Synthesis of Cyclic Hexapeptides Based on the Antibiotic Cyclic Decapeptide Loloatin C by an in situ Indirect Cyclization Method

Article abstract of DOI:10.1002/ejoc.200300477

Three cyclic hexapeptide units based on the parent loloatin C scaffold have been identified by a 'sliding window' method as part of an expeditious SAR search for the basis of the antibiotic activity of the lolatins.Modified Fmoc-based solid-phase synthesis was used to prepare cyclic(L-valyl-l-ornithyl-D-phenylalanyl-l-asparaginyl-L-aspartyl-L-tryptophanyl) and cylic(L-valyl-L-ornithyl-l-leucyl-l-tryptophanyl-D-phenylalanyl-L-asparaginyl) in overall yields of 42 percent-47 percent. A new solution method combined with an in situ indirect cyclization was specifically developed to prepare cyclic(L-ornithyl-L-leucyl-D-tyrosyl-L-prolyl-L-tryptophanyl-D-phenylalanyl), involving cyclization of the linear peptide through the amino group in leucine, liberated selectively from the Fmoc-protected amine in situ, with the activated p-nitrophenyl ester of ornithine. The method was also effectively used for cyclization of the linear precursors of the first two cyclic hexapeptides. NOE analyses coupled with peptide backbone modelling were used to establish conformations of the target compounds. All have helix-like structures with γ-turns.

Full text of DOI:10.1002/ejoc.200300477

FULL PAPER
Synthesis of Cyclic Hexapeptides Based on the Antibiotic Cyclic Decapeptide
Loloatin C by an in situ Indirect Cyclization Method
Heru Chen,[a] Richard K. Haynes,*[a] and Jürgen Scherkenbeck*[a,b]
Keywords: Cyclic peptides / Cyclizations / Conformation analysis
Three cyclic hexapeptide units based on the parent loloatin
C scaffold have been identified by a ‘sliding window’ method
D
-phenylalanyl), involving cyclization of the linear peptide
through the amino group in leucine, liberated selectively
as part of an expeditious SAR search for the basis of the anti- from the Fmoc-protected amine in situ, with the activated
biotic activity of the loloatins. Modified Fmoc-based solid-
phase synthesis was used to prepare cyclic( -valyl-
-trypto-
p-nitrophenyl ester of ornithine. The method was also effec-
tively used for cyclization of the linear precursors of the first
two cyclic hexapeptides. NOE analyses coupled with peptide
L
L-
ornithyl-
phanyl)
D
-phenylalanyl-
and cyclic(
-phenylalanyl- -asparaginyl) in overall yields of
L
-asparaginyl-
L
-aspartyl-
L
L
-valyl- -ornithyl-
L
L-leucyl-L-trypto- backbone modelling were used to establish conformations of
phanyl-
D
L
the target compounds. All have helix-like structures with γ-
42%−47%. A new solution method combined with an in situ
indirect cyclization was specifically developed to prepare
turns.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
cyclic(
L
-ornithyl-
L
-leucyl-
D
-tyrosyl-
L
-prolyl-
L
-tryptophanyl-
Introduction
oxygens, thereby forming a hydrophilic surface (Figure 1).
This apparent amphiphilicity, due to lipophilic side chains
projecting on one side and hydrophobic chains projecting
from the other, is reminiscent of the ‘sidedness’ displayed
by gramicidin S, and it is this feature that appears to play
a role in maintaining high levels of antibiotic activity in
analogues of gramicidin S.[5] The synthesis and structural
examination of analogues of loloatin C are therefore re-
quired.
The problem of bacterial resistance to antibiotics is a
major issue in human health today,[1] and the construction
of new therapeutic agents to combat micro-organisms re-
sistant to traditional antibiotics is an urgent task. From
previously reported bioactivity data, loloatin C (1) serves as
one possible template for this task. It was originally iso-
lated, together with other cyclic decapeptides, from labora-
tory cultures of a tropical marine bacterium collected from
the Great Barrier Reef off the southern coast of Papua New
Guinea, and was reported to display potent antibiotic ac-
tivity against serial strains of Gram-positive and Gram-
negative bacteria.[2] We have synthesized this compound[3]
and have established its solution conformation by use of
NMR and CD techniques coupled with molecular simu-
lation.[4] In particular, we have established that the solution-
state conformation adopted in 70:30 trifluoroethanol/water
corresponds to a dumbbell-like shape with an ‘intersection’
point at Orn2 and -Phe7. All the hydrophobic side chains
project upward on one side, forming a hydrophobic surface,
while the hydrophilic side chains of Orn2, Asp9, and Asn8
project to the other side, together with most of the carbonyl
In the first instance, we used this three-dimensional struc-
ture of loloatin C to design structurally simpler analogues.
The easiest approach to identification of the pharmaco-
phore is to apply the ‘sliding window’ approach to this pep-
tide. This technique has been successfully used to predict
the secondary structures of proteins[6] and to search for gen-
omic regions with special functions.[7Ϫ9] It must be appreci-
ated that the method is not rigorous, and because of the
restrictions imposed on the overall structure by the smaller
sizes of the windowed fragments, it may actually fail in
identifying the pharmacophoric unit. In the case of loloatin
C, however, the striking dumbbell structure in the TFE/
water solvent system (70:30, Figure 1), with its ‘waist’ at the
Orn2 and -Phe7, makes the ‘sliding window’ analysis
rather more attractive. That is, the molecule is subdivided
into two hexapeptides, the first incorporating the Orn2 and
-Phe7 as the ‘southern’ face, and the other the Orn2 and
-Phe7 as the ‘northern face’; in this way, the opposite faces
are left relatively intact with respect to the parent cyclic
decapeptide. A further subdivision, as dictated by the ‘slid-
ing window’, places the Orn2 and -Phe7 at each of the
[a]
Department of Chemistry, The Hong Kong University of
Science and Technology,
Clear Water Bay, Kowloon, Hong Kong, PR China
Fax: (internat.) ϩ 852-2358-1594
E-mail: haynes@ust.hk
[b]
Research Global Chemistry Insecticides, Bayer CropScience,
Alfred-Nobel-Strasse 50, 40789 Monheim, Germany,
Fax: (internat.) ϩ 49-2173-383342
E-mail: juergen.scherkenbeck@bayercropscience.com
38
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200300477
Eur. J. Org. Chem. 2004, 38Ϫ47

Synthesis of Cyclic Hexapeptides
FULL PAPER
valyl--ornithyl--phenylalanyl--asparaginyl--aspartyl--
tryptophanyl) (SCH) and the ‘central’ cyclic hexapeptide,
cyclic(-valyl--ornithyl--leucyl--tryptophanyl--phenyl-
alanyl--asparaginyl) (CCH).
It appears likely that the ‘northern’ face will contribute
most to the bioactivity of loloatin C. There are two reasons
for this. Firstly, as discussed previously, there is a turn struc-
ture (type I β-turn in DMSO, inverse γ-turn in 70:30 TFE/
water) in this region, which may contribute to biological
activity. Secondly, almost all the aromatic residues lie in this
area, and this may be important in relation to binding of
the substrate to the active site.
We now describe the synthesis of the cyclic hexapeptides
identified by the sliding window operation.
