Exploring the conformational and biological versatility of β-turn-modified gramicidin s by using sugar amino acid homologues that vary in ring size

Article abstract of DOI:10.1002/chem.201002895

Monobenzylated sugar amino acids (SAAs) that differ in ether ring size (containing an oxetane, furanoid, and pyranoid ring) were synthesized and incorporated in one of the β-turn regions of the cyclo-decapeptide gramicidin S (GS). CD, NMR spectroscopy, mo

Full text of DOI:10.1002/chem.201002895

FULL PAPER
DOI: 10.1002/chem.201002895
Exploring the Conformational and Biological Versatility of b-Turn-Modified
Gramicidin S by Using Sugar Amino Acid Homologues that Vary
in Ring Size
Annemiek D. Knijnenburg,[a] Adriaan W. Tuin,[a] Emile Spalburg,[b]
Albert J. de Neeling,[b] Roos H. Mars-Groenendijk,[c] Daan Noort,[c] Josꢀ M. Otero,[d, e]
Antonio L. Llamas-Saiz,[f] Mark J. van Raaij,[d, g] Gijs A. van der Marel,[a]
Herman S. Overkleeft,[a] and Mark Overhand*[a]
Abstract:
Monobenzylated
sugar
rated SAA moieties determines the
spatial positioning of their cis-oriented
carboxyl and aminomethyl substituents,
thereby subtly influencing the amide
linkages with the adjacent amino acids
in the sequence. Unlike GS itself, the
conformational behavior of the SAA-
containing peptides is solvent depen-
dent. The derivative containing the pyr-
anoid SAA is slightly less hydrophobic
and displays a diminished haemolytic
activity, but has similar antimicrobial
properties as GS.
amino acids (SAAs) that differ in ether
ring size (containing an oxetane, fura-
noid, and pyranoid ring) were synthe-
sized and incorporated in one of the b-
turn regions of the cyclo-decapeptide
gramicidin S (GS). CD, NMR spectros-
copy, modeling, and X-ray diffraction
reveal that the ring size of the incorpo-
Keywords: amino acids · antimicro-
bial peptides
·
conformation
analysis · gramicidin S · b-turn
Introduction
[a] A. D. Knijnenburg, Dr. A. W. Tuin, Prof. Dr. G. A. van der Marel,
Prof. Dr. H. S. Overkleeft, Dr. M. Overhand
Leiden Institute of Chemistry
Sugar amino acids (SAAs) have found wide attraction in
bioorganic chemistry research in the past decades.[1] SAAs
are broadly defined as carbohydrate-derived structures fea-
turing both an amine and a carboxylate. They can be viewed
as carbohydrate mimics or as amino acid mimics and as such
are used in the design of oligomeric structures resembling
oligosaccharides or oligopeptides, respectively.[2] The sugar
ring of SAAs can confer conformational rigidity onto an oli-
gomer,[3] and additional conformational constraints can be
introduced by making use of bicyclic SAA building blocks,
whereas linear SAAs can be used to achieve the opposite.[4]
The number and positioning of functional groups inherent
to the parent monosaccharides give rise to a wealth of dis-
tinct SAA building blocks, in which specific features can be
introduced by derivatization.[5] SAAs mimicking a- or b-
amino acids are available next to SAAs in which the amine
and carboxylate are positioned more distal, providing g- or
d-amino acids, giving rise to dipeptide isosteres.[6]
Over the past few years we have focused on the study of
gramicidin S (GS) analogues featuring SAA dipeptide iso-
steres in one or both turn regions.[7] GS is a C2-symmetric
cyclo-decapeptide ((Pro-Val-Orn-Leu-d-Phe)2) produced by
the soil bacterium Aneurinibacillus migulanus (formerly Ba-
cillus brevis)[8,9] and consists of two b-strands interconnected
by two type II’ b-turns (1 in Scheme 1). The two b-strands
contain hydrophilic (Orn) and hydrophobic (Val and Leu)
amino acids, and the type II’ b-turns contain d-Phe-Pro se-
quences. Four intramolecular hydrogen bonds stabilize the
antiparallel b-sheet with the hydrophilic and hydrophobic
Leiden University, P.O. Box 9502
2300 RA Leiden (The Netherlands)
Fax : (+31)715274307
E-mail: overhand@chem.leidenuniv.nl
[b] E. Spalburg, Dr. A. J. de Neeling
National Institute of Public Health and the Environment
Laboratory for Infectious Diseases
P.O. Box 1, 3720 BA Bilthoven (The Netherlands)
[c] R. H. Mars-Groenendijk, Dr. D. Noort
TNO, Prins Maurits Laboratory
P.O. Box 45, 2280 AA Rijswijk (The Netherlands)
[d] Dr. J. M. Otero, Dr. M. J. van Raaij
Departamento de Bioquꢀmica y Biologꢀa Molecular
Facultad de Farmacia, Edificio CACTUS
Campus Sur, Universidad de Santiago
15782 Santiago de Compostela (Spain)
[e] Dr. J. M. Otero
Laboratoire des Proteines Membranaires
Institut de Biologie Structurale J. P. Ebel
41 Rue Jules Horowitz, 38027 Grenoble (France)
[f] Dr. A. L. Llamas-Saiz
Unidad de Rayos X (RIAIDT), Laboratorio Integral
de Dinꢁmica y Estructura de Biomolꢂculas “Josꢂ R. Carracido”
Edificio CACTUS, Campus Sur, Universidad de Santiago
15782 Santiago de Compostela (Spain)
[g] Dr. M. J. van Raaij
Departamento de Estructura de Macromolꢂculas
Centro Nacional de Biotecnologꢀa, Campus Cantoblanco
c/Darwin 3, 28049 Madrid (Spain)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201002895.
Chem. Eur. J. 2011, 17, 3995 – 4004
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3995

curious to find out what effect
this has on the overall structur-
al integrity and biological activ-
ity of GS analogues thus de-
signed. Here we present the
synthesis of the monobenzyloxy
oxetane- and pyranoid building
blocks 14 and 22, their use in a
solid-phase peptide synthesis
approach and the structural and
biological evaluation of the
peptides GS4–6.
Results and Discussion
Synthesis: In the first instance,
we turned our attention to the
construction of the three suita-
bly protected SAA building
blocks 4–6. Furanoid SAA 5
was prepared as its azide deriv-
Scheme 1. GS (1) and GS analogues 2–6 with a sugar amino acid modified turn region containing a monoben-
zyloxy oxetane- (4), furanoid- (2, 3, and 5), or pyranoid (6) ring. The arrows indicate the increasing distance
between the carboxyl and aminomethyl substituents.
side chains positioned on opposing faces, thereby creating
an amphiphatic structure.[10] GS is an effective bactericidal
agent against several Gram-positive and -negative bacte-
ria.[11] However it also lyses human red blood cells and is,
therefore, only used as an antibiotic to treat topical infec-
tions.[12]
ative by following a previously reported procedure.[15] To get
access to the azide of SAA 4 and 6, we developed new pro-
cedures as outlined below.
Oxetane SAAs can be obtained through a ring contrac-
tion of an a-triflyloxy g-lactone[17] and our synthesis towards
the benzylated SAA 4 is based on this approach. Benzyla-
tion of the C-3 hydroxyl of the known protected d-arabinose
7 (Scheme 2)[18] gave compound 8. Desilylation, tosylation,
and subsequent substitution with sodium azide gave com-
pound 11 in 74% yield over three steps. The isopropylidene
protective group was removed and oxidation of the inter-
mediate anomeric acetal with bromine provided g-lactone
12. Triflation of g-lactone 12 and ring contraction in basic
methanol yielded 2,4-cis-oxetane 13 in 37% yield as a 1:8
trans/cis mixture of isomers. Hydrolysis of the cis-methyl
In our previous studies, we found that replacing one of
the d-Phe-Pro sequences with furanoid SAA 2 (Scheme 1)
led to a GS analogue (GS2) featuring a b-strand structure
highly reminiscent to that of the parent compound itself. In-
terestingly, X-ray diffraction studies of this compound, GS2,
revealed that six GS2 monomers assemble in a b-barrel-like
structure in a fashion that invited speculation on the mode
of action by which GS (derivatives) penetrate cell membra-
nes.[13a] However, GS2 displayed minimal biological activity.
