Cyclic β-tetra- and pentapeptides: Synthesis through on-resin cyclization and conformational studies by X-ray, NMR and CD spectroscopy and theoretical calculations

Article abstract of DOI:10.1002/chem.200500249

The solution-phase synthesis of the simplest cyclic β-tetrapeptide, cyclo(β-Ala)₄ (4), as well as the solid-phase syntheses through side chain anchoring and on-resin cyclization of the cyclic β³- tetrapeptide cyclo(-β³hPheβ³3hLeu- β³hLys-β³hGln-) (14) and the first cyclic β³-pentapeptide cyclo(-β³hVal- β³hPhe-β³hLeu-β³hLys- β³hLys-) (19) are reported. Extensive computational as well as spectroscopic studies, including X-ray and NMR spectroscopy, were undertaken to determine the preferred conformations of these unnatural oligomers in solution and in the solid state. cyclo(β-Ala)₄ (4) with no chiral side chains is shown to exist as a mixture of rapidly interchanging conformers in solution, whereas inclusion of chiral side chains in the cyclo- β³-tetrapeptide causes stabilization of one dominating conformer. The cyclic β³-pentapeptide on the other hand shows larger conformational freedom. The X-ray structure of achiral cyclo(β-Ala)₄ (4) displays a C_i-symmetrical 16-membered ring with adjacent C=O and N-H atoms pointing pair wise up and down with respect to the ring plane. CD spectroscopic examinations of all cyclic β-peptides were undertaken and revealed results valuable as starting point for further structural investigations of these entities.

Full text of DOI:10.1002/chem.200500249

FULL PAPER
DOI: 10.1002/chem.200500249
Cyclic b-Tetra- and Pentapeptides: Synthesis through On-Resin Cyclization
and Conformational Studies by X-Ray, NMR and CD Spectroscopy and
Theoretical Calculations
Frank Büttner,[a] Anna S. Norgren,[a] Suode Zhang,[a] Samran Prabpai,[b]
Palangpon Kongsaeree,[b] and Per I. Arvidsson*[a]
Abstract: The solution-phase synthesis
of the simplest cyclic b-tetrapeptide,
cyclo(b-Ala)4 (4), as well as the solid-
phase syntheses through side chain an-
choring and on-resin cyclization of the
cyclic b3-tetrapeptide cyclo(-b3hPhe-
b3hLeu-b3hLys-b3hGln-) (14) and the
first cyclic b3-pentapeptide cyclo-
(-b3hVal-b3hPhe-b3hLeu-b3hLys-b3hLys-)
(19) are reported. Extensive computa-
tional as well as spectroscopic studies,
including X-ray and NMR spectrosco-
py, were undertaken to determine the
preferred conformations of these un-
natural oligomers in solution and in the
solid state. cyclo(b-Ala)4 (4) with no
chiral side chains is shown to exist as a
mixture of rapidly interchanging con-
formers in solution, whereas inclusion
of chiral side chains in the cyclo-b3-tet-
rapeptide causes stabilization of one
dominating conformer. The cyclic b3-
pentapeptide on the other hand shows
larger conformational freedom. The X-
ray structure of achiral cyclo(b-Ala)4
(4) displays a Ci-symmetrical 16-mem-
bered ring with adjacent C=Oand N-H
atoms pointing pair wise up and down
with respect to the ring plane. CD
spectroscopic examinations of all cyclic
b-peptides were undertaken and re-
vealed results valuable as starting point
for further structural investigations of
these entities.
Keywords: conformation analysis ·
cyclization · NMR spectroscopy ·
peptides
Introduction
solid state.[1a,2] On the other hand, b-peptides have also been
shown to display various biological activities, such as anti-
bacterial activity[3] and inhibition of cholesterol absorp-
tion.[4]
The investigation of b-peptides is one fundamental part of
the growing research field of “foldamers” (oligomers capa-
ble to form well-defined secondary structures) that has at-
tracted the attention of several research groups during the
last years.[1] On the one hand, structural investigations have
revealed interesting results that these artificial oligomers are
capable of forming secondary structures such as helices, b-
sheetlike strands and other structures in solution and in the
In contrast to the detailed work on linear b-peptides,
cyclic b-peptides have gained less interest, possibly because
of their challenging synthesis. The main reason of the small
number of reports is in our opinion the low solubility of the
protected intermediates involved in the solution-phase syn-
thesis. Usually laborious procedures, for example, the addi-
tion of the chaotropic salts, are necessary to dissolve the
cyclic peptide for final deprotection and to obtain water
soluble derivatives.[5] Progress in the synthesis of cyclic b-
peptides seems to be possible by on-resin cyclization. By
carrying out this crucial step on solid phase one can take ad-
vantage of the pseudo-dilution effect of a resin with a low
level of substitution. One publication reported several ex-
amples where the synthesis was executed through backbone
anchoring.[6] In contrast to this, we have focused on the
methodology through side chain anchoring and subsequent
cleavage from the resin/side chain deprotection at the same
time.[7] In this synthetic approach we can reduce the num-
[a] Dr. F. Büttner, A. S. Norgren, Dr. S. Zhang, Prof. Dr. P. I. Arvidsson
Department of Chemistry, Organic Chemistry
Uppsala University, 75124 Uppsala (Sweden)
Fax : (+46)18-471-3818
E-mail: per.arvidsson@kemi.uu.se
[b] S. Prabpai, Prof. Dr. P. Kongsaeree
Department of Chemistry, Faculty of Science
Mahidol University, Rama VI Road
Bangkok 10400 (Thailand)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author: Theoretical cal-
culations of 20, NMR data and Monte Carlo conformational sampling
of 14 an 19.
Chem. Eur. J. 2005, 11, 6145 – 6158
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6145

bers of the orthogonal protecting groups needed to three
and obtain water soluble peptides.
Results and Discussion
Motivation for the synthesis of cyclic b-peptide arises not
only from a fundamental chemical point of view, but also
from the interesting biological activities of these new enti-
ties found up to now. One cyclic b3-tetrapeptide was shown
to act as a somatostatin analogue,[5b,8] and a number of
cyclic b-tripeptides were found to display antiproliferative
activity and growth inhibition of human cancer cell lines.[9]
Additionally, it was reported that cyclic b-peptides can form
hollow, tubular structures, so called nanotubes, in the solid
state[10] and inside biomembranes, causing formation of
transmembrane ion channels.[5a]
Synthesis: To gain a starting point in the structural investiga-
tions of cyclic b-tetrapeptides, we decided to begin our stud-
ies with the simplest derivative cyclo(b-Ala)4 (4), that is, the
derivative without chiral side chain. The synthesis of 4 was
reported previously in the literature: one procedure includ-
ed solution-phase synthesis with nowadays uncommon re-
agents;[15] a recent method used backbone amide anchoring
and on-resin cyclization,[6] but no additional data such as
NMR- or X-ray structural data have been published so far.
For reasons of convenience we developed a simple solution-
phase method, using the standard protecting groups Boc
and Bzl and conventional coupling reagents (Scheme 1).
Considering that the majority of the few hitherto synthe-
sized cyclic b-peptides display biological activities, these re-
sults imply that cyclic b-peptides may share the property of
cyclic a-peptides to act as privileged structures due to their
inherent reduced conformational flexibility.[11] This con-
strained geometry is the main reason for the display of in-
creased and manifold biologic activities of cyclo-a-peptides
as for example, hormones, antibiotics, antimycotics, and
toxins. Conformational and molecular modeling studies on
key secondary structural elements, that is, b-turns, are also
widely undertaken with cyclic peptides, especially in early
stages of drug discovery.[12] Additionally, b-peptides show an
innate and total stability towards proteolytic degradation
caused by the unnatural building blocks.[13] For these reasons
we are interested in further detailed physicochemical inves-
tigations on cyclic b-peptides in order to rationalize the
structural aspects and open this field for more studies in the
fields of medicinal chemistry, material science, and catalysis.
In previous work it was shown by Gademann and See-
bach,[5c] and by us,[7] that cyclic b3-tripeptides exist in aque-
ous solution in one predominant conformation with uni-
formly oriented amide bonds and all side chains in lateral
positions. This predominant conformation was also found in
the structure of cyclo(b-Ala)3, the only high resolution X-
ray structure reported so far for cyclic b-tripeptides.[14] Now
we have extended the scope of our experiments and want to
report herein the solution phase synthesis, NMR structure,
and the X-ray structure of cyclo(b-Ala)4 (4) which has no
side chains. A theoretical study on the model peptide cyclo-
(b3hAla)4 (20) with four methyl
Scheme 1. a) TEA, isobutyl chloroformate, H-b-Ala-OBzl·HCl, THF, 3 d,
RT, 81%; b) 10% Pd/C, H2, THF/MeOH 1:1, 12 h, RT; c) TEA, isobutyl
chloroformate, THF, ꢀ158C, 15 min, then H-b-Ala-b-Ala-OBzl·TFA, 3 d,
RT, 63%; d) 10% Pd/C, H2, THF/H2O2:1, 12 h, RT; e) pentafluoro-
phenol, DIEA, DMF, RT for 16 h; f) TFA/ CH2Cl2 1:1; g) DIEA, THF,
32 h, 608C, 58%.
Thus, the linear precursor Boc-(b-Ala)4-OBzl (3) was syn-
thesized starting from the suitable protected monomer 1
and the dimer 2, respectively, by coupling via the mixed an-
hydride formed with isobutyl chloroformate. The Bzl group
of Boc-(b-Ala)4-OBzl (3) was removed by catalytic hydroge-
nation; the active pentafluorophenyl ester was formed,
which after Boc deprotection by TFA underwent cyclization
to 4 under high dilution conditions (2.5 mm) in a good yield
of 58% (conditions not optimized).
In the syntheses of the cyclic b-peptides with proteinogen-
ic amino acid side chains through on-resin cyclization we
utilized the method of side chain anchoring to gain the de-
sired products. A water soluble b-tripeptide was recently ob-
tained by us in excellent yield applying this technique.[7] To
extend the scope of this methodology we attempted to
anchor the side chains of two different amino acids utilizing
substituents will provide a basis
for understanding the confor-
mational behavior of the cyclic
b-tetrapeptides.
Additionally,
we will describe the syntheses
with on-resin cyclization and
the NMR-solution structures of
a cyclic b3-tetrapeptide and of
the first cyclic b3-pentapeptide, both with proteinogenic side
chains. CD spectroscopical data for all cyclic b-peptides are
reported as well.
6146
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6145 – 6158