Results and Discussion
Synthesis
Southern and Central Cyclic Hexapeptides ؊ Solid-Phase
Syntheses
The synthesis both of the southern (SCH) and of the cen-
tral (CCH) cyclic hexapeptides (Figure 2) was successfully
carried out by application of the scheme used in the syn-
thesis of the loloatins, reported earlier by our group. Modi-
fied Fmoc (fluorenylmethoxycarbonyl)[10] solid-phase pep-
tide synthesis was used. The disconnection is between Asp9
and Asn8 (Figure 2) for SCH, and between Val1 and Asn8
for CCH. Either TentaGel S RAM or Polystyrene AM
RAM were used as solid carriers (Scheme 1). (Benzotriazol-
1-yl)oxytripyrrolidinophosphonium hexafluorophosphate
(PyBOP[11]) and diisopropylethylamine (DIEA) were used
as coupling reagents. Use of a small amount (10 mol %) of
1-hydroxy-7-azabenzotriazole (HOAt)[12] enhances the
coupling yield and suppresses racemization. Piperidine
(25%) in N,N-dimethylformamide (DMF) was used to re-
move the Fmoc groups. DMF (20%) in 1,2-dichloroethane
(DCE) was used as solvent for coupling reactions; the effec-
tiveness of this has been noted previously. Alternatively,
DMF (40%) in dichloromethane could be used. A new
modification was introduced into the synthesis of CCH,
with the use of the reagent Pd(PPh3)4/PhSiH3 to remove the
allyl group. PhSiH3 is a more powerful trapping reagent
than N-methylmorpholine- (NMM-) acetic acid (HOAc),
and is superior to triethylsilane.[13,14] With 12 equivalent of
PhSiH3 and a catalytic amount of Pd(PPh3)4 (10Ϫ20
mol %), the deprotection time required for the allyl ester
was reduced to around 2 h, whilst over 12 h was required
with the usual Pd(PPh3)4/HOAc/NMM reagent. The overall
yields of the solid-phase synthesis were 47% for the sou-
thern cyclic hexapeptide (Scheme 1, a) and 42% for the cen-
tral cyclic hexapeptide (Scheme 1, b).
Figure 1. Structure of loloatin C and solution-state conformation
in trifluoroethanol/water (70:30) as established by NMR spec-
troscopy and molecular simulation (ref.[4]
)
Figure 2. ‘Sliding window’ analysis of loloatin C to provide three
cyclic hexapeptides, and retrosynthetic analysis of the cyclic hexa-
peptides
‘southern’ and ‘northern’ faces (Figure 2). Thus, by the slid-
ing window analysis, the three smaller cyclic hexapeptides Northern Cyclic Hexapeptide ؊ Solid- and Solution-Phase
generated are the ‘northern’ cyclic hexapeptide (NCH), cyc- Syntheses
lic(-ornithyl--leucyl--tyrosyl--prolyl--tryptophanyl--
The synthesis of the northern cyclic hexapeptide, cyclic(-
phenylalanyl), the ‘southern’ cyclic hexapeptide, cyclic(- ornithyl--leucyl--tyrosyl--prolyl--tryptophanyl--
Eur. J. Org. Chem. 2004, 38Ϫ47
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
39

H. Chen, R. K. Haynes, J. Scherkenbeck
FULL PAPER
zation, unlike in the previous cases. Modifications in which
the side chains of ornithyl or tyrosyl were linked to suitable
resins were unsuccessful in the ornithyl case, and only par-
tially successful in the tyrosyl case. Here, the Mitsunobu
reaction[15] was used to link the tyrosyl hydroxy group to
the Wang resin (Scheme 2); however, only 15% of the
hydroxy tags of the resins were occupied by use of this reac-
tion. Nevertheless, the modified resin was used to assemble
the linear peptide under the conditions given in Scheme 1.
Scheme 2. Modification of Wang resin
The final cyclization was between Tyr4 and Pro5 (Fig-
ure 2), with an excess of HATU and HOAt in DIEA. After
separation and purification by HPLC, however, the north-
ern hexapeptide was obtained only in an overall yield of
around 4%. The reasons for the low yield lie not only in the
low efficiency of coupling to the resin in the first step, but
also in the difficulty of carrying out the cyclization between
the C-terminal tyrosyl and the secondary N-terminal prolyl
residues due to steric effects.
The use of a solution method to assemble the linear hexa-
peptide precursor (Scheme 3) combined with an in situ in-
direct cyclization (Scheme 4) was therefore developed for
the synthesis of the northern cyclic hexapeptide.
As shown in Scheme 3, the C-terminus of the first amino
acid, ornithine, was protected as the allyl ester, and side
chains were protected by tert-butyl-based protecting groups.
The indole nitrogen of tryptophan was left unprotected,
and indeed no side product involving reactions through this
centre was found. In all of the coupling reactions, PyBOP
in the presence of catalytic amount of HOAt was employed
as coupling reagent. Et2NH (25%) in THF was used to re-
move the Fmoc group. In the first three couplings, the ful-
vene-diethylamine adduct was not separated from the mix-
ture, as this did not affect the ensuing coupling reactions.
Scheme 1. Fmoc-based solid-phase synthesis of: (a) southern cyclic
hexapeptide SCH (resin ϭ Tentagel S RAM) and (b) central cyclic
hexapeptide CCH (resin ϭ polystyrene AM RAM)
phenylalanyl) (NCH), required a rather different strategy. In the next three coupling steps the peptides were separated
In the absence of an Asn- or Asp- residue in the sequence from the fulveneϪdiethylamine adducts by column chroma-
(Figure 2), no use can be made of a strategy involving a tography. Yields in each coupling step were good to excel-
side chain linked to the resin coupled with an on-resin cycli- lent, and were generally above 95%.
40
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 38Ϫ47

Synthesis of Cyclic Hexapeptides
FULL PAPER
intramolecular cyclization by displacement of the group
comprising the activated ester, to form the cyclic peptide.
Clearly, the nucleophilic reagent used to deprotect the am-
ine must not react with the activated ester.
The general idea of exposing a nucleophilic amino group
in the presence of an activated ester in a linear peptide pre-
cursor of a cyclic peptide is not new. A robust method in-
volves hydrogenolysis of a Z-protected amino group over a
heterogeneous catalyst (Pd-C) in the presence of a penta-
fluorophenyl ester. Release of the amino group results in
cyclization in situ, which is reliant on the presence of the
Pd-catalyst surface, upon which the amino group is ad-
sorbed upon its release.[19] A second related method that
has been used for construction of cyclic lactams embedded
within cyclic peptides relies on the selective acid-mediated
deprotection of a 4-methoxytrityl-protected lysine primary
ε-amino group in a resin-bound linear peptide containing
either a pentafluorophenyl or p-nitrophenyl activated ester,
followed by basification of the mixture with diisopropyle-
thylamine.[20] The newly released, unhindered, highly nucle-
ophilic primary amino group reacts with the activated ester
to form a lactam. The method works very well in the forma-
tion of lactams of different ring sizes. In neither method is
there a possibility that the reagents used in the deprotection
will react with the activated ester.