In a subsequent study we capitalized on the presence of the
hydroxyl groups in SAA 2 to introduce an additional func-
tionality. By simply replacing SAA 2 by its monobenzylated
counterpart 3 a more hydrophobic GS analogue (GS3) was
obtained with structural and conformational features closely
related to GS and GS2, but with biological activities largely
restored (both bactericidal activity against an array of bacte-
rial strains, as well as haemolytic activity).[13b]
Subtle structural alteration of the incorporated SAA
while leaving other features intact can thus have a major
impact on the biological properties of a cyclic peptide.[14]
Encouraged by this observation we here set out to study the
influence of ring-size of the incorporated SAA building
block on the structure and activity of GS analogues. Going
from a four-membered ether ring (oxetane SAA 4) through
a five-membered ether ring (furanoid, SAA 5)[15] to a six-
membered ether ring (pyranoid, SAA 6), while leaving all
other features intact, leads to a gradual increase in the dis-
tance between the cis-oriented carboxyl and the aminometh-
yl substituents of the SAA in the b-turn region.[16] We were
Scheme 2. a) BnBr, NaH, DMF, 08C, 71%; b) TBAF in THF, 99%;
c) TsCl, pyridine, 608C, 87%; d) NaN3, DMF, 958C, 86%; e) 1) 70%
AcOH, 2) Br2, H2O, 82% over 2 steps; f) 1) Tf2O; 2) MeOH, K2CO3,
37% over 2 steps; g) NaOH, H2O, 96%. TBAF=tetrabutylammonium
fluoride.
3996
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3995 – 4004

b-Turn-Modified Gramicidin S
FULL PAPER
ester 13 produced the desired 2,4-cis-oxetane 14 in an over-
all yield of 15% (over seven steps).
The 2,6-cis-pyranoid SAA 22 was synthesized by modifi-
cation of a reported route.[19] By starting from known tri-O-
acetyl-d-glucal (15; Scheme 3), the cyanide functionality was
introduced at the anomeric position through a Ferrier rear-
NMR spectroscopic analysis: The peptides GS4–6 show well
dispersed 1D 1H NMR spectra in water (H2O/D2O 16:1) and
methanol (CD3OH), respectively. 2D NMR spectra (COSY,
TOCSY, and cROESY[22]) recorded in both solvent systems
1
were used to fully assign the amino acid residues. H homo-
nuclear techniques[23] were applied to obtain additional in-
formation concerning the local environment of an amino
acid residue in the cyclic sequences. Excluding the SAA
3
moieties, the JHNa coupling constants of analogues GS4–6
resemble the typical GS cyclic b-hairpin structure with two
strand regions (7–9 Hz for Val, Orn, and Leu residues) and
a turn region (4 Hz for d-Phe; Figure 1A).[23,24] This charac-
teristic hairpin motif is also observed from the chemical
shift perturbation values of GS4–6 (Figure 1B), which indi-
cates that the Val, Orn, and Leu residues (DdHa >0.1 ppm)
are part of a b-strand and the Pro and d-Phe residues
(DdHa <0 ppm) are part of a b-turn.[25] Generally lower cou-
pling constant values and chemical shift perturbations of
GS4–6 are observed in water when compared to the ob-
tained data with methanol as the solvent (Figure 1A and
B).[26]
Scheme 3. a) BF3·OEt2, TMSCN, CH2Cl2, 16a (trans): 52%, 16b (cis):
41%; b) 10% Pd/C, 0.02m EtOH, 79%; c) DOWEX H+, MeOH, 608C,
88%; d) 1) SOCl2, pyridine, EtOAc, 2) RuCl3, NaIO4, H2O/ACN/CH2Cl2,
93%; e) 1) NaN3, DMF, 808C, 2) H2SO4 in MeOH/THF, 95%; f) HCl,
1008C, 86%; g) BnBr, 2 equiv NaH, DMF, 38%.
The amide proton exchange rates were measured by tem-
perature coefficient experiments (Figure 1C) and the high
exchange rates of the amide protons of the Orn and d-Phe
residues indicate that these are solvent exposed and thus
not involved in hydrogen-bonding interactions.[27] Both the
Val and Leu residues (with the exception of Leu4 in GS4)
are not solvent exposed, which indicates their involvement
in intramolecular hydrogen bonding. Comparison of the
temperature coefficients measurement for peptides GS4–6
with those of GS indicates a similar intramolecular hydro-
gen-bonding pattern. A gradual increase in solvent exposure
of the amide proton of the SAA moieties is found in the
series oxetane, furanoid, and pyranoid SAA (Figure 1C).
The well-resolved 2D cROESY spectra revealed numer-
ous sequential and long-range cross-peaks (Table 1, see also
rangement to give compounds 16a and 16b as a mixture of
trans/cis anomers that could be separated by using silica gel
chromatography. The double bond in the cis-compound 16b
was reduced, followed by deacetylation and introduction of
a sulfonyl group to provide compound 19 in 65% yield over
the three steps.[20] Cyclic sulphate 19 was treated with
sodium azide and subsequently with acid to give compound
20.[21] Intermediate 20 was hydrolyzed to provide carboxylic
acid 21. Finally a benzyl group was introduced at the C-4 hy-
droxyl functionality to give 2,6-cis-pyranoid SAA 22 (20%
overall yield by starting from 16b, Scheme 3).
i
i+1
Previously we reported[15] the synthesis of 3-benzyloxy-
2,5-cis-furanoid azido acid and the use of this building block
in the synthesis of GS5. By following a similar protocol, the
sugar azido acids 14 and 22 were used to obtain the cyclic
peptides GS4 and GS6 (Scheme 1). This procedure involved
the use of highly acid-sensitive HMPB-BHA resin to give
two fully protected immobilized linear oligomers,[7a,15] mild
acidic cleavage from the solid support, and cyclization under
high dilution with a coupling re-
the Supporting Information). The absence of Ha /Ha
i
i+1
cross-peaks and the presence of Ha /HN cross-peaks (with
the exception of the Pro residue) confirmed that all the
amide bonds are in the trans conformation.[23] The NOE sig-
nals observed in the SAA-modified b-turn are depicted in
Figure 2. In all three cyclic peptides a long-range NOE be-
tween the amide protons of Val2 and Leu9 is observed (Fig-
ure 2A–C, arrow a), which indicates a hydrogen-bonding in-
Table 1. Cytotoxic (haemolytic) and antimicrobial activity (MIC, mgL)ꢀ1 of GS and GS analogues.
agent cocktail of benzotriazol-
1-yl-oxytripyrrolidinophospho-
Analogue Retention times[a]
[min]
Erythro
cytes[b]
S. aure- S. epider-
us[c] midis[c]
E. faeca- B. cer- P. aerugi-
E.
R
T
lis[c]
eus[c]
nosa[d]
ACHTUNGTRENNUNG
coli[d]
nium
hexafluorophosphate
(PyBOP), 1-hydroxybenzotri-
azole (HOBt), and N,N-diiso-
GS
8.38
6.21
6.96
7.22
31.2
>500
125
8
8
8
64
32
16
8
16
16
8
64
>64
>64
>64
32
64
64
32
GS4
GS5
GS6
32
16
16
16
16
8
propylethylamine
(DIPEA).
500
After complete deprotection
with strong acid (trifluoroacetic
acid (TFA)) the peptides GS4
and GS6 were purified by prep-
arative HPLC in 21 and 33%
overall yields, respectively.
[a] Retention times from RP-HPLC in minutes (see the Experimental Section). [b] Peptide concentration re-
quired for 100% lysis of erythrocytes in mm. The peptides were tested in triplicate. [c] Gram-positive bacteria,
MIC (MIC=minimal inhibitory concentration) in mgLꢀ1. [d] Gram-negative bacteria, MIC mgLꢀ1 MW in-
cluding 2ꢄTFA anion GS: 1369.49; GS3: 1344.44; GS4: 1358.47; GS5: 1372.49. MIC values were determined
after 24 h of incubation with an experimental error of one MIC interval (a factor two, see the Experimental
Section).
