1H-azepine-4-amino-4-carboxylic acid: A new α,α-disubstituted ornithine analogue capable of inducing helix conformations in short Ala-Aib pentapeptides

Article abstract of DOI:10.1002/chem.201104023

A very efficient synthesis of orthogonally protected 1H-azepine-4-amino-4- carboxylic acid, abbreviated as Azn, a conformationally restricted analogue of ornithine, was realized. It was obtained on a gram scale in good overall yield in five steps, three of which did not require isolation of the intermediates, starting from the readily available 1-amino-4-oxo-cyclohexane-4-carboxylic acid. Both enantiomers were used for the preparation of pentapeptide models containing Ala, Aib, and Azn. Conformational studies using both spectroscopic techniques (NMR, CD) and molecular dynamics on model 5-mer peptides showed that the (R)-Azn isomer possesses a marked helicogenic effect. Copyright

Full text of DOI:10.1002/chem.201104023

FULL PAPER
DOI: 10.1002/chem.201104023
1H-Azepine-4-amino-4-carboxylic Acid: A New a,a-Disubstituted Ornithine
Analogue Capable of Inducing Helix Conformations in Short Ala-Aib
Pentapeptides
Sara Pellegrino,* Alessandro Contini,* Francesca Clerici, Alessandro Gori,
Donatella Nava, and Maria Luisa Gelmi[a]
Abstract: A very efficient synthesis of
orthogonally protected 1H-azepine-4-
amino-4-carboxylic acid, abbreviated as
Azn, a conformationally restricted ana-
logue of ornithine, was realized. It was
obtained on a gram scale in good over-
all yield in five steps, three of which
did not require isolation of the inter-
mediates, starting from the readily
available 1-amino-4-oxo-cyclohexane-
4-carboxylic acid. Both enantiomers
were used for the preparation of penta-
peptide models containing Ala, Aib,
and Azn. Conformational studies using
both spectroscopic techniques (NMR,
CD) and molecular dynamics on model
5-mer peptides showed that the (R)-
Azn isomer possesses a marked helico-
genic effect.
Keywords: conformation analysis ·
helical structures · molecular dy-
namics · peptidomimetics · protein
folding
Introduction
omers that adopt well-defined conformations reminiscent of
protein secondary structure.[3]
The production of marketed synthetic peptide drugs is in-
creasing yearly.[1] Indeed, their role as mediators of key bio-
logical functions and their unique intrinsic properties, such
as high biological activity, low toxicity, and high specificity,[2]
make them particularly attractive therapeutic agents. Never-
theless, the use of peptides containing natural amino acids is
often associated with limitations related to their metabolic
instability and low bioavailability. These drawbacks can be
overcome by preparing peptidomimetics, molecules de-
signed to mimic both the structural and biological features
of peptides. As is widely known, many biological functions
involve protein–protein interactions (PPIs), in which helical
structured motifs often play a fundamental role as recogni-
tion elements. They are characterized by a central polypep-
tide backbone that serves to project, along individual faces,
residues that mediate selective and specific recognition. The
a-helix therefore offers potential as a template for the elab-
oration of “rule-based” small-molecule PPI inhibitors as ar-
tificial peptides and “foldamers”, synthetic non-natural olig-
Typically, preparations of backbone mimetics that repro-
duce the local topography of the helical pattern take advant-
age of a,a-disubstituted amino acids that dramatically
reduce the available conformational space of the peptide
backbone.[4] The main representative amino acid of this class
is a-aminoisobutyric acid (Aib), which, like similar amino
acids,[5] promotes the formation of a- or 310-helices,[6,7a] these
being the two most prevalent helices found in proteins.
In the last decade, ever-increasing numbers of scientific
contributions on the synthesis of a,a-disubstituted amino
acids have appeared. In this context, numerous 1-amino-1-
cycloalkanecarboxylic acids (Acnc, n=3–12) have been pre-
pared, and have proved to be valuable in the preparation of
conformationally constrained peptide backbones.[4,6,7a] On
the contrary, few examples of a,a-disubstituted azacyclic
amino acids, such as the achiral piperidine-4-amino-4-car-
boxylic acid (Api) and its derivatives, have been used in the
preparation of peptides.[4b,e,7g,8,9]
Focusing on the azepino scaffold,[10] some 1H-azepine-3-
amino-3-carboxylic acid derivatives have been prepared,
most of them containing the lactam function. In three
patent applications, 2-oxo derivatives and their benzo-con-
densates have been claimed for the preparation of oligopep-
tides useful as melanoma inhibitors,[11a] brain function im-
provers,[11b] or psycho analeptics.[12] Furthermore, the prepa-
ration of antimicrobial agents by using 1H-azepine-4-amino-
4-carboxylic acid derivatives has been mentioned in another
patent application,[13] while the synthesis of the correspond-
ing 3-oxo derivatives has recently been reported.[14]
[a] Dr. S. Pellegrino, Dr. A. Contini, Prof. F. Clerici, Dr. A. Gori,
D. Nava, Prof. M. L. Gelmi
DISMAB, Sezione di Chimica Organica “A. Marchesini”
Facoltꢀ di Farmacia, Universitꢀ degli Studi Milano
Via Venezian 21, 20133 Milano (Italy)
Fax : (+39)02-50314476
E-mail: sara.pellegrino@unimi.it
alessandro.contini@unimi.it
Our interest in the preparation of constrained amino
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201104023.
acids is well documented,[15] especially carbocyclic a-amino
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

acids,[15b,c,e,f,i–o] and in this work we report the preparation of
the conformationally constrained ornithine analogue 1H-
azepine-4-amino-4-carboxylic acid 3, abbreviated as Azn,
and its use for the preparation of model peptidomimetics.
The key steps for obtaining 3 are shown in the retrosyn-
thetic analysis (Scheme 1). Starting from the readily avail-
able amino acid 1, containing a cyclohexanone scaffold,
amino acid 3 was obtained via lactam intermediate 2. Com-
ing amino form was not trivial. Different procedures were
investigated according to methods reported in the litera-
ture,[17–19] but a mixture of compounds or the starting materi-
al[20] was recovered. Positive results were achieved by selec-
tively transforming the lactam 2 into thiolactam 4 using the
Lawesson reagent (0.5 equiv) in CH2Cl2 at 608C.[21] Com-
pound 4, isolated in 81% yield, was then treated with
Raney Ni in MeOH at reflux to afford the azepino com-
pound 5 (83%). The reaction was found to be efficient
when operating on a small scale (200 mg), but on a multi-
gram scale (10 g) the yields were not reproducible. For this
reason, we shifted toward the preparation of methyl iso-
thiouronium salt 6, obtained by reaction of 4 with MeI in
THF,[22] which was not isolated but directly reduced with
NaBH3CN in MeOH/AcOH (pH 5) to afford pure amine 5
(64%) (Scheme 3).
Scheme 1. Retrosynthetic scheme for 1H-azepine-4-amino-4-carboxylic
acid.
pound 3 was synthesized on a gram scale and orthogonally
protected at its two nitrogen atoms, thus enabling its use for
the preparation of model peptidomimetics. As azepino de-
rivative 3 is chiral, its preparation represented an added
value with respect to Api since different effects on the pep-
tide conformation could be expected depending on the abso-
lute configuration of the starting amino acid.[7d]
Conformational studies using both spectroscopic tech-
niques (NMR, CD) and molecular dynamics on model pen-
tapeptides containing Ala, Aib, and Azn are also reported
herein.
Results and Discussion
Scheme 3. a) The Lawesson reagent, CH2Cl2, D, 81%; b) on small scale:
Raney Ni, MeOH, reflux, 83%; c) MeI, THF; d) NaCNBH3, MeOH,
AcOH, 64% over two steps; e) Ac2O, TEA, THF, 90%; f) PhCHO, 1,2-
dichloroethane, AcOH, NaBHACTHNUTRGNE(NUG OAc)3, 61%; g) c) then d) then f), 50%;
h) 6n HCl, reflux, quantitative yield; i) MeOH, propylene oxide, quanti-
Synthesis of Azn: The starting material for the preparation
of amino acid derivative 3 was the known keto compound
1,[16] which was prepared according to the standard protocol
on a multigram scale (40 g) and in good yield (80%). Start-
ing from the keto derivative 1, compound 2 was efficiently
obtained by a modified Schmidt procedure that involved the
addition of sodium azide (3 equiv) to a cooled solution
(ꢀ108C) of 1 in CHCl3 and in the presence of concentrated
H2SO4 under stirring. Performing the reaction in a sealed
round-bottomed flask was found to be crucial for achieving
a good yield, as was stopping the stirring after 30 min and
allowing the reaction mixture to warm to ambient tempera-
ture over 18 h. Pure lactam 2 could be obtained on a gram
scale in 87% yield (Scheme 2).
tative yield.
Analytical and spectroscopic data confirmed the selective
reduction of the lactam function, although the 1H NMR
spectrum was complicated. With the aim of clarifying the
possible presence of conformers, the N-acetyl derivative 7
was prepared (yield 90%) by reaction of 5 with acetic anhy-
dride in the presence of triethylamine (TEA) in THF
(Scheme 3).
1
The H NMR spectrum of 7 in CDCl3 at 258C revealed
Since compound 2 contains three carbonyl functions, se-
lective reduction of the lactam function to the correspond-
the presence of dual signals at d=6.53/6.39 (NH), d=3.74/
3.71 (OMe), and d=2.05/1.99 (Me), suggesting the presence
of E/Z isomers about the tertiary amide bond. Dynamic
NMR experiments (from 25 to 808C) in DMSO revealed co-
alescence of these pairs of signals at 808C, confirming the
presence of E/Z isomers at room temperature.