Scheme 3. Assembly of the linear hexapeptide precursor and its
conversion into the northern cyclic hexapeptide NCH by the direct
cyclization method
In contrast, our method, to be conducted in solution, was
to utilise the hindered, albeit nucleophilic, secondary amine
N,N-diisopropylamine. In principle, the basic reagent N-
The execution of the intramolecular cyclization in solu- methylpyrrolidine might also be used, as it is also able to
tion was a challenge. The reaction should normally be car- remove Fmoc groups,[21] thus exposing the α-amino group
ried out under highly diluted solutions (10Ϫ3 to 10Ϫ4 mol/ in leucine. This group is relatively very hindered, however,
L) in order to inhibit the intermolecular reaction, due to since it is attached to a secondary carbon atom bearing an
the entropy problem associated with formation of the mac- isobutyl group, and so it is not possible to assess a priori
rocycle.[16]
how well the planned cyclization would proceed.
The method is shown in Scheme 4. The allyl group was
The direct cyclization method[17,18] was used to make the
northern cyclic hexapeptide (Scheme 3). It was found that removed from the linear hexapeptide by use of Pd(PPh3)4/
partial loss of the Boc group from the ornithyl δ-amino PhSiH3 as before, and the catalyst and other impurities
group took place during chain-elongation, so the peptide were removed by chromatography through a short silica gel
was treated with Boc2O/DMAP to prevent ensuing side re- column. The free C-terminus was then converted into the
actions. After the removal of the allyl group with activated p-nitrophenyl ester by use of DCC. The mixture
Pd(PPh3)4/PhSiH3, cyclization was initiated immediately was then diluted directly with 25% diisopropylamine
after removal of the Fmoc group by use of 25% Et2NH (DIPA) in THF at room temperature (11 h). The nucleo-
in THF. Deprotection of the Boc- and tert-butyl-protected philic DIPA induces deprotection of the Fmoc group, and
groups was then carried out to leave the crude product, with the release of the free amino group of leucine, intra-
which gave the target northern cyclic hexapeptide in 85% molecular cyclization is initiated in situ. Because DIPA is
yield after purification by semipreparative reversed-phase a sterically hindered nucleophile, it did not react with the
HPLC.
activated ester. The yield of the cyclization was 73%. Given
the hindered nature of the amino group, the facility of the
intramolecular nucleophilic addition-elimination involving
the activated ester to form the cyclic peptide is noteworthy;
Northern, Southern, and Central Cyclic Hexapeptides ؊
Indirect in situ Solution-Phase Syntheses
An excess of coupling reagents is usually used for cycliza- it proceeds easily at room temperature.
tion in the direct cyclization method in solution, with the
The generality of this indirect in situ cyclization was also
consequences that side reactions may compete, and purifi- demonstrated in the synthesis of the southern and central
cation is rendered difficult. To overcome these problems, we hexapeptides, in which the linear hexapeptide precursors
examined the possibility of carrying out a ‘indirect’ in situ were made by the solid-phase method (cf. Scheme 1). As-
cyclization. This involves selective nucleophile-mediated de- sembly of the linear hexapeptide was carried out as de-
protection of a protected amino group in the presence of scribed above. For the southern cyclic hexapeptide SCH,
an activated ester, and the exposed amino group undergoing the linear peptide was cleaved from the resin with Et3SiH-
Eur. J. Org. Chem. 2004, 38Ϫ47
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
41

H. Chen, R. K. Haynes, J. Scherkenbeck
FULL PAPER
Scheme 4. Synthesis of the northern cyclic hexapeptide NCH by
indirect in situ cyclization in solution
TFA in dichloromethane (Scheme 5). The free carboxyl
moiety of the aspartyl residue and the free amino group of
the ornithyl were protected. The allyl group was removed
from the protected aspartate with Pd(PPh3)4/PhSiH3, and
the liberated carboxyl group was converted into the p-nitro-
phenyl ester with DCC. Without further purification, the
mixture was brought to high dilution with 25% N-methyl-
pyrrolidine (NMP) in THF, such that removal of the Fmoc
group and cyclization was spontaneously initiated. The
yield of this cyclization method was 80%, and the sub-
sequent steps were carried out as shown in Scheme 5.
The in situ solution cyclization as applied to the central
cyclic peptide CCH was conducted as follows. The linear
peptide was cleaved from the resin with Et3SiH/TFA in di-
chloromethane (Scheme 6). The ornithyl residues were pro-
tected, and the allyl group was removed from the protected
aspartate by use of Pd(PPh3)4/PhSiH3 as above. The free
carboxyl was converted into the p-nitrophenyl ester, and the
Fmoc group was removed with NMP in THF. Again, the
remarkable efficiency of the procedure is apparent. Al-
though the yield for the cyclization reaction (74%) is slightly
less than in the above cases, this is probably due to the steric
effect caused by the isopropyl group in the cyclization be-
tween Val1 and Asn8, which is not apparent in the cycliza-
tion between Asp9 and Asn8 in the previous case.
Scheme 5. Synthesis of the southern cyclic hexapeptide SCH by
indirect in situ cyclization in solution
be established. As indicated in Figure 3, there are a series
of strong NOEs involving αN(i, iϩ1), αN(i, i) in the resi-
dues of the southern cyclic hexapeptide. In addition, three
NN(i, iϩ1)-type NOEs were unambiguously identified.
Most of the chemical shifts of the α-methine protons are
upfield (Ϫ0.1 to Ϫ0.3 ppm) of the standard values, except
for that of Val, which is close to the standard value.[25]
A
helix-like conformation combined with a γ-turn secondary
structure between Val, Orn and -Phe is therefore suggested
for the southern cyclic hexapeptide.
Important NOEs were also clearly identified in the cen-
tral cyclic hexapeptide. The characteristic inter-residue
NOEs include a series of strong NOEs for αN(i, iϩ1), and
a weak NOE for αN(6, 1), medium to strong NOEs for
αN(i, i), where the αN(1, 1), αN(2, 2), and αN(5, 5) are
Preliminary NMR Studies
Through the use of 1H, 1H-1H DQF-COSY, TOCSY, and medium, whilst the αN(3, 3), αN(4, 4), αN(6, 6) are strong,
NOESY experiments it was possible to assign all the spin medium to weak inter-residues NN(i, iϩ1) NOEs, where
systems of the three cyclic hexapeptides. Table 1 gives the NN(2, 3) and NN(1, 6) are medium, and NN(1, 3) is very
chemical shift data.
weak. The chemical shift of CαH for Val is 0.05 ppm down-
Through analysis of NOEs[22Ϫ24] in combination with field from the standard value, and all other chemical shift
peptide backbone modelling, preliminary conformational values are upfield of the standard values. The differences
and steric effects in the cyclic hexapeptides in DMSO could range between Ϫ0.2 to Ϫ0.3 ppm. Thus, the conformation
42
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 38Ϫ47

Synthesis of Cyclic Hexapeptides
FULL PAPER
Table 1. 1H NMR spectroscopic data (500 MHz, [D6]DMSO for
cyclic hexapeptides)
SCH
CCH
NCH
Val
NH
7.73
4.19
2.16
0.83
0.92
8.07
4.23
2.30
1.03
1.03
α-CH
β-CH
γ-CH3
γ -CH3
Orn
NH
7.71
8.32
8.10
4.18
1.80, 2.10
1.52, 1.65
2.95, 3.11
7.86, 8.46
α-CH
β-CH2
γ-CH2
δ-CH2
δ-NH
Leu
4.05
1.67, 1.48
1.32
2.70, 2.79
7.64
4.01
1.91
1.75
2.81, 2.92
7.90
NH
7.90
4.20
1.65
1.65
0.91
1.01
7.42
4.40
1.52
1.65
0.89
0.89
α-CH
β-CH2
γ-CH
δ-CH3
δ-CH3
D-Tyr
NH
8.23
α-CH
β-CH2
o-CH
m-CH
p-COH
Pro
α-CH
β-CH2
γ-CH2
δ-CH2
D-Phe
NH
4.56
2.74, 2.86
7.05
6.74
not observed
4.20
2.10
1.30
3.43
Scheme 6. Synthesis of the central cyclic hexapeptide CCH by non
direct in situ cyclization in solution
8.81
8.68
8.18
α-CH
4.49
2.82, 3.29
7.31
7.31
7.31
4.31
2.82, 3.08
7.35
7.35
7.35
4.21
2.94, 3.04
7.35
7.35
7.35
of the central cyclic hexapeptide in DMSO may be also he- β-CH2
o-CH
lix-like, with a γ-turn between Val, Orn, and Leu.