Chem. Eur. J. 2011, 17, 3995 – 4004
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3997

M. Overhand et al.
Figure 1. A) 3JHNa (Hz) in H2O (6% D2O) and CD3OH at 298 K. B) DdHa (=dHa observed—dHa random coil) in H2O and CD3OH at 298 K. C) Tempera-
ture coefficient reported as ꢀDd/DT in ppb/K measured in DMSO. Numbering of amino acids: cyclo-(SAA1-Val2-Orn3-Leu4-d-Phe5-Pro6-Val7-Orn8-
Leu9).
Figure 2. NH/NH cross-peak area of GS4 (A), GS5 (B), and GS6 (C). Arrows indicate the long-range cross-peaks. Spectra were recorded in H2O/D2O
16:1.
3998
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3995 – 4004

b-Turn-Modified Gramicidin S
FULL PAPER
teraction between the carbonyl of Val2 and the amide
proton of Leu9.[23] This is corroborated by the temperature
coefficient experiments shown in Figure 1C. A NOE signal
is observed between the protons of the SAA amide of GS4
and GS5 with the amide of the Val2 residue (Figure 2A and
B, arrow b), which indicates that the incorporated oxetane
and furanoid SAAs slightly alter the positioning of the
amide linkage with Leu9. Importantly in GS6 this NOE
signal is not observed.
ture in water and in methanol. However analogues GS4–6
exhibit an enhancement in excitation, changing from water
to methanol as the solvent, indicating a more pronounced
secondary structure in methanol.
Computational structure determination: The NMR spectro-
scopic data of GS4–6 were used as restraints to generate
computational models of the cyclic peptides (see Tables S2
and S3 in the Supporting Information, Figure 4). Sequential
and long-range NOE cross-peaks were quantified by using
the isolated spin-pair approximation.[29] For additional dihe-
dral angles and hydrogen-bond constraints, respectively,
amide coupling constants, chemical shift perturbations, and
amide exchange values were used.[30] The previously repor-
ted[13a] X-ray structure of GS2 was used as the backbone
starting structure in a distant geometry (DG) calculation.
DG calculations with a modified[31] version of the program
DISGEO[32] were used to generate an assembly of struc-
tures. The five structures with the smallest total error (viola-
tion <0.1 ꢅ, see Table S3 in the Supporting Information)
were further refined by using the CVFF force field[33]
(Biosym) and an average structure was generated (Fig-
ure 4A–C). The generated structures of GS4–6 all exhibit a
twisted structure (twisting angle q=20-458), which is
common for these types of cyclic peptides.[34] The oxetane
SAA in the b-turn (GS4) induces the highest sheet twisting
(twisting angle q at around ~458).[35] In the calculated struc-
tures of GS4 and GS5 the amide linkage between Leu9 and
the SAA moieties is reoriented: the y angle of Leu9 as indi-
cated in Figure 4D and E is between 90–1008. The corre-
sponding y dihedral angle of Leu9 in GS6 is ~1308 and
Circular dichroism: The typical b-sheet/b-turn structure of
all analogues is corroborated by measuring the CD spectra
of GS and GS4–6 at 0.1 mm in water and in methanol
(Figure 3). Two negative minima around 205 and 222 nm are
visible which confirms the b-sheet and b-turn present in the
secondary structure.[28] GS adopts the same secondary struc-
Figure 3. CD spectra at 0.1 mm in water (0.01m NaOAc, pH 5.3) and at
*
0.1 mm in methanol of GS and analogues GS4–6. : GS4, H2O; b:
^
GS4, CH3OH; : GS5, H2O; b: GS5, CH3OH; &: GS6, H2O; b:
~
GS6, CH3OH; : GS, H2O; a: GS, CH3OH.
Figure 4. A)–C) Side view of average DG structures of GS4–6, side chains are omitted for clarity. D)–F) Top view of average structures of GS4–6. G)–
I) Superimposed structures of GS4–6 (r.m.s.d. GS4: 0.37ꢁ0.05; GS5: 0.23ꢁ0.06; GS6: 0.27ꢁ0.04).
Chem. Eur. J. 2011, 17, 3995 – 4004
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3999

M. Overhand et al.
Figure 5. A) Top view of the X-ray structure of GS4, y=13.58, side chains are ommitted for clarity. Intramolecular hydrogen bonds are indicated by
dashed lines. B) Side view of the crystal structure of GS4. C) Intermolecular and intramolecular hydrogen bridges observed in the crystal lattice. D) Top
view of the crystal packing along the a axis. E) Side view of the crystal packing along the c axis.
more closely resembles a regular b-strand dihedral y angle
(~1208).[23]
lysing erythrocytes (Table 1, Figure 6). Interestingly, GS5
and GS6 show a four- and 15-fold decrease in haemolytic ac-
tivity, respectively, as judged by the determined concentra-
tions required for 100% haemolysis (Table 1). GS4 exhibits
the least haemolytic activity with a 100-fold decrease in hae-
molytic activity relative to GS.
X-ray analysis: High-quality crystals of GS4 were grown
from a methanol/water mixture. The crystal structure was
resolved and refined to 1.0 ꢅ resolution (the diffraction
limit of the crystals, see Table S1 in the Supporting Informa-
tion). The crystal structure of GS4 confirms the overall b-
sheet/b-turn structure. In total, three hydrogen bridges are
observed, involving the amides of Leu4, Val7, and Leu9
(Figure 5A). Some differences with the computational struc-
ture were observed. The X-ray structure exhibited a less
twisted pleated-sheet structure (q=188, Figure 5B) and the
amide reorientation is more pronounced (Leu9 y=13.58;
Figure 5A indicated in the dashed box). The crystallographic
analysis of GS4 revealed a shorter b-sheet shape than native
GS, with three intramolecular hydrogen bonds and parallel
disposition of the b-sheets forming infinite “cylindrical”
structures with all the hydrophobic residues in the surface
and the hydrophilic residues pointing to the centre (Fig-
ure 5C–E). The size of the pore in this aggregate is smaller
than in native GS.[34]
In previous studies[36] with related antimicrobial peptides,
it was found that an optimum an optimal hydrophobicity (as
monitored by their retention times) was observed with re-
spect to their biological profile. It was found that com-
pounds that are slightly less hydrophobic than GS are also
less toxic, while retaining potent antimicrobial activity.[36a]
Therefore, determination of retention time under controlled
conditions on RP-HPLC was used as a measure of the pep-
Physical and biological properties: Cyclic peptides GS4–6
were screened for their antimicrobial activity against several
Gram-positive and -negative strains (Table 1). GS6 exhibits
comparable activity to native GS towards the screened
micro-organisms. GS5 displays marginally less activity and
GS4 shows a substantially decreased antimicrobial activity.
All the compounds were also evaluated on their activity in
Figure 6. Haemolytic activity of analogues GS4–6 and GS Experiments
were carried out in duplicate with a maximum of 10% experimental
~
&
*
*
error. : GS4; : GS5; : GS6; : GS.
4000
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3995 – 4004

b-Turn-Modified Gramicidin S
FULL PAPER
tide hydrophobicity. GS4 is the least hydrophobic in this
series of compounds, whereas GS6 is slightly less hydropho-
bic than GS, on the basis of their retention times (Table 1).
antimicrobial peptides:[36] that compounds that are slightly
less hydrophobic than GS are also less toxic while retaining
antimicrobial activity.[36a] It should be noted that this general
trend is followed at low peptide concentration, but that at
higher peptide concentration the less hydrophobic GS5 is
more haemolytic than GS6 (see Figure 6). A possible ex-
planation for this observation might be the subtle changes in
secondary structure.
The relation of antimicrobial activity and haemolytic ac-
tivity is complex due to the diversity in membrane structures
of microorganisms. Several models[37] have been proposed
for the membrane disruption mechanism of GS and it would
be of interest to investigate whether the conformationally
changed peptides (GS4–6) exhibit the same membrane dis-
rupting characteristics. Despite the exhaustive conforma-
tional analysis as described here we do not fully compre-
hend the correlation between the flexibility and biological
activity of the peptides. Whether the conformational flexibil-
ity of the modified peptides GS4–6 can be used for the
future design of compounds with enhanced biological prop-
erties remains subject for further research.
