuming an SDS micelle
aggregation number of 56, this corresponds to a micelle
concentration of 1.8 mM. The H2O/D2O ratio was 90/10.43 For
spin label experiments, the 5- and 16-doxylstearic acids were
solubilized in methanol-d4 and then added to the samples.
The composition of the water/HFIP mixtures for NMR
measurements was respectively 20/80 and 80/20 v/v. The
concentration of the peptide in these solutions was 1.0 mM.
NMR spectra were recorded on a Bruker DRX-600 spec-
trometer. One-dimensional (1D) NMR spectra were recorded
in the Fourier mode with quadrature detection, and the water
Ap p en d ix
Abbr evia tion s. Abbreviations used for amino acids
and designation of peptides follow the rules of the
IUPAC-IUB Commission of Biochemical Nomenclature
in J . Biol. Chem. 1972, 247, 977-983. Amino acid
symbols denote L-configuration unless otherwise indi-
cated. In addition, the following additional abbreviations
are used: AD, Alzheimer disease; SDS, sodium dodecyl
sulfate; CNS, central nervous system; SAR, structure-

4238 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 17
D’Ursi et al.
(20) Wu¨thrich, K. NMR of Proteins and Nucleic Acids; J ohn Wiley
& Sons: New York, 1986.
(21) Piantini, U.; Soerensen, O. W.; Ernst, R. R. Multiple quantum
filters for elucidating NMR coupling networks. J . Am. Chem.
Soc. 1982, 104, 6800-6801.
(22) Bax, A.; Davis, D. G. Mlev-17-based two-dimensional homo-
nuclear magnetization transfer spectroscopy. J . Magn. Reson.
1985, 65, 335-360.
(23) J eener, J .; Meyer, B. H.; Bachman, P.; Ernst, R. R. Investigation
of exchange processes by two-dimensional NMR spectroscopy.
J . Chem. Phys. 1979, 71, 4546-4553.
(24) Goddard, T. D.; Kneller, D. G. SPARKY 3, University of
California, San Francisco.
(25) Guntert, P.; Mumenthaler, C.; Wu¨thrich, K. Torsion angle
dynamics for NMR structure calculation with the new program
DYANA. J . Mol. Biol. 1997, 273, 283-298.
activity relationship; NMR, nuclear magnetic resonance;
CD, circular dichroism; Fmoc, 9-fluorenylmethoxycar-
bonyl; HPLC, high-performance liquid chromatography;
HFA, 1,1,1,3,3,3,-hexafluoroacetone; DQF-COSY, double
quantum filtered correlated spectroscopy; TOCSY, total
correlated spectroscopy; NOESY, nuclear Overhauser
enhancement spectroscopy; NOE, nuclear Overhauser
effect; MD, molecular dynamics; 1D, 2D, and 3D, one-,
two-, and three-dimensional; TFA, trifluoracetic acid;
HFIP, hexafluoroisopropanol.
Refer en ces
(26) J arvet, J .; Zdunek, J .; Damberg, P.; Graslund, A. Three-
dimensional structure and position of porcine motilin in sodium
dodecyl sulfate micelles determined by 1H NMR. Biochemistry
1997, 36, 8153-8163.
(27) Lindberg, M.; J arvet, J .; Langel, U.; Graslund, A. Secondary
structure and position of the cell-penetrating peptide transportan
in SDS micelles as determined by NMR. Biochemistry 2001, 40,
3141-3149.
(1) Sunde, M.; Blake, C. From the globular to the fibrous state:
protein structure and structural conversion in amyloid forma-
tion. Q. Rev. Biophys. 1998, 31, 1-39.
(2) Rochet, J . C.; Lansbury, P. T., J r. Amyloid fibrillogenesis:
themes and variations. Curr. Opin. Struct. Biol. 2000, 10, 60-
68.
(3) Sipe, J . D.; Cohen, A. S. Review: history of the amyloid fibril.
J . Struct. Biol. 2000, 130, 88-98.
(28) J ohnson, W. C., J r. Protein secondary structure and circular
dichroism: a practical guide. Proteins 1990, 7, 205-214.
