Synthesis and biological evaluation of novel turn-modified gramicidin S analogues

Article abstract of DOI:10.1016/S0968-0896(03)00219-0

The synthesis of novel gramicidin S analogues having additional functionalities in the turn region by employing a biomimetic approach is described. The preservation of β-sheet character in all analogues was established by NMR and the biological activity was evaluated.

Full text of DOI:10.1016/S0968-0896(03)00219-0

Bioorganic & Medicinal Chemistry 11 (2003) 2835–2841
Synthesis and Biological Evaluation of Novel Turn-Modified
Gramicidin S Analogues
Gijsbert M. Grotenbreg,a Emile Spalburg,b Albert J. de Neeling,b
Gijsbert A. van der Marel,a Herman S. Overkleeft,a Jacques H. van Booma
and Mark Overhanda,*
aLeiden Institute of Chemistry, Gorlaeus Laboratories, PO Box 9502, 2300 RALeiden, The Netherlands
bNational Institute of Public Health and the Environment, Research Laboratory for Infectious Diseases, PO Box 1,
3720 BABilthoven, The Netherlands
Received 3 January 2003; accepted 25 March 2003
Abstract—The synthesis of novel gramicidin S analogues having additional functionalities in the turn region by employing a bio-
mimetic approach is described. The preservation of b-sheet character in all analogues was established by NMR and the biological
activity was evaluated.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
over the last decades.2,9 These examples include mod-
ulation of the amino acid composition as well as enlar-
ging the b-strand region (i.e., the synthesis of GS
homologues).10,11 Perusal of the latter studies revealed a
number of factors influencing the bactericidal efficiency
of b-strand modified GS analogues such as amphiphili-
city, hydrophobicity, nature and number of cationic
residues, backbone size and conformation. However,
the wealth of information gathered has, to date, not
resulted in the generation of a synthetic, clinically
applicable antibiotic based on GS.
The cationic antimicrobial peptide gramicidin S (GS, 1,
Fig. 1), isolated from Bacillus brevis,1 is active against a
wide range of bacteria and fungi.2,3 The accumulation
of GS in cytoplasmic membranes leads to loss of lipid
bilayer integrity and increased cell permeability, ultimately
resulting in bacterial cell death. GS resides in the polar–
apolar interfacial regions of lipid bilayers.4 Presumably,
electrostatic and hydrophobic peptide–membrane inter-
actions are predominantly responsible for antibacterial
activity.
Modulation of the turn regions of GS is a relatively
unexplored area of research. It is likely that substitution
of the turn region amino acids, especially the proline
residue, in most cases will lead to loss of b-sheet char-
acter and concomitant loss of antimicrobial activity.3
We envisaged that the development of a strategy that
allows the introduction of additional functionalities to
the b-turn region of GS, without interfering with its
intrinsic b-sheet character, would provide a potential
entrance towards new GS-based antibiotics. Here, we
report our results on the generation of turn-modified
GS analogues, in which the two dPhe residues are
replaced by benzylated d-tyrosine (i.e., dTyr(Bn), pep-
tide 3, Fig. 1), as well as derivatives where Pro is sub-
stituted for either 2S,4R-azidoproline (R-Azp, peptide
4) or 2S,4S-azidoproline (S-Azp, peptide 5). Further-
more, the transformation of the azide residues into sec-
ondary amines (6 and 7, having additional cationic
GS is a C2-symmetric cyclic decamer that adopts a rigid
b-sheet structure,5ꢀ7 in which the Val-, Orn- and Leu-
residues align to form the antiparallel b-strands and
Pro-dPhe induce the type II’ b-turns.8 The side chains of
the hydrophilic and hydrophobic residues are posi-
tioned on opposite faces of the molecule, thus creating
the amphiphilic arrangement that is thought to be at the
basis of the biological properties of GS.
There is a wide interest in the generation of new GS
analogues with improved antimicrobial activities. In this
respect, a plethora of GS derivatives, in which the
b-strand region has been modified, has been described
*Corresponding author. Tel.: +31-715-274483; fax: +31-715-274307;
e-mail: overhand@chem.leidenuniv.nl
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00219-0

2836
G. M. Grotenbreg et al. / Bioorg. Med. Chem. 11 (2003) 2835–2841
Figure 1. Gramicidin S and envisaged analogues.
Scheme 1. Synthesis of GS derivatives; a Xaa1=dPhe, Xaa2=Pro; b
Xaa1=dTyr, Xaa2=Pro; c Xaa1=dTyr(Bn), Xaa2=Pro; d Xaa1=d-
Phe, Xaa2=R-Azp; e Xaa1=dPhe, Xaa2=S-Azp; (i) HMPB, BOP,
DiPEA, NMP; (ii) Fmoc-Leu-OH, DIC, DMAP (5 mol%), DCM (iii)
Repetitive deprotect: 20% piperidine in NMP, condense; Fmoc-aa-
OH or Boc-Xaa1-OH, BOP, HOBt, DiPEA, NMP; (iv) 1% TFA in
DCM; (v) HONSu, EDC, DCM; (vi) 50% TFA in DCM; (vii) 15a–e
in DMF, add to pyridine.
functionality in the turn region), benzyloxycarbamates
(8 and 9, with bulky hydrophobic residues) and succi-
nylamides (10 and 11, featuring carboxylic acid turn
region elements) is presented. The secondary structure
as well as the capacity to arrest proliferation of various
Gram-positive and -negative bacterial strains of GS
analogues 3–11 were compared with GS and the known
GS analogue 2.
support the Boc-protected pentamers were condensed
with HOSu under the agency of EDC to give their
respective N-succinimic esters. After removal of the
Boc-protective groups (TFA/DCM, 1/1, v/v) the acti-
vated pentapeptides 14a–e were subjected to cyclodi-
merization by slow addition to a pyridine solution up to
a final concentration of 3 Â 10ꢀ3 M at 25 ꢁC.13 The
resulting cyclic decamers were identified using LC/MS
and purified by semi-preparative HPLC to give GS and
2–5 in yields of around 5%, based on the initial loading
of resin 13. The crude cyclodimerization mixture con-
tained, besides the expected products (i.e., the cyclic
monomer and dimer described by Tamaki et al.), several
other unidentified fragments that displayed an identical
ESI-MS profile. Apparently, the formation of cycli-
zation products involving the ornithine side chain can
also occur once the linear decameric active ester is
formed. Notwithstanding the formation of these unde-
sired side products, the biomimetic synthetic sequence
provides an easy and rapid access to the construction of
C2-symmetric GS analogues 2–5.