In summary, we have identified a GS derivative (GS6)
with reduced haemolytic activity and largely retained the
antimicrobial activity. Thus, the strategy to design and syn-
thesize a series of homologues with decreasing hydrophobic-
ity and subtle conformational changes presented here
proves to be fruitful to obtain a compound with improved
properties. Currently, we are investigating the haemolytic
properties of GS6 exceeds its slightly decreased antimicrobi-
al activity in more advanced biological assays and animal in-
fection models.
Conclusion
In this report we present a study to evaluate the effect of
ring size of SAAs as type II’ b-turn mimics on the structural
and biological properties of GS-derived cyclic peptides. Ho-
mologues GS4–6, containing a monobenzylated oxetane, fur-
anoid, and pyranoid SAA, respectively, exhibit well-defined
cyclic hairpin structures in solution. However, as the struc-
tures of the peptides GS4–6 show solvent dependence, they
are not as rigid as GS. Sugar amino acid dipeptide isosteres
are generally considered more rigid than the corresponding
dipeptide moieties they replace.[1,6] In cyclic peptides, as is
shown here, this may not always be the case because of the
strict geometrical requirements of the secondary struc-
ture.[30d]
As anticipated the distance between the cis-oriented car-
boxyl and the aminomethyl substituents of the ether rings
becomes larger going from a four-membered (oxetane SAA
4) through a five-membered (furanoid, SAA 5) to a six-
membered ether ring (pyranoid, SAA 6; Figure 7A). The
Experimental Section
Figure 7. A) Overlay of the different ring-size sugar amino acids (4–6) in-
corporated in gramicidin S, oxetane in light grey, furanoid in dark grey
and pyranoid in black. B) Overlay of SAA 6 (black) with the X-ray struc-
ture of the d-Phe-Pro sequence of GS (grey).
General: Light petroleum ether (Petet) with a boiling range of 40–608C
was used. All other solvents used under anhydrous conditions were
stored over 4 ꢅ molecular sieves except for methanol, which was stored
over 3 ꢅ molecular sieves. Solvents used for workup and column chroma-
tography were of technical grade and distilled before use. All other sol-
vents were used without further purification. Reactions were monitored
by TLC analysis. Elemental analyses were carried out with a Perkin-
Elmer Series II CHNS/O analyzer 2400. The linear peptides were cleaved
from resin, cyclized, and purified by RP-HPLC (Gilson GX-281) with a
preparative Gemini C18 column (Phenomenex 150ꢄ21.2 mm, 5 mg parti-
cle size). The applied eluents were A: 0.1% aq. TFA, B: MeCN. The
linear peptides and cyclized peptides were analyzed with LCMS (detec-
tion simultaneously at 214 and 254 nm) equipped with an analytical C18
column (4.6 mmDꢄ250 mmL, 5 mg particle size). The applied buffers
were A: H2O, B: MeCN, and C: 1.0% aq. TFA. HRMS were recorded
by direct injection (2 mL of a 2 mm solution in H2O/MeCN 50:50 v/v and
0.1% formic acid) on a mass spectrometer Thermo Finnigan LTQ Orbi-
trap equipped with an electrospray ion source in positive mode (source
voltage 3.5 kV, sheath gas flow 10, capillary temperature 523 K) with res-
olution R=60000 at m/z: 400 (mass range m/z: 150–2000) and dioctyl-
phthalate (m/z: 391.28428) as lock mass. Optical rotations were measured
on a Propol automatic polarimeter (sodium d-line, l=589 nm). Specific
rotations [a]D are given in degree per centimeter and the concentration c
is given in mgmLꢀ1 in the specific solvent. IR spectra were recorded on a
Perkin–Elmer Paragon 1000 FTIR spectrometer.
overall effect of the difference in ring size is that the amide
linkage between the adjacent leucine residue with the SAA
moiety is subtly reoriented in the oxetane (GS4) and fura-
noid case (GS5) leaving the rest of the hairpin structure
largely intact, whereas the pyranoid (of GS6) is the best
mimic of the d-Phe-Pro two-residue turn motif (Figure 7B).
The difference in the calculated structure of GS4 and its X-
ray structure is attributed to packing in the crystal. From a
biological point of view, SAA6 can be viewed as an im-
proved version of the d-Phe-Pro sequence of GS. The anti-
microbial activity of GS6 (containing SAA 6) is similar to
that of GS and markedly better than GS5 and GS4. Of inter-
est is that GS6 displays a significantly decreased haemolytic
activity relative to GS. As GS6 is slightly less hydrophobic
than GS (as empirically determined by their retention times
(Table 1)), it follows the general trend observed with related
Chem. Eur. J. 2011, 17, 3995 – 4004
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4001

M. Overhand et al.
1
NMR spectroscopy: H and 13C NMR spectra for all intermediates (7–22)
this inoculum was added to 100 mL growth medium, Nutrient Broth from
Difco (ref. nr. 234000, lot nr. 6194895) with yeastextract (Oxoid LP 0021,
lot nr. 900711, 2 g/400 mL broth), in microtiter plates (96 wells). The pep-
tides GS and GS4–6 were dissolved in ethanol (4 gL)ꢀ1 and diluted in
distilled water (1000 mgL)ꢀ1, and two-fold diluted in the broth (64, 32,
16, 8, 4, and 1 mgL)ꢀ1. The plates were incubated at 308C (24–96 h) and
the MIC was determined as the lowest concentration inhibiting bacterial
growth. The experiments were conducted once; the experimental error is
one MIC interval (a factor two).
were recorded on a Bruker AV-400 (400/100 MHz). The spectra of the
peptides (GS4–6) were recorded on a Bruker DMX 600 equipped with a
pulsed field gradient accessory and a cryo-probe. For the 2D cROESY
spectra (200 msec mixing time) the peptides were dissolved 7 mgmLꢀ1 in
a
6% D2O/H2O mixture. Standard DGF-COSY (512cꢄ2084c) and
TOCSY (400cꢄ2048c) spectra were recorded by using presaturation for
solvent suppression. cROESY[22] spectra (400cꢄ2048c, tmix =180 ms)
were recorded by using the presat solvent suppression. All spectra were
recorded in phase-sensitive mode, by using either the TPPI or states-
TPPI for quadrature detection in the indirect dimension. Homonuclear
coupling constants were determined from the corresponding 1H spectra.
Haemolytic assays: Freshly drawn heparinized blood was centrifuged for
10 min at 1000 g at 108C. Subsequently, the erythrocyte pellet was
washed three times with 0.85% saline solution and diluted with saline to
a 1/25 packed volume of red blood cells. The peptides to be evaluated
(GS4–6 and GS) were dissolved in a 30% DMSO/0.5 mm saline solution
to give a 1.5 mm solution of peptide. If a suspension was formed, the sus-
pension was sonicated for a few seconds. A 1% Triton-X solution was
prepared. Subsequently, 100 mL of saline solution was dispensed in col-
umns 1–11 of a microtiter plate and 100 mL of 1% Triton solution was
dispensed in column 12. To wells A1-C1, 100 mL of the peptide was
added and mixed properly. 100 mL of wells A1–C1 was dispensed into
wells A2–C2. This process was repeated until wells A10–C10, followed
by discarding 100 mL of wells A10–C10. These steps were repeated for
the other peptides. Subsequently, 50 mL of the red blood cell solution was
added to the wells and the plates were incubated at 378C for 4 h. After
incubation, the plates were centrifuged at 1000 g at 108C for 4 min. In a
new microtitre plate, 50 mL of the supernatant of each well was dispensed
into a corresponding well. The absorbance at 405 nm was measured and
the percentage of hemolysis was determined. The experiment was con-
ducted once and carried out in triplo. A maximum of 10% experimental
error was found.
Structure generation: The computational structure determinations were
performed on a Silicon Graphics O2 by using the program Insight-II. Dis-
tant geometry (DG) calculations were carried out with the DISCOVER
program by using the CVFF force field[33] and a dielectric constant of 1.