As compound 5 was the key intermediate for the prepara-
tion of the target model peptide, orthogonal protection of
its two nitrogen atoms was needed. By following a reductive
amination approach, the N-benzylamine 8 was obtained in
Scheme 2. a) H2SO4, NaN3, CHCl3, from ꢀ10 to 258C, 87%.
&
2
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Induction of Helical Conformations in Pentapeptides
FULL PAPER
61% yield by reacting compound 5 with benzaldehyde in
1,2-dichloroethane in the presence of acetic acid (pH 5,
30 min) and then adding NaBHACHTNUGRTENUNG(OAc)3 and allowing the re-
action to proceed for 12 h (Scheme 3). In order to improve
the overall yield of 8 starting from 4, we evaluated the possi-
bility of directly obtaining the N-benzyl derivative 8 without
the isolation of intermediates. The two reductive steps were
then performed in a “one-pot” fashion, directly affording N-
benzylamine 8 (50% instead of 38% yield).
Compound 8 was then transformed into amino acid 9a by
hydrolysis with 6n HCl (reflux, 8 h, quantitative yield). Free
amine 9b was isolated by treating 9a with propylene oxide
in MeOH (Scheme 3).
Synthesis of model pentapeptides: The use of the seven-
membered-ring azepino amino acid 9 in peptide synthesis
was explored. We planned the preparation of pentapeptide
models containing Ala and Aib, in which the Azn residue
was located at position i+1, that is, Fmoc-l-Ala-(R)-Azn-l-
Ala-Aib-l-AlaNH2 (11a) and Fmoc-l-Ala-(S)-Azn-l-Ala-
Aib-l-AlaNH2 (11b), differing only in the presence of the
two Azn stereoisomers.
With the aim of separating the Azn enantiomers, we plan-
ned the synthesis of the dipeptide Ala-Azn by using l-ala-
nine as both reagent and resolving agent following the
methodology of Carpino et al.,[23] who exploited the high re-
activity of acyl fluorides. First, the hydrochloride salt of rac-
emic Azn 9a was reacted with triethylsilyl chloride (2 equiv,
CH2Cl2, reflux) allowing simultaneous protection of the car-
boxylic acid and activation of the nitrogen atom.[24] After
2 h, the reaction mixture was cooled and diisopropylethyl-
Scheme 4. a) BSA, molecular sieves in anhydrous CH2Cl2 for 15 h, then
Fmoc-l-Ala-F for 10 min, 70%; b) HOAT, EDC in NMP for 1 h, then
H2N-l-Ala-Aib-l-Ala-CONH2 for 15 h, 30%; c) H2N-l-Ala-Aib-l-Ala-
CONH2, DCC in CH2Cl2, 48 h; 11a: 50%, 11b: 30%.
(20 min) before performing the coupling (Aib: 3 h; Ala:
overnight). The tripeptide was then cleaved from the resin
(TFA/H2O/triisopropylsilane (TIS), 90:5:5) and purified by
crystallization (Et2O).
Finally, pentapeptide Fmoc-l-Ala-(R)-Azn-l-Ala-Aib-l-
AlaNH2 (11a) was prepared by solution-phase synthesis.
Enantiopure Fmoc-l-Ala-(R)-AznOH (10a) was reacted
with Ala-Aib-AlaNH2 (3 equiv) according to a standard pro-
tocol (2 equiv 1-hydroxy-7-azabenzotriazole (HOAT),
1 equiv
N-(3-dimethylaminopropyl)-N’-ethylcarbodiimide
A
hydrochloride (EDC) in N-methylpyrrolidone (NMP), 258C,
15 h). Pentapeptide 11a was isolated in moderate yield
(30%) together with recovered dipeptide 10a (40%) after
purification by RP-HPLC. Interestingly, the diastereoisomer
Fmoc-l-Ala-(S)-AznOH (10b) was found to be dramatically
less reactive in the coupling reaction. Indeed, by following
the same protocol as described above, only the starting di-
peptide was recovered and no product formation was detect-
ed. Various attempts were then made to obtain the desired
pentapeptide 11b. First, the coupling was attempted by
using the already mentioned conditions with the aid of mi-
crowave irradiation (558C, 400 W, 30 min), but no positive
results were obtained. According to a known procedure,[7d]
a mixture of activating reagents (HOAT, 2 equiv; HBTU,
1 equiv) in the presence of DIPEA (4 equiv) was used, but
only after 5 days was a trace amount of 11b obtained. A fur-
ther attempt was made by converting the dipeptide 10b into
the corresponding more reactive acyl fluoride but, curiously,
no reaction occurred. The conversion of 10b into its acyl
chloride equivalent with SOCl2 also failed. Finally, better re-
sults were obtained when compound 10b was reacted with
the tripeptide Ala-Aib-AlaNH2 using N,N’-dicyclohexylcar-
bodiimide (DCC) (1.5 equiv) as the coupling activator. The
desired pentapeptide 11b was thereby obtained in 30%
yield. When the same procedure was applied to dipeptide
10a, the product 11a was obtained in 50% yield. From
added. A mixture of Fmoc-l-Ala-(R)-Azn (10a) and Fmoc-
l-Ala-(S)-Azn (10b) was isolated in 45% yield. With the
aim of avoiding the use of DIPEA, which hampers the pu-
rification of our basic dipeptides, we turned to the use of
free amino acid 9b in conjunction with bis-trimethylsilylace-
tamide (BSA, 4 equiv) as silylating agent.[25] The reaction
was performed at 258C (15 h) in CH2Cl2 in the presence of
activated molecular sieves. The coupling reaction with
Fmoc-l-Ala-F was immediate (10 min), affording Fmoc-l-
Ala-(R)-Azn (10a) and Fmoc-l-Ala-(S)-Azn (10b) in an im-
proved yield (70%). The two diastereoisomers were separa-
ble by flash column chromatography (Scheme 4).
The tripeptide Ala-Aib-AlaNH2 was prepared on a solid
phase using Rink amide resin (loading 0.8 mmolgꢀ1) as the
solid support. The first Ala was loaded onto the resin under
standard coupling conditions (5 equiv Ala, 5 equiv 1-hydrox-
ybenzotriazole (HOBt), 5 equiv O-(benzotriazol-1-yl)-
N,N,N’,N’-tetramethyluronium
(HBTU), 10 equiv DIPEA). The Fmoc protecting group was
removed by using 20% piperidine in DMF. Due to the
steric hindrance of the methyl groups on Ca, the coupling ef-
ficiency of Aib and the residue at the Aib+1 position in the
stepwise assembly of peptide chains is usually poor.[26] For
this reason, we pre-activated both Aib and the adjacent Ala
using N,N’-diisopropylcarbodiimide (DIC) and HOBt
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

S. Pellegrino, A. Contini et al.
these results, the difference in reactivity of the two diaste-
reomeric dipeptides is clear, which can most probably be at-
tributed to different degrees of exposure of the carboxylic
group as a result of different hydrophobic interactions be-
tween the Fmoc and benzyl groups.
ing peptides have shown that a right-handed 310-helical
structure usually first appears at the 7-mer level in
MeOH.[5a] In the present case, we observed a helical confor-
mation with a pentamer, suggesting that Azn has a strong
helicogenic effect.
Unfortunately, all attempts to prepare crystals suitable for
structure determination and an unequivocal assignment of
the (R)/(S) configuration of Azn were unsuccessful. Al-
though an NOE experiment on peptide 11a provided a clear
indication of (R)-Azn stereochemistry (see the section
headed NMR spectroscopy studies), a proposed assignment
should be considered in the following discussion.
NMR spectroscopy studies: Compared with CD, NMR spec-
troscopy techniques offer the advantage of monitoring pep-
tide conformations at the level of the individual residues. In
particular, dynamic NMR experiments can give information
on H-bonded protons, while 2D experiments can reveal the
spatial disposition of the peptide backbone.[28] As a further
confirmation of the CD data, which indicated a prevalent
right-handed 310-helix conformation for peptide 11a, dynam-
ic NMR experiments were also performed.
A full assignment of the protons of peptide 11a (see
Table S1 in the Supporting Information) was accomplished
by 500 MHz NMR analysis (1H, 13C, COSY, HSQC, and
HMBC NMR experiments). To obtain detailed information
on the conformational preferences, 2D-NOESY at different
mixing times, temperature-dependent chemical shift varia-
tions (from 0 to 608C), and deuterium exchange experi-
ments were performed in CD3CN solution.
Solution conformational analysis: To evaluate whether the
Azn scaffold could stabilize a helical conformation, we per-
formed conformational studies by CD and NMR.
CD measurements: The CD spectra of peptides 11a and 11b
were recorded in MeOH (300 mm solutions) in the far-UV
region (195–250 nm). As shown in Figure 1, in both cases,
the presence of a positive band below 200 nm, a minimum
in the region 205–210 nm (amide p!p* transition), and
a shoulder at around 222 nm (amide n!p* transition) indi-
As shown in Figure S3 (A and B) (in the Supporting In-
formation), various NOEs are operative in pentapeptide
11a. Medium to weak a,N
ACHUTGTNREN(NUG i,i+1) and a,NACHTNUGTRENNUNG
ure 2(A)) as well as a complete set of N,NACHTUNGTRENNNUG
cross-peaks (except for Ala3, the NH signal of which falls in
the aromatic region), which are characteristic of a helical
conformation, were observed (Figure 2(B)). Medium b,N-
AHCTUNGTREG(NNNU i,i+1) NOEs between residues 3/4 and 4/5 reinforce the hy-
pothesis of a helical structure (Figure S3 (A), in the Sup-
porting Information).[28]
Longer range a,NACTHNUGTRENNUG(i,i+2) and a,NACHTUTGNREN(NUGN i,i+4) NOEs allow a dis-
tinction of the helix types: the former is typical of a 310-
helix, while the latter is typical of an a-helix.[28] Our data
support the prevalence of a 310-helix since the a,NACHTUNGTRENNUNG(i,i+2)
NOE is present in the structured helical fragment corre-
sponding to Ala3/Ala5 (Figure 2A). Interestingly, a positive
Overhauser effect was found between the methyl group of
Ala5 and H2 of Azn (Figure S3 (B), in the Supporting Infor-
mation), confirming not only the formation of a helical
structure, but also that the absolute configuration of the
amino acid Azn is (R). In fact, this NOE is forbidden in the
case of the (S)-isomer (see Figure 5). Furthermore, the
Fmoc moiety was seen to be in spatial proximity with the
methyl group of Ala3 but not with the Azn moiety, thus
confirming the formation of a hydrogen bond between NH-
3 and the Fmoc carbonyl function that helps to orientate
this group.