Cyclic Peptides
FULL PAPER
two different linkers. For the synthesis of the cyclic b3-tetra-
peptide 14 we chose attachment of a b3-glutamic acid deriv-
ative to the Rink amide linker, whereas the cyclic b3-penta-
peptide 19 was originally anchored via a b3-lysine residue to
an activated carbonate resin. A three dimensional orthogo-
nal protecting group strategy, including Fmoc, Boc and All
groups, was utilized as general synthetic route, based on the
pioneering work of Trzeciak and Bannwarth,[16] and Alberi-
cio and co-workers.[17]
The required building blocks were synthesized by Arndt–
Eistert homologation of the corresponding Fmoc-protected
a-l-amino acid derivatives according to procedures publish-
ed by Seebach and co-workers,[18] and later Sewald and co-
workers.[19] For the side chain anchoring to the resin, and
the final cyclization, it was necessary to synthesize new
Fmoc-b3-amino acid derivatives with unprotected side
chains and allyl protected carboxyl functions. Thus, the diaz-
oketones Fmoc-Glu(OtBu)-CHN2 (5)[18] and Fmoc-Lys-
(Boc)-CHN2 (6)[18] were subjected to the conditions of the
Wolff rearrangement with allyl alcohol and N-methylmor-
pholine (NMM)[20] to give the corresponding allyl esters 7
and
8 in good yields of 67 and 77%, respectively
(Scheme 2). Cleavage of the Boc and the tBu protecting
Scheme 2. a) Allyl alcohol, silver benzoate, NMM, THF, ꢀ158C to RT,
5 h; 67% for 7, 77% for 8; b) TFA, TIPS, CH2Cl2, RT, 91% for 9, >
95% for 10.
Scheme 3. a) Fmoc-Rink linker, HBTU, HOBt, DIEA, CH2Cl2, RT, 3 h,
81–92%; b) 2% DBU/2% piperidine in DMF, 5”5 min; c) Fmoc-
b3hGlu-OAll (9), HBTU, HOBt, DIEA, DMF, RT, 3 h; d) Fmoc-b3hPhe-
OH, HBTU, HOBt, DIEA, DMF, RT, 3 h; e) Fmoc-b3hLeu-OH, HBTU,
HOBt, DIEA, DMF, RT, 3 h; f) Fmoc-b3hLys(Boc)-OH, HBTU, HOBt,
DIEA, DMF, RT, 3 h; g) [Pd(PPh3)4], DMSO/THF/0.5m HCl/NMM
4:4:2:1, RT, 4.5 h; h) HBTU, HOBt, DMF, RT, 17 h; i) TFA/H2O/TIPS
95:2.5:2.5, RT, 2.5 h.
groups of both derivatives was accomplished with TFA and
[21]
TIPS in CH2Cl2 and the special protected amino acid de-
rivatives Fmoc-b3-hGlu-OAll (9) and Fmoc-b3hLys-OAll
(10) were obtained in 91% and almost quantitative yield,
respectively.
Two types of TentaGel S resins were chosen for the solid-
phase synthesis of the peptides, as these resins combine
good swelling properties and a low level of substitution. The
synthesis of cyclo-b3-tetrapeptide 14 (Scheme 3) started with
attachment of Rink amide linker to TentaGel NH2 resin 11
to give 12.[22] The fully protected resin bound linear tetra-
peptide 13 was obtained by standard Fmoc solid-phase pep-
tides synthesis protocols[23] and included side chain anchor-
ing of Fmoc-b3hGlu-OAll (9) and successive coupling of the
other three b3-amino acid derivatives.
The following allyl deprotection was accomplished with
[Pd(PPh3)4] following a method reported by Bloomberg
et al.[24] The final Fmoc deprotection yielded the precursor
for on-resin cyclization, which was accomplished with
HBTU/HOBt and DIEA in DMF at room temperature.
After the reaction mixture was stirred over night the TNBS
test showed completeness of the reaction. The resin was
treated with TFA, which resulted in the cleavage from the
resin and Boc deprotection at the same time. The crude mix-
ture of products had to be purified extensively by reversed-
phase HPLC, as a variety of unidentified products was
formed. No known side products, which usual can occur
during cyclization with HBTU/HOBt (e.g. oligomers or tet-
ramethylguanidinium derivatives[25]), could be identified
through LC-MS analysis. The cyclic b3-tetrapeptide 14 was
obtained in 10% yield (based on anchored Rink amide
linker, conditions not optimized).
TentaGel S PHB resin 15, which includes a Wang linker,
was used for the synthesis of the cyclo-b3-pentapeptide 19
(Scheme 4). Reaction of the solid support with p-nitrophen-
yl chloroformate gave the active carbonate resin 16[26] and
Chem. Eur. J. 2005, 11, 6145 – 6158
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6147

P. I. Arvidsson et al.
afterwards the e-amino group of Fmoc-b3hLys-OAll·TFA
(10) was anchored to the resin in 60% yield.[27]
with slight better conversions to the desired cyclic product
and with full consumption of the starting material after 12 h
in contrast to the reaction executed with PyBOP/HOBt
where substantial amounts of the linear precursor could be
detected after the same time. Cleavage from the resin and
simultaneous Boc deprotection gave the cyclic product 19 in
crude form, which after extensive purification by reversed-
phase HPLC was obtained in pure form in only 9% yield.
Just as in the case of tetrapeptide 14, the assembly of the
linear, resin bound fully protected b-peptide 18 was carried
out following standard peptide synthesis procedures.[23] The
subsequent allyl deprotection was executed with [Pd(PPh3)4]
following a methodology of GuibØ and Albericio and co-
workers who introduced phenylsilane as an allyl group scav-
enger into the solid-phase peptide synthesis.[28] The final
Fmoc deprotection yielded the precursor for on-resin cycli-
zation. Two cyclization experiments with different coupling
reagents were carried out. It was found that, based on
HPLC chromatograms, the reaction with PyAOP proceeded
NMR spectroscopy of cyclo(b-Ala)4 (4): Both proton and
carbon NMR spectra of the cyclic b-tetrapeptide 4 in water
were found to be very simple, as expected, due to the lack
of side chains and the symmetry of the molecule. Three sig-
1
nals were recorded in the H NMR spectrum (two triplets
for the eight methylene groups and one broad singlet for all
NH resonances) and also three signals in the 13C NMR spec-
trum. These data clearly indicate the occurrence of rapid in-
terchanging conformers in solution; therefore mean values
of the signals were measured, as it is very unlikely that only
one fixed, symmetrical conformation of 4 in solution would
occur.
X-ray structure of cyclo(b-Ala)4 (4): Colorless crystals suita-
ble for X-ray analysis were grown by slow evaporation from
a MeOH/H2Osolution at room temperature. The crystal
structure of the 16-membered ring of 4 (Figure 1) forms a
flat Ci-symmetrical rhombic-shaped ring of dimensions
4.05”6.57 Š and does not adopt a hollow tube as in other
cyclic b-tripeptides and cyclic b-tetrapeptides.[10a,14]
Figure 1. An ORTEP drawing of the crystal structure of 4.
Instead only four of the eight intermolecular hydrogen
bonds of one molecule stabilize a “column-like” supra-
molecular structure which is “filled” with hydrogen bonds
(Figure 2a and b). The other four intermolecular hydrogen
bonds connect the four neighboring columns. The planes of
the cyclic subunits are not perpendicular to the column axis.
All the amide planes are in trans geometry; all four C=O
bonds are aligned almost in a parallel fashion relative to the
ring plane with two adjacent amide groups pointing up and
the other two pointing in the opposite direction, similarly
Scheme 4. a) p-Nitrophenyl chloroformate, NMM, CH2Cl2, RT, 14 h;
b) Fmoc-b3hLys-OAll·TFA (10), DIEA, DMF, RT, 14 h, 60%; c) 2%
DBU/2% piperidine in DMF, 5”5 min; d) Fmoc-b3hVal-OH, HBTU,
HOBt, DIEA, DMF, RT, 3 h; e) Fmoc-b3hPhe-OH, HBTU, HOBt,
DIEA, DMF, RT, 3 h; f) Fmoc-b3hLeu-OH, HBTU, HOBt, DIEA, DMF,
RT, 3 h; g) Fmoc-b3hLys(Boc)-OH, HBTU, HOBt, DIEA, DMF, RT, 3 h;
h) [Pd(PPh3)4], phenylsilane, CH2Cl2, RT, 2”10 min; i) PyAOP, DIEA,
DMF, RT, 12 h; j) TFA/H2O/TIPS 95:2.5:2.5, RT, 3 h.
6148
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6145 – 6158