m-CH
In the case of the northern cyclic hexapeptide, fewer
p-CH
NOEs are found in the spectrum. Only three αN(i, iϩ1)
Asn
NH
NOEs are strong — these are between αN(2, 3), αN(5, 6),
and αN(1, 2). The other two are medium NOEs; these are
between αN(4, 5) and αN(6, 1), respectively. There are three
αN(i, i) NOEs, of which the αN(1, 1), αN(5, 5) is medium,
and the αN(2, 2) is weak. NOEs between amide protons are
8.70
4.59
2.72
7.05, 7.61
8.52
4.59
2.61, 2.79
7.11, 7.58
α-CH
β-CH2
NH2
Asp
NH
8.22
also observed, the NN(1, 2) and NN(2, 3) being medium, α-CH
4.52
2.72, 2.81
and the NN(5, 3) very weak. The chemical shifts of CαH
β-CH2
Trp
are, as usual, upfield relative to standard values for a coil
structure. The northern cyclic hexapeptide thus adopts a
helix-like structure, and a γ-turn structure between Leu,
Orn and -Phe.
NH
8.46
4.42
3.21
7.22
7.75
7.13
7.18
7.43
10.93
7.95
4.55
3.09
7.09
7.65
6.95
7.12
7.46
10.91
6.95
4.52
3.08
6.80
7.65
6.90
7.15
7.36
10.85
α-CH
β-CH2
2H
4H
5H
6H
7H
NH
Conclusion
We have developed an efficient new in situ cyclization
method involving the preparation of activated esters of the
free C-termini of three linear hexapeptides, followed by peptide loloatin C in very good yields. The successful out-
selective deprotection of the Fmoc-protected N-terminal come of the in situ cyclization reactions is particularly note-
amino groups with concomitant cyclization to provide three worthy in the case of the northern cyclic hexapeptide, cyc-
model cyclic hexapeptides based on the parent cyclic deca- lic(-ornithyl--leucyl--tyrosyl--prolyl--tryptophanyl--
Eur. J. Org. Chem. 2004, 38Ϫ47
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
43

H. Chen, R. K. Haynes, J. Scherkenbeck
FULL PAPER
HPLC was carried out on a Waters 717/600/2996 machine with
XterraTM RP18 (4.6 ϫ 250 mm for analysis; 7.8 ϫ 300 mm for semi-
preparative scale) by use of isocratic or gradient methanol (A)/0.1%
TFA in water (B). Mass spectra were acquired in FAB or CI modes
with a Finnigan Mat TSQ 7000 instrument. NMR spectra were
acquired either on a Varian Unity INOVA 500-MHz instrument or
on a JNM EX-400 instrument.
Southern Cyclic Hexapeptide [Cyclic(L-valyl-L-ornithyl-D-phenyl-
alanyl- -asparaginyl- -aspartyl- -tryptophanyl)] (SCH)
L
L
L
a. Solid-Phase Synthesis: The procedure was similar to that re-
ported previously.[3] The amount of TentaGel S RAM resin was
0.53 gram (capacity 0.22 mmol/g), measured mole quantity was
0.1132 mmol. PyBOP/HOAt/DIEA (0.4 mmol) were employed as
coupling reagents in each step, and 25% piperidine in DMF was
used to remove the Fmoc groups in each step. Sequential coupling
and deprotection was carried out in the sequence Fmoc-Asp(OH)-
O-All, -Fmoc-Phe-OH, Fmoc-Orn(Boc)-OH, Fmoc-Val-OH,
Fmoc-Trp(Boc)-OH, and Fmoc-Asp(tBu)-OH, and 0.3 mmol
Pd(PPh3)4/NMM/HOAc were used to remove the allyl group.
HATU/HOAt/DIEA (0.5 mmol) were used for cyclization of the
peptide. RP-HPLC purification [eluent: methanol (A), 0.1% TFA
in water (B); gradient: 25% A to 65% A within 60 minutes; flow
rate: 3.0 mL/min] gave a white solid (41.4 mg), purity 99.2%, (47%).
[α]2D5 ϭ Ϫ2.34 (96% ethanol, c ϭ 0.32). UV (96% ethanol): λmax
(ε) ϭ 219.8 (26 771.84), 283.5 (2997.57), 340.5 (800.97). FAB-MS:
1
m/z ϭ 776.2 (calcd. for M ϩ H/C38H50N9O9, 776.864). H NMR
spectroscopic data are given in Table 1.
b. Chain-Elongation by Solid-Phase Synthesis and Cyclization by
the in situ Solution Method: Chain-elongation was carried out as
described in a. above by commencing with 0.5 gram TentaGel S
RAM resin (capacity 0.22 mmol/g), measured quantity
0.109 mmol. When the chain-assembly was finished, the linear pep-
tide was cleaved from the resin by double application of TFA/
Et3SiH/dichloromethane (95:5:5). The crude linear peptide was
purified by RP-HPLC (gradient 10% to 30% methanol/ 0.1% TFA
in water within 50 minutes) to give a white solid (80.6 mg), purity
98.3% (71%), FAB-MS 1040.4 (calcd. for M ϩ H/C56H65N9O11,
1040.234). The linear peptide was dissolved in THF (5 mL),
(Boc)2O (30.6 mg, 0.14 mmol) and DMAP (1 mg) were then added,
and the solution was stirred at room temperature for 2 h. The THF
was removed under vacuum, and the residue was subjected to col-
umn chromatography over a short column of silica gel (3 ϫ 10 cm)
with CHCl3/methanol (10:1, v/v) to give the protected linear pep-
tide (65.8 mg, 82.5%). A solution of the linear peptide (65.0 mg,
0.063 mmol) in THF (5 mL) was cooled in an ice bath and then
treated sequentially with tBuOH (14.8 mg, 0.2 mmol), HOAt
(0.8 mg, 0.006 mmol), and DMAP (0.4 mg). After that, DCC
(23.9 mg, 0.106 mmol) in THF (2 mL) was added by syringe. The
resulting mixture was stirred overnight with warming to room tem-
perature. The THF was removed by evaporation under reduced
pressure, and the residue was purified by chromatography over sil-
ica gel with CHCl3/methanol (20:1) to give the linear protected pep-
tide (60.5 mg, 87%). A solution of the linear peptide (60.5 mg) in
chloroform (5 mL) was treated sequentially with Pd(PPh3)4
(11.78 mg, 0.01 mmol) and PhSiH3 (33.11 mg, 0.31 mmol). The co-
urse of the deprotection reaction was followed by TLC (chloro-
form/methanol 25:1), which indicated that the allyl group had been
completely removed after 2 h. The mixture was filtered through
silica gel to remove palladium and silicon-containing by-products,
and the chloroform was then evaporated under reduced pressure.