The temperature dependences of the NH proton chemical
shifts, which provide information on inaccessible or intramo-
lecular H-bonds,[29] were evaluated as ꢀ2 ppbKꢀ1 for Ala3,
ꢀ2.1 ppbKꢀ1 for Aib, ꢀ1.7 ppbKꢀ1 for Ala5, and
ꢀ2.1 ppbKꢀ1 for one of the lower-field protons of NH2,
which fall within the typical ranges for intramolecularly H-
bonded protons (Figure 3). Conversely, an equilibrium be-
Figure 1. CD spectra in the 195–250 nm region of peptides 11a and 11b.
The ellipticity is expressed as mean residue molar ellipticity, [q]R/
degcm2 dmolꢀ1
.
cate the presence of right-handed helical structures. The R
ratios (q222/q208) suggest that the dominant helical structure
is the 310-helix (11a: R=0.55; 11b: R=0.5), although for
11b a blue shift of the p!p* band indicates the presence of
different conformations as well as a disordered fraction. The
lower intensity of the Cotton effect in peptide 11b than in
11a could be ascribed to the presence of a left-handed helix
contribution, as suggested by the computational study de-
scribed below. Furthermore, the Fmoc group can contribute
in the far-UV region with a positive peak below 214 nm,
suggesting that the helical contents of both peptides could
be underestimated.[27] It should also be noted that previous
studies on the conformational behavior of Ala-Aib-contain-
&
4
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Induction of Helical Conformations in Pentapeptides
FULL PAPER
Figure 3. Plot of the variations of NH proton chemical shifts in the
1H NMR spectra of 11a as
a function of increasing temperature
(10.41 mm solution in CD3CN, 300 MHz, 273–333 K).
From the above experiments, it is evident that the chemi-
cal shifts of two protons, NH-1 and NH-2, are remarkably
sensitive to heating, and that these protons readily undergo
exchange with deuterium. On the contrary, the observations
for NH-3, NH-4, and NH-5 are consistent with overwhelm-
ingly intramolecularly H-bonded protons. As regards the
NH2 group, the variable-temperature experiment revealed
the involvement of the proton resonating at lower field in
a hydrogen-bond, even though the same proton was ex-
changed very quickly.
All of the above data suggest that a 310-helix construct is
present at the C-termini and that the hydrogen bonds in-
volving NH-3, NH-4, NH-5, and NH-6 are preserved up to
608C.[7d,29]
Computational analysis: To gain a deeper knowledge about
the folding preferences of the (R)- and (S)-Azn-based pep-
tides 11a,b, we planned a molecular dynamics (MD) study
of pentapeptides containing l-Ala at positions 1, 3, and 5,
(R)-Azn or (S)-Azn at position 2, and Aib at position 4,
using the Amber11 package.[30]
The replica exchange molecular dynamics (REMD)
method,[31] a generalized-ensemble algorithm performing
random walks in energy space, thus helping a system to
escape from local energy traps, has been successfully adopt-
ed in secondary structure predictions due to its improved
sampling capabilities over standard MD runs. Recent litera-
ture evidences several successful applications of REMD for
the prediction of the native structures of small proteins such
the Villin headpiece domain HP35,[32] the pin WW
domain,[33] Trp-cage,[34] and GB1 peptide.[35] There have
been fewer instances of the use of MD to study the folding
of peptides containing non-natural amino acids.[36] Among
non-natural amino acids, the best studied is Aib, the tenden-
cy of which to induce 310-helix formation has been correctly
evidenced by MD simulations.[36b,c] When dealing with pep-
tide folding, the choice of an appropriate combination of
force-field type and solvent model appears to be critical.[37]
For instance, the popular Amber force fields ff03 and f99
have shown a strong bias toward helical structures,[38a,b]
while ff99SB has proved to be promising in ab initio folding
experiments,[38c,d] due to its lower a-helical propensity.[38a]
Figure 2. Sections of the NOESY spectrum of 11a (10.41 mm solution in
CD3CN, 500 MHz, RT). A) CH/NH region and B) NH region.
tween an intramolecularly H-bonded and a non-H-bonded
state is suggested for the NH moieties of Ala1 and Azn on
the basis of temperature dependences of ꢀ5.6 and
ꢀ6.8 ppbKꢀ1, respectively. These results were confirmed by
NH/ND exchange experiments (0–70887 s; 10.4 mm). The
exchange rate of H-5/D upon addition of D2O proved to be
very slow, with the signal of this proton still being present at
the end of the experiment (about 40% deuterated). Due to
partial overlap with the aromatic proton signals, the deuter-
ation of NH-3 and NH-4 could not be quantitatively ana-
lyzed, although the relative integrals of the aromatic region
at the beginning and end of the experiment did not show
a significant decrease. In marked contrast, NH-1, NH-2, and
both protons of NH2 exchanged immediately.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

S. Pellegrino, A. Contini et al.
The force-field behavior can be strongly influenced by the
adopted solvent model. In this framework, Shell and co-
workers[37] recently reported a study aiming to identify the
force field/solvation model combination best able to repro-
duce the a-helix or b-hairpin native structure of
reported crystal structure cannot be observed. Moreover,
a cluster analysis, performed by grouping coordinate frames
by mass-weighted root-mean-square-deviation (rmsd)
(Table 2, Table S2, and Figure S5 in the Supporting Informa-
four 12- to 16-mer proteins, and concluded that the
Table 2. Cluster analysis[a] of the final 25 ns of the 308.5 K REMD trajectory for P1
ff96 force field and the Generalized Born (GB) sol-
vation model with the atomic radii of Onufriev,
Bashford, and Case (igb=5)[39,40] provided a correct
balance of a-helical and b-hairpin tendencies of the
tested peptides. The ff99SB/igb=5 combination
also performed well in predicting helical conforma-
tions, but was less accurate for b-hairpins. However,
in the aforementioned studies, no peptides adopting
different helical structures, such as the 310- and pol-
yproline (PPII) helices, were investigated. More-
over, the relative performances of force-field-based
methods might be affected by the presence of non-
with ff96/igb=5 and ff99SB/igb=5 combinations.
ff96
ff99SB
f, yavg
[b]
[b]
#
pop [%]
f, yavg
pop [%]
1
2
3
4
35.7
22.7
20.7
17.8
ꢀ45.7ꢁ74.1, 54.7ꢁ155.1
13.9ꢁ77.7, 1.8ꢁ157.1
ꢀ53.8ꢁ19.1, 7.1ꢁ114.0
ꢀ13.0ꢁ80.4, 66.2ꢁ110.3
44.0
31.0
13.2
6.9
ꢀ50.6ꢁ9.6, ꢀ33.7ꢁ11.7
ꢀ22.4ꢁ87.2, ꢀ20.9ꢁ45.9
18.5ꢁ65.7, 85.6ꢁ77.0
46.0ꢁ8.1, 56.8ꢁ49.9
[a] Results for the most populated clusters, covering above 95% of the total popula-
tion, are reported. [b] Average f and y dihedrals (8) are obtained for each cluster rep-
resentative structure by averaging the corresponding values for all residues, without
the capping Ac and NHMe groups.
natural amino acids. For these reasons, and considering that
due to computational limitations we preferred implicit as
opposed to explicit solvation, we decided to evaluate both
the ff96 and the ff99SB force fields coupled with the igb=5
solvent model. Several sets of experimental data were avail-
able for Aib- and Ala-containing peptides,[5a,6c,d] but the 5-
mer Ala-Aib-Ala-Aib-Aib was particularly well suited for
our testing because its crystal structure, evidencing a 310-hel-
ical conformation, has been solved.[6c] The N- and C-termini
were capped with acetyl (Ac) and NHMe groups instead of
tBoc and OMe, respectively, as parameters for these resi-
dues are not present in the adopted force fields; the result-
ing model peptide is hereinafter referred to as P1. REMD
simulations, consisting of 12 replicas with temperatures ex-
ponentially spaced between 260 and 660 K, were performed
starting from unfolded conformations (see the Supporting
Information for details). The peptide was simulated for
50 ns, and H-bond, geometrical, and clustering analyses
were performed on the final 25 ns of the 308.5 K trajectory,
this temperature most closely resembling the experimental
conditions. Representative structures resulting from cluster-
ing analyses of the ff96 and ff99SB trajectories are com-
pared in Figure S5 in the Supporting Information.
tion), showed that with the ff99SB force field the most
populated cluster (pop.=44.0%) corresponded to the ex-
pected 310-helix. Two partially folded conformations (clus-
ters #2 and #3, pop.=31.0 and 13.2%, respectively) and
a small amount of left-handed helix (cluster #4, pop.=
6.9%) were also identified. On the contrary, predominantly
unfolded conformations were obtained with ff96, with the
most populated cluster (pop.=35.7%) corresponding to an
almost completely extended conformation. These calcula-
tions showed that, at least for the case reported herein, only
the ff99SB force field coupled with the igb=5 solvation
model provided results consistent with the reported crystal
structure, and hence this combination was adopted for the
following calculations.
