A convergent solution-phase synthesis of the macrocycle Ac-Phe-[Orn-Pro-D-Cha-Trp-Arg], a potent new antiinflammatory drug

Article abstract of DOI:10.1021/jo034228r

Relatively few cyclic peptides have reached the pharmaceutical marketplace during the past decade, most produced through fermentation rather than made synthetically. Generally, this class of compounds is synthesized for research purposes on milligram scales by solid-phase methods, but if the potential of macrocyclic peptidomimetics is to be realized, low-cost larger scale solution-phase syntheses need to be devised and optimized to provide sufficient quantities for preclinical, clinical, and commercial uses. Here, we describe a cheap, medium-scale, solution-phase synthesis of the first reported highly potent, selective, and orally active antagonist of the human C5a receptor. This compound, Ac-Phe[Orn-Pro-D-Cha-Trp-Arg], known as 3D53, is a macrocyclic peptidomimetic of the human plasma protein C5a and displays excellent antiinflammatory activity in numerous animal models of human disease. In a convergent approach, two tripeptide fragments Ac-Phe-Orn-(Boc)-Pro-OH and H-D-Cha-Trp(For)-Arg-OEt were first prepared by high-yielding solution-phase couplings using a mixed anhydride method before coupling them to give a linear hexapeptide which, after deprotection, was obtained in 38% overall yield from the commercially available amino acids. Cyclization in solution using BOP reagent gave the antagonist in 33% yield (13% overall) after HPLC purification. Significant features of the synthesis were that the Arg side chain was left unprotected throughout, the component Boc-D-Cha-OH was obtained very efficiently via hydrogenation of D-Phe with PtO₂ in TFA/water, the tripeptides were coupled at the Pro-Cha junction to minimize racemization via the oxazolone pathway, and the entire synthesis was carried out without purification of any intermediates. The target cyclic product was purified (> 97%) by reversed-phase HPLC. This convergent synthesis with minimal use of protecting groups allowed batches of 50-100 g to be prepared efficiently in high yield using standard laboratory equipment. This type of procedure should be useful for making even larger quantities of this and other macrocyclic peptidomimetic drugs.

Full text of DOI:10.1021/jo034228r

A Con ver gen t Solu tion -P h a se Syn th esis of th e Ma cr ocycle
Ac-P h e-[Or n -P r o-D-Ch a -Tr p -Ar g], a P oten t New An tiin fla m m a tor y
Dr u g
Robert C. Reid,† Giovanni Abbenante,† Stephen M. Taylor,‡ and David P. Fairlie*
Institute for Molecular Bioscience, The University of Queensland, Brisbane 4072, Australia, and
Department of Physiology and Pharmacology, The University of Queensland, Brisbane 4072, Australia
d.fairlie@imb.uq.edu.au
Received February 20, 2003
Relatively few cyclic peptides have reached the pharmaceutical marketplace during the past decade,
most produced through fermentation rather than made synthetically. Generally, this class of
compounds is synthesized for research purposes on milligram scales by solid-phase methods, but
if the potential of macrocyclic peptidomimetics is to be realized, low-cost larger scale solution-
phase syntheses need to be devised and optimized to provide sufficient quantities for preclinical,
clinical, and commercial uses. Here, we describe a cheap, medium-scale, solution-phase synthesis
of the first reported highly potent, selective, and orally active antagonist of the human C5a receptor.
This compound, Ac-Phe[Orn-Pro-D-Cha-Trp-Arg], known as 3D53, is a macrocyclic peptidomimetic
of the human plasma protein C5a and displays excellent antiinflammatory activity in numerous
animal models of human disease. In a convergent approach, two tripeptide fragments Ac-Phe-Orn-
(Boc)-Pro-OH and H-D-Cha-Trp(For)-Arg-OEt were first prepared by high-yielding solution-phase
couplings using a mixed anhydride method before coupling them to give a linear hexapeptide which,
after deprotection, was obtained in 38% overall yield from the commercially available amino acids.