Results and Discussion
Chemistry
Several different synthetic strategies towards GS and its
analogues have been reported in literature.12 Tamaki
and coworkers followed a biomimetic approach in
which the specific pentameric sequence H2N-dPhe-Pro-
Val-Orn-Leu-ONSu was found to cyclodimerize and
produce GS. The ability of this particular sequence to
form GS was attributed to a preorganization of the
activated decameric linear peptide forming a b-hairpin
structure.13 Importantly, the replacement of specific
amino acid residues in the synthetic sequence (e.g.,
d
dPhe(4-Br) instead of Phe) was shown not to interfere
with the dimerization-cyclization reaction.14 We there-
fore chose to study the efficiency of this biomimetic
synthesis for the construction of GS analogues 2–5.
Commercially
available
4-methylbenzhydrylamine
As the next research objective, we set out to functiona-
lize R/S-Azp containing GS analogues 4 and 5. Treat-
ment of azides 4 and 5 (Scheme 2) with 10% Pd/C under
hydrogen atmosphere in the presence of CHCl3 directly
furnished the positively charged aminoproline (Amp)
derivatives 6 and 7 in a respective yield of 58 and 60%,
after HPLC purification. Alternatively, protection of
the ornithine side chains (Boc2O, 4–16 and 5–17, 89 and
86%, respectively) allowed the selective modification of
the azidoproline derivatives, as follows. Staudinger
reduction20 of the azides in 16 and 17, followed by con-
densation of the resulting secondary amines with either
benzyl chloroformate or succinic anhydride, and final
acidic removal of the Boc protective groups afforded
(MBHA) resin (12, Scheme 1) was equipped with the acid-
labile 4-(4-hydroxymethyl-3-methoxyphenoxy)-butyric
acid (HMPB) linker15 under the agency of Castro’s
reagent16 and N,N0-diisopropylethylamine (DiPEA) and
condensed with Fmoc-Leu-OH using N,N0-diisopro-
pylcarbodiimide (DIC) and a catalytic amount of
4-dimethylaminopyridine (DMAP) to give the function-
alized resin 13. The immobilized pentapeptide sequences
14a–e were synthesized via standard peptide chemistry,
employing, next to standard amino acid building blocks,
the readily available Fmoc-2S,4R-azidoproline (in 4)
and Fmoc-2S,4S-azidoproline (in 5), respectively.17ꢀ19
After mild cleavage (1% TFA/DCM) from the solid

G. M. Grotenbreg et al. / Bioorg. Med. Chem. 11 (2003) 2835–2841
2837
The perturbation of chemical shift is defined by Wishart
and co-workers23 as the difference between the mea-
sured chemical shift for the Ha of an amino acid and the
Ha chemical shift value of the same residue reported for
a random coil peptide. The presence of three or more
consecutive residues having ÁdHa >0.1 ppm signifies
an extended b-strand conformation. As can been seen
from the data displayed in Figure 3, the a-protons of the
Leu-Orn-Val sequence of all presented peptides (i.e., 1–
11) clearly show idiosyncratic secondary chemical
shifts.24 The dPhe and dTyr residues location (i.e., in the
turn region) is illustrated by a negative value in these
peptides. In summary, the vicinal coupling constants
and perturbation of chemical shift of compounds 2–11
agree well with GS. Therefore, peptides 2–11 have a
b-sheet structure which is most likely similar to that of
GS.
Scheme 2. Reaction conditions: (i) Boc2O, DiPEA, MeCN; (ii) 10%
Pd/C, H2, CHCl3/MeOH (1/1 v/v); (iii) (a) PMe3, 1,4-dioxane/MeCN/
H2O (20/20/1 v/v/v); (b) Z-Cl, DiPEA, DMF; (vi) 50% TFA in DCM;
(v) (a) PMe3, 1,4-dioxane/MeCN/H2O (20/20/1 v/v/v); (b)
(CH2CO)2O, TEA, DMF.
Biological activity
The assessment of antibacterial activity of peptides 2–11
and GS was performed using a standard minimal inhi-
bitory concentration (MIC) test employing several
Gram-positive and Gram-negative bacterial strains. The
results, listed in Table 1, show activity for GS and pep-
tide 2 that are in agreement with the literature data.3
Azides 4 and 5 as well as peptide 3, containing benzy-
target compounds 8–11 in good yields (8: 42%, 9: 36%,
10: 78%, 11: 66%, respectively, after HPLC purifica-
tion).