A modified[31] version of the program DISGEO[32] in combination with
3
interproton distances, dihedral angles (f out of JNHa and using the Kar-
plus[30a] equation; y out of DdHa by using y=1208 for values>0.1 ppm
and ꢀ608 for values ꢀ0.1 ppm)[30d] and quantified hydrogen bridges[30b] of
the natural amino acids was employed for the DG. For the interproton
distances the isolated spin-pair approximation[27] was used, setting the in-
tegrated intensity of the NOE cross-peak derived from a set of geminal
protons of the proline (dCH2) to a distance of 1.78 ꢅ. Regions on both
sides of the diagonal were analyzed. Pseudoatom corrections were taken
into account. The upper and lower bond restraints were set as distance
ꢁ10%, accounting for the experimental error. The DG calculation was
performed by using the X-ray structure of compound GS2 (TM5025)[13a]
as the starting structure.
CD spectroscopy: CD spectra were recorded at 298 K on a Jasco J-815
spectropolarimeter by using 0.1 cm path length quartz cells. The CD
spectra are averages of four scans, collected at 0.1 nm intervals between
190 and 250 nm with a scanning speed of 50 nmminꢀ1. The peptides were
prepared at concentrations of 0.1 mm in CH3OH and H2O (0.01m
NaOAc, pH 5.3). Ellipticity is reported as mean residue ellipticity [q],
with approximate errors of ꢁ10% at 220 nm.
Synthesis: The general synthetic procedure of GS analogues and the in-
termediates towards SAA 14 and SAA 22 are reported in the Supporting
Information.
2,4-Anhydro-5-azido-5-deoxy-3-O-benzyl-d-ribonic acid (14): Compound
13 (0.71 mmol, 193 mg) was dissolved in THF/H2O (16 mL, 7:1) and 1m
of aq. NaOH was added (4 equiv 2.84 mL). The mixture was stirred for
16 h and TLC analysis showed full conversion into product (Rf =0.19, 1:1
EtOAc/Petet, 1% AcOH). The solvent was evaporated and the crude
was ion-exchanged with Amberlite H+ in H2O to yield 14 as a colorless
oil (0.68 mmol, 180 mg, 96%). [a]2D0 =+196.4 (c=0.5 in CHCl3); 1H NMR
(400 MHz, CDCl3): d=8.72 (brs, 1H; OH), 7.41–7.28 (m, 5H; CH Ar),
5.06 (d, J=5.2 Hz, 1H; H2), 4.76 (m, 1H; H4), 4.73 (d, J=12.0 Hz, 1H;
CH2 Benzyl), 4.48 (d, J=12.0 Hz, 1H; CH2 Benzyl), 4.51 (m, 1H; H3),
3.36 ppm (ddd, J=3.6, 13.9, 121.2 Hz, 2H; H5, H5’); 13C NMR (101 MHz,
CDCl3): d=173.8 (C=O), 136.5 (Cq Ar), 128.6, 128.4, 128.1 (CH Ar), 84.8
(C4), 81.4 (C2), 76.3 (C3), 72.1 (CH2 Ar), 52.3 ppm (C5); IR n˜ =2923.4
(w), 2104.6 (s), 1735.5 (s), 1262.3 (s), 1126.7 cmꢀ1 (s); HRMS (ESI): m/z:
Crystallization and crystal structure determination of compound GS4:
Colorless needle-shaped crystals were obtained after slow evaporation of
droplets (2 mL) of GS4 (10 mgmLꢀ1) in a 40% solution of MeOH in H2O
plus NaOH in MeOH (2 mL, 0.5m) under paraffin oil in a Terasaki plate.
A 0.5ꢄ0.01ꢄ0.01 mm crystal was mounted in air and then rapidly placed
under liquid nitrogen. Screening of several crystals and preliminary dif-
fraction tests were performed at beamline BM30A and high-resolution
synchrotron data were collected at beamline ID23–2 at the ESRF (Gre-
noble, France). Images were collected by using DNA software,[38] pro-
cessed with XDS,[39] and scaled with XPREP.[39] The structure was solved
by direct methods with the program SHELXD[40] and refined with no in-
tensity cutoff by using the full-matrix least-squares method on F2 imple-
mented in SHELXL[40] and included in the WinGX package.[41] Through-
out the refinement, bond length, bond angle, planarity, and stereochemi-
cal restraints were imposed. All non-H atoms were refined anisotropical-
ly with suitable rigid bond and similarity restraints. All hydrogen posi-
tions were calculated and refined by using a riding atom model. There is
a crystallographically independent molecule per asymmetric unit with
several disordered amino acids. Selected crystallographic data is reported
in Table S1 in the Supporting Information. Final figures were created by
using PyMOL (Delano Scientific, Palo Alto CA, USA). CCDC-792843
contains the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallograph-
ic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
calcd for NaC12H13N3O4: 286.07983; found: 286.07986 [M+Na]+
.
6-Azido-4-O-benzyl-2,3-dideoxy-b-d-erythro-hexopyranosecarboxylic
acid (22): Compound 21 (95 mg, 0.47 mmol) was co-evaporated with tolu-
ene (3ꢄ 50 mL) and dissolved in DMF (30 mL). The solution was cooled
to 08C and 60% mineral oil NaH (2.5 equiv, 1.18 mmol, 47.2 mg) was
added. After 20 min, benzyl bromide (1.1 equiv 0.52 mmol, 61 mL) was
added and the reaction was stirred overnight. The reaction was quenched
with H2O and the volatiles were removed in vacuo. The residue was ex-
tracted with CH2Cl2 (20 mL) and 1m HCl (100 mL). The aqueous layer
was extracted trice with CH2Cl2. The combined organic layers were dried
with Na2SO4 and the volatiles evaporated. The residue was column puri-
fied (52 mg, 0.19 mmol, 38%, 1:5 EtOAc/Petet, 1% AcOH). [a]2D0 =+
114.2 (c
=
1, CHCl3); 1H NMR (400 MHz, CDCl3): d=1H NMR
Antibacterial assays: The following bacterial strains were used: Staphylo-
coccus aureus (ATCC 29213), Staphylococcus epidermidis (ATCC 12228),
Enterococcus faecalis (ATCC 29212), Bacillus cereus (ATCC 11778), Es-
cherichia coli (ATCC 25922), and Pseudomonas aeruginosa (ATCC
27853). Bacteria were stored at ꢀ708C and grown at 308C on Columbia
Agar with sheep blood (Oxoid, Wesel, Germany) suspended in physio-
logical saline until an optical density of 0.1 AU (at 595 nm, 1 cm cuvette).
The suspension was diluted (10ꢄ) with physiological saline, and 2 mL of
(400 MHz, CDCl3): d=8.21 (brs, 1H; COOH), 7.65–6.93 (m, 5H; CH
Benzyl), 4.63 (d, J=11.4 Hz, 1H; CH2 Benzyl), 4.44 (d, J=11.4 Hz, 1H;
CH2 Benzyl), 4.06 (d, J=11.6 Hz, 1H; H1), 3.62 (d, J=12.0 Hz, 1H;
H6,6’), 3.48 (d, J=12.8 Hz, 2H; H6,6’), 3.54–3.45 (m, 1H; H5), 3.36 (m, 1H;
H4), 2.39 (m, 1H; H3), 2.18 (d, J=12.6 Hz, 1H; H2), 1.68 (qd, J=13.6,
3.0 Hz, 1H; H2’), 1.53 ppm (td, J=13.4, 3.6 Hz, 1H; H3’); 13C NMR
(101 MHz, CDCl3): d=173.94 (COOH), 137.51 (Cq benzyl), 128.45,
127.94, 127.78 (CH benzyl), 79.79 (C5), 75.27 (C1), 72.77 (C4), 70.86 (CH2
4002
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3995 – 4004

b-Turn-Modified Gramicidin S
FULL PAPER
benzyl), 51.47 (C6), 28.46 (C3), 27.54 ppm (C2); IR=n˜ =2869.9, 2098.3,
1729.0, 1455.0, 1441.4, 1353.9, 1288.9, 1216.9, 1139.3, 1104.3, 739.5, 699.6,
629.1, 544.3 cmꢀ1; HRMS (ESI): m/z: calcd for NaC14H17N3O4: 314.11113;
found: 314.11126 [M+Na]+.