Cyclization in solution using BOP reagent gave the antagonist in 33% yield (13% overall) after
HPLC purification. Significant features of the synthesis were that the Arg side chain was left
unprotected throughout, the component Boc-D-Cha-OH was obtained very efficiently via hydrogena-
tion of D-Phe with PtO2 in TFA/water, the tripeptides were coupled at the Pro-Cha junction to
minimize racemization via the oxazolone pathway, and the entire synthesis was carried out without
purification of any intermediates. The target cyclic product was purified (>97%) by reversed-phase
HPLC. This convergent synthesis with minimal use of protecting groups allowed batches of 50-
100 g to be prepared efficiently in high yield using standard laboratory equipment. This type of
procedure should be useful for making even larger quantities of this and other macrocyclic
peptidomimetic drugs.
In tr od u ction
distress syndrome.2 Medical conditions that arise from
excessive complement activation are known to affect
hundreds of millions of people and represent annual
multibillion dollar pharmaceutical market opportunities
in the USA alone.3
We recently recognized the importance of a turn
conformation in the recognition of the C-terminus of C5a
by its G protein-coupled receptor. From the structure4 of
a truncated hexapeptide derivative, we derived a cyclic
antagonist 1 to stabilize the putative turn structure
which we believed to be involved in receptor binding.5
Compound 1, featuring an i f i + 4 side chain (ornithine-
δNH2) to main chain (arginine-CO2H) amide bond link-
age, was originally created as a molecular probe to
An important part of the human immune system is a
set of blood proteins termed complement. One of these
proteins, known as C5a, regulates many types of human
immune and other cells by binding to a specific receptor
on the cell surface, triggering cellular immune responses,
and release of numerous inflammatory mediators.1 How-
ever, overexpression or underregulation of C5a is impli-
cated in the pathogenesis of many immunoinflammatory
conditions, such as rheumatoid and osteoarthritis, Alzhe-
imer’s disease, cystic fibrosis, tissue graft rejection,
ischaemic heart disease, psoriasis, gingivitis, atheroscle-
rosis, lung injury, fibrosis, systemic lupus erythematosus,
reperfusion injury, and major systemic disturbances such
as septic and anaphylactic shock, burns, and major
trauma or infection that leads to adult respiratory
(2) Taylor, S. M.; Fairlie, D. P. Exp. Opin. Ther. Patents 2000, 10,
449-458.
(3) Webpage for AVANT Immunotherapeutics, Inc. Complement
Inhibitors, http://www.avantimmune.com/technology.htm
(4) Wong, A. K.; Finch, A. M.; Pierens, G. K.; Craik, D. J .; Taylor,
S. M.; Fairlie, D. P. J . Med. Chem. 1998, 41, 3417-25.
(5) Finch, A. M.; Wong, A. K.; Paczkowski, N. J .; Wadi, S. K.; Craik,
D. J .; Fairlie, D. P.; Taylor, S. M. J . Med. Chem. 1999, 42, 1965-74.
* To whom correspondence should be addressed. Tel: +61-733651268.
Fax: +61-73365 1990.
† R.C.R. and G.A. are joint first authors.
‡ The University of Queensland.
(1) Makrides, S. C. Pharmacol. Rev. 1998, 50, 59-87.
10.1021/jo034228r CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/06/2003
4464
J . Org. Chem. 2003, 68, 4464-4471

Solution-Phase Synthesis of Ac-Phe-[Orn-Pro-D-Cha-Trp-Arg]
identify key features needed for construction of a non-
peptidic drug candidate. This macrocyclic compound
proved to be the first potent, selective, and orally active
antagonist of the human C5a receptor with potent
inhibition in vitro6-9 of C5a binding to human cells and
C5a-mediated activation of neutrophils and macrophages,
chemotaxis, and cytokine release from polymorphonu-
clear leukocytes. Since it also showed potent inhibition
in vivo in many rat models of human disease, including
neutropenia/sepsis,10 arthritis,11 immune-complex dermal
inflammation,12 arthus and endotoxic shock,13 and is-
chemia-reperfusion injury,14,15 it was decided to more
extensively evaluate this compound for efficacy in vivo.