Cyclic Peptides
FULL PAPER
side chain conformations. Conformer B is highly symmetri-
cal with all side chains in equatorial positions on the peptide
ring. Conformer A, on the other hand, is approximately C2
symmetrical with two of the side chains bent upwards, rela-
tive to the plane of the ring. Conformer C is characterized
by three uniformly oriented amide groups, while one amide
NH now is pointing in the opposite orientation. This leads
to an energetically disfavored NH–CH(b) staggered interac-
tion, but allows one intramolecular hydrogen bond to be
formed.
Conformers D and E have two amide groups oriented
“up” and two “down” relative to the plane of the ring. The
orientation of the amide groups in conformer D is alternat-
ing, while the orientation in conformer E is non-alternating
(cf. the orientation of the amide groups in the crystal struc-
ture of the unsubstituted cyclo(b-Ala)4 4). Both conformers
are destabilized by two staggered NH–CH(b) interactions,
which forces the side chains to adopt an axial orientation
with resulting 1,3-strain to CH2(a) and the carbonyl oxygen
of the following residue. Conformer D is stabilized by two
intramolecular hydrogen bonds, while conformer E forms
only one hydrogen bond between the neighboring amide
groups.
The relative energies of these conformers in the gas phase
may be readily explained by the stabilizing hydrogen bonds
and the destabilizing steric interactions discussed above. In
the gas phase, conformer D is most stable due to the lack of
a net dipole moment, while conformers A and B are strong-
ly disfavored by their large macrodipole moments. This sit-
uation is completely reversed when the relative energies are
calculated in water using the SCRF solvent model. Now, the
symmetrical conformer B is the most stable due to a mini-
mum of repulsive steric interactions, and an efficient stabili-
zation of the peptides macrodipole by the solvent. Accord-
Figure 2. Crystal packing of 4: a) view from the top of one column of
molecules, b) side view of one column. Three cyclic subunits of one
column are presented. Four of eight intermolecular H-bonds from one
molecule 4 are shown, and the others are omitted for clarity.
observed in a structure of (R,R,S,S)-cyclotetra-b-homoala-
nine determined from powder diffraction data.[10a] The ob-
served torsional angles suggest that a b-alanine unit in the
crystal structure of 4 adopts two distinct conformations. The
first conformation, with f, q and y dihedral angles of
101.46, ꢀ78.04 and 106.378, orients the C=Oand the N-H
groups of the same monomer pointing to the opposite direc-
tions whereas the second conformation with f, q and y di-
hedral angles of 80.03, 78.46 and 153.978, orients the two
groups into the same direction (where f is defined as CO-
N-C-C, q as N-C-C-COand y as C-C-CO-N torsional
angle[29]). With no side chains, the two methylene carbons in
each monomer, not restricted by the A1,3 strains,[30] contrib-
ute to the ring flexibility and thus allows more densely pack-
ing to maximize favorable van der Waals interactions. In ad-
dition, the crystal packing is stabilized through intermolecu-
lar hydrogen-bonding patterns similar to an antiparallel b-
sheetlike strand with an intersubunit N–Odistance of 2.88
and 2.92 Š.
ing to this DFT calculation, conformer
B is about
10 kcalmolꢀ1 more stable than the other low energy confor-
mations investigated. These calculations suggest that all the
conformers investigated are populated at room temperature,
assuming a low activation enthalpy for rotation of the amide
groups, but that conformer B should be the most predomi-
nant conformation in solution.
Theoretical studies of all-(S)-cyclo(b3hAla)4 (20): Prior to
the NMR spectroscopic analysis of the cyclo-b-peptides syn-
thesized, we initiated a theoretical investigation of cyclo(b3-
hAla)4 (20) as a simplified model for b3-tetrapeptide 14. The
objective of this study was to gain more quantitative infor-
mation on the conformers expected for this peptide than
provided by the conformational averaged structure which
would be derived from the NMR spectroscopic measure-
ments.
Initially, the conformational space of this model peptide
was sampled through an unrestrained Monte Carlo simula-
tion and the resulting approximately 2000 structures were
clustered into five families. One member from each family
was extracted and optimized with a density functional
method (B3LYP/6-31G(d,p)). The optimized structures, as
shown in Figure 3, are characterized by different backbone
conformations. Two conformers, A and B, have a uniform
orientation of all amide groups, but differ in backbone and
The five theoretical conformers found for cyclo(b3hAla)4
(20) were evaluated one at a time in order to get a higher
comprehension of which NOE is arising from which con-
former. With this information we can more easily determine
the major conformer of the cyclic b3-tetrapeptide when
looking at a recorded ROESY spectrum of 14.
Starting with the two conformers, where all the amide
bonds are directed in the same orientation, the residues
would either adopt a structure with completely symmetrical
shape B, or one that gives an unsymmetrical oval cyclic pep-
tide A. The four amino acid residues in conformer B would
give rise to identical NOE signals; all NH- and axial a-CH
protons would show NOE signals to neighboring axial a-
protons. The NOE signals expected for conformer A would
be a special case of conformer B in which some effects
would be stronger, while others would be weaker.
Chem. Eur. J. 2005, 11, 6145 – 6158
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6149

P. I. Arvidsson et al.
signals would arise between the
same residues. For the b-pro-
tons, only one of these would
be expected to have a NOE to
an a-proton positioned on an-
other residue. Further, since
one of the NH–CH(b) angles
reaches an angle close to 08, the
NOE arising between these two
protons are expected to be
strong. The other residues
would be expected to give rise
to weak inter-residue NOE sig-
nals.
The discussion regarding con-
former D and E are quite simi-
lar each other, since both con-
formers have two b3hAla resi-
due amide bonds in an up-ori-
entation and two in a down-ori-
entation. The NOE signals
arising between the CH(b) and
the two a-protons on the neigh-
boring carbon are expected to
be different whether they are
positioned on the up- or down-
oriented residues. The first sce-
nario will show stronger NOE
between CH2(a) pro-S and
CH(b) than between CH2(a)
pro-R and CH(b), while the
latter one gives rise to two
NOE signals of similar strength.
The NH–CH(b) protons on the
down-oriented residues are in
almost 08 relative each other,
and the NOE signals arising are
therefore expected to be strong.
As a last observation, the possi-
ble NH–CH(b) angles of con-
formation D and E are either
close to 180 or 08, which both
would result in large coupling
Figure 3. Calculated relative energies [kcalmolꢀ1] for the model peptide cyclo(b3hAla)4 (20) in the gas phase
(B3LYP/6-31G(d,p)) and in aqueous solution (B3LYP/6-311+G(d,p); SCRF solvation model).
Regarding the amide bonds in these two conformers, they
are all positioned in an almost 1808 angle relative to the b-
proton, differing slightly between the four amino acids. Con-
former B would give rise to symmetrical coupling constant
values, whereas the coupling constants of conformer A
would fluctuate within a small range.
One of the amino acid residues in conformer C has one of
its amide bonds in a different orientation, compared to the
other three. The four NH–CH(b) angles would be signifi-
cantly different in this case, since one of the angles is close
to 908, which would lead to one amide proton having a cou-
pling constant close to 0 Hz.
constants.
Also included in Figure 3 is a structure resembling that
found in solution by Seebach and co-workers for a somato-
statin mimicking cyclic b3-tetrapeptide,[5b] that is, C’. This
molecule was not found in the original molecular mechanics
based Monte Carlo sampling, but was included for compari-
son. The optimized geometry shown in Figure 3 is of compa-
rable energy as the other conformers investigated (by DFT
calculation); all attempts to build a “flat” structure, more
similar to the one reported, resulted in the structure shown
or structures similar to C, that is, with hydrogen bonds be-
tween neighboring residues, after unrestrained optimization.
The structures C and C’ are also expected to have similar
NOE signals.
Two of the amide protons are rather close, which should
give an NOE signal between those two. NH–CH2(a) NO E
6150
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6145 – 6158

Cyclic Peptides
FULL PAPER
NMR Spectroscopic investigations: A detailed 2D-NMR
spectroscopic study was undertaken to obtain high-resolu-
tion data on the conformational performance of the cyclic b-
peptides 14 and 19. The studies were made in methanol for
both peptides, whereas only the conformation of the penta-
peptide 19 could be established in water, due to solubility
problems of 14 in this solvent.
All spectra were recorded at 500 MHz by using 80% non-
deuterated solvent and solvent suppression by using the
WET pulse sequence.[31] A complete assignment of all 1H
resonances was possible by using TOCSY,[31] P.E. COSY,[32]
and ROESY[33] experiments. The assignment of 13C spectra
was either fully or partially established by using gHMBC[34]
technique. The sequence of 19 could also be established by
the use of this experiment, distinguishing b3hLys4 and
b3hLys5 from each other.
Solution structure of cyclic b3-tetrapeptide 14: The amide
1
region of the H NMR spectrum of 14 in methanol is shown
in Figure 4a. Traces of extraneous signals are visible which
probably stem from alternative conformers, indicating the
presence of one major conformer. But we can not complete-
ly exclude the option that impurities are causing these sig-
nals, although the peptide was pure according to LC-MS
analysis. All 3J(NH,CH(b)) coupling constants show high
values (7.6, 8.4, 8.6 and 9.4 Hz, respectively), which suggests
an anti arrangement between the amide and the b-protons
of the amino acid residues. The coupling constants also indi-
cate an unsymmetrical shape of 14, as the size of the cou-
pling constants and thereby the torsional angles differ signif-
icantly.
Figure 4. a) Expansion of the amide region of the 1H NMR spectrum of
14 recorded in methanol. b) Expansion of the amide region of the
ROESY spectrum of 14 recorded in methanol. Some of the NOE signals
discussed in the text are highlighted.
The diastereotopic a-protons could be assigned from the
P.E. COSY, assuming that the coupling constants between
the b-proton and the pro-R (equatorial) a-protons are small
and that the pro-S (axial) a-protons have a large coupling
constant to CH(b). The cross-peaks in the P.E. COSY are
thus assigned to be the pro-S protons. Even though the pro-
S and pro-R protons could be distinguished from each other,
the coupling constants for CH(b)–CH2(a) could not be de-
between NH and pro-S a-proton, and strong NOE signals
between NH and pro-R a-proton. This, in combination with
the opposite observations for b3hLys in position 3, indicates
that one conformation possesses an unsymmetrical shape.
The region of NH–CH(b) on the other hand shows signs of
dynamics, as NOE signals are observed between two amide
protons and their neighboring b-protons.
The conclusions made from the comparison of the five
theoretically calculated conformers and the observed NOE
signals of 14 are that the b3-tetrapeptide most likely adopts
an oval cyclic conformer having all amide bonds pointing in
the same direction, that is, similar to conformer A. O f
course, it should be noted that these five possible conforma-
tions of cyclo(b3hAla)4 (20) do not differ significantly in
energy, and some observed intra-residue NOE signals con-
tradict the conformer supported by the other NOE signals.
A total of 29 NOE signals were classified into three cate-
gories with the following upper bound distance limits:
strong 3.0 ꢁ 1 Š, medium 4.0 ꢁ 1 Š and weak 5.0 ꢁ 1 Š,
and used as such together with four (NH, CH(b)) dihedral
1
termined, neither from the 1D H NMR nor from the P.E.
COSY spectra.
The indication that 14 possesses an unsymmetrical shape
was not only assumed from the NH–CH(b) coupling con-
stants mentioned above, but also emerged when analyzing
the ROESY spectra. The observed NOE signals differ too
much from the ones expected for a symmetric molecule (see
above); this leads to the exclusion of conformer B as the
major conformer, even though this is lowest in energy ac-
cording to the theoretical calculations. The requirement to
have not only different but also large values of the NH–
CH(b) coupling constants make conformer C a non-likely
option as the major conformer, since this would require an
NH–CH(b) coupling constant close to 0 Hz.
That the major conformer would have all amide bonds
pointing at the same directions seems most likely when ana-
lyzing the ROESY spectrum of 14 (Figure 4b). In the spec-
trum, b3hGln in position 4 gives rise to weak NOE signals
Chem. Eur. J. 2005, 11, 6145 – 6158
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6151