Analogous REMD simulations were then run for Ac-Ala-
(R/S)-Azn-Ala-Aib-Ala-NMe peptides (hereafter referred
to as P2 and P3, respectively). Trajectories were analyzed in
terms of H-bonds (Table 3), 3D-Ramachandran plots, in
Table 3. H-bond[a] analysis of the final 25 ns of the 308.5 K ff99SB
REMD trajectory for P2 and P3 pentapeptides.
P2 (R)-Azn2
P3 (S)-Azn2
H-bonds
Occupancy [%]
By analyzing the occupancies of selected H-bonds
(Table 1), it can be observed that H-bonds characteristic of
the expected 310-helix (i+3!i) are only observed with the
ff99SB force field, together with a modestly persistent H-
bond typical of a-helices (1+4!i). On the other hand, with
the ff96 force field, an H-bond pattern compatible with the
Ala1 C=O···HN Aib4
Ala1 C=O···HN Ala5
Azn2 C=O···HN Ala5
39.4
7.2
13.6
26.5
3.5
12.5
[a] Only H-bonds with an occupancy >1% are reported. H-bonds involv-
ing the capping Ac and NHMe groups are not reported.
which the Cartesian z-coordinate represents the frequency
of occurrence of a particular combination of f and y dihe-
drals (Figure 4 and Figure S7 in the Supporting Informa-
tion), and cluster analyses (Table 4 and Figure 5 and
Table S2 and Figure S8 in the Supporting Information).
As is evident from the H-bond analysis (Table 3), the (R)-
Azn-containing peptide is characterized by two i+3!i H-
bonds, which are typical of 310-helices, between residues 1–4
(Ala and Aib, respectively) and 2–5 (Azn and Ala, respec-
tively). The third observed H-bond is of the i+4!i type, in-
Table 1. H-bond[a] analysis of the final 25 ns of the 308.5 K REMD tra-
jectory for P1 with ff96/igb=5 and ff99SB/igb=5 combinations.
ff96
ff99SB
H-bonds
Occupancy [%]
Ala1 C=O···HN Aib4
Ala1 C=O···HN Aib5
Aib2 C=O···HN Aib5
3.1
10.6
2.6
41.4
7.5
21.7
[a] Only H-bonds with an occupancy >1% are reported. H-bonds involv-
ing the capping Ac and NHMe groups are not reported.
&
6
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Induction of Helical Conformations in Pentapeptides
FULL PAPER
Figure 5. Cluster #1 representative structure as obtained from cluster
analysis of the 308.5 K REMD trajectory of Ac-Ala-(R)-Azn-Ala-Aib-
Ala-NMe peptide. C=O···HN distances are shown in angstroms. Selected
geometrical parameters (8): y1 =ꢀ22.1, f1 =ꢀ30.7, y2 =ꢀ52.7, f2 =
ꢀ73.3, y3 =ꢀ24.5, f3 =ꢀ57.7, y4 =ꢀ32.2, f4 =ꢀ104.1.
taining peptide P3 a minor tendency for adopting helical
configurations was observed, although definitive conclusions
could not be drawn from this analysis.
More details were provided by an analysis of the 3D
Ramachandran plots. Concerning the peptide P1 used as the
reference (see Figure S6 in the Supporting Information), it
is important to note that the Ramachandran plots obtained
here for the Aib residues are consistent with those reported
by Schweitzer-Stenner and co-workers for the tripeptide Ac-
Ala-Aib-Ala-OMe.[36c]
Figure 4. 3D Ramachandran plots for Ac-Ala-(R)-Azn-Ala-Aib-Ala-
NMe (P2) and Ac-Ala-(S)-Azn-Ala-Aib-Ala-NMe (P3). The z-axis rep-
resents the frequency of occurrence. A: 3D plots for average f and y di-
hedrals for the whole peptide. B: 3D plots for f and y dihedrals for the
sole residue 2 ((R)-Azn or (S)-Azn).
Analysis of 3D Ramachandran plots for penta-
peptide P2 (Figure 4 and Figure S7 in the Support-
Table 4. Cluster analysis[a] of the final 25 ns of the 308.5 K ff99SB REMD trajectory
for P2 and P3 pentapeptides.
ing Information) showed that, for the chiral (R)-
Azn residue, the angles f and y were mostly con-
fined to the right-handed helix region (ꢀ808ꢂ fꢂ
ꢀ308, ꢀ608ꢂyꢂ08, with sharp peaks at ꢀ558ꢂ fꢂ
ꢀ658 and ꢀ308ꢂyꢂꢀ408, Figure 4B1). This is in
contrast to what was observed for P1-Aib2, which
sampled both the right- and left-handed helices to
similar extents. The 3D Ramachandran plot ob-
tained for the whole peptide (Figure 4A1) also sup-
ported this observation, showing a clear preference
for the right-handed over the left-handed helix; the
latter is also present, but in a much lower amount
P2 (R)-Azn2
f, yavg
P3 (S)-Azn2
f, yavg
[b]
[b]
#
pop [%]
pop [%]
1
2
3
44.7
42.9
7.4
ꢀ66.5ꢁ30.7, ꢀ32.9ꢁ13.9
ꢀ42.8ꢁ64.2, ꢀ8.0ꢁ25.5
ꢀ53.2ꢁ67.9, 42.7ꢁ78.0
54.2
35.6
7.6
ꢀ58.6ꢁ72.0, 75.8ꢁ90.9
ꢀ41.5ꢁ58.7, ꢀ0.5ꢁ57.4
18.0ꢁ60.6, 51.3ꢁ46.4
[a] Results for the most populated clusters, covering above 95% of the total popula-
tion, are reported. [b] Average f and y dihedrals (8) are obtained for each cluster rep-
resentative structure by averaging the corresponding values for all residues, without
the capping Ac and NHMe groups.
dicating an a-helix, and has an occupancy of 7.2%, suggest-
ing that this secondary structure is also occasionally sam-
pled. For the peptide P3, having (S)-Azn at the 2-position,
a lower occupancy was observed for both the 1–4 and 1–5
H-bonds, while a comparable occupancy was observed for
the 2–5 H-bond. In conclusion, H-bond analysis revealed
a similar behavior for peptides P1 and P2, both of which
show a preference for 310-helices, while for the (S)-Azn-con-
than in the case of the peptide P1 (Figure S6 in the Support-
ing Information).
The behavior of P3 is substantially different. The confor-
mational space explored by (S)-Azn (Figure 4B2) is indeed
predominately in the left-handed helix region, with sharp
peaks at 358ꢂ fꢂ458 and 208ꢂyꢂ308, and only a minor
contribution from the right-handed helix region. These find-
ings further confirm that the behavior of Azn, and hence its
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

S. Pellegrino, A. Contini et al.
ability to induce a certain secondary structure, is strongly
dependent on its stereochemical configuration.
peptides Fmoc-l-Ala-(R)-Azn and Fmoc-l-Ala-(S)-Azn,
which were transformed into the pentapeptides Fmoc-l-Ala-
(R)-Azn-l-Ala-Aib-l-AlaNH2 (11a) and Fmoc-l-Ala-(S)-
Azn-l-Ala-Aib-l-AlaNH2 (11b).
The findings are in excellent accordance with the experi-
mental CD results, which indicated a prevalent helical sec-
ondary structure for the (R)-Azn2-containing peptide 11a,
but mixed or unordered conformations for the (S)-Azn2-
containing peptide 11b. Following on from the above discus-
sion, we can conclude that (R)-Azn behaves as an l-amino
acid and best exerts its helix-inducing capability when in-
serted into a sequence containing predominantly l-amino
acids, such as that of 11a. On the other hand, (S)-Azn be-
haves as a d-amino acid and its capability of inducing left-
handed helices can only be properly exerted in a peptide
containing d-amino acids, otherwise a prevalence of unor-
dered structures or mixtures of right- and left-handed heli-
ces will be produced, as observed experimentally for 11b.
A final confirmation was obtained by a cluster analysis of
the REMD trajectories. As indicated in Table 4 and
Table S2 in the Supporting Information, the first three most
populated clusters represented over 95% of the total popu-
lation for both peptides P2 and P3.
It has been shown that the stereochemistry at Ca plays an
important role in determining the folding preferences of
Azn-containing peptides. Both computational and spectro-
scopic data revealed the capability of (R)-Azn to direct the
folding of short peptide 11a toward a well-defined 310-helix
secondary structure. For this reason, we believe that Azn
might be used for the rational design of non-natural pep-
tides or peptidomimetic tools, leading to a finer and highly
specific modulation of biochemical pathways. Further stud-
ies will be oriented towards evaluating the behavior of (S)-
Azn in d-amino acid peptides, as it might prove to be
a potent tool for the design of left-handed-helix peptides or
peptidomimetics.
Experimental Section
Concerning P2, the angles f and y for the representative
structure of cluster #1 (pop.=44.7%) fall within the 310-
helix region, with a significant deviation only for the f4 di-
hedral angle of the C-terminal Ala5 (Figure 5). Interestingly,
cluster #2 (Table S2 and Figure S8 in the Supporting Infor-
mation), which is also well populated (pop.=42.9%), corre-
sponds to a partially folded structure in which the f and y
values for residues 1–3 are appropriate for a right-handed
helix, while the f and y dihedrals for Aib4 fall in the left-
handed helix region, analogously to what was observed for
the reference peptide P1 (Table S2 and Figure S5 in the Sup-
porting Information). Cluster #3 is only populated to
a minor extent (pop.=7.4%) and corresponds to an unfold-
ed conformation.
For the (S)-Azn2-containing peptide P3, cluster #1 (pop.=
54.2%) corresponds to an unfolded conformation, while
a structuring into a poorly defined helical conformation can
be observed for cluster #2 (pop.=35.6%). As also observed
for P1, a minor contribution from the left-handed helix con-
formation is observed for the least populated cluster #3
(pop.=7.6%).