Having all desired cyclic peptides in hand, attention was
focused on their structural evaluation by NMR. The
resonance assignment of compounds 1–11 was unam-
biguously accomplished using two-dimensional NMR
experiments (i.e., COSY, TOCSY). Several methods for
d
lated Tyr, have activity profiles comparable to grami-
cidin S. However, peptides 6–11 display a considerable
loss of activity. The exocyclic amines of Amp, adding
positive charge to the turn regions of 6, 7 retain a small
activity for Staphylococcus epidermidis. Supplementary
negative charge, introduced by the succinyl-group, leads
to some activity for peptide 10 against S. epidermidis,
Enterococcus faecalis and Escherichia coli and for pep-
tide 11 against Pseudomonas aeruginosa. In contrast to
the introduction of hydrophobic moieties on the tyro-
sine residues, as in 3, supplementary hydrophobicity on
the proline residue (i.e., 8 and 9) largely abolishes all
activity. These findings underscore the speculation of
1
the interpretation of the acquired H NMR spectra can
be applied to establish secondary structure elements in
peptides.
d
d
The presence of the Phe and Tyr residues in the turn
regions was indicated by the small vicinal spin-spin
coupling constants (3JHNa <4 Hz), as was postulated by
Ramachandran et al.21 As can be seen by tabulating the
3
coupling constants for peptides 1–11 (Fig. 2) the JHNa
d
of the Leu, Orn and Val (ranging between 8.5 and
9.0 Hz) corresponds to a b-sheet structure.22 The value
of the coupling constants for all residues are largely
comparable with the corresponding values in GS.
Izuyama et al.,2 that large groups on the Phe-position
have a stabilizing effect on the b-turn, thereby enhan-
Figure 3. Chemical shift perturbation: ÁdHa=observed dHa
ꢀrandom coil dHa. For entries 2 and 3, the random coil value for
tyrosine was used. As the substitution pattern on the pyrrolidine moi-
ety influences the chemical shift of the Ha, no reference values for the
Azp residues were available and were therefore omitted. The reported
random coil value for lysine was used for ornithine.25
Figure 2. Coupling constants (3JHNa) are given in Hertz (Hz). In the
1H spectrum of peptide 7, no splitting pattern of amide resonance sig-
nals could be observed and was therefore excluded. Peptides 8 and 10
showed no observable 3JHNa for the dPhe residues.

2838
G. M. Grotenbreg et al. / Bioorg. Med. Chem. 11 (2003) 2835–2841
Table 1. Antimicrobial activity (MIC in mg/mL)
Compds
S. aureusa
25Wc MTd
S. epidermidisa
25Wc MTd
E. faecalisa
25Wc MTd
E. colib
P. aeruginosab
25Wc MTd
B. cereusa
25Wc
MTd
25Wc
MTd
1
2
3
4
5
6
7
8
4
32
8
4
4
>64
>64
>64
>64
>64
>64
8
32
16
8
4
8
4
2
4
4
8
8
4
8
8
32
8
8
16
>64
64
>64
64
32
64
16
32
32
8
16
>64
>64
>64
64
32
64
>64
32
>64
>64
>64
>64
>64
>64
64
32
>64
>64
32
>64
>64
>64
>64
64
—
—
—
—
>64
>64
>64
>64
>64
>64
>64
>64
>64
>64
32
—
—
—
—
8
8
16
>64
16
>64
>64
>64
>64
64
8
—
—
>64
>64
64
>64
>64
>64
32
64
>64
32
>64
64
32
64
>64
64
32
>64
>64
>64
>64
>64
>64
64
>64
>64
>64
>64
>64
64
9
10
11
32
>64
32
>64
64
Measurements were executed using standard agar dilution techniques (—=not determined).
aGram-positive.
bGram-negative.
c3 mL/25 well plates.
d100 mL/96 microtiter plates.
cing the antimicrobial activity. However, as the exact
process of membrane disruption is not fully under-
stood,4 the existence of b-sheet structure alone in the
presented GS analogues can not be used for the predic-
tion of potential antimicrobial activity.
analysis was performed on a Jasco HPLC-system
(detection simultaneously at 214 and 254 nm) coupled to
a Perkin Elmer Sciex API 165 mass instrument with a
custom-made Electronspray Interface (ESI). An analy-
tical Alltima C18 colomn (Alltech, 4.6 mmD
Â
250 mmL, 5 m particle size) was used in combination
with buffers A: H2O, B: MeCN and C: 0.5% aqTFA.
For RP-HPLC purification of the peptides, a BioCAD
‘Vision’ automated HPLC system (PerSeptiveBiosys-
tems, inc.) supplied with a semi-preperative Alltima C18
column (Alltech, 10.0 mmD Â 250 mmL, 5 m particle
size) was used. The applied buffers were A: H2O, B:
MeCN and C: 1.0% aqTFA.
Conclusions
The biomimetic synthesis of GS and analogues 2–5 was
successfully employed. Modification of the fully assem-
bled azide containing peptides 4 and 5 led to several GS
analogues (6–11) having hydrophobic and hydrophilic
functionalities in the turn region. The conservation of b-
sheet character was confirmed for all peptides (1–11)
utilizing several NMR techniques. Examination of the
antimicrobial activity of the aforementioned peptides
showed a lowered bactericidal effect for compounds 6–
11. It may be concluded that modification of the proline
residue is counterproductive with respect to anti-
bacterial activity even when the b-sheet character is
preserved. The highest antimicrobial activity was for
azides 4 and 5, reflecting the small tolerance for turn
D
cyclo( Phe-Pro-Val-Orn-Leu)2 (1). RP-HPLC purifica-
tion (linear gradient of 3.5 CV; 40!75% B; Rt 3.2 CV)
followed by lyophilization gave GS in a yield of 5.9 mg
(5.2 mmol, 5%). LC/MS analysis: Rt 20.48 min (linear
gradient 50!90% B in 20 min); m/z=1142.0 [M+H]+,
1
571.6 [M+H]2+. For H NMR, see Table 2.