We anticipated requiring much larger quantities (50-
100 g) of 1 than could be obtained inexpensively and
rapidly by solid-phase approaches.
supressant),16 caspofungan (fungicidal),17 eptifibatide
(antithrombotic),18 dalfopristin and quinupristine (anti-
bacterials),18 atosiban (tocolytic),19 lepirudin (anticoagu-
lant),20 lanreotide and octreotide (acromegaly).21 Although
most cyclic peptides synthesized for research purposes
are made on a small scale using conventional Merrifield-
based solid-phase peptide synthesis methods, larger
quantities needed for preclinical and clinical investiga-
tions need to be obtained more cheaply. Until now, this
has usually been through fermentation, but sometimes
via solution-phase syntheses. Relatively few large-scale
solution syntheses of cyclic peptides have been previously
reported in the literature,22,23 most using a mixed-
anhydride method that appears optimal for large-scale
peptide couplings in solution. The procedure is efficient
and inexpensive and gives high yields with low racem-
ization at each step. The one report23 that dealt with an
arginine-containing peptide used the tosyl protecting
group for the arginine side chain. This required the use
of trifluoromethanesulfonic acid, a very corrosive agent,
in the final deprotection step.
Our original synthesis4,5 of 1 involved conventional
assembly of the linear hexapeptide in small quantities
by solid-phase peptide synthesis using Fmoc protocols on
Arg(Pmc)-Wang resin,24,25 followed by cyclization in solu-
tion using BOP. To scale-up the synthesis via solution
phase, we needed a plan to realize high yields from
inexpensive reagents, to minimize purification steps, and
to avoid racemization. We decided that the synthesis of
1 would be most efficient via a convergent approach,
involving synthesis and coupling of the component tri-
peptides Ac-Phe-Orn(Boc)-Pro-OH 2 and H-D-Cha-Trp-
(For)-Arg-OEt 3 to give the linear hexapeptide, which
could then be cyclized. We envisaged that coupling of the
two tripeptides by activation at the central peptidylpro-
line residue would be less prone to racemization via the
oxazolone pathway than at other sites in the peptide
sequence.26 We chose a mixed anhydride method, using
the inexpensive ethyl chloroformate as coupling reagent
where possible, and attempted a continuous synthesis
without purification of intermediates by chromatography.
Due to the low nucleophilicity of the positively charged
guanidino group and the likelihood that it would not be
acylated during peptide couplings, we decided to leave
the arginine side chain unprotected throughout the
synthesis. This strategy has the benefit of avoiding the
use of hazardous triflic acid and minimizes deprotection
A number of other cyclic peptides have entered the
marketplace as drugs, including cyclosporin (immuno-
(6) Paczkowski, N. J .; Finch, A. M.; Whitmore, J . B.; Short, A. J .;
Wong, A. K.; Monk, P. N.; Cain, S. A.; Fairlie, D. P.; Taylor, S. M. Br.
J . Pharmacol. 1999, 128, 1461-6.
(7) Woodruff, T. M.; Strachan, A. J .; Sanderson, S. D.; Monk, P. N.;
Wong, A. K.; Fairlie, D. P.; Taylor, S. M. Inflammation 2001, 25, 171-
7.
(8) Haynes, D. R.; Harkin, D. G.; Bignold, L. P.; Hutchens, M. J .;
Taylor, S. M.; Fairlie, D. P. Biochem. Pharmacol. 2000, 60, 729-33.
(9) Cain, S. A.; Woodruff, T. M.; Taylor, S. M.; Fairlie, D. P.;
Sanderson, S. D.; Monk, P. N. Biochem. Pharmacol. 2001, 61, 1571-
9.