P. I. Arvidsson et al.
angle restraints, for a restrained Monte Carlo conformation-
al search in Macromodel 7.0 (Figure 5).[35] The four H-N-
CH(b) torsional angles were constrained according to a
modified Karplus curve,[36] the b3hPhe in position 1 and the
b3hGln in position 4 were assigned the angle of 151 ꢁ 208,
while the b3hLeu in position 2 and the b3hLys in position 3
were assigned 144 ꢁ 208 and 162 ꢁ 208, respectively.
Figure 6. a) Expansion of the amide region of the 1H NMR spectrum of
19 recorded in methanol. b) Expansion of the amide region of the
ROESY spectrum of 19 recorded in methanol. Some of the NOE signals
discussed in the text are highlighted.
Figure 5. Solution structure of the cyclic b3-tetrapeptide 14 in MeOH rep-
resented as a bundle of six lowest-energy structures obtained from
Monte Carlo search, using NMR dihedral-angles and NOE distance re-
straints. View from the top.
From this conformational search it could be concluded
that even though 14 most probably fluctuate between oval
and very symmetrical conformations, it preferably reaches
the boat-shaped oval structure (Figure 5). We could also
conclude that even though the peptide possesses a relatively
high flexibility, the amide bonds do not have such degree of
free rotation to form the conformation where the amide
bonds are pointing in different directions. This finding is
also supported by a preliminary molecular dynamics simula-
tion, which shows that the molecule eventually adopts a par-
allel orientation of the amide groups independently of the
starting conformation.
CH(b) protons. The difference in size of the coupling con-
stants for the amide protons is similar to those of cyclopep-
tide 14. This indicates, as expected, that 19 possesses an un-
symmetrical shape as well.
The assignment of the protons of 19 was made using the
same experiments as for 14; a full assignment of the carbons
could not be done. By using the ROESY spectra obtained in
methanol (Figure 6b), the first Monte Carlo conformational
search of 19 was made. A total of 32 NOE signals were ex-
tracted and classified in the same way as the NOE signals of
14.
The amide region of 19 is shown in Figure 7 as an expan-
1
sion of the recorded H NMR spectrum in water. The chem-
Solution structure of cyclic b3-pentapeptide 19: An expan-
ical shifts have changed somewhat for all amino acids
except for b3hPhe and b3hLeu; the chemical shift of the
signal belonging to b3hVal slightly shifted to lower field,
while the signals for b3hLys4 and b3hLys5 shifted to higher
field. Not only chemical shifts but also the coupling con-
stants are different in water compared with methanol.
Except for the coupling constant of b3hLeu, which has de-
creased in size, all amino acids have an increased value. Sim-
ilar to the amide region of 19 recorded in methanol, the
1
sion of the amide region of the H NMR spectrum of 19 in
methanol is shown in Figure 6a. Broad unresolved signals
are seen in the amide region similar to the spectrum of 14.
The major signals are caused by one conformer or confor-
mational family, the minor ones either by disfavored con-
formers or by impurities (although the peptide was pure ac-
cording to LC-MS data). The size of the coupling constants
(8.3–9.9 Hz) indicates an anti arrangement of the NH–
6152
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6145 – 6158

Cyclic Peptides
FULL PAPER
Figure 7. Expansion of the amide region of the 1H NMR spectrum of 19
recorded in water.
broad baseline close to the signal of b3hLys5 indicates the
existence of more than one major conformer.
The assignment of the protons of 19 in water was also
made. The flexibility observed for the ROESY spectrum in
methanol could be seen in water as well. Again, a number
of NOE signals (47) were extracted from the ROESY spec-
trum, this time obtained in water. Running the Monte Carlo
conformational search according to the same principles used
earlier, a possible representative structure of the mean con-
formation of 19 in water was found (Figure 8c).
From the Monte Carlo search of the cyclic b3-pentapep-
tide in methanol, three different families of conformations
were found (Figure 8a and b). All these families possess
high flexibility in their backbone structure, and the amide
bonds do no longer have a unidirectional arrangement, as
was the case for the b3-tetrapeptide 14. Between the two
amino acids b3hVal in position 1 and b3hLys in position 5 a
turn similar to the b-turn arisen from a b2/b3-sequence[37] is
observed. Although it seems likely that such turns could
occur between any two consecutive amino acids in the ring,
the data only showed a turn between the b3hVal and the
b3hLys; this suggests that the a-branching at the side chain
b3hVal may contribute somehow.[38]
In water, the number of different conformational families
has been reduced by one. These structures also seem to be
more similar to each other in the backbone conformation.
The high flexibility observed in methanol remains, which is
clearly indicated by the completely opposite direction of the
amide bonds at one position (Figure 8c, position marked by
an arrow). Reflecting on the amide bonds in the two fami-
lies, one has them in rather defined down- and up-orienta-
tions, whereas the other has its bonds in more irregular ori-
entations.
Figure 8. Solution structure of the cyclic b3-pentapeptide 19 in MeOH (a
and b) and in H2O(c). a) The result of the conformational search of the
b-pentapeptide in methanol as a bundle of 12 low-energy structures.
Three different families were detected. b) Superposition of the backbones
of the three conformational families. c) The results of the investigation in
water as a bundle of six low-energy structures, where two different con-
formational families were found. The arrow indicates the position of a
completely reversed direction of one amide bond.
CD Spectroscopy: CD spectra (Figure 9) were measured for
all cyclo-b-peptides synthesized. Although circular dichro-
ism is a powerful tool for the determination of secondary
structures of b-peptides, it is still an empirical method and
assured data exist only for linear b-peptides capable to form
helices[1a] and turns.[39] The only reports on CD spectra of
Chem. Eur. J. 2005, 11, 6145 – 6158
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6153

P. I. Arvidsson et al.
supermolecular structures in the solid state, due to extensive
intermolecular hydrogen bonding. One molecule of cyclo(b-
Ala)4 adopts in the crystal state a conformation with two C=
Obonds pointing down and two up with respect to the aver-
age ring plane, similar to the structure of (R,R,S,S)-cyclo-
(b3hAla)4. We also demonstrate that the scope of the on-
resin cyclization protocol could be broadened to obtain
cyclic b-peptides by anchoring the side chains of Fmoc-
b3hGlu-OAll and Fmoc-b3hLys-OAll to the solid support via
Rink-amide linker and an active carbonate linker, respec-
tively. The target compounds cyclo(-b3Phe-b3hLeu-b3hLys-
b3hGln-) (14) and cyclo(-b3hVal-b3hPhe-b3hLeu-b3hLys-
b3hLys-) (19) could be obtained in overall yields of 10 and
9%, respectively. The conformationally averaged structures
of the synthesized cyclic b3-tetrapeptide 14 in aqueous solu-
tion were calculated based on NMR investigations. The con-
formational search resulted in a set of predominant similar
structures with predictable up-orientation of the amide
bonds and a pseudo-equatorial orientation of the side
chains. Although 14 shows flexibility, it seems limited
enough to render it potentially useful as a molecular scaf-
fold. In contrast to this, the cyclic b3-pentapeptide 19 dis-
plays a more flexible backbone ring structure including a b-
turn both in water and in methanol.
Figure 9. CD spectra of cyclic b-peptides (0.1 mm, 248C), c: cyclic b3-
pentapeptide 19 in MeOH, a: cyclic b3-pentapeptide 19 in water,
cyclic b3-tetrapeptide 14 in MeOH, g: cyclo(b-Ala)4 (4) in water.
:
~
some cyclic b-peptides are that of Seebach and co-work-
ers.[5b,40,41]
The cyclic b3-pentapeptide 19 displays distinct Cotton ef-
fects in both methanol and water. One intense maximum at
196 nm and an additional maximum at 211 nm are observed
in methanol. But a drastic change in the spectrum of 19 in
water is noted: the first maximum at 196 nm is sustained,
but after one zero crossing at 203 nm two minima occur:
one at 208 nm and the other at 222 nm. The differences in
the spectra clearly indicate a major change in the secondary
structure of 19 dependent on the solvent and may reflect the
reduction from three to two conformational families when
switching from methanol to water as found in the NMR
studies.
Experimental Section
Abbreviations: All: allyl, Boc: tert-butoxycarbonyl, Bzl: benzyl, DBU:
1,8-diazabicyclo[5.4.0]undec-7-ene, DIEA: N,N-diisopropylethylamine,
DMAP: 4-(dimethylamino)pyridine, EDC: 1-[3-(dimethylamino)propyl]-
3-ethylcabodiimide hydrochloride, Fmoc: (9H-fluoren-9-yl)methoxycar-
bonyl, HBTU: 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexa-
fluorophosphate, HOBt: 1-hydroxy-1H-benzotriazole, NMM: 4-methyl-
morpholine, PyAOP: 7-azabenzotriazol-1-yl-oxytris(pyrrolidino)phospho-
nium hexafluorophosphate, PyBOP: benzotriazole-1-yl-oxytris(pyrrolido-
no)phosphonium hexafluorophosphate, TFA: trifluoroacetic acid, TFE:
trifluoroethanol, TIPS: triisopropyl silane, TNBS: 2,4,6-trinitrobenzene-
sulfonic acid.
The cyclic b3-tetrapeptide 14 shows one maximum at
198 nm which is less pronounced than for 19, and one weak
minimum at 221 nm after a zero crossing at 213 nm. This
pattern is quite similar to that of the tetrapeptide cyclo-
(b3hAla)4 (20).[41] Both spectra resemble that obtained from
linear b3-peptides forming (M)-314 helices.[42] Contrarily, the
CD pattern of cyclo(-b3hPhe-b3hThr-b3hLys-b3hTrp-),[5b]
shows a CD spectrum similar to that observed for the 12/10-
helix,[1a,3d] or for the b-peptide hairpin,[39] because of an in-
tramolecular hydrogen bond which divides the ring[5b] (cf.
structure C’ in Figure 3). These observed differences suggest
that cyclo-b3-tetrapeptides can fold in different conforma-
tions, maybe depending if their side chains are branched or
not.
Solvents and reagents: All a-amino acid derivatives, HOBt and HBTU
were purchased from Senn Chemicals AG, Switzerland. PyAOP was from
Applied Biosystems (USA). DMF (peptide synthesis grade) was used
from Scharlau Chemie S.A. Dichloromethane was distilled under nitro-
gen from powdered CaH2. Methanol and acetonitrile were HPLC grade
and were purchased from VWR International Ltd., England. TentaGel
NH2 resin (0.24 mmolgꢀ1) and TentaGel S PHB resin (0.29 mmolgꢀ1
were purchased from Rapp Polymere GmbH, Germany.
)
Cyclo(b-Ala)4 (4) exhibits almost no Cotton effect in
agreement with the lack of both chiral groups and a prefer-
red secondary structure, which supports our findings by
NMR spectroscopy.
Instruments: Solid-phase peptide synthesis was performed on a Quest
210 synthesizer, Argonaut Technologies (USA). HPLC purifications of
the peptides were run on a Gilson system (Gilson 215 Liquid Handler,
Gilson UV/VIS-152, Gilson 322 Pump). Analytical runs were performed
on a Phenomenex Luna C8(2) column (100”4.60 mm, 5 mm), preparative
runs on a Phenomenex Luna C8(2) column (250”21.20 mm, 10 mm).
HPLC-MS runs were made on the same Gilson system coupled to a Fin-
nigan AQA Thermo Quest mass spectrometer with electrospray ionisa-
tion. Solvents for HPLC: A=0.1% TFA in water, B=0.1% TFA in ace-
tonitrile. NMR spectra were recorded on a Varian Unity 500 (1H at
500 MHz, 13C at 125.8 MHz) or a Varian Unity 400 (1H at 400 MHz, 13C
Conclusion
The synthesis of cyclo(b-Ala)4 (4) including conventional so-
lution-phase synthesis and cyclization in solution is reported.
Both the NMR and CD spectra of this simple cyclic b-pep-
tide points to a mixture of conformers in solution, whereas
the X-ray structure of 4 shows the formation of column-like
at 100.5 MHz) or
a
Varian Mercury plus (1H at 300 MHz, 13C at
75.4 MHz) spectrometer. Measurements were made at ambient tempera-
ture (unless otherwise stated) using the residual solvent signal as internal
reference. Peptides were analyzed by Matrix-assisted Laser Desorption
6154
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6145 – 6158