D
cyclo( Tyr-Pro-Val-Orn-Leu)2 (2). RP-HPLC purifica-
tion (linear gradient of 4.0 CV; 30!65% B; Rt 3.5 CV)
furnished, after lyophilization, unprotected peptide 2 in
a yield of 3.8 mg (3.2 mmol, 3%). LC/MS analysis: Rt
10.15 min (linear gradient 25!90% B in 20 min); m/z
=1173.8 [M+H]+, 587.5 [M+H]2+. For 1H NMR, see
Table 2.
d
region modifications. Surprisingly, the benzylated Tyr
analogue 3 is more active than its unprotected counter-
part 2. Thus, the introduction of large hydrophobic
d
entities in the Phe region holds promise for the future
development of GS-based antibiotics.
D
cyclo( Tyr(Bn)-Pro-Val-Orn-Leu)2 (3). RP-HPLC pur-
ification (linear gradient of 4.0 CV; 40!85% B; Rt
4.0 CV) furnished, after lyophilization, benzyl protected
peptide 3 in a yield of 2.4 mg (1.7 mmol, 2%). LC/MS
analysis: Rt 23.73 min (linear gradient 50!90% B in
20 min); m/z=1354.1 [M+H]+, 677.8 [M+H]2+. For
1H NMR, see Table 2.
Experimental
Chemistry
The SPPS was performed on an ABI 433A (Applied
Biosystems) automated peptide synthesizer supplied
with the FastMoc1 peptide synthesis protocol. 1H
NMR spectra were recorded on a Bruker AV-400
(400 MHz) spectrometer at 298 K. Chemical shifts (d)
are tabulated in ppm, relative to the solvent peak of
CD3OH (3.30 ppm), unless stated otherwise. LC/MS
D
cyclo( Phe-2S,4R-Azp-Val-Orn-Leu)2 (4). RP-HPLC
purification (linear gradient of 3.5 CV; 50!70% B; Rt
2.9 CV) gave, after lyophilization, azide 4 in a yield of
3.8 mg (3.1 mmol, 3%). LC/MS analysis: Rt 20.43 min

G. M. Grotenbreg et al. / Bioorg. Med. Chem. 11 (2003) 2835–2841
2839
Table 2. Chemical shift (d in ppm)
aNH bd bu
(linear gradient 50!90% B in 20 min); m/z=1223.9
1
[M+H]+, 612.6 [M+H]2+. For H NMR, see Table 2.
a
gd
gu
dd
du NH
D
cyclo( Phe-2S,4S-Azp-Val-Orn-Leu)2 (5). RP-HPLC
GS
Leu
Orn
Val
Pro
dPhe
8.80 4.66 1.55 1.41 1.50
—
0.87
3.05
—
3.73
0.87 —
2.91 7.80
purification (linear gradient of 4.0 CV; 40!75% B; Rt
3.8 CV) gave, after lyophilization, azide 5 in a yield of
5.1 mg (4.2 mmol, 4%). LC/MS analysis: Rt 21.15 min
(linear gradient 50!90% B in 20 min); m/z=1224.0
8.70 4.97 2.05 1.64 1.79 1.79
7.73 4.17 2.26 0.96 0.86
4.35 2.00 1.67 1.71 1.59
—
—
2.48
—
—
—
8.90 4.50 3.10 2.96
7.33–7.24
1
[M+H]+, 612.6 [M+H]2+. For H NMR, see Table 2.
Peptide 2
Leu
Orn
Val
Pro
dTyr
8.72 4.65 1.53 1.39 1.53
—
0.88
2.99
—
3.74
0.88 —
2.81 7.81
8.68 4.97 2.01 1.64 1.74 1.74
7.71 4.15 2.26 0.95 0.88
4.36 1.98 1.54 1.64 1.52
D
cyclo( Phe-2S,4R-Amp-Val-Orn-Leu)2 (6). The unpro-
—
—
2.56
—
—
tected azide 4 (23 mg, 19 mmol) was dissolved in
chloroform (2 mL) and methanol (2 mL) and a cataly-
tic amount of 10% Pd/C was added. The resulting
mixture was placed under an atmosphere of hydrogen
and stirred 16 h. The suspension was filtered over a
plug of Hyflow Super Gel1 and concentrated in
vacuo. RP-HPLC purification (linear gradient of 3.5 CV;
30!55% B; Rt 1.6 CV) and freeze-drying of the com-
bined collected fractions, furnished 12.93 mg of peptide 6
(11 mmol, 58%). LC/MS analysis: Rt 13.64 min (linear
gradient 20!60% B in 20 min); m/z =1172.0 [M+H]+,
586.6 [M+H]2+. For 1H NMR, see Table 2.
—
8.86 4.42 3.02 2.86
7.06–6.71
Peptide 3
Leu
Orn
8.70 4.64 1.52 1.39 1.52
—
0.88
3.01
—
3.68
7.12–6.90
7.42–7.25
0.88 —
2.84 7.81
8.67 4.98 2.02 1.60 1.74 1.74
7.68 4.13 2.23 0.93 0.86
4.28 1.92 1.58 1.67 1.52
8.87 4.42 3.02 2.87
4.87
Val
—
—
2.40
—
—
Pro
—
dTyr
Bn
—
—
—
Peptide 4
Leu
Orn
Val
Azp
dPhe
8.74 4.65 1.51 1.38 1.51
8.72 4.94 2.06 1.65 1.75 1.75
7.67 4.12 2.25 0.96 0.87
4.41 2.25 1.88 3.95
—
0.87
3.02
—
3.95
0.87 —
2.90 7.80
—
—
2.54
—
—
—
—
8.92 4.50 3.13 2.96
7.34–7.24
D
cyclo( Phe-2S,4S-Amp-Val-Orn-Leu)2 (7). The unpro-
Peptide 5
Leu
Orn
Val
Azp
dPhe
8.68 4.66 1.52 1.39 1.52
—
1.01
3.04
—
3.56
1.01 —
2.89 7.82
tected azide 5 (17 mg, 14 mmol) was treated similarly to
4, to give, after RP-HPLC purification (linear gradient
of 3.5 CV; 30!55% B; Rt 1.4 CV) and freeze-drying of
all collected fractions, 9.89 mg of peptide 7 (8.4 mmol,
60%). LC/MS analysis: Rt 12.85 min (linear gradient
20!60% B in 20 min); m/z=1172.0 [M+H]+, 586.6
8.64 4.98 2.04 1.61 1.77 1.77
7.72 4.26 2.24 1.01 0.89
4.48 2.25 1.87 3.98
—
—
2.53
—
—
—
—
8.86 4.43 3.09 2.94
7.31–7.23
Peptide 6
Leu
Orn
Val
Amp
dPhe
8.72 4.62 1.47 1.37 1.47
—
0.85
3.03
—
4.31
0.85 —
2.91 7.81
8.70 4.94 2.02 1.65 1.74 1.74
7.68 4.17 2.25 0.97 0.88
4.57 2.48 1.99 3.76
1
[M+H]2+. For H NMR, see Table 2.