(10) Short, A.; Wong, A. K.; Finch, A. M.; Haaima, G.; Shiels, I. A.;
Fairlie, D. P.; Taylor, S. M. Br. J . Pharmacol. 1999, 126, 551-4.
(11) Woodruff, T. M.; Strachan, A. J .; Dryburgh, N.; Shiels, I. A.;
Reid, R. C.; Fairlie, D. P.; Taylor, S. M. Arthritis Rheum. 2002, 46,
2476-85.
(12) Strachan, A. J .; Shiels, I. A.; Reid, R. C.; Fairlie, D. P.; Taylor,
S. M. Br. J . Pharmacol. 2001, 134, 1778-86.
(13) Strachan, A. J .; Woodruff, T. M.; Haaima, G.; Fairlie, D. P.;
Taylor, S. M. J . Immunol. 2000, 164, 6560-5.
(16) Bernardelli, P.; Gaudilliere, B.; Vergne, F. Ann. Rep. Med.
Chem. 2002, 37, 257-277.
(17) Gaudilliere, B.; Berna, P. Ann. Rep. Med. Chem. 2000, 35, 331-
356.
(18) Gaudilliere, B.; Bernardelli, P.; Berna, P. Ann. Rep. Med. Chem.
2001, 36, 293-318.
(19) Galatsis, P. Ann. Rep. Med. Chem. 1997, 32, 305-326.
(20) Cheng, X. M. Ann. Rep. Med. Chem. 1996, 31, 337-355.
(21) Stewart, P. M. Trends Endocrinol. Metab. 2000, 11, 128-132.
(22) Brady, S. F.; Freidinger, R. M.; Paleveda, W. J .; Colton, C. D.;
Homnick, C. F.; Whitter, W. L.; Curley, P.; Nutt, R. F.; Veber, D. F. J .
Org. Chem. 1987, 52, 764-769.
(23) Zhang, L. H.; Pesti, J . A.; Costello, T. D.; Sheeran, P. J .; Uyeda,
R.; Ma, P.; Kauffman, G. S.; Ward, R.; McMillan, J . L. J . Org. Chem.
1996, 61, 5180-5185.
(24) Chan, W. C.; White, P. D. Fmoc solid-phase peptide synthesis:
a practical approach; Oxford University Press: New York, 2000.
(25) Fields, G. B.; Noble, R. L. Int. J . Pept. Protein Res. 1990, 35,
161-214.
(26) J ones, J . The chemical synthesis of peptides; Clarendon Press:
New York: Oxford University Press: Oxford, 1991.
(14) Arumugam, T. V.; Shiels, I. A.; Strachan, A. J .; Abbenante, G.;
Fairlie, D. P.; Taylor, S. M. Kidney Int. 2003, 63, 134-42.
(15) Arumugam, T. V.; Shiels, I. A.; Woodruff, T. M.; Reid, R. C.;
Fairlie, D. P.; Taylor, S. M. J . Surg. Res. 2002, 103, 260-7.
J . Org. Chem, Vol. 68, No. 11, 2003 4465

Reid et al.
SCHEME 1
conveniently separated from neutral impurities by ex-
traction into an aqueous phase as the hydrochloride salt
and washing with ether. After basification, the amine 11
was treated with Boc2O to give 12. The C-terminal
methyl ester of 12 was hydrolyzed with NaOH, and after
acidification and extraction, the required tripeptide 2 was
obtained as a colorless brittle glass (90% overall yield)
that was easily ground to a powder that stored well.