Cyclic Peptides
FULL PAPER
(MALDI) with an Ultraflex Tof/Tof instrument (Bruker, Germany) oper-
ated in positive and reflectron mode. a-Cyano-4-hydroxycinnamic acid
was used as matrix and the instrument was calibrated using matrix ions.
All electrospray mass spectra were acquired by using a Bruker Daltonics
BioAPEX-94e superconducting 9.4 T FTICR mass spectrometer (Bruker
Daltonics, Billerica, MA) in broadband mode. A home-built apparatus
controlled the direct infusion of sample. The sample was delivered using
a helium gas container at a pressure of 1.3 bar, pushing the sample
through a 30 cm fused silica capillary of inner diameter 20 mm. The
sample end of the capillary was lowered into the sample tube inside the
pressurized container and the electrospray end was coated by a conduct-
ing graphite/polymer layer and connected to ground.[43,44] No sheath flow
or nebulizing gas was used and the flow rate was approximately
100 nLminꢀ1. The ion source was coupled to an Analytica atmosphere/
vacuum interface (Analytica of Branford, CT) and a potential difference
of 2–4 kV was applied across a distance of approximately 5 mm between
the spraying needle and the inlet capillary. Infrared spectra were ob-
tained from a Perkin–Elmer 1760 IR FT spectrometer. CD spectra were
measured with a Jasco J-810 spectropolarimeter (c=0.1 mm, 248C). UV
spectra were recorded on a Varian Cary 3 Bio spectrophotometer. X-ray
mate (1.3 mL, 10 mmol) were added and stirring was continued for
15 min. All of the H-b-Ala-b-Ala-OBzl·TFA obtained before was added
and the reaction mixture was stirred for 3 d at room temperature. The
solvent was evaporated and the residue was purified by flash chromatog-
raphy (CH2Cl2/CH3OH/AcOH 90:8:2). The linear tetrapeptide 3 was ob-
tained as colorless solid (3.12 g, 6.3 mmol, 63%). Rf =0.36 (CHCl3/
CH3OH/AcOH 90:8:2); 1H NMR (500 MHz, CDCl3): d = 1.42 (s, 9H,
Boc), 1.86 (brs, 4H, CH2), 2.34 (brs, 6H, CH2), 3.51 (m, 6H, CH2), 5.13
(s, 2H, CH2-Bzl), 5.33 (brs, 1H, NH), 6.52 (brs, 1H, NH), 6.68 (m, 2H,
NH), 7.32–7.39 (m, 5H, Bzl).
Cyclo(b-Ala)4 (4): Boc-(b-Ala)4-OBzl (3) (1.23 g, 2.50 mmol) and 10%
Pd/C (120 mg) were added to a mixture of THF and H2O(2:1, 150 mL)
and hydrogenated for 12 h at atmospheric pressure. Control of the reac-
tion by TLC (Boc-(b-Ala)4-OBzl, Rf =0.6, Boc-(b-Ala)4-OH, Rf =0.3,
CHCl3/CH3OH/AcOH 20:4:1) revealed the completeness of the depro-
tection. Pd/C was removed by filtration and the solvent was evaporated.
Boc-(b-Ala)4-OH was obtained as a colorless solid and further used with-
out purification (991 mg, 2.46 mmol, 98%).
Boc-(b-Ala)4-OH (402 mg, 1.0 mmol) was dissolved in DMF (100 mL)
and the solution was cooled under an atmosphere of argon to 08C. Penta-
fluorophenol (193 mg, 1.05 mmol) and DIEA (18 mL, 1.1 mmol) were
added. The reaction mixture was allowed to warm to room temperature
and stirred for further 16 h. The DMF was evaporated and the residue
was taken up in dichloromethane and washed with 1m HCl and brine.
The solvent was evaporated to give the activated ester as a colorless
solid.
Crystallographic Analysis was carried out on
a Bruker-Nonius kap-
paCCD diffractometer. For TLC, Alugram SIL G/UV254 silica sheets
from Macherey–Nagel were used. Flash Chromatography was carried out
on silica gel (35–70 mm) from Millipore Corp. Melting points were deter-
mined with a melting point apparatus SMP 10 from Stuart Scientific/
Bibby Sterilin Ltd. and are uncorrected.
The Fmoc-b3-amino acid derivatives Fmoc-b3hGlu(OtBu)-OAll, Fmoc-
b3hLeu-OH, Fmoc-b3hLys(Boc)-OAll, Fmoc-b3hPhe-OH, Fmoc-b3hVal-
OH were prepared via Arndt–Eistert homologation as described in refer-
ences.[18,19] The Wolff rearrangement to the carboxylic acids was facilitat-
ed with a High Intensity Ultrasonic Processor VCX 500 from Sonics Ma-
terials Inc.
This intermediate was deprotected by stirring for 1 h at 08C and subse-
quently for 1 h at room temperature in a mixture of CH2Cl2 and TFA
(1:1, 2 mL). After evaporation the residue was dissolved in dry THF
(20 mL) and was slowly added via syringe pump over 8 h to a mixture of
DIEA (260 mL, 1.6 mmol) in dry THF (400 mL) at 608C. The stirring was
continued for further 24 h at room temperature. The resulting precipitate
H-b-Ala-OBzl·HCl: H-b-Ala-OH (8.90 g, 100 mmol), benzyl alcohol
(50 mL) and chlorotrimethylsilane (20 mL) were heated to 1008C for 4 h.
The reaction mixture was cooled to room temperature and was poured
into diethyl ether (2 L) and subsequently cooled in an ice bath for 24 h.
The resulting precipitate was isolated by filtration. The benzyl protected
b-alanine was obtained as hydrochloric salt (20.5 g, 95 mmol, 95%) and
used without further purification.
was filtered and washed with THF to give cyclic b-tetrapeptide 4
1
(164 mg, 0.58 mmol, 58%). H NMR (500 MHz, D2O): d = 2.28 (dd, 8H,
3J=5.5 Hz; CH2), 3.32 (brq, 8H, 3J=5.5 Hz; CH2), 7.92 (brs, 4H; NH);
13C NMR (125 MHz, D2O): d = 174.5 (C=O), 35.8 (CH2), 35.6 (CH2).
Colorless crystals of 4 suitable for X-ray crystallography were grown
from a mixture of MeOH/H2Oat room temperature (Table 1).
CCDC-264893 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Boc-b-Ala-b-Ala-OBzl (2): Boc-b-Ala-OH (1) (7.57 g, 40.0 mmol) was
dissolved in dry THF (400 mL) under an argon atmosphere. The solution
was cooled to ꢀ158C and TEA (11.7 mL, 83 mmol) and isobutyl chloro-
formate (5.2 mL, 40.0 mmol) were added. The mixture was stirred for
15 min and H-b-Ala-OBzl (8.62 g, 40.0 mmol) was added. The reaction
mixture was allowed to warm to room temperature and was stirred for
3 d. The solvent was evaporated and the residue was dissolved in ethyl
acetate (800 mL). The organic phase was washed successively with H2O,
1n citric acid, H2O, sat. NaHCO3 and brine, subsequently dried with an-
hydrous MgSO4 and the solvent was evaporated. The residue was recrys-
tallized from ethyl acetate and heptane to give dipeptide 2 as colorless
crystals (11.4 g, 3.24 mmol, 81%). 1H NMR (500 MHz, CDCl3): d = 1.45
(s, 9H; Boc), 2.35 (t, 2H, J=7.5 Hz; CH2), 2.61 (t, 2H, J=7.5 Hz; CH2),
3.40 (q, 2H, J=7.5 Hz; CH2), 3.56 (q, 2H, J=7.5 Hz; CH2), 5.16 (s, 3H;
CH2-Bzl, NH), 6.08 (s, 1H; NH), 7.2–7.5 (m, 5H; Bzl).
(S)-3-(9H-Fluoren-9-ylmethoxycarbonylamino)-hexanedioic acid 1-allyl
ester 6-tert-butyl ester (Fmoc-b3hGlu(OtBu)-OAll) (7): Fmoc-Glu-
(OtBu)-CHN2 (5)[18] (660 mg, 1.47 mmol) was dissolved in dry THF
(10 mL) and allyl alcohol (0.15 mL, 2.25 mmol) was added. The mixture
was cooled to ꢀ158C and
a solution of silver benzoate (34 mg,
0.15 mmol) in NMM (0.33 mL, 3 mmol) was added. The reaction mixture
was stirred and warmed to room temperature. After 5 h the mixture was
filtered through a pad of Celite and the THF was removed under re-
duced pressure. The residue was dissolved in ethyl acetate and succes-
sively washed with saturated NaHCO3, water and 5% HCl. After drying
with MgSO4 the solvent was evaporated under reduced pressure. The res-
idue was purified by flash chromatography (pentane/ethyl acetate 4:1)
and yielded allyl ester 7 (456 mg, 0.98 mmol, 67%). Rf =0.68 (pentane/
ethyl acetate 3:1); 1H NMR (500 MHz, CDCl3): d = 1.44 (s, 9H; tBu),
1.86 (m, 2H; CH2), 2.31 (m, 2H; CH2COOAll), 2.61 (d, 2H, 2J=5 Hz;
Boc-(b-Ala)4-OBzl (3): This compound was prepared previously by a
similar method.[45] Boc-b-Ala-b-Ala-OBzl (2) (3.52 g, 10.0 mmol) was
added to a mixture of TFA and CH2Cl2 (1:1, 10 mL) at 08C and the reac-
tion mixture was stirred for 1 h at 08C and additionally for 1 h at room
temperature. The solvent was evaporated and b-Ala-b-Ala-OBzl·TFA
was obtained and used without further purification. Boc-b-Ala-b-Ala-
OBzl (2) (3.52 g, 10.0 mmol) and 10% Pd/C (0.35 g) were added to a
mixture of THF and CH3OH (1:1, 200 mL) and hydrogenated for 12 h at
atmospheric pressure. Control of the reaction by TLC (Boc-b-Ala-b-Ala-
OBzl, Rf =0.73, Boc-b-Ala-b-Ala-OH, Rf =0.38, CHCl3/CH3OH/AcOH
90:8:2) revealed the completeness of the deprotection. Pd/C was re-
moved by filtration and the solvent was evaporated. The residue was dis-
solved in dry THF (100 mL) under an argon atmosphere and the solution
was cooled to ꢀ158C. TEA (4.68 mL, 33 mmol) and isobutyl chlorofor-
3
CH2COOtBu), 4.00 (m, 1H; CHNH), 4.20 (t, 1H, J=7.5 Hz; CHCH2O),
2
4.37 (m, 2H; CHCH2O), 4.60 (d, 2H, J=6 Hz; CHOCH2), 5.23 (dd, 1H,
J=10.5, 1 Hz; CH=CH2), 5.30–5.35 (m, 3H; CH=CH2, NH), 5.90 (m,
1H; CH=CH2), 7.31 (dt, 2H, 3J=7.5, 4J=1 Hz; Fmoc), 7.40 (t, 2H, 3J=
7.5 Hz; Fmoc), 7.59 (d, 2H, 3J=8 Hz; Fmoc), 7.76 (d, 2H, 3J=7.5 Hz;
Fmoc); 13C NMR (100.5 MHz, CDCl3): d = 28.1 (tBu), 29.2, 32.3, 39.1,
47.3 (CH), 47.9 (CH), 65.4 (CH2), 66.7 (CH2), 80.6 (tBu), 118.6 (CH2
allyl), 119.9, 125.1, 127.0, 127.7 (4”CH Fmoc), 131.8 (CH allyl), 141.3,
143.9 (2”C Fmoc), 155.8 (OCONH), 171.0 (CO), 172.5 (CO); IR (KBr):
n˜ = 3363 (m), 2975 (m), 1734 (s), 1689 (s), 1521 (s), 1452 (m), 1368 (m),
1246 (m), 1155 (s), 761 (m), 739 cmꢀ1 (m); HR ESI-MS: m/z: calcd for
C28H33NO6: 479.231; found: 480.238 [M+H]+, 502.220 [M+Na]+.
Chem. Eur. J. 2005, 11, 6145 – 6158
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6155