—
—
2.84
—
—
—
—
D
cyclo( Phe-2S,4R-Amp(Z)-Val-Orn-Leu)2 (8). To
8.86 4.49 3.03 3.03
7.35–7.26
a
solution of peptide 16 (17 mg, 12 mmol) in 1,4-dioxane
(2 mL) and MeCN (2 mL) was added 100 mL (100 mmol)
of PMe3 (1 M in toluene). The mixture was stirred for
3 h after which H2O (200 mL) was added and allowed to
stir 16 h. All volatiles were evaporated under reduced
pressure, followed by addition of benzylchoroformate
(7 mL, 48 mmol) and N,N0-diisopropylethylamine (12 mL,
72 mmol) in DMF (2 mL). The solution was stirred 6 h,
concentrated, redissolved in DCM (2 mL) and cooled to
0 ꢁC after which TFA (2 mL) was added. The resulting
mixture was warmed to room temperature over a period
of 30 min. Evaporation of all solvents and RP-HPLC
purification (linear gradient of 3.0 CV; 60!90% B; Rt
1.9 CV) gave, after lyophilization, the Z-protected peptide
8 in 7.2mg (5.0 mmol, 42%). LC/MS analysis: Rt 13.96 min
(linear gradient 50!90% B in 20 min); m/z =1440.2
[M+H]+, 720.7 [M+H]2+. For 1H NMR, see Table 2.
Peptide 7
Leu
Orn
Val
Amp
dPhe
8.69 4.60 1.50 1.37 1.50
—
0.87
3.02
—
3.34
0.87 —
2.94 7.81
8.64 4.92 2.03 1.68 1.77 1.77
7.71 4.10 2.24 1.00 0.89
4.38 2.67 1.77 3.50
—
—
3.34
—
—
—
—
8.96 4.64 3.12 2.94
7.38–7.16
Peptide 8
Leu
Orn
8.73 4.67 1.56 1.41 1.56
—
0.91
3.05
—
4.13
7.35–7.22
7.35–7.22
0.91 —
2.90 7.80
— —
2.29 7.03
8.68 4.97 2.04 1.63 1.76 1.76
7.72 4.17 2.29 1.01 0.91
4.40 2.29 1.75 4.06
8.88 4.46 3.08 2.93
4.90
Val
—
Amp
dPhe
Z
—
—
—
—
—
Peptide 9
Leu
Orn
8.72 4.64 1.52 1.38 1.52
—
0.88
3.06
—
3.37
7.42–7.16
7.42–7.16
0.88 —
2.88 7.79
— —
3.02 6.99
8.63 4.96 2.06 1.62 1.76 1.76
7.64 4.09 2.22 0.93 0.85
4.31 2.24 1.90 3.37
8.87 4.51 3.10 2.92
4.88
Val
—
Amp
dPhe
Z
—
—
—
—
—
Peptide 10
Leu
Orn
Val
Amp
dPhe
Succinyl
8.70 4.65 1.46 1.38 1.46
—
0.88
3.04
—
4.20
7.32–7.23
—
0.88 —
2.89 7.82
— —
2.27 7.96
D
cyclo( Phe-2S,4S-Amp(Z)-Val-Orn-Leu)2 (9). To
a
8.66 4.95 2.02 1.61 1.74 1.74
7.74 4.16 2.25 0.98 0.88
4.42 2.27 1.69 4.20
8.88 4.42 3.08 2.92
2.52 2.52 2.36 2.36
solution of peptide 17 (4 mg, 2.8 mmol) in 1,4-dioxane
(2 mL) and MeCN (2 mL) was added 23 mL (23 mmol) of
PMe3 (1 M in toluene). The mixture was stirred for 3 h
after which H2O (200 mL) was added and allowed to stir
16 h. All volatiles were evaporated under reduced pres-
sure followed by addition of benzylchoroformate (2 mL,
12 mmol) and N,N0-diisopropylethylamine (3 mL,
17 mmol) in DMF (2 mL). The solution was stirred 6 h,
concentrated, redissolved DCM (2 mL) and cooled to 0 ꢁC
after which TFA (2 mL) was added. The resulting mixture
—
—
—
—
—
—
—
Peptide 11
Leu
Orn
Val
Amp
dPhe
Succinyl
8.74 4.64 1.51 1.38 1.51
8.64 4.96 2.05 1.63 1.76 1.76
7.64 4.09 2.25 1.00 0.87
4.30 2.40 1.76 3.47
8.86 4.51 3.10 2.91
2.52 2.52 2.37 2.37
—
0.89
3.04
—
3.23
7.35–7.21
—
0.89 —
2.88 7.79
— —
3.23 7.95
—
—
—
—
—
—
—

2840
G. M. Grotenbreg et al. / Bioorg. Med. Chem. 11 (2003) 2835–2841
was warmed to room temperature over a period of 30min.