General: Chemicals were obtained from Sigma-Aldrich and used without
further purification. Fmoc-protected amino acids and Fmoc-Rink amide
resin were purchased from Iris Biotech. RP-HPLC analyses were carried
out on JASCO BS-997–01 equipment. Preparative RP-HPLC analyses
were performed using a DENALI C-18 column from GRACE VYDAC
(10 mm, 250ꢂ22 mm). Two mobile phases were used: A=94.9% water,
5% MeCN, 0.1% TFA; B=95% MeCN, 4.9% water, 0.1% TFA. ESI
mass spectra were recorded on an LCQ Advantage spectrometer from
Thermo Finnigan. CD spectra were recorded using a JASCO J-810 spec-
tropolarimeter.
NMR spectroscopic analysis: NMR spectroscopic experiments were car-
ried out on either a Varian MERCURY 200 MHz (200 and 50 MHz for
1H and 13C, respectively) or a Bruker Avance 500 MHz spectrometer
1
(500 and 125 MHz for H and 13C, respectively). To take advantage of the
magnetic field value, measurements that required temperatures higher
than room temperature for observing coalescence were performed in an
apparatus with a 1H resonance frequency of 300 MHz (Bruker Avance
300 MHz spectrometer). 2D-NOESY experiments on peptide 11a were
performed at different mixing times (300, 600, 900 ms). Chemical shifts d
are given in ppm relative to CHCl3 or CD3CN as internal standards, and
coupling constants J are reported in hertz (Hz).
Methyl 5-benzoylamino-2-oxa-azepan-5-carboxylate (2): Keto compound
1 (3 g, 10.9 mmol) was dissolved in CHCl3 (45 mL) and the solution was
cooled to ꢀ108C under vigorous stirring. 98% H2SO4 (1.75 mL) was
added dropwise and then NaN3 (2.2 g, 32.7 mmol) was added in small
portions over a period of 20 min. The reaction mixture was allowed to
warm to 258C over 30 min, then the stirring was stopped and the mixture
was left for 18 h (TLC: CH2Cl2/MeOH, 10:1). The reaction was then
quenched with ice (5 g) and the mixture was neutralized to pH 7 with 2n
NaOH. The layers were separated and the aqueous phase was extracted
with CH2Cl2 (3ꢂ30 mL). The combined organic layers were dried over
Na2SO4. After evaporation of the solvent, the crude product crystallized
to afford pure compound 2 (2.8 g, 87%) as a white solid. M.p. 90–928C
(Et2O); 1H NMR (200 MHz, CDCl3, 258C): d=7.78–7.75 (m, 2H), 7.59–
7.41 (m, 3H), 6.28 (s, 1H, exch.), 5.30 (t, J=7.2 Hz, 1H, exch.), 3.77 (s,
3H), 3.46–3.36 (m, 2H; H-7), 2.75–2.40 (m, 2H; H-3), 2.75–2.35 (m, 2H;
H-6), 2.35–2.25 ppm (m, 2H; H-4); 13C NMR (50 MHz, CDCl3, 258C):
d=178.0, 174.3, 168.3, 134.4, 132.6, 129.2, 127.8, 61.9 (C-5), 53.4 (OMe),
37.8 (C-7), 36.4 (C-3), 31.0 (C-6), 30.4 ppm (C-4); IR (CH2Cl2): n˜ =3291,
1737, 1650 cmꢀ1; MS (APCI): m/z: 291.1 (M+1)+; elemental analysis
calcd (%) for C15H18N2O4: C 62.06, H 6.25, N 9.65; found: C 61.80, H
6.51, N 9.37.
Taken together, in accordance with the reported CD ex-
perimental results, the above data confirm that only (R)-
Azn has a strong helicogenic effect when inserted in a short
Ala- and Aib-containing sequence, while its enantiomer (S)-
Azn, under the same conditions, is unable to induce a well-
defined secondary structure.
Conclusion
Starting from the known keto amino acid 1, a very efficient
synthesis of orthogonally protected racemic amino acid 9,
a constrained analogue of ornithine, has been accomplished.
The use of l-alanine as reagent and resolving agent allowed
the preparation and isolation of pure diastereoisomeric di-
&
8
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Induction of Helical Conformations in Pentapeptides
FULL PAPER
Methyl 5-benzoylamino-2-thioxa-azepan-5-carboxylate (4): Compound 2
(10 g, 35 mmol) was dissolved in CH2Cl2 (100 mL) and the Lawesson re-
agent (7 g, 17.4 mmol) was added. After sealing in a tube, the mixture
was kept at 608C under stirring for 12 h (TLC: CH2Cl2/MeOH, 10:1).
After evaporation of the solvent, the crude mixture was purified by chro-
matography on silica gel (cyclohexane/AcOEt, 100:1 to 2:1). Pure com-
pound 4 (8.8 g, 81%) was isolated as a pale-yellow solid. M.p. 2268C
(dec.; Et2O); 1H NMR (200 MHz, CDCl3, 258C): d=8.60 (s, 1H, exch.),
7.81–7.77 (m, 2H), 7.55–7.44 (m, 3H), 6.45 (s, 1H, exch.), 3.77 (s, 3H),
3.57–3.53 (m, 2H; H-7), 3.14–3.08 (m, 2H; H-3), 2.47–2.35 (m, 2H; H-6),
2.35–2.20 ppm (m, 2H; H-4); 13C NMR (50 MHz, CDCl3, 258C): d=
208.5, 173.6, 167.9, 133.9, 132.6, 129.1, 127.5, 61.5 (C-5), 53.3 (OMe), 42.0
(C-7), 38.7 (C-3), 34.4 (C-6), 32.4 ppm (C-4); IR (KBr): n˜ =3436, 1735,
1645 cmꢀ1; MS (APCI): m/z: 307.1 (M+1)+; elemental analysis calcd (%)
for C15H18N2O3S: C 58.80, H 5.92, N 9.14; found: C 58.53, H 6.12, N 8.87.
MS (APCI): m/z: 367.2 (M+1)+; elemental analysis calcd (%) for
C22H26N2O3: C 72.11, H 7.15, N 7.64; found: C 71.73, H 7.40, N 7.31.
4-Amino-1-benzyl-azepan-4-carboxylic acid (9): Compound
8 (2.2 g,
6.0 mmol) was suspended in HCl (6n, 25 mL). After sealing in a tube,
the mixture was stirred at 1208C for 8 h. After cooling, the precipitated
solid was filtered off. The aqueous phase was washed with Et2O (2ꢂ
20 mL) and then concentrated to dryness to afford 9a (2 g) in quantita-
tive yield. Salt 9a was then suspended in MeOH (50 mL), and propylene
oxide (500 mL, 7.2 mmol) was added. The mixture was stirred and heated
under reflux for 1 h. The solvent was then evaporated and the crude resi-
due was crystallized from Et2O to afford free amino acid 9b (1.5 g) in
quantitative yield. 9a: Mixture of conformers: M.p. 1958C (MeOH/
Et2O); 1H NMR (200 MHz, D2O, 25 8C): d=7.47 (brs, 5H), 4.34 (s, 2H;
PhCH2), 3.60–3.35 (m, 3H; H-2, H-7), 3.17–3.10 (m, 1H; H-7), 2.69–
1.97 ppm (m, 6H; H-3, H-5, H-6); 1H NMR (D2O, 90 8C): d=8.20 (s,
5H), 5.06 (s, 2H; PhCH2), 4.30–4.14 (m, 2H; H-2), 4.14–4.00 (m, 2H; H-
7), 3.35–3.26 (m, 1H; H-3), 3.15–2.96 (m, 2H; H-3, H-5), 2.84–2.72 ppm
(m, 3H; H-6, H-5); 13C NMR (50 MHz, D2O, 25 8C): d=173.6, 131.4
(131.2), 130.5, 129.5, 129.0 (128.9), 61.4 (61.3, CPh), 61.2 (60.8, C-4), 56.3
(C-7), 49.4 (49.3, C-2), 33.3 (33.8, C-5), 30.1 (30.6, C-3), 19.6 ppm (20.6,
C-6). Detected signals for the second conformer: 1H/13C NMR (D2O,
278C): d=3.53, 3.13 (d=56.3, CH2-7), 3.60, 3.39 (d=49.3, CH2-2), 2.64,
2.30 (d=30.6, CH2-3), 2.40, 1.98 (d=33.8, CH2-5), 2.02 ppm (d=20.6, C-
6); IR (KBr): n˜ =3435, 1726, 1631 cmꢀ1; ESI: m/z: 249.5 (M+1)+; ele-
mental analysis calcd (%) for C14H22Cl2N2O2: C 52.34, H 6.90, N 8.72;
Methyl 4-benzoylamino-azepan-4-carboxylate (5): Method A: Compound
4 (200 mg, 0.693 mmol) was dissolved in MeOH (4 mL) and Raney Ni
(7.7 g of a methanolic slurry) was added. The mixture was heated at
reflux for 2 h (TLC: CH2Cl2/MeOH, 10:1). After filtration through a bed
of Celite, the solvent was removed and the crude mixture was purified by
chromatography on silica gel (CH2Cl2/MeOH, 50:1 to 5:1
+ 0.1%
NH4OH) to afford pure 5 (150 mg, 83%). Method B: MeI (2.1 mL,
30.8 mmol) was added to a solution of compound 4 (4.8 g, 15.4 mmol) in
anhydrous THF (70 mL). The mixture was stirred at 258C for 8 h to form
the solid isothiouronium salt 6. After evaporation of the solvent, the
crude mixture was used directly in the reductive step without further pu-
rification. MeOH (50 mL), AcOH (0.9 mL, to attain pH 5), and
NaBH3CN (650 mg, 10.2 mmol) were added sequentially, and stirring was
continued at 258C for 10 h, after which 6 had dissolved (TLC: CH2Cl2/
MeOH, 10:1). 2n HCl was added dropwise until pH 1 was attained, and
the aqueous layer was washed with CH2Cl2 (20 mL). 2n NaOH was
added to the aqueous layer until pH 8 was attained. After evaporation of
the solvent, the crude mixture was purified by chromatography on silica
gel (CH2Cl2/MeOH, 50:1 to 5:1 + 0.1% NH4OH). Pure amine 5 (2.8 g,
64%) was obtained as a colorless solid after crystallization. M.p. 1908C
(dec.; MeOH/Et2O); 1H NMR (200 MHz, CD3OD, 258C): d=7.87–7.81
(m, 2H), 7.60–7.40 (m, 3H), 3.72 (s, 3H), 3.42–3.29 (m, 4H; H-2, H-7),
2.90–2.70 (m, 1H), 2.65–2.44 (m, 1H), 2.44–2.22 (m, 2H), 2.18–1.80 ppm
(m, 2H; H-6); 13C NMR (500 MHz, [D6]DMSO, 258C): d=174.2, 169.6,
133.8, 131.9, 128.4, 127.5, 61.5 (C-4), 52.1 (OMe), 47.0 (C-7), 41.5 (C-2),
34.9 (C-5), 31.8 (C-3), 20.5 ppm (C-6); IR (KBr): n˜ =3435, 1733,
1635 cmꢀ1; MS (APCI): m/z: 277.3 (M+1); elemental analysis calcd (%)
for C15H20N2O3: C 65.20, H 7.30, 10.14; found: C 65.00, H 7.48, N 9.95.