SAA6), 3.58 (m, 1H; Hdd Pro6), 3.44 (m, 1H; H5 SAA6), 3.32 (m, 1H; H6u
SAA6), 3.19 (td, J=10.3, 4.4 Hz, 1H; H4 SAA6), 3.03 (dd, J=12.7,
5.9 Hz, 1H; Hbd d-Phe5), 2.94 (m, 3H; Hbu d-Phe5, Hdd,u Orn3), 2.84 (m,
2H; Hdd,u Orn8), 2.60 (q, J=8.4 Hz, 1H; Hdu Pro6), 2.35 (m, 1H; H2d
SAA6), 2.12–2.00 (m, 3H; H3d SAA6, Hb Val2,7), 1.89 (m, 3H; Hbd,u Pro6,
Hbd Orn3), 1.80 (m, 2H; Hbd Orn8), 1.67–1.51 (m, 8H; Hbu,gd,u Orn, Hgd,u
Pro6), 1.48–1.34 (m, 6H; Hbd,u,g Leu), 1.12 (m, 1H; H3u SAA6), 0.88 (d,
J=7.0 Hz, 3H; CH3 Val2), 0.86 (d, J=7.1 Hz, 3H; CH3 Val2), 0.84 (d, J=
7.0 Hz, 3H; CH3 Val7), 0.82 (d, J=5.3 Hz, 9H; CH3 Leu4, Val7), 0.79 (d,
J=5.3 Hz, 6H; CH3 Leu9); 13C NMR (151 MHz, H2O+D2O): d=174.92,
174.75, 174.43, 173.88, 173.41, 173.29, 173.10, 163.99, 163.75, 138.23,
136.34, 130.21, 129.90, 129.84, 129.58, 128.57, 118.27, 116.32, 78.89, 75.98,
72.96, 71.73, 67.54, 61.65, 59.70, 59.05, 55.22, 54.13, 53.39, 51.89, 48.15,
41.44, 40.97, 40.28, 40.23, 39.90, 36.76, 32.27, 32.14, 31.25, 30.36, 29.88,
28.87, 28.38, 25.39, 25.17, 24.47, 24.36, 23.27, 23.18, 22.74, 22.48, 19.69,
19.35, 18.82, 18.46 ppm; HRMS (ESI): m/z: calcd for C60H94N11O11:
1144.71288; found: 1144.71441 [M+H]+; LCMS: Rf =7.22 min, linear gra-
dient 10!90% B in 13.5 min; m/z: 1145.0 [M+H]+.
cyclo-[SAA4-Val-Orn-Leu-d-Phe-Pro-Val-Orn-Leu]·2TFA (GS4): The
cyclized deprotected peptide was RP-HPLC purified (linear gradient of
37–46%, 3CV) and yielded 42.0 mg, 31.2 mmol, 21%. 1H NMR
(600 MHz, H2O+D2O): d=8.63 (d, J=3.7 Hz, 1H; NH d-Phe5), 8.47 (d,
J=7.3 Hz, 1H; NH Orn3), 8.41 (d, J=8.7 Hz, 1H; NH Leu4), 8.34 (d, J=
6.8 Hz, 2H; NH Leu9, Orn8), 7.96 (dd, J=3.7, 6.7 Hz, 1H; NH SAA4),
7.81 (d, J=7.6 Hz, 1H; NH Val2), 7.67 (d, J=7.8 Hz, 1H; NH Val7), 7.54
(brs, 4H; NH2 Orn), 7.41–7.17 (m, 10H; CH Ar), 5.00 (d, J=5.4 Hz, 1H;
H2 SAA), 4.85 (m, 1H; H4 SAA), 4.63 (m, 1H; Ha d-Phe5), 4.78 (d, J=
11.2 Hz, 1H; CH2 Benzyl), 4.30 (d, J=11.2 Hz, 1H; CH2 Benzyl), 4.50
(m, 2H; Ha Orn3, Leu4), 4.33 (m, 1H; Ha Pro6), 4.30 (m, 2H; Ha Leu9,
Orn8), 4.12 (t, J=7.2 Hz, 1H; Ha Val2), 4.09 (t, J=5.4 Hz, 1H; H3 SAA),
3.98 (t, J=8.0 Hz, 1H; Ha Val7), 3.61 (m, 1H; Hdd Pro6), 3.54 (m, 1H;
H5d SAA4), 3.23 (d, J=15.5 Hz, 1H; H5u SAA4), 3.01 (m, 1H; Hbd,u d-
Phe5), 2.92 (m, 2H; Hdd,u Orn3), 2.84 (m, 2H; Hdd,u Orn8), 2.75 (m, 1H;
Hdu Pro6), 2.09 (q, J=7.0 Hz, 2H; Hb Val2,7), 1.88 (m, 3H; Hbd,u Pro6, Hbd
Orn3), 1.81 (s, 1H; Hbd Orn8), 1.74–1.50 (m, 8H; Hbu,gd,u Orn, Hgd,u Pro6),
1.50–1.36 (m, 4H; Hbd,u Leu), 1.32 (m, 2H; Hg Leu), 0.92 (d, J=6.7 Hz,
3H; CH3 Val2), 0.89 (d, J=6.7 Hz, 3H; CH3 Val2), 0.86 (d, J=6.7 Hz,
3H; CH3 Val7), 0.83 (d, J=6.9 Hz, 3H; CH3 Val7), 0.82 (d, J=6.1 Hz,
3H; CH3 Leu9), 0.81 (d, J=6.0 Hz, 3H; CH3 Leu9), 0.79 (d, J=6.5 Hz,
3H; CH3 Leu4), 0.77 ppm (d, J=6.4 Hz, 3H; CH3 Leu4); 13C NMR
(151 MHz, CD3OH): d=175.01, 173.91, 173.71, 173.55, 173.45, 173.29,
172.89, 172.70, 138.76, 137.01, 130.50, 129.78, 129.59, 129.14, 128.87,
128.58, 86.50, 83.40, 77.84, 72.50, 62.12, 60.87, 60.40, 55.97, 53.96, 53.43,
52.90, 51.97, 49.58, 49.43, 49.29, 49.15, 49.01, 48.87, 48.72, 48.02, 42.33,
41.88, 40.89, 40.76, 37.48, 32.05, 31.75, 30.70, 30.42, 30.36, 25.90, 25.25,
24.83, 24.60, 23.89, 23.18, 23.14, 22.24, 19.88, 19.73, 19.39, 19.30 ppm:
HRMS (ESI): m/z: calcd for C58H90N11O11: 1116.68158; found:
1116.68316 [M+H]+; LCMS: Rf =6.21 min, linear gradient 10!90% B in
13.5 min.; m/z: 1117.1 [M+H]+.
Acknowledgements
This work was supported by the Leiden Institute of Chemistry. Assis-
tance in performing NMR spectroscopic measurements and computation-
al studies by C. Erkelens, A. W. M. Lefeber, and K. Babu is kindly ac-
knowledged. We thank H. van den Elst for recording HRMS data and
maintaining LCMS systems. We thank FIP and ESRF-Grenoble for pro-
vision of beam-time on beamlines BM30A and ID23–2, respectively.
J.M.O. thanks the Xunta de Galicia and Spanish Ministry of Science and
Innovation for “ꢆngeles AlvariÇo” and “Josꢂ Castillejo” fellowships, re-
spectively. This work was funded by grant BFU2008-01588 to M.J.v.R.
from the Spanish Ministry of Science and Innovation.