Evaporation of all solvents and RP-HPLC purification
(linear gradient of 3.0 CV; 50!90% B; Rt 2.4 CV) gave,
after lyophilization, the Z-protected peptide 9 in 1.5 mg,
(1.0 mmol, 36%). LC/MS analysis: Rt 16.02 min (linear
gradient 50!90% B in 20min); m/z=1440.0 [M+H]+,
720.7 [M+H]2+. For 1H NMR, see Table 2.
1 min; (ii) Coupling: the appropriate amino acid
(0.5 mmol) was dissolved in NMP (1 mL) and subse-
quently 0.5 mmol (0.5 M BOP/0.5 M HOBt in NMP/
DMF 1/1, v/v) and 1.5 mmol of DiPEA (1.25 M in
NMP) were added. The resulting mixture was trans-
ferred to the reaction vessel and shaken for 90 min; (iii)
Capping: the resin was subjected to 1 min of shaking in
a solution of 0.5 M acetic anhydride, 0.125 M DiPEA
and 0.015 M HOBt in NMP (2 mL). The applied amino
acids were Boc-dPhe-OH, Boc-dTyr-OH, Fmoc-2S,4R-
Azp-OH, Fmoc-2S,4S-Azp-OH, Fmoc-Leu-OH, Fmoc-
Orn(Boc)-OH, Fmoc-Pro-OH and Fmoc-Val-OH.
Double couplings were executed for Val.
D
cyclo( Phe-2S,4R-Amp(Su)-Val-Orn-Leu)2 (10). Azide
16 (17 mg, 12 mmol) was dissolved in 1,4-dioxane (2 mL)
and MeCN (2 mL) and to the solution was added tri-
methylphosphine (100 mL, 1.0 M in toluene). The result-
ing mixture was stirred for 3 h, followed by addition of
H2O (200 mL) and stirred 16 h. All volatiles were
removed by evaporation and succinic anhydride
(4.8 mg, 48 mmol) and triethylamine (6.8 mL, 48 mmol) in
N,N0-dimethylformamide (4 mL) were added. After stir-
ring for 2 h, the reaction was quenched with H2O
(200 mL) and concentrated. The resulting peptides were
dissolved in DCM (5 mL) and cooled to 0 ꢁC, after
which TFA (5 mL) was added. The mixture was allowed
to warm to room temperature, stirred for 30 min and
concentrated in vacuo. RP-HPLC purification (linear
gradient of 3.0 CV; 30!50% B; Rt 2.7 CV) followed by
lyophilization furnished 12.8 mg of 10 (9.3 mmol, 78%).
LC/MS analysis: Rt 15.39 min (linear gradient 20!60%
Cleavage from the resin. The resin (14a–e) was treated
with 1% TFA in DCM (4 Â 10 mL). All fractions were
collected and after addition of toluene (50 mL) con-
centrated under reduced pressure. The mass of the fully
protected peptides was established by mass spectro-
scopy (ESI-MS):
a
m/z=830.4 [M+H]+, 852.6
[M+Na]+ b m/z=830.7 [M+H]+, 852.6 [M+Na]+
c
m/z=895.7 [M+H]+, 917.7 [M+Na]+ d m/z=805.7
[M+H]+, 827.5 [M+Na]+ e m/z=789.9 [M+H]+,
811.7 [M+Na]+
Activation of pentapeptides. The crude pentamers
(100 mmol) were dissolved in DMF (2 mL) and cooled to
0 ꢁC. To this mixture were added N-hydroxy succini-
mide (23 mg, 200 mmol) in DCM (1 mL) and N-(3-
B in 20 min.); m/z=1371.9 [M+H]+, 686.6 [M+H]2+
For H NMR, see Table 2.
.
1
cyclo( Phe-2S,4S-Amp(Su)-Val-Orn-Leu)2 (11). Azide
17 (15 mg, 11 mmol) was subjected to the same reaction
conditions as described for peptide 10. RP-HPLC pur-
ification (linear gradient of 3.0 CV; 30!50% B; Rt
2.8 CV) followed by lyophilization furnished 9.4 mg of
11 (6.9 mmol, 66%). LC/MS analysis: Rt 14.30 min (lin-
ear gradient 20!60% B in 20 min); m/z=1372.0
Dimethylaminopropyl)-N0-ethylcarbodiimide
hydro-
D
chloride (38 mg, 200 mmol). The reaction mixture was
then allowed to warm to room temperature and stirred
16 h. The resulting solution was concentrated and par-
titioned between DCM (50 mL) and H2O (10 mL). The
organic layer was dried (MgSO4) and concentrated.
1
[M+H]+, 686.7 [M+H]2+. For H NMR, see Table 2.
Deprotection. The Boc-protection groups were removed
by dissolving the peptide active esters in DCM (2 mL)
followed by addition of TFA (2 mL) at 0 ꢁC. The
reaction mixture was allowed to warm to room tem-
perature, stirred for 30 min after which toluene (10 mL)
was added and the solvents were removed in vacuo.
The crude peptides (15a–e) were used in the following
cyclodimerization reaction without further purifica-
tion.