1
found: C 51.95, H 7.30, N 8.41. 9b: M.p. 1888C (MeOH/iPr2O); H NMR
(200 MHz, CD3OD, 258C): d=7.55–7.47 (m, 5H), 4.36 (s, 2H; PhCH2),
3.60–3.50 (m, 2H), 3.50–3.30 (m, 2H), 2.60–2.40 (m, 1H; H-3), 2.40–2.20
(m, 3H; H-3, H-5), 2.20–2.00 ppm (m, 3H; H-5, H-6); 13C NMR
(75 MHz, CD3OD, 258C): d=174.3, 131.1, 130.7, 129.8, 129.4, 62.0 (CPh),
61.2 (C-4), 55.3 (C-7), 50.5 (C-2), 34.3 (C-5), 31.3 (C-3), 22.0 ppm (C-6);
IR (KBr): n˜ =3435, 1642 cmꢀ1; MS (ESI+): m/z: 249.1 (M+1)+.
Fmoc-l-Ala-(R)- and Fmoc-l-Ala-(S)-Azn (10a,b): Operating under ni-
trogen atmosphere and in the presence of powdered activated molecular
sieves, free amino acid 9b (750 mg, 3 mmol) was dissolved in anhydrous
CH2Cl2 (20 mL), and BSA (2.65 g, 12 mmol) was added to the stirred so-
lution. The mixture was kept overnight at 258C, and then Fmoc-l-Ala-F
(1.9 g, 6.0 mmol) was added. After 10 min, the mixture was filtered and
the filtrate was concentrated. The crude residue was purified by chroma-
tography on silica gel (CH2Cl2/MeOH, 50:1 to 5:1). The resulting mixture
of two diastereoisomers 10a,b (1.2 g, 70%) was separated by flash chro-
matography (GraceResolv silica cartridges; CH2Cl2/MeOH, 10:1) to
afford pure 10a (0.4 g, 25%; [a]D25 =43.28 (c=0.25, CHCl3)) and 10b
(0.35 g, 20%; [a]D25 =ꢀ107.28 (c=0.25, CHCl3)), along with unresolved
10a,b (0.37 g, 22%).
Methyl 1-benzyl-4-benzoylamino-azepan-4-carboxylate (8): Method C:
Amine 5 (1.4 g, 4.9 mmol) was added to 1,2-dichloroethane (25 mL) and
a few drops of MeOH were added until complete dissolution of the re-
agent. AcOH (0.44 mL, 4.9 mmol) and benzaldehyde (578 mg, 6.4 mmol)
were added and the mixture was stirred for 30 min, after which NaBH-
Preparation of H2N-l-Ala-Aib-l-Ala-CONH2 on a solid phase: Fmoc-
Rink amide resin (250 mg, loading 0.8 mmolgꢀ1) was swollen in CH2Cl2
for 20 min. The Fmoc group was removed with piperidine (20% in DMF,
2ꢂ1 mL, 15 + 5 min). Fmoc-Ala-OH (313 mg, 1 mmol) was pre-activated
with HOBt/HBTU (0.5m in DMF, 5 equiv) for 15 min. This solution and
DIPEA (1m in NMP, 10 equiv) were added to the resin and the mixture
was stirred for 1 h. Thereafter, the resin was washed with DMF (3 times)
and Fmoc cleavage was performed (20% piperidine in DMF, 15 min +
5 min). The resin was subsequently washed with DMF (6 times). Fmoc-
AibOH (325 mg, 1 mmol) was pre-activated with HOBt (135.1 mg,
1 mmol) and DIC (126.2 mg, 1 mmol) in DMF for 20 min. This solution
was added to the resin and the mixture was stirred for 3 h. The resin was
washed with DMF (3 times) and then the Fmoc group was removed with
20% piperidine in DMF (15 min + 5 min). The resin was subsequently
washed with DMF (6 times). Fmoc-Ala-OH (313 mg, 1 mmol) was pre-
activated with HOBt (135.12 mg, 1 mmol) and DIC (126.2 mg, 1 mmol)
in DMF for 20 min. This solution was added to the resin and the mixture
was stirred for 12 h. The resin was washed with DMF (3 times) and then
the Fmoc group was removed with 20% piperidine in DMF (15 min +
5 min). The resin was subsequently washed with DMF (6 times). The
peptide was cleaved from the resin with a mixture of TFA/H2O/TIS
(2 mL/66 mL/88 mL) under stirring for 2 h. The peptide (40 mg, 90%) was
precipitated with a cold mixture of petroleum ether/t-butyl methyl ether
ACHTUNGTRENNUNG(OAc)3 (1.9 g, 8.9 mmol) was gradually added over a period of 20 min.
The reaction mixture was stirred overnight (TLC: CH2Cl2/MeOH, 10:1).
2n HCl was added dropwise until pH 1 was attained and then 2n NaOH
was added until pH 8 was attained. The phases were separated and the
aqueous layer was extracted with AcOEt (3ꢂ25 mL). The combined or-
ganic extracts were dried over Na2SO4. After evaporation of the solvent,
the crude mixture was purified by chromatography on silica gel (cyclo-
hexane/AcOEt, 10:1 to 1:2). N-Benzylamine 8 was obtained as a colorless
oil (1.1 g, 61%). Method D: Compound 8 (1.21 g, 50% overall yield) was
directly obtained from amine 5, which was obtained from 4 (2.4 g,
7.7 mmol) according to Method B. After evaporation of the solvent, the
crude mixture was subjected to Method C to afford N-benzylamine 8,
which was isolated as a colorless oil after purification on silica gel. Mix-
ture of conformers: 1H NMR (200 MHz, CDCl3, 258C): d=7.82–7.78 (m,
2H), 7.57–7.40 (m, 3H), 7.25 (s, 5H), 7.15 (s, 1H, exch.), 3.72 (s, 3H),
3.60 (s, 2H), 2.92–2.70 (m, 2H; H-7, H-2), 2.70–2.00 (m, 6H; H-7, H-2,
H-5, H-3), 1.80–1.60 ppm (m, 2H; H-6); 13C NMR (50 MHz, CDCl3,
258C): d=174.6, 167.1, 134.3, 131.9, 129.4, 129.2, 128.7, 128.6, 127.5,
127.4, 63.8 (CH2Ph), 61.7 (C-4), 56.4 (C-7), 52.7 (OMe), 51.4 (C-2), 34.1
(C-5), 32.1 (C-3), 24.7 ppm (C-6); IR (KBr): n˜ =3435, 1738, 1639 cmꢀ1
;
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

S. Pellegrino, A. Contini et al.
(1:1) and purified by RP-HPLC (95% A 2 min, then to 30% A in
30 min, 20 mLminꢀ1 flow); 1H NMR (CD3OD, 258C): d=4.42–4.25 (m,
1H; CHMe), 3.98–3.83 (m; CHMe), 1.52 (d, J=6.9 Hz, 3H; Me), 1.50 (s,
6H; Me), 1.31 ppm (d, J=6.9 Hz, 3H; Me); MS (ESI+): m/z: 245.0
(M+1).
G. L. Verdine, J. E. Bradner, Nature 2009, 462, 182–188; e) E. F.
Lee, I. D. Sadowsky, B. J. Smith, P. E. Czabotar, K. J. Peterson-Kauf-
man, P. M. Colman, S. H. Gellman, W. D. Fairlie, Angew. Chem.
2009, 121, 4382–4386; Angew. Chem. Int. Ed. 2009, 48, 4318–4322;
f) E. A. Harker, D. S. Daniel, D. A. Guarracino, A. Schepartz,
Bioorg. Med. Chem. 2009, 17, 2038–2046; g) W. D. Wu, S. H. Gell-
man, Acc. Chem. Res. 2008, 41, 1231–1232; h) W. S. Horne, S. H.
Gellman, Acc. Chem. Res. 2008, 41, 1399–1408; i) W. S. Horne,
M. D. Boersma, M. A. Windsor, S. H. Gellman, Angew. Chem. 2008,
120, 2895–2898; Angew. Chem. Int. Ed. 2008, 47, 2853–2856; j) F.