The residue was dissolved in THF (5 mL), and the solution was
treated sequentially with p-nitrophenol (14.19 mg, 0.102 mmol),
Figure 3. Important NOEs of the southern cyclic hexapeptide
(SCH), central cyclic hexapeptide (CCH), and northern cyclic hexa-
peptide (NCH) (s: strong; m: medium; w: weak)
phenylalanyl), given the hindered nature of the nucleophilic
leucine amino group, which undergoes the ring-forming re-
action with the activated p-nitrophenyl ester of ornithine.
Experimental Section
General Remarks. Peptide Syntheses: N,N-Dimethylformamide
˚
(DMF), peptide synthesis grade (Millipore) was dried over 4A mo-
lecular sieves. Chloroform and dichloromethane were freshly dis-
˚
tilled from CaH2 and stored over 4A molecular sieves, piperidine
was dried with KOH and distilled prior to use, and 1,2-dichloroe-
thane (DCE) was distilled from over phosphorus pentoxide prior
to use. Ethyl acetate and n-hexane were freshly distilled from anhy-
drous Na2SO4, and THF was freshly distilled from granular so-
dium. Other commercial reagents and solvents were used as re-
ceived. Fmoc-protected amino acids were obtained from BA-
CHEM Inc., (Boc)2O, DIPCI, 85% DEAD, DCC, Pd(PPh3)4,
HBTU, HATU, HOAt, and HOBt from Aldrich, PyBOP, TBTU
from Nova Biochem, TentaGel
S RAM resin (capacity:
0.22Ϫ0.25 mmol/g), Polystyrene AM RAM resin (capacity:
0.78 mmol/g) and Wang resin (1.0 mmol/g) from RAPP Polymere
Gmbh, Tübingen, Germany. All the resins used in the synthesis
were dried at room temperature in vacuo for several hours.
Instrumentation: UV spectra were acquired on a Milton ROY 3000
Spectronic spectrophotometer, and optical rotation values were de-
termined on a PerkinϪElmer 241 Polarimeter. Reversed-phase
44
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 38Ϫ47

Synthesis of Cyclic Hexapeptides
FULL PAPER
HOAt (0.7 mg, 0.005 mmol), DMAP (0.3 mg), and DCC (21.0 mg, silicon-containing by-products. The solvent was evaporated under
0.102 mmol). The resulting mixture was stirred for 2 h, and a solu- reduced pressure, and the residue was dissolved in THF (5 mL).
tion (100 mL) of 25% N-methylpyrrolidine (NMP) in THF (v/v) The solution was treated sequentially with p-nitrophenol (28.38 mg,
was then added to initiate the cyclization. The reaction was con-
tinued for 11 h. The DIPA and THF were then removed by evapor-
ation in vacuo at room temperature. The residue was dissolved in
0.2 mmol), HOAt (1.4 mg, 0.01 mmol), DMAP (0.6 mg), and
DCCI (41.26 mg, 0.2 mmol), and the resulting mixture was stirred
for 2 h. N-Methylpyrrolidine (NMP) in THF (25%, 100 mL) was
CHCl3 (50 mL) and filtered to remove DCU. The filtrate was then added to initiate the cyclization. The reaction mixture was
washed with 5% aqueous Na2CO3 and water, and the organic layer stirred for 11 h, and the NMP and THF were then removed by
was then evaporated under reduced pressure. The residue was
treated carefully with TFA/Et3SiH/dichloromethane (95:5:5, taken up in chloroform (50 mL) and filtered to remove DCU. The
10 mL), and the resulting mixture was stirred for 1.5 h. The TFA filtrate was washed with 5% aqueous Na2CO3 and then with water,
and dichloromethane were removed by evaporation under reduced and the organic layer was separated. The chloroform was removed
evaporation under vacuum at room temperature. The residue was
pressure, and the residue was purified by semipreparative RP-
HPLC [eluent: methanol (A), 0.1% TFA in water (B); gradient:
25% A to 65% A within 60 minutes; flow rate: 3.0 mL/min] to give
the cyclic hexapeptide as a white solid (31.8 mg, 80%), purity
by evaporation under reduced pressure, and TFA/Et3SiH/dichloro-
methane (95:5:5, 10 mL) was then added slowly to the residue. The
resulting mixture was stirred for 1.5 h, and the TFA and dichloro-
methane were removed by evaporation under reduced pressure. The
98.6%, FAB-MS: m/z ϭ 776.5 (calcd. for M ϩ H/C38H50N9O9, residue was purified by semipreparative RP-HPLC [eluent meth-
776.864).
anol (A), 0.1% TFA in water (B); gradient 20% A to 60% A within
60 minutes; flow rate 3.0 mL/min] to give the cyclic peptide as a
white solid (53.2 mg, 73.8%), purity 97.9%. FAB-MS: 775.2 (calcd.
for M ϩ H/C40H56N9O7, 775.944).
Central Cyclic Hexapeptide [Cyclic(L-Valyl-L-ornithyl-L-leucyl-L-
tryptophanyl-D-phenylalanyl-L-asparaginyl)] (CCH)
a. Solid-Phase Synthesis: The procedure was similar to that devel-
oped for the synthesis of loloatin C. The amount of polystyrene
AM RAM resin (capacity: 0.78 mmol/g) was 0.2047 gram, meas-
ured mole quantity 0.16 mmol. PyBOP/HOAt/DIEA (0.4 mmol)
was employed as the coupling reagent in each step, and 25% piperi-
dine in DMF was used to remove Fmoc groups in each step.
Sequential coupling and deprotection was carried out in the se-
quence Fmoc-Asp(OH)-O-All, Fmoc--Phe-OH, Fmoc- Trp(Boc)-
OH, Fmoc-Leu-OH, Fmoc-Orn(Boc)-OH, and Foc-Val-OH.
Pd(PPh3)4 (0.032 mmol) and PhSiH3 (0.96 mmol) were used to re-
move the allyl group; a reaction time of 2 h was required. HATU/
HOAt/DIEA (0.5 mmol) was used to cyclize the peptide. Purifi-
cation by RP-HPLC [eluent: methanol (A), 0.1% TFA in water (B);
gradient: 20% A to 60% A within 60 minutes; flow rate: 3.0 mL/
min] gave a white solid, purity 99.7%, 52.5 mg (42%). [α]2D5 ϭ Ϫ5.0
(96% ethanol, c ϭ 0.35). UV (96% ethanol) λmax (ε) ϭ 220.2 (19
379.84), 280.9 (2746.40), 373.1 (465.12). FAB-MS: m/z ϭ 774.3
Northern Cyclic Hexapeptide [Cyclic(
L-Ornithyl-L-leucyl-D-tyrosyl-
L
-prolyl- -tryptophanyl- -phenylalanyl)] (NCH)
L
D
Solution-Phase Synthesis of Fmoc-Leu-D-Tyr(tBu)-Pro-Trp-D-
PheOrn(Boc)-O-All
i. Fmoc-Orn(Boc)-O-All: Fmoc-Orn(Boc)-OH (0.4545 g, 1.0 mmol),
allyl alcohol (69.7 mg, 1.2 mmol), HOAt (11.66 mg, 0.1 mmol) and
DMAP (6.1 mg, 0.05 mmol) were dissolved in THF (10 mL). The
solution was cooled to 0 °C in an ice bath, and DCC (0.2476 g,
1.2 mmol) in dichloromethane solution was then added slowly by
syringe. The solution was stirred overnight, and the THF was then
removed by evaporation under reduced pressure. The residue was
dissolved in EtOAc (50 mL) and the solution was washed sequen-
tially with water (2 ϫ 15 mL), 5% aqueous NaHCO3 (2 ϫ 15 mL),
water, and saturated aqueous CaCl2. The organic layer was then
dried with anhydrous Na2SO4. After filtration, the solvent was re-
moved under reduced pressure, and the residue was purified by
flash chromatography over silica gel with 1:1 EtOAc/n-hexane to
1
(calcd. for M ϩ H/C40H56N9O7, 775.944). H NMR spectroscopic
data are given in Table 1.