P. I. Arvidsson et al.
Table 1. Crystallographic data for 4.
3067 (m), 2939 (m), 1732 (s), 1699 (s), 1540 (s), 1451 (m), 1418 (m), 1278
(s), 1141 (m), 1087 (m), 1051 (m), 985 (m), 931 (m), 759 (m), 738 cmꢀ1
(s); HR ESI-MS: m/z: calcd for C24H25NO6: 423.168; found: 424.175
[M+H]+, 446.157 [M+Na]+.
empirical formula
Mr
radiation
crystal system
l [Š]
space group
a [Š]
b [Š]
c [Š]
a [8]
b [8]
g [8]
C12H20N4O4
284.316
MoKa
monoclinic, colorless prism
0.71073
P21/c
5.291(2)
10.4820(7)
12.6910(8)
90.00
106.391(4)
90.00
0.998–29.57
675.24(7)
2
298
0.106
(S)-7-tert-Butoxycarbonylamino-3-(9H-fluoren-9-ylmethoxycarbonyl-
amino)-heptanoic acid allyl ester (Fmoc-b3hLys(Boc)-OAll) (8): Fmoc-
Lys(Boc)-CHN2 (6)[18] (980 mg, 2.0 mmol) was dissolved in dry THF
(10 mL) and allyl alcohol (0.2 mL, 3.0 mmol) was added. The mixture
was cooled to ꢀ158C and a solution of silver benzoate (48 mg, 0.2 mmol)
in NMM (0.44 mL, 4.0 mmol) was added. The reaction mixture was stir-
red and warmed to room temperature. After 5 h the mixture was filtered
through a pad of Celite and the THF was removed under reduced pres-
sure. The residue was dissolved in ethyl acetate and successively washed
with saturated NaHCO3, water and 5% HCl. After drying with MgSO4
the solvent was evaporated under reduced pressure. The residue was pu-
rified by flash chromatography (pentane/ethyl acetate 3:1 ! 2:1) and
yielded allyl ester 8 (805 mg, 1.54 mmol, 77%), which was recrystallized
from CH2Cl2/pentane to obtain colorless crystals. M.p. 948C; Rf =0.18
(pentane/ethyl acetate 3:1); 1H NMR (300 MHz, CDCl3): d = 1.21–1.66
(m, 15H; 3”CH2, Boc), 2.58 (d, 2H, 2J=7 Hz; CH2CO), 3.10 (d, 2H,
2J=10 Hz; CH2NHBoc), 3.97 (m, 1H; CHNHFmoc), 4.21 (t, 1H, 3J=
11.5 Hz; CHCH2O), 4.37 (d, 2H, 2J=11.5 Hz; CH2O), 4.58 (d, 2H, 2J=
9.5 Hz; CHOCH2), 5.22–5.34 (m, 3H; CH=CH2, NH), 5.90 (m, 1H; CH=
CH2), 7.31 (dt, 2H, 3J=12.5, 4J=2 Hz; Fmoc), 7.40 (t, 2H, 3J=12 Hz;
Fmoc), 7.59 (d, 2H, 3J=12.5 Hz; Fmoc), 7.76 (d, 2H, 3J=12.5 Hz;
Fmoc); 13C NMR (75.4 MHz, CDCl3): d = 23.2 (CH2), 28.4 (Boc), 29.6
(CH2), 33.9 (CH2), 39.0 (CH2), 40.1 (CH2), 47.2 (CH), 48.0 (CH), 65.3
(CH2), 66.6 (CH2), 79.1 (Boc), 118.6 (CH2 allyl), 119.9, 125.0, 127.0, 127.6
(4”CH Fmoc), 131.8 (CH allyl), 141.3, 143.9 (2”C Fmoc), 155.9
(OCONH), 156.0 (OCONH), 171.2 (COOAll); IR (KBr): n = 3362 (s),
3338 (s), 2939 (m), 1735 (s), 1684 (s), 1534 (s), 1451 (m), 1369 (m), 1281
(s), 1253 (s), 1173 (m), 1090 (m), 987 (m), 736 cmꢀ1 (m); HR ESI-MS:
m/z: calcd for C30H38N2O6: 522.273; found: 523.280 [M+H]+, 545.262
[M+Na]+.
q [8]
V [Š3]
Z
T [K]
m [mmꢀ1
]
1calcd [gcmꢀ3
qmax [8]
]
1.3982
29.57
1843
cell parameters
(1485 with I > 4s(I))
parameters refined
final R
Rw
GOF
data collection:
absorption correction
qmax
92
0.0455
0.1325
1.045
KappaCCD
none
29.57
ꢀ7ꢂhꢂ7
ꢀ14ꢂkꢂ14
ꢀ17ꢂlꢂ16
7397
measured reflections
independent reflections
observed reflections
refinement on F2
calcd weights D/smax
D1max [eŠ3]
1869
1485
(S)-7-Amino-3-(9H-fluoren-9-ylmethoxycarbonylamino)-heptanoic acid
allyl ester trifluoroacetic acid salt (Fmoc-b3hLys-OAll·TFA) (10): Fmoc-
b3hLys(Boc)-OAll (8) (164 mg, 0.31 mmol) was dissolved in dry CH2Cl2
(10 mL). After addition of TFA (3 mL, 67.8 mmol) and TIPS (0.2 mL,
1 mmol) the reaction mixture was stirred at room temperature. After 2 h
stirring at room temperature (no more starting material detectable by
TLC) the solvent was removed under reduced pressure. The residue was
purified by flash chromatography (CH2Cl2/methanol 9:1) if necessary.
The primary amine was obtained as the trifluoroacetic acid salt 10 and
appeared as yellowish oil, which turned solid after several weeks. Rf =
0.56 (CH2Cl2/MeOH 9:1); 1H NMR (500 MHz, CDCl3): d = 1.35–1.49
(m, 4H; 2CH2), 1.64 (m, 2H; CH2), 2.45–2.54 (m, CH2CO), 2.90 (m, 2H;
full-matrix least squares
0.000
0.120
D1min [eŠ3]
R(all)
R(gt)
wR(ref)
wR(gt)
S(ref)
extinction correction
ꢀ0.138
0.0580
0.0455
0.1325
0.1201
1.0045
none
3
CH2), 3.89 (m, 1H, CHN), 4.15 (t, 1H, J=7 Hz; CHCH2O), 4.32 (d, 2H,
2
(S)-3-(9H-Fluoren-9-ylmethoxycarbonylamino)-hexanedioic acid 1-allyl
ester (Fmoc-b3hGlu-OAll) (9): Fmoc-b3hGlu(OtBu)-OAll (7) (2.50 g,
5.2 mmol) was dissolved in dry CH2Cl2 (30 mL). After addition of TFA
(5.1 mL, 67.8 mmol) and TIPS (2.7 mL, 13 mmol) the reaction mixture
was stirred at room temperature. After 2 h much starting material re-
mained and additional TFA (5.1 mL) was added. After stirring over night
the solvent was removed under reduced pressure and the residue was pu-
rified by flash chromatography (ethyl acetate/pentane 1:2 ! 1:0). The
product obtained (2.0 g, 4.72 mmol, 91%) was re-crystallized from ethyl
acetate/pentane. Rf =0.64 (ethyl acetate); m.p. 1058C; 1H NMR
J=6.5 Hz; CH2), 4.54 (d, 2H, J=5.5 Hz; CH2), 5.20 (d, 1H, J=10.5 Hz;
3
CH=CH2), 5.27 (m, 1H; CH=CH2), 5.37 (d, 1H, J=9 Hz; NH), 5.87 (m,
1H; CH=CH2), 7.27 (t, 2H, 3J=7.5 Hz; Fmoc), 7.36 (t, 2H, 3J=7.5 Hz;
Fmoc), 7.54 (d, 2H, 3J=7.5 Hz; Fmoc), 7.72 (d, 2H, 3J=7.5 Hz; Fmoc),
7.92 (brs, 3H, NH3); 13C NMR (100.5 MHz, CDCl3): d = 22.6 (CH2),
26.7 (CH2), 33.5 (CH2), 39.0 (CH2), 39.6 (CH2), 47.1 (CH), 47.7 (CH),
65.4 (CH2), 66.7 (CH2), 118.6 (CH2 allyl), 119.9, 125.0, 127.0, 127.7 (4”
CH Fmoc), 131.7 (CH allyl), 141.2, 143.7 (2”C Fmoc), 156.2 (OCONH),
161.9 (q, 2J(C,F)=36 Hz, CF3COOH), 171.3 (COOAll); IR (KBr): n =
3070 (s), 2946 (s), 1685 (s), 1540 (m), 1451 (m), 1203 (s), 1133 (s), 987
(m), 836 (m), 799 (m), 761 (m), 739 (m), 721 cmꢀ1 (m); HR ESI-MS: m/
z: calcd for C25H30N2O4: 422.221; found: 423.228 [M+H]+, 445.2100
[M+Na]+.
(500 MHz, CDCl3): d
CH2COOAll), 2.60 (m, 2H; CH2COOtBu), 4.03 (m, 1H, CHNH), 4.20 (t,
=
1.88 (m, 2H; CH2), 2.41 (t, 2H, 2J=5 Hz;
3
2
1H, J=7 Hz; CHCH2O), 4.35–4.44 (m, 2H, CHCH2O), 4.59 (d, 2H, J=
5.5 Hz, CHOCH2), 5.23 (dd, 1H, J=10.5, 1 Hz, CH=CH2), 5.31 (d, 1H,
J=17.5 Hz; CH=CH2), 5.36 (d, 1H, 3J=9.5 H; NH), 5.89 (m, 1H, CH=
CH2), 7.31 (dt, 2H, 3J=7.5, 4J=1.5 Hz; Fmoc), 7.39 (t, 2H, 3J=7.5 Hz;
Fmoc), 7.58 (d, 2H, 3J=7.5 Hz; Fmoc), 7.75 (d, 2H, 3J=8 Hz; Fmoc);
13C NMR (75.4 MHz, CDCl3): d = 29.1, 30.7, 39.0 (3”CH2), 47.2 (CH),
47.6 (CH), 65.4 (CH2), 66.7(CH2), 118.7 (CH2 allyl), 120.0, 125.0, 127.0,
127.7 (4”CH Fmoc), 131.7 (CH allyl), 141.3, 143.8 (2”C Fmoc), 156.0
(CONH), 171.0 (COOAll), 177.8 (COOH); IR (KBr): n = 3321 (m),
Cyclo(b3hGln-b3hPhe-b3hLeu-b3hLys) (14)
Attachment of the Rink linker: TentaGel NH2 resin 11 (0.24 mmolgꢀ1
,
720 mg, 0.17 mmol) was swollen in dry CH2Cl2 for 30 min. A solution of
Fmoc-Rink linker (466 mg, 0.86 mmol), HBTU (328 mg, 0.86 mmol),
HOBt (117 mg, 0.86 mmol) and DIEA (0.15 mL, 0.86 mmol) in dry
CH2Cl2 (6 mL) were added and the suspension was stirred at room tem-
perature for 3 h (TNBS test negative). The resin was washed with CH2Cl2
(5”6 mL), DMF (5”6 mL) and was finally shrunk down with diethyl
6156
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6145 – 6158