Fmoc-Leu-HMPB-MBHA resin (13). 4-Methylbenzhy-
drylamine resin 12 (806 mg, 0.5 mmol) was suspended in
1,4-dioxane, evaporated to dryness (3 Â 50 mL) and
resuspended in NMP (25 mL). To the mixture, HMPB
(360 mg, 1.5 mmol), BOP (663 mg, 1.5 mmol) and
DiPEA (0.523 mL, 3.0 mmol) were added. The suspen-
sion was shaken 16 h, filtered and the resin was con-
secutively washed with DCM (2 Â 20 mL), MeOH
(20 mL) and DCM (2 Â 20 mL). The resin was sus-
pended in 1,4-dioxane, evaporated to dryness (3 Â
50 mL) and resuspended in DCM (25 mL). Subsequent
condensation of the first amino acid was effected by
addition of Fmoc-Leu-OH (530 mg, 1.5 mmol), DIC
(0.258 mL, 1.65 mmol) and DMAP (10 mg, 82 mmol)
after which the reaction mixture was shaken for 2 h.
Washing of the resin and a second esterification cycle
was performed as described above. The loading of the
General cyclodimerization procedure
The crude active esters 15a–e were taken up in DMF
(2 mL) and added dropwise to pyridine (30 mL). After
stirring for 24 h, the resulting mixture was concentrated,
analysed by LC/MS and purified by RP-HPLC to give
peptides 1–5.
D
cyclo( Phe-2S,4R-Azp-Val-Orn(Boc)-Leu)2 (16). To a
resin was determined to be 0.48 mmol  gꢀ1
.
solution of cyclic decamer 4 (53 mg, 37 mmol) in MeCN
(5 mL) were added DiPEA (34 mL, 194 mmol) in CHCl3
(1 mL) and di-tert-butyl dicarbonate (21 mg, 97 mmol).
After stirring for 3 h, the solvents were removed in
vacuo and the resulting residue directly subjected to
silica gel colomn chromatography (0!5% MeOH in
EtOAc) to furnish 16 as a white amorphous solid
(47 mg, 33 mmol, 89%). MS (ESI): m/z=1424.2
General procedure for peptide synthesis
Stepwise elongation. The immoblized peptides (14a–e)
were synthesized using 210 mg (0.1 mmol) of resin 13.
The consecutive steps in each coupling cycle were: (i)
Deprotection: 20% piperidine in NMP (2 mL) 5 Â

G. M. Grotenbreg et al. / Bioorg. Med. Chem. 11 (2003) 2835–2841
2841
[M+H]+, 1445.9 [M+Na]+ 1H NMR (DMSO-d6): d
References and Notes
8.94 (d, 1H, NH dPhe, J=2.6 Hz), 8.55 (d, 1H, NH Orn,
J=8.8 Hz), 8.36 (d, 1H, NH Leu, J=8.6 Hz), 7.26 (bs,
6H, Harom dPhe, NH Val), 6.77 (m, 1H, dNH Orn), 4.70
(m, 1H, Ha Orn), 4.54 (m, 2H, Ha Leu, Amp), 4.39 (m,
2H, Ha dPhe, Val), 4.00 (m, 1H, Hg Amp), 3.82 (m, 1H,
Hd Amp), 2.99 (m, 1H, Hb dPhe), 2.88 (m, 3H, Hd Orn,
Hb dPhe), 2.48 (m, 1H, Hd Amp), 2.30 (m, 1H, Hb Amp),
2.01 (m, 1H, Hb Val), 1.67 (m, 1H, Hb Orn), 1.59 (m, 1H,
Hb Amp), 1.43–1.15 (m, 15H, 3 Â CH3 Boc, 1 Â Hb Orn,
2 Â Hg Orn, 2 Â Hb Leu, 1 Â Hg Leu), 0.89–0.77 (m,
12H, 3 Â Hd Leu, 3 Â Hg Val).
1. Gause, G. F.; Brazhnikova, M. G. Nature 1944, 154, 703.
2. Izuyima, N.; Kato, T.; Aoyagi, H.; Waki, M.; Kondo, M.
Synthetic Aspects of Biologically Active Cyclic Peptides–Gra-
micidin S and Tyrocidines; Halstead (Wiley): New York, 1979.
3. Kondejewski, L. H.; Farmer, S. W.; Wishart, D. S.; Han-
cock, R. E. W.; Hodges, R. S. Int. J. Peptide Protein Res.
1996, 47, 460 and references cited therein.
4. Prenner, E. J.; Lewis, R. N. A. H.; McElhaney, R. N. Bio-
chim. Biophys. Acta 1999, 1462, 201.
5. Stern, A.; Gibbons, W. A.; Craig, L. C. Proc. Natl. Acad.
Sci. U.S.A. 1968, 61, 734.
6. Hull, S. E.; Karlsson, R.; Main, P.; Woolfson, M. M.;
Dodson, E. J. Nature 1978, 275, 206.
7. Yamada, K.; Unno, M.; Kobayashi, K.; Oku, H.; Yama-
mura, H.; Araki, S.; Matsumoto, H.; Katakai, R.; Kawai, M.
J. Am. Chem. Soc. 2002, 124, 12684.
8. Gibbs, A. C.; Bjorndahl, T. C.; Hodges, R. S.; Wishart, D. S.
J. Am. Chem. Soc. 2002, 124, 1203 and references cited therein.
9. Ovchinnikov, Y. A.; Ivanov, V. T. In The Proteins; Neurath,
H., Hill, R. Eds.; Academic: New York, 1979; Vol. 5, p 391.
10. (a) Jelokhani-Niaraki, M.; Kondejewski, L. H.; Farmer,
S. W.; Hancock, R. E. W.; Kay, C. M.; Hodges, R. S. Bio-
chem. J. 2000, 349, 747. (b) Kondejewski, L. H.; Lee, D. L.;
Jelokhani-Niaraki, M.; Farmer, S. W.; Hancock, R. E. W.;
Hodges, R. S. J. Biol. Chem. 2002, 277, 67 and references cited
therein.
11. Gibbs, A. C.; Kondejewski, L. H.; Gronwald, W.; Nip,
A. M.; Hodges, R. S.; Sykes, B. D.; Wishart, D. S. Nat. Struct.
Biol. 1998, 5, 284.