give the peptide as a white solid (0.481 g, 97.3%), Rf ϭ 0.59. [α]2D5
ϭ
b. Chain-Elongation by Solid-Phase Synthesis and Cyclization by
the in situ Solution Method: Chain-elongation was carried out as
described in a. above. Polystyrene AM RAM resin (capacity:
0.78 mmol/g), 0.2 g, measured mole quantity 0.15 mmol was used.
When the chain-assembly was finished, the linear peptide was
cleaved from the resin by double treatment with TFA/Et3SiH/di-
chloromethane. The crude linear peptide was purified by semipre-
1
ϩ3.9 (c ϭ 0.74, CHCl3). H NMR (400 MHz, CDCl3): δ [ppm] ϭ
7.4Ϫ7.8 (m, 4 H, Ar-H), 7.2Ϫ7.4 (m, 4 H, Ar-H), 5.88 (1 H,
-CHϭ), 5.53 (t, H), 5.24Ϫ5.35 (m, 2 H, ϭCH2), 4.1Ϫ4.63 (m,
5 H), 3.12 (m, 2 H), 1.5Ϫ2.0 (m, 4 H), 1.4 (s, 9 H, -CH3).
ii. Fmoc-
D-Phe-Orn(Boc)-O-All: Fmoc-Orn(Boc)-O-All (0.2091 g,
0.45 mmol) was dissolved in 25% diethylamine in THF solution
parative RP-HPLC (gradient 10% to 30% methanol/0.1% TFA in (10 mL) and the system was stirred for 2 h. When analysis by TLC
water within 50 minutes) to give a white solid (113.89 mg, 73%), (silica gel; 3:7 EtOAc/n-hexane) indicated that the Fmoc group had
purity 99.3%. FAB-MS 1038.6 (calcd. for M ϩ H/C58H71N9O9, been completely removed, the THF and excess diethylamine were
1038.26). A solution of the linear peptide (113.2 mg) in THF removed by evaporation under vacuum at room temperature. The
(5 mL) was treated sequentially with (Boc)2O (48.02 mg, residue was kept under vacuum at room temperature for 2 h, and
0.22 mmol) and DMAP (1 mg), and the resulting mixture was
stirred at room temperature for 2 h. The THF was then removed
by evaporation under reduced pressure, and the residue was puri-
fied by passage through a short column of silica gel (3 ϫ 15 cm)
then dissolved in THF (2 mL) for the following reaction. Fmoc--
Phe-OH and PyBOP (1.2 equivalents each) were dissolved in THF
(10 mL), and DIEA was then added. After being stirred for 15
minutes, the mixture was cooled to 0 °C in an ice bath, and was
with CHCl3/methanol (10:1) to give the linear peptide (105.97 mg, then treated with above amine in THF solution. The ice bath was
85%). A solution of the linear peptide in CHCl3 (5 mL) was treated removed, and the temperature of the reaction mixture was allowed
sequentially with Pd(PPh3)4 (23.11 mg, 0.02 mmol) and PhSiH3 to warm up to room temperature. The mixture was stirred for an-
(60.6 mg, 0.56 mmol). The course of the ensuing reaction was other 30Ϫ60 minutes. The solvent was then removed in vacuo, and
monitored by TLC in chloroform/methanol (25:1), which indicated the residue was dissolved in chloroform (50 mL). The solution was
that the allyl group had been completely removed after 2 h. The
mixture was filtered through silica gel to remove the palladium and
washed with water (2 ϫ 15 mL), 5% aqueous NaHCO3 (3 ϫ
15 mL), and water (2 ϫ 15 mL), and then dried with anhydrous
Eur. J. Org. Chem. 2004, 38Ϫ47
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
45

H. Chen, R. K. Haynes, J. Scherkenbeck
FULL PAPER
Na2SO4. After filtration, the chloroform was evaporated under re-
duced pressure at room temperature, and the residue was purified
in THF (25%, 10 mL) to remove the Fmoc group. The details were
the same as described in v. above. The compound was purified by
by flash chromatography with chloroform/methanol (50:1 v:v), to flash chromatography with chloroform/methanol (10:1, with pH
give the protected dipeptide as a white solid (0.2873 g, 99.5%), Rf ϭ
value adjusted to 8Ϫ9 with triethylamine) to give the peptide as a
white solid (0.2183 g, 84.6%), Rf ϭ 0.35. FAB-MS: m/z ϭ 922.5
(100) [M]ϩ, 822.4 (90) [M ϩ H Ϫ Boc]ϩ; calcd. M/C51H67N7O9:
922.14. 1H NMR (400 MHz, CDCl3): δ [ppm] ϭ 8.84 (s, 1 H,
0.48. FAB-MS: m/z ϭ 642.2 (20, [M ϩ H]ϩ), 542.2 (100, [M ϩ H
1
Ϫ Boc]ϩ). H NMR (400 MHz, CDCl3): δ [ppm] ϭ 7.75 (d, 2 H,
Ar-H), 7.38Ϫ7.55 (m, 2 H, Ar-H), 7.22Ϫ7.32 (m, 9 H, Ar-H), 6.54
(br., 1 H, -NH), 5.81Ϫ5.91 (m, 1 H, -CHϭ), 5.45 (br., 1 H, -NH), NinH), 7.46Ϫ7.52 (m, 2 H, Ar-H), 7.40 (d, 1 H, -NH-), 7.17Ϫ7.39
5.22Ϫ5.32 (m, 2 H, ϭCH2), 4.58 (d, 2 H), 4.31Ϫ4.51 (m, 5 H),
(m, 7 H, Ar-H), 6.85Ϫ7.09 (m, 4 H, Ar-H), 6.62 (d, 1 H,
4.10Ϫ4.18 (m, 1 H), 3.03Ϫ3.09 (m, 4 H), 1.52Ϫ1.71 (m, 4 H), 1.44 -NinCH-), 5.88Ϫ5.95 (m, 1 H, -CHϭ), 5.25Ϫ5.36 (m, 2 H), 4.75
(s, 9 H, -CH3).
iii. Fmoc-Trp-
(br., 1 H), 4.62 (d, 4 H), 4.43Ϫ4.46 (m,1 H), 4.07Ϫ4.11 (m,1 H),
3.28 (m, 1 H), 2.98Ϫ3.25 (m, 10 H), 2.75 (d, 2 H), 2.42 (m, 1 H),
1.74Ϫ1.83 (m, 7 H), 1.40 (m, 2 H), 1.33 (s, 9 H, -CH3), 1.28 (s, 9
H, -CH3).