The
second
tripeptide
H-D-Cha-Trp(For)-Arg-
OEt.HSO4 3 presented several challenges due to side-
chain functionality. The C-terminal arginine residue
traditionally requires protection of the side chain guani-
dine group, as a sulfonamide (e.g. tosyl) or perhaps by
nitration, to prevent unwanted acylation during the
several subsequent coupling reactions resulting in a
complex mixture. We were concerned however that such
protecting groups may prove difficult to remove from the
final product under mild conditions. For example, the use
of trifluoromethane sulfonic acid or HF was not consid-
ered practical for removal of the tosyl group due to the
difficulty of handling and the often protracted hydroge-
nations needed to cleave nitroarginine might not
proceed to completion or may effect reduction of the
tryptophan indole nucleus. For these reasons, we chose
to develop conditions that would allow side chain unpro-
tected H-Arg-OEt‚2HCl 17 to be employed. It was rea-
soned that the much greater basicity of the guanidine
group (pKa 13) verses amino groups (pKa 9) should permit
the desired regioselective couplings. Boc-Trp(For)-OH 16
was considered to be a suitably protected tryptophan
derivative for the initial coupling reaction. The coupling
of 16 and 17 was best achieved via the mixed anhydride
method with ethyl chloroformate using DMF as solvent.
The reaction works well if the ethyl ester of arginine (17)
is totally dissolved in the DMF solvent before addition
to the mixed anhydride of 16; otherwise, substantial acyl-
ation does occur at the guanidino side chain. Pouring the
reaction mixture into a solution of 10% KHSO4 causes
precipitation of the dipeptide 18 as a gel which can be
filtered, washed, and air-dried. To quicken the drying
process, the wet gel can also be azeotroped with 1-butanol
on a rotary evaporator to remove water, precipitated from
the oil formed by the addition of ether, and then air-dried.
The reaction was performed several times on 25-50 g
batches of Boc-Trp(For)-OH with yields typically in the
70-80% range and of >93% purity.
steps in the procedure. We now report a solution-phase
synthesis of 1 that uses cheap reagents, no purification
of intermediates, and delivers reasonable yields (12.5%
overall yield from commercially available amino acids)
of the required product in 50-100 g quantities and in
high purity. This method is suitable for the synthesis of
1 and derivatives in medium to large scale.
D-Cyclohexylalanine 14 was a reasonably expensive
amino acid if sourced commercially, and we sought to
develop improved conditions for its preparation from
D-phenylalanine 13 by hydrogenation. The use of a
solvent consisting of TFA/water 1:1 was found to be the
single key factor in improving the efficiency of the PtO2
catalyzed hydrogenation of the aromatic ring over re-
ported literature procedures.27 The improvement was
largely due to the much greater solubility of the amino
acids (13 and 14) in this solvent mixture which enabled
concentrations to be kept much higher and avoided
poisoning of the catalyst due to precipitation. There is
however literature precedence to suggest that TFA
increases the overall rate of hydrogenation as compared
Resu lts a n d Discu ssion
Syn th esis of Tr ip ep tid e F r a gm en ts 2 a n d 3. Trip-
eptide Ac-Phe-Orn(Boc)-Pro-OH 2 was prepared as shown
in Scheme 1. Boc-Orn(Cbz)-OH 4 was first coupled to
H-Pro-OMe 5. Following removal of the Boc protecting
group with TFA, the product 7 was then coupled to Boc-
Phe-OH 13 to give the tripeptide unit 8. After subsequent
removal of the Boc group, the N-terminus was acetylated
with Ac2O to give 10. It was necessary to replace the
ornithine side chain Cbz group with Boc for compatibility
with subsequent steps. This was accomplished easily by
hydrogenation to give the free amine 11 that was
(27) Schuda, P. F.; Greenlee, W. J .; Chakravarty, P. K.; Eskola, P.
J . Org. Chem. 1988, 53, 873-875.