12. Wishart, D. S.; Kondejewski, L. H.; Semchuk, P. D.;
Sykes, B. D.; Hodges, R. S. Lett. Pep. Sci. 1996, 3, 53.
13. Tamaki, M.; Akabori, S.; Muramatsu, I. J. Am. Chem.
Soc. 1993, 115, 10492.
D
cyclo( Phe-2S,4S-Azp-Val-Orn(Boc)-Leu)2 (17). Start-
ing from 5 (51 mg, 35 mmol) peptide 17 was obtained as
described for 16, as a white amorphous solid (42 mg,
30 mmol, 86%). MS (ESI): m/z=1424.2 [M+H]+,
1445.9 [M+Na]+ 1H NMR (DMSO-d6): d 8.81 (d, 1H,
d
NH Phe, J=2.8 Hz), 8.57 (d, 1H, NH Orn, J=8.9 Hz),
8.34 (d, 1H, NH Leu, J=8.9 Hz), 7.27 (bs, 5H, Harom
dPhe), 7.18 (d, 1H, NH Val, J=8.9 Hz), 6.80 (m, 1H, dNH
Orn), 4.69 (m, 1H, Ha Orn), 4.55 (m, 3H, Ha Leu, Ha Val,
Ha Amp), 4.35 (m, 1H, Ha dPhe), 4.11 (m, 1H, Hg Amp),
3.41 (d, 1H, 1 Â Hd Amp, J=11.5 Hz), 2.92 (m, 4H, 2 Â
Hd Orn, 2 Â Hb dPhe), 2.73 (m, 1H, 1 Â Hd Amp), 2.20 (d,
1H, 1 Â Hb Amp, J=13.3 Hz), 1.96 (m, 1H, Hb Val), 1.71
(m, 2H, 1 Â Hb Amp, 1 Â Hb Orn), 1.48–1.16 (m, 15H, 3 Â
CH3 Boc, 1 Â Hb Orn, 2 Â Hg Orn, 2 Â Hb Leu, 1 Â Hg
Leu), 0.87–0.80 (m, 12H, 3 Â Hd Leu, 3 Â Hg Val).
Biological activity
The following bacterial strains were used: Staphylo-
coccus aureus (ATCC 29213), S. epidermidis (ATCC
12228), E. faecalis (ATCC 29212), E. coli (ATCC
25922), P. aeruginosa (ATCC 27853) and Bacillus cereus
(ATCC 11778). Bacteria were stored at ꢀ70 ꢁC and
grown at 35 ꢁC on Columbia Agar with sheep blood
(Oxoid, Wesel, Germany) overnight and diluted in 0.9%
NaCl. Large plates (25 wells of 3 mL) as well as micro-
titre plates (96 wells of 100 mL) were filled with Mueller
Hinton II Agar (Becton Dickinson, Cockeysvill, USA)
containing serial 2-fold dilutions of peptides 1–11. To
the wells was added 3 mL of bacteria, to give a final
inoculum of 104 colony forming units (CFU) per well.
The plates were incubated overnight at 35 ꢁC and the
MIC was determined as the lowest concentration inhi-
biting bacterial growth.
14. Aimoto, S. Bull. Chem. Soc. Jpn. 1988, 61, 2220.
15. Florsheimer, A.; Riniker, B. In Peptides, 1990. Giralt, E.,
¨
Andreu, D., Eds. ESCOM: 1991; p 131.
16. Castro, B.; Dormoy, J. R.; Evin, G.; Selve, C. Tetrahedron
Lett. 1975, 16, 1219.
17. Peterson, M. L.; Vince, R. J. Med. Chem. 1991, 34, 2787.
18. Gangamani, B. P.; Kumar, V. A.; Ganesh, K. N. Tetra-
hedron 1991, 34, 2787.
19. Klein, L. L.; Li, L. P.; Chen, H. J.; Curty, C. B.; DeGoey,
D. A.; Grampovnik, D. J.; Leone, C. L.; Thomas, S. A.;
Yeung, C. M.; Funk, K. W.; Kishore, V.; Lundell, E. O.;
Wodka, D.; Meulbroek, J. A.; Alder, J. D.; Nilius, A. M.;
Lartey, P. A.; Plattner, J. J. Bioorg. Med. Chem. 2000, 8, 1677.
20. Reduction of azides 16 and 17 by Pd-catalysed hydro-
genolysis resulted in undesirable loss of Boc-protecting groups
under several reaction conditions.
21. Ramachandran, G. N.; Chandrasekaran, R.; Kopple,
K. D. Biopolymers 1971, 10, 2113.
22. Wuthrich, K. NMR of Proteins and Nucleic Acids; John
Wiley & Sons: New York, 1986.
Acknowledgements
23. Wishart, D. S.; Sykes, B. D.; Richards, F. M. Biochem-
istry 1992, 31, 1647.
This work was financially supported by the Council for
Chemical Sciences of the Netherlands Organization for
Scientific Research (CW-NWO), the Netherlands Tech-
nology Foundation (STW) and DSM Research. We
thank Nico Meeuwenoord and Hans van der Elst for
their technical assistance and are grateful to Kees Erke-
lens and Fons Lefeber for their help with recording the
NMR spectra.
24. It was previously reported that the chemical shifts of these
residues did not significantly alter when using methanol
instead of water as solvent system. Therefore, to enhance
solubility, CD3OH was employed. Krauss, E. M.; Chan, S. I.
J. Am. Chem. Soc. 1982, 104, 6953.
25. Stanger, H. E.; Syud, F. A.; Espinosa, J. F.; Giriat, I.;
Muir, T.; Gellman, S. H. Proc. Natl. Acad. Sci. U.S.A. 2001,
98, 12015.