D-Phe-Orn(Boc)-O-All: Fmoc--Phe-Orn(Boc)-O-All
(0.2873 g, 0.45 mmol) was treated sequentially with 1.2 equivalents
each of Fmoc-Trp-OH, PyBOP, and DIEA, according to the pro-
cedure in part b above. The product was purified by flash chroma-
tography with chloroform/methanol (25:1) to give the protected tri-
peptide as a white solid, Rf ϭ 0.51. FAB-MS: m/z ϭ 828.1 (68) [M
ϩ H]ϩ, 728.0 (100) [M ϩ H Ϫ Boc]ϩ. The compound did not dis-
solve in CDCl3 and no 1H NMR spectroscopic data were obtained;
it was characterized by conversion into the pentapeptide described
in section v. below.
A stirred solution of Fmoc-Leu-OH (99.0 mg, 0.28 mmol) and
PyBOP (0.1457 g, 0.28 mmol) in THF (10 mL) was treated with
DIEA (0.05 mL) and then with a solution of the above pentapep-
tide (0.21 g, 0.228 mmol) in THF (4 mL). The reaction was con-
tinued for 2 h. A residue was obtained by the usual workup pro-
cedure, and was purified by flash chromatography with chloroform/
methanol (10:1, with pH value adjusted to 8Ϫ9 with triethylamine)
to give the protected linear hexapeptide as a white solid (0.2544 g,
88.7%), Rf ϭ 0.57. FAB-MS: m/z ϭ 1258.3 (10) [M ϩ H]ϩ, 1157.8
(100) [M ϩ H Ϫ Boc]ϩ. It was characterized by cyclization as de-
scribed below.
iv. Fmoc-Pro-Trp-D-Phe-Orn(Boc)-O-All: Fmoc-Trp--Phe-Orn-
(Boc)-O-All (0.2958 g, 0.5 mmol) was treated sequentially with 1.2
equivalents each of Fmoc-ProOH, PyBOP, and DIEA, according
to the procedure in part b above. The residue was purified by flash
chromatography with chloroform/methanol (10:1) to give the pro-
tected tetrapeptide (0.4615 g, 99.8%), Rf ϭ 0.56. FAB-MS: m/z ϭ
947.4 (40) [M ϩ Na]ϩ, 924.4 (100) [M Ϫ H]ϩ.
vii. In situ Solution Cyclization: The protected hexapeptide
Fmoc-Leu--Tyr(tBu)-Pro-Trp--Phe-Orn(Boc)-O-All (0.1258 g,
0.1 mmol) was dissolved in dry chloroform (2 mL, freshly distilled
from P2O5). Under a nitrogen atmosphere, Pd(PPh3)4 (11.56 mg,
0.01 mmol) was added to the solution, followed by PhSiH3
(0.074 mL, 0.6 mmol). The course of the reaction was monitored
by TLC (chloroform/methanol, 10:1), which indicated that the re-
action was complete after 2 h. The chloroform and some of the
PhSiH3 were partially removed by evaporation under reduced
pressure at room temperature, and the concentrated solution was
passed through a short silica gel column (10 ϫ 3 cm) to remove
palladium and silicon-containing by-products. The crude product
was isolated by evaporation of the eluate, and was used directly
without further purification. A solution of the crude product, p-
nitrophenol (16.7 mg, 0.12 mmol), HOAt (1.4 mg, 0.012 mmol),
and DMAP (0.7 mg) in THF (10 mL) was treated with DCC
(24.8 mg, 0.12 mmol) in dried THF (10 mL). The reaction mixture
was allowed to stand for 4 h, and was then filtered under nitrogen
gas. The filtrate was then diluted with 25% diisopropylamine
(DIPA) in THF (100 mL) to initiate the cyclization. Monitoring of
the cyclization by RP-HPLC indicated that the reaction was com-
plete after 11 h. The excess of DIPA and THF was removed by
evaporation under reduced pressure at room temperature, and the
residue was dissolved in chloroform (50 mL). The solution was
washed successively with water (2 ϫ 15 mL), 5% K2CO3 (2 ϫ
15 mL), and water (2 ϫ 15 mL), and the chloroform was then re-
moved under reduced pressure. The residue was purified by RP-
HPLC [eluent: methanol (A) and 0.1% TFA in water (B); gradient;
from 40% to 80% A within 60 minutes] to give the cyclized pro-
tected hexapeptide as a white solid (70.9 mg, 72.6%), purity 97.5%.
FAB-MS: m/z ϭ 978.3 (25) [M ϩ H]ϩ, 878.2 (100) [M ϩ H Ϫ
Boc]ϩ, 822.1 (65) [M ϩ 2H Ϫ Boc Ϫ tBu]ϩ; calcd. M ϩ H/
C54H72N8O9: 977.22. A solution of the compound in ethanol
(2 mL) at 0 °C was treated with TFA (8 mL). After 10 minutes, the
solution was allowed to warm to room temperature, and was then
stirred for another 1 h. The excess of TFA and ethanol were re-
moved by evaporation under vacuum at room temperature. The
v. Fmoc-D-Tyr(tBu)-Pro-Trp-D-Phe-Orn(Boc)-O-All: Fmoc-Pro-
Trp--Phe-Orn(Boc)-O-All (0.296 g, 0.32 mmol) was dissolved in
25% Et2NH in THF (10 mL). The solution was stirred for 2 h, and
excess Et2NH and THF were then removed by evaporation under
vacuum at room temperature. The residue was dissolved in chloro-
form (50 mL), and the solution was washed with water (2 ϫ
15 mL), 5% aqueous K2CO3 (2 ϫ 15 mL), and then water (2 ϫ
15 mL). The organic layer was dried with anhydrous Na2SO4. After
filtration, the chloroform was removed by evaporation under re-
duced pressure, and the residue was purified by flash chromatogra-
phy with chloroform/methanol (10:1, with pH value adjusted to
8Ϫ9 by triethylamine) to give the partially deprotected tetrapeptide
as a white solid (0.2207, 98.1%), Rf ϭ 0.29. FAB-MS: m/z ϭ 703.3
(100) [M ϩ H]ϩ, 603.2 (28) [M ϩ H-Boc]ϩ, calcd. [M ϩ H]ϩ/
C38H51N6O7: 703.86. 1H NMR (400 MHz, CDCl3): δ [ppm] ϭ
10.16 (sh, 1 H, NinH), 8.09 (d, 2 H, -NH-), 7.75 (m, 2 H, Ar-H),
7.54 (d, 1 H, -NH-), 7.35 (d, 1 H, -NH-), 6.90Ϫ7.23 (m, 7 H, Ar-
H), 6.89 (s, 1 H, -NH-), 5.84Ϫ5.92 (m, 2 H), 5.21Ϫ5.34 (m, 2 H),
4.36Ϫ4.66 (m,5 H), 3.60 (t, 1 H), 2.83Ϫ3.07 (m, 9 H), 2.58 (h, 2
H), 1.41Ϫ2.0 (m, 5 H), 1.39 (s, 9 H, -CH3).