(29) Weiner, S. J .; Kollman, P. A.; Case, D. A.; Singh, U. C.; Chio,
C.; Alagona, G.; Profeta, S.; Weiner, P. A new force field for
molecular mechanical simulation of nucleic acids and proteins.
J . Am. Chem. Soc. 1984, 106, 765-784.
(30) Case, D. A.; Pearlman, D. A.; Caldwell, J . W.; Cheatham, T. E.,
III; Ross, W. S.; Simmerling, C. L.; Darden, T. A.; Merz, K. M.;
Stanton, R. V.; Cheng, A. L.; Vincent, J . J .; Crowley, M.;
Ferguson, D. M.; Radmer, R. J .; Seibel, G. L.; Singh, U. C.;
Weiner, P. K.; Kollman, P. A. AMBER 5; University of California,
San Francisco, 1997.
(4) Kirschner, D. A.; Abraham, C.; Selkoe, D. J . X-ray diffraction
from intraneuronal paired helical filaments and extraneuronal
amyloid fibers in Alzheimer disease indicates cross-â conforma-
tion. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 503-507.
(5) Glenner, G. G.; Wong, C. W. Alzheimer’s disease and Down’s
syndrome: sharing of a unique cerebrovascular amyloid fibril
protein. Biochem. Biophys. Res. Commun. 1984, 120, 885-890.
(6) Selkoe, D. J . Translating cell biology into therapeutic advances
in Alzheimer’s disease. Nature 1999, 399 (Suppl.), A23-A31.
(7) Selkoe, D. J . Amyloid â-protein and the genetics of Alzheimer’s
disease. J . Biol. Chem. 1996, 271, 18295-18298.
(8) Temussi, P. A.; Masino, L.; Pastore, A. From Alzheimer to
Huntington: why is a structural understanding so difficult?
EMBO J . 2003, 22, 355-361.
(31) Hutchinson, E. G.; Thornton, J . M. PROMOTIFsa program to
identify and analyze structural motifs in proteins. Protein Sci.
1996, 2, 212-220.
(9) Terry, R. D. The neuropathology of Alzheimer disease and the
structural basis of its cognitive alterations. In Alzheimer Disease;
Terry, R. D., Katzman, R., Bick, K. L., Sisodia, S. S., Eds.;
Lippincott Williams and Wilkins: Philadelphia, PA, 1999; pp
187-206.
(10) Lambert, M. P.; Barlow, A. K.; Chromy, B. A.; Edwards, C.;
Freed, R.; Liosatos, M.; Morgan, T. E.; Rozovsky, I.; Trommer,
B.; Viola, K. L.; Wals, P.; Zhang, C.; Finch, C. E.; Krafft, G. A.;
Klein, W. L. Diffusible, nonfibrillar ligands derived from Aâ1-
42 are potent central nervous system neurotoxins. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 6448-6453.
(11) Harper, J . D.; Wong, S. S.; Lieber, C. M.; Lansbury, P. T.
Assembly of Aâ amyloid protofibrils: an in vitro model for a
possible early event in Alzheimer’s disease. Biochemistry 1999,
38, 8972-8980.
(12) Crescenzi, O.; Tomaselli, S.; Guerrini, R.; Salvadori, S.; D’Ursi,
A. M.; Temussi, P. A.; Picone, D. Solution structure of the
Alzheimer amyloid â-peptide (1-42) in an apolar microenviron-
ment. Similarity with a virus fusion domain. Eur. J . Biochem
2002, 269, 5642-5648.
(13) Wojtowicz, W. M.; Farzan, M.; J oyal, J . L.; Carter, K.; Babcock,
G. J .; Israel, D. I.; Sodroski, J .; Mirzabekov, T. Stimulation of
enveloped virus infection by â-amyloid fibrils. J . Biol. Chem.
2002, 277, 35019-35024.
(14) Pike, C. J .; Walencewicz-Wasserman, A. J .; Kosmoski, J .; Cribbs,
D. H.; Glabe, C. G.; Cotman, C. W. Structure-activity analyses
of â-amyloid peptides: contributions of the â 25-35 region to
aggregation and neurotoxicity. J . Neurochem. 1995, 64, 253-
265.
(15) Terzi, E.; Holzemann, G.; Seelig, J . Alzheimer â-amyloid peptide
25-35: electrostatic interactions with phospholipid membranes.