Bernal, A. F. Tyler, S. J. Korsmeyer, L. D. Walensky, G. L. Verdine,
J. Am. Chem. Soc. 2007, 129, 2456–2457; k) J. A. Kritzer, R. Zutshi,
M. Cheah, F. A. Ran, R. Webman, T. M. Wongjiard, A. Schepartz,
ChemBioChem 2006, 7, 29–31; l) S. H. Gellman, Acc. Chem. Res.
1998, 31, 173–180.
General procedures for the preparation of pentapeptides 11a and 11b:
Method E: HOAT (27 mg, 0.2 mmol) and EDC (28.4 mg, 0.1 mmol) were
added to
a solution of dipeptide 10a (50 mg, 0.1 mmol) in NMP
(1.2 mL). The mixture was stirred at 258C for 1 h, and then the tripeptide
H2N-l-Ala-Aib-l-Ala-CONH2 (73 mg, 0.3 mmol) was added. The reac-
tion mixture was stirred overnight (TLC: CH2Cl2/CH3OH, 10:1). There-
after, the solvent was removed under vacuum, and the crude mixture was
first purified by chromatography on silica gel (CH2Cl2/MeOH, 1:0 to
0:1). After RP-HPLC purification (95% A to 30% A gradient over
30 min, 20 mLminꢀ1 flow) the product 11a (20 mg, 30%) and the starting
dipeptide 10a (20 mg, 40%) were collected and freeze-dried. Method F:
The tripeptide H2N-l-Ala-Aib-l-Ala-CONH2 (25 mg, 0.1 mmol) and
DCC (30 mg, 015 mmol) were added to a solution of dipeptide 10a or
10b (50 mg, 0.1 mmol) in CH2Cl2 (3 mL). The mixture was stirred at
258C for 48 h (TLC: CH2Cl2/CH3OH, 10:1). The white solid formed was
filtered off and the solvent was removed under vacuum. After RP-HPLC
purification under the same conditions of Method E, the product 11a
(35 mg, 50%) or 11b (20 mg, 30%) was obtained and freeze-dried. 11a:
MS (ESI+): m/z: 768.3 (M+1)+; 11b: MS (ESI+): m/z: 768.4 (M+1)+.
[4] a) Y. Demizu, M. Doi, M. Kurihara, H. Okuda, M. Nagano, H. Sue-
mune, M. Tanaka, Org. Biomol. Chem. 2011, 9, 3303–3312; b) E.
Gatto, A. Porchetta, L. Stella, I. Guryanov, F. Formaggio, C. Tonio-
lo, B. Kaptein, Q. B. Broxterman, M. Venanzi, Chem. Biodiversity
2008, 5, 1263–1278; c) M. Tanaka, Chem. Pharm. Bull. 2007, 55,
349–358; d) C. Toniolo, M. Crisma, F. Formaggio, C. Peggion, Q. B.
Broxterman, B. Kaptein, Biopolymers 2004, 76, 162–176; e) C. To-
niolo, M. Crisma, F. Formaggio, C. Peggion, Biopolymers 2001, 60,
396–419 and references cited therein.
Circular dichroism (CD): The CD spectra of compounds 11a and 11b
were recorded in MeOH (300 mm solutions) at 258C (0.1 cm quartz cell,
Hellma Suprasil). Spectra were obtained from 195 to 250 nm with a step
of 0.1 nm and a collection time per step of 1 s, averaging over three de-
terminations. The spectrum of the solvent was subtracted to eliminate in-
terference from the cell, solvent, and optical equipment. The CD spectra
were plotted as mean residue ellipticity q (degree ꢂ cm2 ꢂ dmolꢀ1) versus
wavelength l (nm). Noise reduction was applied using a Fourier-trans-
form filter program from JASCO.
[5] a) E. Longo, A. Moretto, F. Formaggio, C. Toniolo, Chirality 2011,
23, 756–760 and references cited therein; b) M. Alꢄas, S. Ayuso-Te-
jedor, J. Fernꢅndez-Recio, C. Cativiela, J. Sancho, Org. Biomol.
Chem. 2010, 8, 788–792; c) I. L. Karle, Biopolymers 2001, 60, 351–
365 and references cited therein; d) G. Basu, A. Kuki, Biopolymers
1993, 33, 995–1000; e) P. Balaram, Curr. Opin. Struct. Biol. 1992, 2,
845–851.
[6] a) C. Toniolo, A. Polese, F. Formaggio, M. Crisma, J. Kamphuis, J.
Am. Chem. Soc. 1996, 118, 2744–2745; b) P. Enkhbayar, K. Hikichi,
M. Osaki, R. H. Kretsinger, N. Matsushima, Proteins Struct. Funct.
Bioinf. 2006, 64, 691–699; c) A. K. Francis, M. Iqbal, P. Balaram, M.
Vijayan, Biopolymers 1983, 22, 1499–1505.
[7] a) Y. Demizu, M. Tanaka, M. Doi, M. Kurihara, H. Okuda, H. Sue-
mune, J. Pept. Sci. 2010, 16, 621–626; b) K. Wright, J.-F. Lohier, M.
Wakselman, J.-P. Mazaleyrat, C. Peggion, F. Formaggio, C. Toniolo,
Biopolymers 2007, 88, 797–806; c) P. Maity, B. Kçnig, J. Pept. Sci.
2008, 90, 8–27; d) M. Crisma, W. M. De Borggraeve, C. Peggion, F.
Formaggio, S. Royo, A. I. Jimꢆnez, C. Cativiela, C. Toniolo, Chem.
Eur. J. 2006, 12, 251–260; e) Y. Ohfune, T. Shinada, Bull. Chem.
Soc. Jpn. 2003, 76, 1115–1129; f) C. Toniolo, M. Crisma, F. Formag-
gio, E. Benedetti, A. Santini, R. Iacovino, M. Saviano, B. Di Blasio,
C. Pedone, J. Kamphuis, Biopolymers 1997, 41-44, 519–522; g) J.
Venkatraman, S. C. Shankaramma, P. Balaram, Chem. Rev. 2001,
101, 3131–3152.
Computational methods: Replica exchange molecular dynamic (REMD)
simulations were performed with the Sander module of the Amber11
program package.[30] The parameter sets for the Aib residue were ob-
tained from the RED database,[41] while those for (R)/(S)-Azn were de-
rived using REDIII software.[42] For the REMD simulation, the number
of replicas and the temperature range were selected, on the basis of the
number of atoms to be simulated, through the T-REMD server.[43] VMD
1.8.7 software[44] was used for visualizing trajectories, for H-bond analy-
ses, and for obtaining 3D Ramachandran plots. Cluster analyses were
performed with ptraj (Amber Tools 1.5)[30] by using the average-linkage
algorithm, as suggested by Shao and co-workers,[45] by sampling once
every two frames and using the pairwise mass-weighted rmsd between
frames as a metric.
[8] a) S. R. Haynes, S. D. Hagius, M. M. Juban, P. H. Elzer, R. P.
Hammer, J. Peptide Res. 2008, 66, 333–347; b) C. L. Wysong, T. S.
Yokum, G. A. Morales, R. L. Gundry, M. L. McLaughlin, R. P.
Hammer, J. Org. Chem. 1996, 61, 7650–7651.
[9] Z. Zhang, H. A. Remmer, D. D. Thomas, C. B. Karim, Biopolymers
2007, 88, 29–35.
[10] a) M. Bialer, Epilepsia 2007, 48, 46–48; b) M. N. Ponnuswamy,
M. M. Gromiha, S. M. M. Sony, K. Saraboji, Topics Het. Chem. 2006,
3, 81–147; c) J.-F. Liegeois, J. Bruhwyler, F. Rogister, J. Delarge,
Curr. Med. Chem. 1995, 2, 471–501.
Acknowledgements
Funding for this work was provided MIUR (PRIN 2008). Alessandro
Contini is grateful to the Consorzio Interuniversitario Lombardo per lꢃE-
laborazione Automatica (CILEA, Via R. Sanzio 4, Segrate, MI) for com-
putational facilities and to Dr. Emanuele Galletta for appreciated collab-
oration.
[11] a) W. J. Fairbrother, K. Deshayes, S. Fischer, J. A. Flygare, M. C.
Franklin, D. Vucic, PCT Int. Appl., PIXXD2 WO 2004072641A1,
2004; [Chem. Abstr. 2004, 141, 218943]; b) O. Murakami, T. Ikebe, I.
Hakamata, K. Anami, Jpn. Kokai Tokkyo Koho, Heisei, JKXXAF
JP 03034977A 19910214, 1991; [Chem. Abstr. 1991, 115, 71603].
[12] S. Murakami, T. Ikebe, I. Hakamada, K. Anami, Eur. Pat. Appl.,
EPXXDW EP 322779A2 19890705, 1989; [Chem. Abstr. 1990, 112,
77239].
[1] J. Tulla-Puche, F. Albericio, in The Power of Functional Resins in
Organic Synthesis (Eds.: J. Tulla-Puche, F. Albericio), Wiley-VCH,
2008, 495–527.
[2] C. Pichereau, C. Allary, Eur. Bio Pharm. Review 2005, 88–91.
[3] a) P. Jana, S. Maity, S. K. Maity, D. Haldar, Chem. Commun. 2011,
47, 2092–2094; b) T. A. Edwards, A. J. Wilson, Amino Acids 2011,
41, 743–754; c) W. S. Horne, L. M. Johnson, T. J. Ketas, P. J. Klasse,
M. Lu, J. P. Moore, S. H. Gellman, Proc. Natl. Acad. Sci. USA 2009,
106, 14751–14756; d) R. E. Moellering, M. Cornejo, T. N. Davis, C.