Cyclic Peptides
FULL PAPER
ether (5”6 mL). Fmoc/UV spectrophotometry with a small sample of the
resin resulted in a calculated substitution level of 82%. Unreacted sites
of the resin were capped with a mixture of acetic anhydride (3 mL) and
CH2Cl2 (4 mL) and stirring for 3 h at room temperature. The resin was
subsequently washed with CH2Cl2 (6”6 mL).
(Boc)-OH. After the last coupling the resin was additionally washed with
CH2Cl2 (5”6 mL) and shrunk down with diethyl ether (5”6 mL).
Allyl deprotection: The resin with the linear, fully protected peptide 18
was dried under high vacuum at 408C for 1 h and swollen in CH2Cl2. To
the damp resin were added CH2Cl2 (1 mL), phenylsilane (0.3 mL,
2.4 mmol) and a solution of [Pd(PPh3)4] (12 mg, 0.01 mmol) in CH2Cl2
(4 mL). Argon was bubbled through the suspension for 10 min and the
resin was filtered and washed with CH2Cl2 (5”6 mL). The deprotection
procedure was repeated once and the resin was washed with CH2Cl2 (5”
6 mL) and shrunk down with diethyl ether (5”6 mL).
Synthesis of the linear peptide: The Fmoc-Rink amide TentaGel resin 12
was swollen in DMF and Fmoc-deprotected with 2% DBU/2% piperi-
dine in DMF (5”5 min). The resin was washed with DMF (5”6 mL). A
solution of Fmoc-b3hGlu-OAll (9) (148 mg, 0.35 mmol), HBTU (128 mg,
0.34 mmol), HOBt (48 mg, 0.35 mmol) and DIEA (0.12 mL, 0.7 mmol) in
DMF (6 mL) was added and the suspension was stirred for 3 h at room
temperature (TNBS test negative). The resin was subsequently washed
with DMF (5”6 mL). The Fmoc-deprotection/coupling procedure was re-
peated three times with 2.5 equiv Fmoc-b3hPhe-OH, Fmoc-b3hLeu-OH
and Fmoc-b3hLys(Boc)-OH. After the last coupling the resin was addi-
tionally washed with CH2Cl2 (5”6 mL) and shrunk down with diethyl
ether (5”6 mL).
Cyclization and cleavage: After Fmoc deprotection (2% DBU/2% piper-
idine in DMF, 5”5 min) and washing (5”6 mL of DMF) a solution of
PyAOP (209 mg, 0.4 mmol) and DIEA (0.14 mL, 0.8 mmol) in DMF
(6 mL) was added to the resin. The suspension was stirred for 12 h at
room temperature (TNBS test negative) and finally washed with DMF
(5”6 mL), CH2Cl2 (5”6 mL) and shrunk down with diethyl ether (5”
6 mL). To the resin were added a mixture of TFA/H2O/TIPS (95:2.5:2.5,
6 mL). After stirring for 3 h at room temperature the resin was filtered,
washed two times with the cleavage mixture (6 mL, 2 min) and the col-
lected filtrates were combined. This mixture was concentrated under re-
duced pressure and the residue was treated with cold ether. The crude
peptide precipitated and the solvent was decanted. This step was repeat-
ed two times and yielded the crude peptide as a white solid. The mixture
was purified by RP-HPLC (10–50% B over 29 min, tr =19.4 min) to yield
cyclic b3-pentapeptide 19 after freeze-drying (6 mg, 9%). Analytical
HPLC (10–70% B over 14 min): tr =9.15 min. MALDI-TOF MS: m/z:
calcd for C37H63N7O5: 685.49; found 686.145 [M+H]+, 708.089 [M+Na]+,
724.042 [M+K]+.
Allyl deprotection: The resin with the linear, fully protected peptide 13
was dried under high vacuum at 408C for 1 h and a solution of [Pd-
(PPh3)4] (81 mg, 0.07 mmol) in a mixture of DMSO/THF/0.5m HCl/
NMM (4:4:2:1, 38.5 mL) was added. The reaction mixture was incubated
at room temperature with occasional shaking for 4.5 h. The resin was
washed consecutively with THF (5”6 mL), 0.5% DIEA in DMF (5”
6 mL), 0.5% sodium diethyl dithiocarbamate in DMF (5”6 mL), CH2Cl2
(5”6 mL) and shrunk down with diethyl ether (5”6 mL).
Cyclization and cleavage: After Fmoc deprotection (2% DBU/2% piper-
idine in DMF, 5”5 min) and washing (5”6 mL of DMF) a solution of
HBTU (127 mg, 0.34 mmol), HOBt (47 mg, 0.35 mmol) and DIEA
(0.12 mL, 0.35 mmol) in DMF (6 mL) was added to the resin. The sus-
pension was stirred for 17 h at room temperature (TNBS test negative)
and finally washed with DMF (5”6 mL), CH2Cl2 (5”6 mL) and shrunk
down with diethyl ether (5”6 mL). To the resin were added a mixture of
TFA/H2O/TIPS (95:2.5:2.5, 6 mL). After stirring for 2.5 h at room tem-
perature the resin was filtered, washed two times with the cleavage mix-
ture (6 mL, 2 min) and the collected filtrates were combined. This mix-
ture was concentrated under reduced pressure and the residue was treat-
ed with cold ether. The crude peptide precipitated and the solvent was
decanted. This step was repeated two times and yielded the crude pep-
tide as a white solid. The mixture was purified by RP-HPLC (15–45% of
B over 44 min, tr =23.1 min) to give cyclic b3-tetrapeptide 14 in pure form
Acknowledgements
This work was supported by Vetenskapsrådet (The Swedish Research
Council), the Carl Trygger Foundation, and Magnus Bergvallꢁs Founda-
tion. F.B. acknowledges the Carl Trygger Foundation for a postdoctoral
fellowship during 2003, and S.Z. is grateful to the Wenner-Gren founda-
tion for a fellowship 2003–2004. P.K. acknowledges the Thailand Re-
search Fund (TRF4780020) for a financial support and S.P. thanks the
Postgraduate Education and Research Program in Chemistry for a schol-
arship. MALDI-Tof MS analysis was done at the Expression Proteomics
Laboratory at Uppsala University supported by the Wallenberg founda-
tion. We are grateful to Jonas Bergquist (Uppsala University, Depart-
ment of Analytical Chemistry) for performing the ESI-MS analyses. The
CD instrument was purchased with funding from the Wallenberg founda-
tion.
as white powder (8 mg, 10%). Analytical HPLC (10–70%
B over
14 min): tr =8.68 min. MALDI-TOF MS: m/z: calcd for C30H48N6O5:
572.369; found 573.306 [M+H]+, 595.249 [M+Na]+, 611.231 [M+K]+.
Cyclo(b3hLys-b3hVal-b3hPhe-b3hLeu-b3hLys) (19)
Attachment of the first amino acid: TentaGel
S PHB resin 15
(0.29 mmolgꢀ1
,
600 mg, 0.17 mmol) was suspended in dry CH2Cl2
(10 mL) and cooled to 08C. 4-Nitrophenyl chloroformate (117 mg,
0.58 mmol) and NMM (0.06 mL, 0.58 mmol) were added. The suspension
was warmed to room temperature and stirred over night. After washing
the resin with CH2Cl2 (5”6 mL) and DMF (2”6 mL) a solution of
Fmoc-b3hLys-OAll (10) (613 mg, 1.45 mmol) and DIEA (0.76 mL,
4.35 mmol) in DMF (6 mL) was added. The suspension was stirred at
room temperature for 14 h. The resin was washed with DMF (5”6 mL)
and the coupling step was repeated once. The resin was finally washed
with DMF (5”6 mL), CH2Cl2 (5”6 mL) and shrunk down with diethyl
ether (5”6 mL).
[1] For reviews see: a) R. P. Cheng, S. H. Gellman, W. F. DeGrado,
Chem. Rev. 2001, 101, 3219–3232; b) D. J. Hill, M. J. Mio, R. B.
Prince, T. S. Hughes, J. S. Moore, Chem. Rev. 2001, 101, 3893–4011.
[2] D. Seebach, J. L. Matthews, Chem. Commun. 1997, 2015–2022.
[3] See for example, a) Y. Hamuro, J. P. Schneider, W. F. DeGrado, J.
Am. Chem. Soc. 1999, 121, 12200–12201; b) P. I. Arvidsson, J.
Frackenpohl, N. S. Ryder, B. Liechty, F. Petersen, H. Zimmermann,
G. P. Camenisch, R. Woessner, D. Seebach, ChemBioChem 2001, 2,
771–773; c) E. A. Porter, B. Weisblum, S. H. Gellman, J. Am. Chem.
Soc. 2002, 124, 7324–7330; d) P. I. Arvidsson, N. S. Ryder, H. M.
Weiss, G. Gross, O. Kretz, R. Woessner, D. Seebach, ChemBioChem
2003, 4, 1345–1347.
[4] M. Werder, H. Hauser, S. Abele, D. Seebach, Helv. Chim. Acta
1999, 82, 1774–1783.
[5] a) T. D. Clark, L. K. Buehler, M. R. Ghadiri, J. Am. Chem. Soc.
1998, 120, 651–656; b) K. Gademann, M. Ernst, D. Seebach, D.
Hoyer, Helv. Chim. Acta 2000, 83, 16–33; c) K. Gademann, D. See-
bach, Helv. Chim. Acta 1999, 82, 957–962.
Assembly of the linear peptide: The Fmoc-b3hLys-OAll TentaGel resin
17 was swollen in DMF and Fmoc-deprotected with 2% DBU/2% piperi-
dine in DMF (5”5 min). The resin was washed with DMF (5”6 mL). A
solution of Fmoc-b3hVal-OH (150 mg, 0.42 mmol), HBTU (155 mg,
0.41 mmol), HOBt (57 mg, 0.42 mmol) and DIEA (0.15 mL, 0.85 mmol)
in DMF (6 mL) was added and the suspension was stirred for 3 h at
room temperature (TNBS test negative). The resin was subsequently
washed with DMF (5”6 mL), CH2Cl2 (5”6 mL) and shrunk down with
diethyl ether (5”6 mL). Fmoc/UV spectrophotometry with
a small
sample of the resin resulted in a calculated substitution level of 60%.
The Fmoc-deprotection/coupling procedure was repeated three times
with 2.5 equiv Fmoc-b3hPhe-OH, Fmoc-b3hLeu-OH and Fmoc-b3hLys-
[6] M. Royo, J. Farrera-Sinfreu, L. SolØ, F. Albericio, Tetrahedron Lett.
2002, 43, 2029–2032.
Chem. Eur. J. 2005, 11, 6145 – 6158
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6157