[1] S. A. W. Gruner, E. Locardi, E. Lohof, H. Kessler, Chem. Rev. 2002,
102, 491–514.
cyclo-[SAA5-Val-Orn-Leu-d-Phe-Pro-Val-Orn-Leu]·2TFA
(GS5):
1H NMR (600 MHz, H2O+D2O): d=8.70 (d, J=4.3 Hz, 1H; NH d-
Phe5), 8.63 (d, J=8.5 Hz, 1H; NH Leu9), 8.52 (d, J=7.5 Hz, 1H; NH
Orn3), 8.50 (d, J=9.1 Hz, 1H; NH Leu4), 8.36 (d, J=8.4 Hz, 1H; NH
Orn8), 8.23 (t, J=6.1 Hz, 1H; NH SAA), 7.77 (d, J=7.7 Hz, 1H; NH
Val2), 7.57 (brs, 4H; NH2 Orn), 7.53 (d, J=8.6 Hz, 1H; NH Val7), 7.44–
7.17 (m, 10H; CH Ar), 4.65–4.61 (m, 3H; Ha d-Phe5, Orn8, H1 SAA),
4.55 (m, 2H; CH2 Benzyl), 4.49 (m, 2H; Ha Orn3, Leu4), 4.40 (m, 1H; Ha
Leu9), 4.35 (dd, J=2.9, 8.4 Hz, 1H; Ha Pro6), 4.28 (s, 1H; H4 SAA), 4.13
(t, J=8.0 Hz, 1H; Ha Val7), 4.09 (d, J=5.3 Hz, 1H; H3 SAA), 4.01 (t, J=
7.7 Hz, 1H; Ha Val2), 3.62 (m, 1H; Hdd Pro6), 3.58 (m, 1H; H5d SAA),
3.25 (m, 1H; H5u SAA), 3.04 (dd, J=6.0, 12.9 Hz, 1H; Hbd d-Phe5), 2.94
(m, 5H; Hbu d-Phe5, Hdd,u Orn), 2.67 (q, J=8.3 Hz, 1H; Hdu Pro6), 2.39
(dd, J=6.2, 13.2 Hz, 1H; H2d SAA), 2.09 (dq, J=6.9, 13.9 Hz, 2H; Hb
Val2,7), 1.92–1.80 (m, 4H; Hbd Orn, Hbd,u Pro6), 1.73–1.29 (m, 14H; Hbd,ug
Leu, Hbu,gd,u Orn, Hgd,u Pro), 0.93, 0.91, 0.90, 0.88 (s, 12H; CH3 Val2,7),
0.82 (d, J=6.1 Hz, 3H; CH3 Leu9), 0.81 (d, J=5.9 Hz, 3H; CH3 Leu9),
0.79 (d, J=6.4 Hz, 3H; CH3 Leu4), 0.78 ppm (d, J=6.4 Hz, 3H; CH3
Leu4); HRMS (ESI): m/z: calcd for C59H92N11O11: 1130.69723; found:
1130.69918 [M+H]+; LCMS: Rf =6.96 min, linear gradient 10!90% B in
13.5 min; m/z: 1131.0 [M+H]+.
[2] a) F. Schweizer, Angew. Chem. 2002, 114, 240–264; Angew. Chem.
Int. Ed. 2002, 41, 230–253; b) T. K. Chakraborty, S. Ghosh, S. Jayap-
rakash, Curr. Med. Chem. 2002, 9, 421–435; c) I. Velter, B. La Ferla,
F. Nicotra, J. Carbohydr. Chem. 2006, 25, 97–138; d) J. L. J. Blanco,
F. Ortega-Caballero, C. O. Mellet, J. M. G. Fernandez, Beilstein J.
Org. Chem. 2010, 6, No. 20.
[3] a) B. Aguilera, G. Siegal, H. S. Overkleeft, N. J. Meeuwenoord, F.
Rutjes, J. C. M. van Hest, H. E. Schoemaker, G. A. van der Marel,
J. H. van Boom, M. Overhand, Eur. J. Org. Chem. 2001, 1541–1547;
b) T. K. Chakraborty, D. Koley, R. Ravi, V. Krishnakumari, R. Na-
garaj, A. C. Kunwar, J. Org. Chem. 2008, 73, 8731–8744; c) B.
Lopez-Ortega, S. F. Jenkinson, T. D. W. Claridge, G. W. J. Fleet, Tet-
rahedron: Asymmetry 2008, 19, 976–983; d) R. Baron, D. Bakowies,
W. F. van Gunsteren, J. Pept. Sci. 2005, 11, 74–84.
[4] B. A. Mayes, R. J. E. Stetz, M. P. Watterson, A. A. Edwards,
C. W. G. Ansell, G. E. Tranter, G. W. J. Fleet, Tetrahedron: Asymme-
try 2004, 15, 627–638.
[5] M. D. P. Risseeuw, M. Overhand, G. W. J. Fleet, M. I. Simone, Tetra-
hedron: Asymmetry 2007, 18, 2001–2010.
[6] E. G. von Roedern, H. Kessler, Angew. Chem. 1994, 106, 684–686;
Angew. Chem. Int. Ed. Engl. 1994, 33, 687–689.
cyclo-[SAA6-Val-Orn-Leu-d-Phe-Pro-Val-Orn-Leu]·2TFA (GS6): The
cyclized deprotected peptide was RP-HPLC purified (linear gradient of
38–42%, 3CV) and yielded 45.8 mg, 33.4 mmol, 33%. 1H NMR
(600 MHz, H2O+D2O): d=8.76 (d, J=3.5 Hz, 1H; NH d-Phe5), 8.54 (d,
J=9.0 Hz, 1H; NH Leu4), 8.51 (d, J=7.1 Hz, 1H; NH Leu9), 8.49 (d, J=
8.6 Hz, 1H; NH Orn3), 8.33 (d, J=8.6 Hz, 1H; NH Orn8), 7.79 (d, J=
8.9 Hz, 1H; NH Val2), 7.47 (d, J=8.5 Hz, 1H; NH Val7), 7.44–7.20 (m,
11H; CH Ar, NH SAA), 4.69 (m, 1H; Ha Orn8), 4.64 (d, J=11.1 Hz,
1H; CH2 Benzyl), 4.59 (m, 1H; Ha d-Phe5), 4.51 (m, 1H; Ha Leu4), 4.48
(d, J=11.4 Hz, 1H; CH2 Benzyl), 4.36 (m, 1H; Ha Pro6), 4.32 (q, J=
7.4 Hz, 1H; Ha Leu9), 4.22 (t, J=7.6 Hz, 1H; Ha Val7), 4.14 (t, J=8.2 Hz,
1H; Ha Val2), 4.02 (d, J=11.7 Hz, 1H; H1 SAA6), 3.65 (m, 1H; H6d
[7] a) G. M. Grotenbreg, M. Kronemeijer, M. S. M. Timmer, F. El Oual-
id, R. M. van Well, M. Verdoes, E. Spalburg, P. A. V. van Hooft,
A. J. de Neeling, D. Noort, J. H. van Boom, G. A. van der Marel,
H. S. Overkleeft, M. Overhand, J. Org. Chem. 2004, 69, 7851–7859;
b) V. V. Kapoerchan, E. Spalburg, A. J. de Neeling, R. H. Mars-
Groenendijk, D. Noort, J. M. Otero, P. Ferraces-Casais, A. L.
Llamas-Saiz, M. J. van Raaij, J. van Doorn, G. A. van der Marel,
H. S. Overkleeft, M. Overhand, Chem. Eur. J. 2010, 16, 4259–4265.
[8] G. F. Gause, M. G. Brazhnikova, Nature 1944, 154, 703.
[9] M. Berditsch, S. Afonin, A. S. Ulrich, Appl. Environ. Microbiol.
2007, 73, 6620–6628.
Chem. Eur. J. 2011, 17, 3995 – 4004
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4003

M. Overhand et al.
[10] a) D. C. Hodgkin, B. M. Oughton, Biochem. J. 1957, 65, 752–756;
b) G. M. J. Schmidt, D. C. Hodgkin, B. M. Oughton, Biochem. J.
1957, 65, 744–750.
chia, J. H. van Boom, M. Overhand, J. Am. Chem. Soc. 2003, 125,
10822–10829; b) T. Cierpicki, J. Otlewski, J. Biomol. NMR 2001, 21,
249–261.
[11] L. H. Kondejewski, S. W. Farmer, D. S. Wishart, R. E. W. Hancock,
R. S. Hodges, Int. J. Pept. Protein Res. 1996, 47, 460–466.
[12] C. J. Wadsten, C. A. Bertilsson, H. Sieradzki, S. Edstrom, Arch. Oto-
Rhino-Laryngol. 1985, 242, 135–139.
[13] a) G. M. Grotenbreg, M. S. M. Timmer, A. L. Llamas-Saiz, M. Ver-
does, G. A. van der Marel, M. J. van Raaij, H. S. Overkleeft, M.
Overhand, J. Am. Chem. Soc. 2004, 126, 3444–3446; b) G. M. Gro-
tenbreg, A. E. M. Buizert, A. L. Llamas-Saiz, E. Spalburg, P. A. V.
van Hooft, A. J. de Neeling, D. Noort, M. J. van Raaij, G. A. van der
Marel, H. S. Overkleeft, M. Overhand, J. Am. Chem. Soc. 2006, 128,
7559–7565.