Biochemistry 1994, 33, 7434-7441.
(16) Kohno, T.; Kobayashi, K.; Maeda, T.; Sato, K.; Takashima, A.
Three-dimensional structures of the amyloid â-peptide (25-35)
in membrane-mimicking environment. Biochemistry 1996, 35,
16094-16104.
(32) Coles, M.; Bicknell, W.; Watson, A. A.; Fairlie, D. P.; Craik, D.
J . Solution structure of amyloid â-peptide(1-40) in a water-
micelle environment. Is the membrane-spanning domain where
we think it is? Biochemistry 1998, 37, 11064-11077.
(33) Shao, H.; J ao, S.; Ma, K.; Zagorski, M. G. Solution structures of
micelle-bound amyloid â-(1-40) and â-(1-42) peptides of Alz-
heimer’s disease. J . Mol. Biol. 1999, 285, 755-773.
(34) Eilers, M.; Shekar, S. C.; Shieh, T.; Smith, S. O.; Fleming, P. J .
Internal packing of helical membrane proteins. Proc. Natl. Acad.
Sci. U.S.A. 2000, 97, 5796-5801.
(35) Rhee, S. K.; Quist, A. P.; Lal, R. Amyloid â-protein-(1-42) forms
calcium-permeable, Zn2+-sensitive channel. J . Biol. Chem. 1998,
273, 13379-13382.
(36) Han, X.; Bushweller, J . H.; Cafiso, D. S.; Tamm, L. K. Membrane
structure and fusion-triggering conformational change of the
fusion domain from influenza hemagglutinin. Nat. Struct. Biol.
2001, 8, 715-720.
(37) Hristova, K.; Dempsey, C. E.; White, S. H. Structure, location,
and lipid perturbations of melittin at the membrane interface.
Biophys. J . 2001, 80, 801-811.
(38) Petkova, A. T.; Ishii, Y.; Balbach, J . J .; Antzutkin, O. N.;
Leapman, R. D.; Delaglio, F.; Tycko, R. A structural model for
Alzheimer’s â-amyloid fibrils based on experimental constraints
from solid state NMR. Proc. Natl. Acad. Sci. U.S.A. 2002, 99,
16742-16747.
(39) Atherton, E.; Sheppard, R. C. In Solid Phase Peptide Synthesis;
Rickwood, D., Hames, B. D., Eds.; IRL Press: Oxford, U.K.,
1989.
(40) Carpino, L. A. 1-Hydroxy-7-azabenzotriazole. An efficient peptide
coupling additive. J . Am. Chem. Soc. 1993, 115, 4397-4398.
(41) King, D. S.; Fields, C. G.; Fields, G. B. A cleavage method which
minimizes side reactions following Fmoc solid phase peptide
synthesis. Int. J . Pept. Protein Res. 1990, 36, 255-266.
(42) Greenfield, N.; Fasman, G. D. Computed circular dichroism
spectra for the evaluation of protein conformation. Biochemistry
1969, 8, 4108-4116.
(43) Lauterwein, J .; Bosch, C.; Brown, L. R.; Wuthrich, K. Physico-
chemical studies of the protein-lipid interactions in melittin-
containing micelles. Biochim. Biophys. Acta 1979, 556, 244-
264.
(17) Del Mar Martinez-Senac, M.; Villalain, J .; Gomez-Fernandez,
J . C. Structure of the Alzheimer â-amyloid peptide (25-35) and
its interaction with negatively charged phospholipid vesicles.
Eur. J . Biochem. 1999, 265, 744-753.
(18) Rajan, R.; Awasthi, S. K.; Bhattachajya, S.; Balaram, P. Teflon-
coated peptides: hexafluoroacetone trihydrate as a structure
stabilizer for peptides. Biopolymers 1997, 42, 125-128.
(19) Vieira, E. P.; Hermel, H.; Mo¨hwald, H. Change and stabilization
of the amyloid-â (1-40) secondary structure by fluorocompounds.
Biochim. Biophys. Acta 2003, 1645, 6-14.
(44) MSI Molecular Simulations, 965 Scranton Road, San Diego, CA
92121-3752.
J M040773O