P. I. Arvidsson et al.
[7] F. Büttner, M. ErdØly, P. I. Arvidsson, Helv. Chim. Acta 2004, 87,
2735–2741.
[26] B. A. Dressmann, L. A. Spangle, S. W. Kaldor, Tetrahedron Lett.
1996, 37, 937–940.
[27] J. Alsina, F. Rabanal, C. Chiva, E. Giralt, F. Albericio, Tetrahedron
1998, 54, 10125–10152.
[28] N. Thieriet, J. Alsina, E. Giralt, F. GuibØ, F. Albericio, Tetrahedron
Lett. 1997, 38, 7275–7278.
[29] T. A. Martinek, F. Fülçp, Eur. J. Biochem. 2003, 270, 3657–3666.
[30] F. Johnson, Chem. Rev. 1968, 68, 375–413.
[31] S. H. Smallcombe, S. L. Patt, P. A. Keifer, J. Magn. Reson. Ser. A
1995, 117, 295–303.
[32] L. Mueller, J. Magn. Reson. (1969) 1987, 72, 191–196.
[33] A. Bax, D. G. Davis, J. Magn. Reson. (1969) 1985, 63, 207–213.
[34] T. Parella, Magn. Reson. Chem. 1998, 36, 467–495.
[35] a) MacroModel 7.0, Schrçdinger, Inc., Portland OR, 1999 b) F. Mo-
hamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M. Lipton, C.
Caufield, G. Chang, T. Hendrickson, W. C. Still, J. Comput. Chem.
1990, 11, 440–467.
[36] S. A. Perera, R. J. Bartlett, Magn. Reson. Chem. 2001, 39, S183–
S189.
[37] D. Seebach, S. Abele, K. Gademann, B. Jaun, Angew. Chem. 1999,
111, 1700–1703; Angew. Chem. Int. Ed. 1999, 38, 1595–1597.
[38] T. L. Raguse, J. R. Lai, S. H. Gellman, J. Am. Chem. Soc. 2003, 125,
5592–5593, and references therein.
[8] K. Gademann, M. Ernst, D. Hoyer, D. Seebach, Angew. Chem.
1999, 111, 1302–1304; Angew. Chem. Int. Ed. 1999, 38, 1223–1226.
[9] K. Gademann, D. Seebach, Helv. Chim. Acta 2001, 84, 2924–2937.
[10] a) D. Seebach, J. L. Matthews, A. Meden, T. Wessels, C. Baerlocher,
L. B. McCusker, Helv. Chim. Acta 1997, 80, 173–182; b) J. L. Mat-
thews, K. Gademann, B. Jaun, D. Seebach, J. Chem. Soc. Perkin
Trans. 1 1998, 3331–3340.
[11] D. A. Horton, G. T. Bourne, M. L. Smythe, Mol. Diversity 2000, 5,
289–304.
[12] N. Sewald, H.-D. Jakubke, Peptides: Chemistry and Biology, Wiley-
VCH, Weinheim, 2002, pp. 311–312, p. 344.
[13] a) J. Frackenpohl, P. I. Arvidsson, J. V. Schreiber, D. Seebach, Chem-
BioChem 2001, 2, 445–455; b) H. Wiegand, B. Wirz, A. Schweitzer,
G. P. Camenisch, M. I. Rodriguez Perez, G. Gross, R. Woessner, R.
Voges, P. I. Arvidsson, J. Frackenpohl, D. Seebach, Biopharm. Drug
Dispos. 2002, 23, 251–262.
[14] V. Tereshko, J. M. Monserrat, J. PØrez-Folch, J. Aymamí, I. Fita,
J. A. Subirana, Acta Crystallogr. Sect. B 1994, 50, 243–251.
[15] M. Rothe, D. Mühlhausen, Angew. Chem. 1979, 91, 79–80; Angew.
Chem. Int. Ed. Engl. 1979, 18, 74–75.
[16] A. Trzeciak, W. Bannwarth, Tetrahedron Lett. 1992, 33, 4557–4560.
[17] S. A. Kates, N. A. SolØ, C. R. Johnson, D. Hudson, G. Barany, F. Al-
bericio, Tetrahedron Lett. 1993, 34, 1549–1552.
[18] G. Guichard, S. Abele, D. Seebach, Helv. Chim. Acta 1998, 81, 187–
206.
[39] a) K. Gademann, T. Kimmerlin, D. Hoyer, D. Seebach, J. Med.
Chem. 2001, 44, 2460–2468; b) D. Seebach, J. V. Schreiber, S. Abele,
X. Daura, W. F. van Gunsteren, Helv. Chim. Acta 2000, 83, 34–57.
[40] M. Rueping, B. Jaun, D. Seebach, Chem. Commun. 2000, 22, 2267–
2268.
[19] A. Müller, C. Vogt, N. Sewald, Synthesis 1998, 837–841.
[20] H. N. Gopi, V. V. Suresh Babu, Lett. Pept. Sci. 2000, 7, 165–169.
[21] A. Mehta, R. Jaouhari, T. J. Benson, K. T. Douglas, Tetrahedron
Lett. 1992, 33, 5441–5444.
[41] J. L. Matthews, M. Overhand, F. N. M. Kühnle, P. E. Ciceri, D. See-
bach, Liebigs Ann. 1997, 1371–1379.
[42] P. I. Arvidsson, J. Frackenpohl, D. Seebach, Helv. Chim. Acta 2003,
86, 1522–1553.
[22] M. Roice, V. N. Rajasekharan Pillai, J. Appl. Polym. Sci. 2003, 88,
2897–2903.
[23] W. C. Chan, P. D. White in Fmoc Solid Phase Peptide Synthesis, A
Practical Approach (Eds.: W. C. Chan, P. D. White), Oxford Univer-
sity Press, New York, 2000, pp. 41–76.
[43] S. Nilsson, M. Wetterhall, J. Bergquist, L. Nyholm, K. Markides,
Rapid Commun. Mass Spectrom. 2001, 15, 1997–2000.
[44] M. Wetterhall, S. Nilsson, K. Markides, J. Bergquist, Anal. Chem.
2002, 74, 239–245.
[45] M. Narita, M. Doi, K. Kudo, Y. Terauchi, Bull. Chem. Soc. Jpn.
1986, 59, 3553–3557.
[24] G. B. Bloomberg, D. Askin, A. R. Gargaro, M. J. A. Tanner, Tetrahe-
dron Lett. 1993, 34, 4709–4712.
Received: March 4, 2005
[25] S. C. Story, J. V. Aldrich, Int. J. Pept. Protein Res. 1994, 43, 292–296.
Published online: July 29, 2005
6158
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6145 – 6158