A solution of Fmoc--Tyr(tBu)-OH (0.1654 g, 0.36 mmol) and
PyBOP (0.1873 g, 0.36 mmol) in THF (10 mL) was treated with
DIEA (0.06 mL, 0.36 mmol). After 15 minutes, the mixture was
cooled to 0 °C with an ice bath, and it was then treated dropwise
with a solution of the above tetrapeptide (0.2126 g, 0.3 mmol) in
THF (3 mL). The reaction was allowed to continue for 2 h, and the
mixture was then worked up as described above to give a residue,
which was purified by flash chromatography with chloroform/
methanol (10:1) to give the protected pentapeptide as a white solid
(0.3412 g, 99.4%), Rf ϭ 0.68. FAB-MS: m/z ϭ 1145.6 (20) [M ϩ
H]ϩ, 1044.5 (100) [M Ϫ Boc]ϩ.
vi. Fmoc-Leu-D-Tyr(tBu)-Pro-Trp-D-Phe-Orn(Boc)-O-All: The pro-
tected pentapeptide (0.3204 g, 0.28 mmol) was dissolved in Et2NH
46
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 38Ϫ47

Synthesis of Cyclic Hexapeptides
FULL PAPER
[4]
[5]
[6]
[7]
[8]
H. Chen, R. K. Haynes, J. Scherkenbeck, K. H. Sze, G. Zhu,
Eur. J. Org. Chem. 2003, preceding paper.
E. J. Prenner, A. H. Lewis, R. N. Elhaney, Biochem. Biophys.
Acta 1999, 1462, 201Ϫ221.
S. C. Schmidler, J. S. Liu, D. L. Brutlag, J. Comput. Biol. 2000,
7, 233Ϫ248.
G. A. Huttley, M. W. Smith, M. Carrington, S. J. O’Brien, Gen-
etics 1999, 152, 1711Ϫ1722.
C. Womack, W. Roth, C. Newman, J. P. Rissing, R. Lovell, D.
Haburchak, M. Essex, V. C. Bond, J. Infect. Diseases 2001,
183, 138Ϫ142.
C. A. Lozupone, R. D. Knight, L. F. Landweber, Current Bi-
ology 2001, 11, 65Ϫ74.
L. A. Carpino, G. Y. Han, J. Org. Chem. 1972, 37, 3404.
J. Coste, D. Le-Nguyen, B. Castro, Synthesis 1990, 31,
205Ϫ208.
L. A. Carpino, J. Am. Chem. Soc. 1993, 115, 4397Ϫ4398.
T. Nathalie, A. Jordi, G. Ernest, G. Francois, A. Fernando,
Tetrahedron Lett. 1997, 38, 7275Ϫ7278.
M. Dessolin, M.-G. Guillerez, N. Thieriet, F. Guibe, A. Loffet,
Tetrahedron Lett. 1995, 36, 5741Ϫ5744.
O. Mitsunobu, Synthesis 1981, 1Ϫ28.
K. D. Kopple, J. Pharm. Sci. 1972, 61, 1345Ϫ1356.
E. Schröder, K. Lübke, in: The Peptides, vols. I and II (Ed.: E.
Gross), Academic Press, New York, 1966.
G. T. Young, in: Specialist Periodical Reports, Amino Acids,
Peptides and Proteins, vols. 1Ϫ3, The Chemical Society, Lon-
don, England, 1969, 1970, and 1971.
U. Schmidt, A. Lieberknecht, H. Griessler, J. Talbiersky, J. Org.
Chem. 1982, 47, 3261Ϫ3264; U. Schmidt, J. Langner, J. Peptide
Res. 1997, 49, 67Ϫ73.
D. Tumelty, M. C. Needels, V. V. Antonenko, P. R. Bovy, in:
Peptides: Chemistry, Structure and Biology (Eds.: T. P. Kau-
maya, R. S. Hodges), Mayflower Scientific Ltd., 1996,
121Ϫ122.
residues were purified by RP-HPLC [eluent: methanol (A) and
0.1% TFA in water (B); gradient; from 30% to 70% A within 60
minutes] to give the cyclic hexapeptide as a white solid (69.2 mg),
purity 98.7%. FAB-MS: m/z ϭ 837.3 (12) [M ϩ H2O ϩ 2H]ϩ, 821.3
(100) [M ϩ H]ϩ; calcd. M ϩ H/C45H56N8O7: 820.992. [α]D20
ϭ
Ϫ45.4 (c ϭ 0.41, 96% ethanol). UV (96% ethanol): λmax (ε) ϭ 223.5
(29 305.41), 273.2 (3634.39). 1H NMR spectroscopic data are given
in Table 1.
Direct Solution Cyclization: Removal of the allyl group from
Fmoc-Leu--Tyr(tBu)-Pro-Trp--Phe-Orn(Boc)-O-All (0.1257 g,
0.1 mmol) was performed as described above. After purification
through a short column of silica gel (10 ϫ 3 cm), the residue ob-
tained after evaporation of the solvent was treated with Et2NH in
THF solution (25%, 10 mL) for 2 h to remove the Fmoc group.
Excess Et2NH and THF were removed by evaporation under re-
duced pressure at room temperature, and the residue was dried un-
der vacuum at room temperature for 4 h. The residue was then
dissolved in dried THF (100 mL) under nitrogen gas. The solution
was then treated with HOAt (0.1166 g, 1.0 mmol), HATU
(0.3803 g, 1.0 mmol) and DIEA (0.15 mL, 1.0 mmol). The resulting
mixture was allowed to stand for 11 h, and then worked up in the
usual way to give the cyclic hexapeptide as a white solid (83.4 mg,
85.4%), purity 96.8%. FAB-MS: m/z ϭ 978.3 (40) [M ϩ H]ϩ, 878.2
(100) [M ϩ H Ϫ Boc]ϩ, 822.1 (54) [M ϩ 2H Ϫ Boc Ϫ tBu]ϩ; calcd.
M ϩ H/C54H72N8O9: 977.22.
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
Acknowledgments
We express our thanks to Bayer AG for their generous financial
support of this work at the Hong Kong University of Science and
Technology. Support from the Hong Kong University of Science
and Technology, Grant HKUST DAG01/02.SC10 is also grate-
fully acknowledged.
[21]
[22]
[23]
[24]
[25]
D. Orain, J. Ellard, M. Bradley, J. Combinational Chem. 2002,
4(1), 1Ϫ16.
H. Günther, in: NMR Spectroscopy: An Introduction, John
Wiley & Sons, Ltd., New York, 1980, pp. 299Ϫ304.
J. H. Noggle, R. E. Schirmer, in: The Nuclear Overhauser Ef-
fect: Chemical Applications, Academic Press, New York, 1971.
J. K. M. Sanders, J. D. Mersh, Prog. Nucl. Magn. Resonance
1982, 15, 353.
[1]
H. C. Neu, Science 1992, 257, 1064Ϫ1073.
J. M. Gerard, P. Haden, M. T. Kelly, R. J. Anderson, J. Nat.
[2]
Prod. 1999, 62, 80Ϫ85.
J. Scherkenbeck, H. Chen, R. K. Haynes, Eur. J. Org. Chem.
[3]
M. P. Williamson, Biopolymers 1990, 29, 1423Ϫ1431.
Received July 30, 2003
2002, 2350Ϫ2355.
Eur. J. Org. Chem. 2004, 38Ϫ47
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
47