[14] a) C. Solanas, B. G. de La Torre, M. Fernandez-Reyes, C. M. Santi-
veri, M. A. Jimenez, L. Rivas, A. I. Jimenez, D. Andreu, C. Cativie-
la, J. Med. Chem. 2010, 53, 4119–4129; b) C. Solanas, B. G. de La
Torre, M. Fernandez-Reyes, C. M. Santiveri, M. A. Jimenez, L.
Rivas, A. I. Jimenez, D. Andreu, C. Cativiela, J. Med. Chem. 2009,
52, 664–674; c) A. D. Knijnenburg, E. Spalburg, A. J. de Neeling,
R. H. Mars-Groenendijk, D. Noort, G. M. Grotenbreg, G. A. van der
Marel, H. S. Overkleeft, M. Overhand, ChemMedChem 2009, 4,
1976–1979.
[28] D. S. Wishart, L. H. Kondejewski, P. D. Semchuk, C. M. Kay, R. S.
Hodges, B. D. Sykes, in Techniques in Protein Chemistry VI (Ed:
J. W. Crabb) Academic Press, Orlando, 1995, pp. 451–458.
[29] D. Neuhaus, M. P. Williamson, The Nuclear Overhauser Effect in
Structural and Conformational Analysis, Wiley, New York, 1989.
[30] a) C. Altona in Encyclopedia of NMR (Eds.: D. M. Grant, R. K.
Harris, Wiley, New York, 1996, p. 4909; b) D. S. Wishart, B. D.
Sykes, F. M. Richards, J. Mol. Biol. 1991, 222, 311–333; c) D. S.
Wishart, B. D. Sykes, F. M. Richards, Biochemistry 1992, 31, 1647–
1651; d) A. C. Gibbs, L. H. Kondejewski, W. Gronwald, A. M. Nip,
R. S. Hodges, B. D. Sykes, D. S. Wishart, Nat. Struct. Biol. 1998, 5,
284–288.
[31] D. F. Mierke, H. Kessler, Biopolymers 1993, 33, 1003–1017.
[32] T. F. Havel, Prog. Biophys. Mol. Biol. 1991, 56, 43–78.
[33] A. T. Hagler, S. Lifson, P. Dauber, J. Am. Chem. Soc. 1979, 101,
5122–5130.
[34] A. L. Llamas-Saiz, G. M. Grotenbreg, M. Overhand, M. J. van Raaij,
Acta Crystallogr. Sect. D 2007, 63, 401–407.
[35] L. Wang, T. Oconnell, A. Tropsha, J. Hermans, J. Mol. Biol. 1996,
262, 283–293.
[15] A. W. Tuin, D. K. Palachanis, A. Buizert, G. M. Grotenbreg, E. Spal-
burg, A. J. de Neeling, R. H. Mars-Groenendijk, D. Noort, G. A. van
der Marel, H. S. Overkleeft, M. Overhand, Eur. J. Org. Chem. 2009,
4231–4241.
[16] A. A. Edwards, G. W. J. Fleet, G. E. Tranter, Chirality 2006, 18, 265–
272.
[17] D. R. Witty, G. W. J. Fleet, K. Vogt, F. X. Wilson, Y. Wang, R.
Storer, P. L. Myers, C. J. Wallis, Tetrahedron Lett. 1990, 31, 4787–
4790.
[18] O. Dahlman, P. J. Garegg, H. Mayer, S. Schramek, Acta Chem.
Scand. Ser. B 1986, 40, 15–20.
[36] a) V. V. Kapoerchan, A. D. Knijnenburg, M. Niamat, E. Spalburg,
A. J. de Neeling, P. H. Nibbering, R. H. Mars-Groenendijk, D.
Noort, J. M. Otero, A. L. Llamas-Saiz, M. J. Van Raaij, G. A. van
der Marel, H. S. Overkleeft, M. Overhand, Chem. Eur. J. 2010, 16,
12174–12181; b) L. H. Kondejewski, M. Jelokhani-Niaraki, S. W.
Farmer, B. Lix, C. M. Kay, B. D. Sykes, R. E. W. Hancock, R. S.
Hodges, J. Biol. Chem. 1999, 274, 13181–13192; c) D. L. Lee, J. P. S.
Powers, K. Pflegerl, M. L. Vasil, R. E. W. Hancock, R. S. Hodges, J.
Pept. Res. 2004, 63, 69–84; d) E. J. Prenner, M. Kiricsi, M. Jelokha-
ni-Niaraki, R. Lewis, R. S. Hodges, R. N. McElhaney, J. Biol. Chem.
2005, 280, 2002–2011; e) A. D. Knijnenburg, V. V. Kapoerchan, E.
Spalburg, A. J. de Neeling, R. H. Mars-Groenendijk, D. Noort, G. A.
van der Marel, H. S. Overkleeft, M. Overhand, Bioorg. Med. Chem.
2010, 18, 8403–8409.
[19] H. S. Overkleeft, S. H. L. Verhelst, E. Pieterman, W. J. Meeuwe-
noord, M. Overhand, L. H. Cohen, G. A. van der Marel, J. H. van
Boom, Tetrahedron Lett. 1999, 40, 4103–4106.
[20] P. A. M. van der Klein, W. Filemon, G. H. Veeneman, G. A. van der
Marel, J. H. van Boom, J. Carbohydr. Chem. 1992, 11, 837–848.
[21] G. C. Chi, B. I. Seo, V. Nair, Nucleosides Nucleotides Nucleic Acids
2005, 24, 481–484.
[22] C. Griesinger, R. R. Ernst, J. Magn. Reson. (1969–1992) 1987, 75,
261–271.
[37] a) E. F. Haney, S. Nathoo, H. J. Vogel, E. J. Prenner, Chem. Phys.
Lipids 2010, 163, 82–93; E. J. Prenner, R. Lewis, R. N. McElhaney,
Biochim. Biophys. Acta Biomembr. 1999, 1462, 201–221; b) Md.
Ashrafuzzaman, O. S. Andersen, R. N. McElhaney, Biochim. Bio-
Phys. Acta Biomembr. 2008, 1778, 2814–2822; c) M. Jelokhani-Niar-
aki, R. S. Hodges, J. E. Meissner, U. E. Hassenstein, L. Wheaton,
Biophys. J. 2008, 95, 3306–3321; d) S. Semrau, M. W. L. Monster, M.
van der Knaap, B. I. Florea, T. Schmidt, M. Overhand, Biochim. Bio-
phys. Acta Biomembr. 2010, 1798, 2033–2039.
[38] A. G. W. Leslie, H. R. Powell, G. Winter, O. Svensson, D. Spruce, S.
McSweeney, D. Love, S. Kinder, E. Duke, C. Nave, Acta Crystallogr.
Sect. D 2002, 58, 1924–1928.
[39] W. Kabsch, J. Appl. Crystallogr. 1993, 26, 795–800.
[40] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[41] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[23] K. Wꢇthrich, Nuclear Magnetic Resonance of Proteins and Nucleic
Acids, John Wiley & Sons, New York, 1986.
[24] A. Pardi, M. Billeter, K. Wꢇthrich, J. Mol. Biol. 1984, 180, 741–751.
[25] a) A. J. Maynard, G. J. Sharman, M. S. Searle, J. Am. Chem. Soc.
1998, 120, 1996–2007; b) S. R. Griffiths-Jones, A. J. Maynard, M. S.
Searle, J. Mol. Biol. 1999, 292, 1051–1069; c) the dHa of the amino
acid residues in GS are not significantly affected when using metha-
nol instead of water as the solvent system. For the Orn residues, the
random coil value of Lys is taken.
[26] M. Kinoshita, Y. Okamoto, F. Hirata, J. Am. Chem. Soc. 2000, 122,
2773–2779.
[27] a) R. M. van Well, L. Marinelli, C. Altona, K. Erkelens, G. Siegal,
M. van Raaij, A. L. Llamas-Saiz, H. Kessler, E. Novellino, A. Lavec-
Received: October 7, 2010
Published online: March 1, 2011
4004
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3995 – 4004