4466 J . Org. Chem., Vol. 68, No. 11, 2003

Solution-Phase Synthesis of Ac-Phe-[Orn-Pro-D-Cha-Trp-Arg]
SCHEME 2
SCHEME 3
to HCl, sulfuric acid, or acetic acid.28 The same charge
of catalyst was used for 5 × 20 g batches of 13 giving
complete conversion to 14 within 4 h with minimal loss
of catalytic activity before recycling. D-Cyclohexylalanine
14 was converted to the Boc-derivative 15 by standard
procedures29 before coupling to H-Trp(For)-Arg-OEt 19
(Scheme 2). The coupling of dipeptide 18 and Boc-D-
cyclohexylalanine 15 was achieved by the mixed anhy-
dride method. The tripeptide product 20 could again be
purified by simply pouring the reaction mixture into a
solution of 10% KHSO4, filtering, washing, and either air-
drying or azeotroping with butanol using the same
conditions as described for the dipeptide 18. This proce-
dure gave tripeptide 20 with typical yields of 80% and
>90% purity.
found to be inefficient using the mixed anhydride method.
There remained approximately 20% of unchanged tri-
peptide 2 together with the ethyl carbamate of tripeptide
3, which was difficult to separate from the desired
product 21. An attempt to use the more hindered
analogue, isobutyl chloroformate, did not improve the
conversion significantly. Ultimately, the coupling was
achieved cleanly using BOP reagent. The fully protected
hexapeptide 21 was found to be very hygroscopic and
difficult to handle and, thus, was never isolated but
deprotected to product 22 which dries to a nonhygroscopic
powder after ether precipitation. The coupling and partial
deprotection with NaOH was repeated three times giving
product 22 in 80% yield and of >94% purity.
Assem bly of Lin ea r Hexa p ep tid e fr om F r a gm en ts
2 a n d 3. Coupling of tripeptides 2 and 3 (Scheme 3) was
(28) Beames, D. J .; Mander, L. N. Aust. J . Chem. 1974, 27, 1257-
68.
(29) Bodanszky, M.; Bodanszky, A. The practice of peptide synthesis,
2nd rev. ed.; Springer-Verlag: Berlin, New York, 1994.
Deprotection of both the C-terminal ethyl ester and the
Boc group on the ornithine side chain of 21 is required
J . Org. Chem, Vol. 68, No. 11, 2003 4467

Reid et al.
TABLE 1. Effect of Cycliza tion Con d ition s on
Ra cem iza tion
prior to the final cyclization to 1, and surprisingly, the
order in which these steps were carried out was abso-
lutely critical for success. Hydrolysis of the C-terminal
ethyl ester with concomitant loss of the formyl group from
tryptophan using NaOH must precede the removal of the
Boc group, otherwise transfer of the formyl group from
tryptophan to the ornithine side chain occurs (∼50%)
giving a byproduct 23 that cannot be cyclized. As the
molecular weight of these isomeric formylated peptides
is the same, the structure of the byproduct 23 was
confirmed by NMR spectroscopy where clear correlations
were observed between the formyl proton and the orni-
thine δ -CH2 in both the HMBC and NOESY spectra.
Removal of the Boc group from the linear hexapeptide
22 was accompanied by a considerable amount (∼10%)
of tert-butylation of the tryptophan residue, as a conse-
quence of the loss of the formyl protecting group, if TFA
was used even in the presence of scavengers such as
water and triisopropylsilane. Efficient deprotection was
achieved using a 1:1 mixture of concentrated HCl/dioxane
which gave only 3.6% of the tert-butylated product as
shown by rpHPLC. As it was not possible to retain the
protection on the indole ring, the side chains of both Trp
and Arg were not protected for the final cyclization (24
to 1, Scheme 3).