Del Bianco, J. C. Aster, S. C. Blacklow, A. L. Kung, D. G. Gilliland,
[13] S. J. Brickner, J. M. Chen, Z. B. Li, A. Marfat, M. J. Mitton-Fry,
M. A. Plotkin, U. D. Reilly, C. Subramanyam, Z. Zhang, S. Robin-
&
10
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Induction of Helical Conformations in Pentapeptides
FULL PAPER
son, U.S. Pat. Appl. Publ., USXXCO US 2008280879A1 20081113,
2008; [Chem. Abstr. 2008, 149, 556600].
[14] D. Nunez-Villanueva, M. A. Bonache, L. Infantes, M. T. Garcia-
Lopez, M. Martin-Martinez, R. Gonzalez-Muniz, J. Org. Chem.
2011, 76, 6592–6603.
hauser, Biochemistry 1995, 34, 3873–3877; d) K. Wꢇthrich, NMR of
Proteins and Nucleic Acids, Wiley, New York, 1986.
[29] J. D. Augspurger, V. A. Bindra, H. A. Scheraga, A. Kuki, Biochem-
istry 1995, 34, 2566–2576.
[30] D. A. Case, T. A. Darden, T. E. Cheatham, III, C. L. Simmerling, J.
Wang, R. E. Duke, R. Luo, R. C. Walker, W. Zhang, K. M. Merz,
B. P. Roberts, B. Wang, S. Hayik, A. Roitberg, G. Seabra, I. Koloss-
vai, K. F. Wong, F. Paesani, J. Vanicek, J. Liu, X. Wu, S. R. Brozell,
T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G.
Cui, D. R. Roe, D. H. Mathews, M. G. Seetin, C. Sagui, V. Babin, T.
Luchko, S. Gusarov, A. Kovalenko, P. A. Kollman (2010), AMBER
11, University of California, San Francisco.
[31] Y. Sugita, Y. Okamoto, Chem. Phys. Lett. 1999, 314, 141–151.
[32] H. Lei, C. Wu, H. Liu, Y. Duan, Proc. Natl. Acad. Sci. USA 2007,
104, 4925–4930.
[33] J. Mittal, R. B. Best, Biophys. J. 2010, 99, L26–L28.
[34] D. Paschek, H. Nymeyer, A. E. Garcꢄa, J. Struct. Biol. 2007, 157,
524–533.
[15] a) S. Pellegrino, F. Clerici, M. L. Gelmi, Tetrahedron 2008, 64, 5657–
5665; b) M. L. Gelmi, C. Cattaneo, S. Pellegrino, F. Clerici, M. Mon-
tali, C. Martini, J. Org. Chem. 2007, 72, 9811–9814; c) F. Caputo, F.
Clerici, M. L. Gelmi, S. Pellegrino, D. Pocar, Tetrahedron: Asymme-
try 2006, 17, 1430–1436; d) F. Caputo, F. Clerici, M. L. Gelmi, S. Pel-
legrino, T. Pilati, Tetrahedron: Asymmetry 2006, 17, 61–67; e) F.
Caputo, C. Cattaneo, F. Clerici, M. L. Gelmi, S. Pellegrino, J. Org.
Chem. 2006, 71, 8467–8472; f) F. Caputo, F. Clerici, M. L. Gelmi, D.
Nava, S. Pellegrino, Tetrahedron 2006, 62, 1288–1294; g) C. Cabrele,
F. Clerici, M. L. Gelmi, R. Gandolfi, F. Molinari, S. Pellegrino, Tet-
rahedron 2006, 62, 3502–3508; h) F. Clerici, M. L. Gelmi, S. Pellegri-
no, T. Pilati, J. Org. Chem. 2003, 68, 5286–5291; i) F. Clerici, M. L.
Gelmi, A. Gambini, J. Org. Chem. 2001, 66, 4941–4944; j) G. Abbia-
ti, F. Clerici, M. L. Gelmi, A. Gambini, T. Pilati, J. Org. Chem. 2001,
66, 6299–6304; k) F. Clerici, M. L. Gelmi, D. Pocar, T. Pilati, Tetra-
hedron: Asymmetry 2001, 12, 2663–2669; l) F. Clerici, M. L. Gelmi,
A. Gambini, D. Nava, Tetrahedron 2001, 57, 6429–6438; m) F. Cleri-
ci, M. L. Gelmi, A. Gambini, J. Org. Chem. 2000, 65, 6138–6141;
n) F. Clerici, M. L. Gelmi, D. Pocar, J. Org. Chem. 1999, 64, 726–
730; o) F. Clerici, M. L. Gelmi, A. Gambini, J. Org. Chem. 1999, 64,
5764–5767.
[16] A. Avenoza, C. Cativiela, J. H. Busto, M. A. Fernꢅndez-Recio, J. M.
Peregrina, F. Rodrꢄguez, Tetrahedron 2001, 57, 545–548.
[17] C. Pedregal, J. Ezquerra, A. Escribano, M. C. CarreÇo, J. L. G.
Ruano, Tetrahedron Lett. 1994, 35, 2053–2056.
[18] Y. Ohfune, M. Tomita, J. Am. Chem. Soc. 1982, 104, 3511–3513.
[19] C. Frauenfelder, H.-J. Borschberg, Helv. Chim. Acta 2000, 83, 1753–
1765.
[20] D. Fokas, Z. Wang, Synth. Commun. 2008, 38, 3816–3822.
[21] M. Jesberger, T. P. Davis, L. Barner, Synthesis 2003, 1929–1958.
[22] R. J. Sundberg, C. P. Walters, J. D. Bloom, J. Org. Chem. 1981, 46,
3730–3732.
[23] L. A. Carpino, D. Sadat-Aalaee, H. G. Chao, R. H. DeSelms, J. Am.
Chem. Soc. 1990, 112, 9651–9652.
[35] M. Andrec, A. K. Felts, E. Gallicchio, R. M. Levy, Proc. Natl. Acad.
Sci. USA 2005, 102, 6801–6806.
[36] a) H. Yu, X. Daura, W. F. van Gunsteren, Proteins Struct. Funct.
Bioinf. 2004, 54, 116–127; b) R. Bꢇrgi, X. Daura, A. Mark, M. Bel-
landa, S. Mammi, E. Peggion, W. J. van Gunsteren, Pept. Res. 2001,
57, 107–118; c) R. Schweitzer-Stenner, W. Gonzales, G. T. Bourne,
J. A. Feng, G. R. Marshall, J. Am. Chem. Soc. 2007, 129, 13095–
13109.
[37] M. S. Shell, R. Ritterson, K. A. J. Dill, J. Phys. Chem. B 2008, 112,
6878–6886.
[38] a) R. B. Best, G. Hummer, J. Phys. Chem. B 2009, 113, 9004–9015;
b) J. Mongan, C. L. Simmerling, J. A. McCammon, D. A. Case, A.
Onufriev, J. Chem. Theory Comput. 2007, 3, 156–169.
[39] G. D. Hawkins, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. 1996,
100, 19824–19839.
[40] A. Onufriev, D. Bashford, D. A. Case, Proteins Struct. Funct. Bioinf.
2004, 55, 383–394.
[41] C. Cꢆzard, E. Vanquelef, J. Pecher, P. Sonnet, P. Cieplak, E. Derat,
F.-Y. Dupradeau, 236th ACS National Meeting, Philadelphia, PA,
USA, August 17 - August 21, 2008.
[42] F.-Y. Dupradeau, A. Pigache, T. Zaffran, C. Savineau, R. Lelong, N.
Grivel, D. Lelong, W. Rosanski, P. Cieplak, Phys. Chem. Chem.
Phys. 2010, 12, 7821–7839.
[43] A. Patriksson, D. van der Spoel, Phys. Chem. Chem. Phys. 2008, 10,
2073–2077.
[24] H. R. Kricheldorf, Justus Liebigs Ann. Chem. 1972, 763, 17–38.
[25] H. Wenschuh, M. Beyermann, R. Winter, M. Bienert, D. Ionescu,
L. A. Carpino, Tetrahedron Lett. 1996, 37, 5483–5486.
[26] Y. Higashimoto, H. Kodama, M. Jelokhani-Niaraki, F. Kato, M.
Kondo, J. Biochem. 1999, 125, 705–712.
[44] W. Humphrey, A. Dalke, K. Schulten, J. Molec. Graphics 1996, 14,
33–38.
[27] A. M. Smith, R. J. Williams, C. Tang, P. Coppo, R. F. Collins, Adv.
Mater. 2008, 20, 37–41.
[45] J. Shao, S. W. Tanner, N. Thompson, T. E. Cheatham III, J. Chem.
Theory Comput. 2007, 3, 2312–2334.
[28] a) G. L. Millhauser, C. J. Stenland, P. Hanson, K. A. Bolin, F. J.
van de Ven, J. Mol. Biol. 1997, 267, 963–974; b) A. Contini, D.
Nava, P. Trimarco, J. Org. Chem. 2006, 71, 159–166; c) G. L. Mill-
Received: December 22, 2011
Revised: March 30, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!

S. Pellegrino, A. Contini et al.
Ahead of the curve: A very efficient
synthesis of orthogonally protected
1H-azepine-4-amino-4-carboxylic acid
(Azn), a conformationally restricted
ornithine analogue, was realized. Azn
was used to prepare two pentapeptides
(l-Ala-(R/S)-Azn-l-Ala-Aib-l-
AlaNH2). Both computational and
spectroscopic data (CD/NMR)
revealed the capability of (R)-Azn to
drive the folding of this short peptide
toward a well-defined 310-helix secon-
dary structure (see figure).
Peptides
S. Pellegrino,* A. Contini,* F. Clerici,
A. Gori, D. Nava,
M. L. Gelmi ................... &&&& — &&&&
1H-Azepine-4-amino-4-carboxylic
Acid: A New a,a-Disubstituted Orni-
thine Analogue Capable of Inducing
Helix Conformations in Short Ala-Aib
Pentapeptides
&
12
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!