T
(°C)
diastereomer ratio
viala
baseb
of 1c (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
DIPEA (4 µL, 2 equiv)
DIPEA (10 µL, 5 equiv)
DIPEA (4 µL, 2 equiv)
DIPEA (10 µL, 5 equiv)
DIPEA (4 µL, 2 equiv)
pyridine (5 µL, 5 equiv)
NMM (6 µL, 5 equiv)
NaHCO3 (5 mg, 5 equiv)
K2CO3 (8 mg, 5 equiv)
DBU (8 µL, 5 equiv)
no base
rt
rt
0
88.9
92.9
93.5
95.6
0
-10 95.9
rt
rt
rt
rt
rt
rt
81.2 (28% conversion)
80.8
88.8
89.2
40.6
81.4 (5% conversion)
82.9 (18% conversion)
89.4
dimethylaniline (5 equiv) rt
TMEDA (2.5 equiv)
no base after 24 h
rt
rt
82.7 (12% conversion)
a
Linear hexapeptide 24 (100 mg) and BOP reagent (50 mg, 1
equiv) were dissolved in DMF (1 mL), and then 10 aliquots of 100
b
µL were taken in separate tubes. The bases in the amounts
indicated in the table were added at the temperature specified.
c After 1 h, 900 µL of 80%A/20% B was added with shaking to each
tube, and then after brief centrifugation, 5 µL was injected into
the HPLC with linear gradient 70% A/30% B to 55% A/45% B over
30 min. Retention times: linear 24, 16.2 min; cyclic 1, 26.5 min;
diastereomer, 28.1 min.
Cycliza tion of th e Hexa p ep tid e. The cyclization
requires the activation and coupling of a peptidyl-Arg
residue (unlike a Boc-Arg residue) and is expected to
proceed with some degree of racemization. Considerable
effort was directed toward minimizing the amount of
racemization, because separation of diastereomers of 1
was known to be difficult. The coupling reagents BOP
and DPPA are reportedly superior to all others in terms
of suppressing racemization when cyclizing or coupling
peptide fragments, although there is clear evidence that
some racemization can take place.30,31 Table 1 sum-
marizes cyclization conditions used here and outcomes
for the preparation of 1. Clearly, the use of BOP in
combination with DIPEA, especially at low temperatures,
can limit racemization to as little as 4% as determined
by rpHPLC. Interestingly, the use of DBU as base caused
extensive epimerization in which the D-Arg containing
macrocycle predominated by 60%, suggesting that it is a
thermodynamically more stable diastereomer. DPPA
caused at least 10% racemization, and the reaction was
also 10 times slower than observed with BOP.
The cyclization of the linear hexapeptide 24 was
carried out in DMF (10-1M) at low temperature (-10 °C)
using DIPEA as the base and monitored by analytical
rpHPLC. The reaction appeared to be strikingly clean.
No trace of linear hexapeptide remained after 2 h, and
the crude product, after extraction and precipitation with
diethyl ether, always appeared to be greater than 90%
pure as indicated by gradient rpHPLC (0-90% MeCN, λ
214 nm). Although there were no low molecular weight
byproducts arising from the unprotected side chains of
Arg and Trp, there was evidence for considerable amounts
of polymeric material that did not elute from the rpHPLC
column even with 100% MeCN. To quantify this, solu-
tions of the analytically pure cyclic peptide 1 and the
crude product were accurately prepared (5 mg/mL in 50%
MeCN) and analyzed by rpHPLC under the same condi-
tions. Although both chromatograms displayed essen-
tially only 1 peak, the integrated peak area for the crude
product was only 40-60% that of the pure product,
suggesting a maximum yield of only 60% w/w could be
expected after purification and this indeed was found to
be the case.
Previous studies have suggested that linear peptides
of similar sequence to 24, and certainly the cyclic product
1, adopt turn conformations in solution that are favored
by the presence of a central proline residue and stabilized
by one or more intramolecular hydrogen bonds.5 This pre-
organization appears to greatly assist the cyclization
reaction, relative to competing polymerization, and high
dilution conditions were not essential for cyclization.
Comparable yields of cyclic product were obtained at
concentrations of 10-1 M (49%) and 10-2 M (51%) and
any benefits from further dilution were offset by the
excessive solvent consumption. Purification of the final
product 1 was achieved by preparative rpHPLC, after an
initial adsorption step to remove polymeric material, in
33% yield and >97% purity.
Con clu sion s
This work describes methods that can be utilized for
the medium-large scale synthesis of cyclic peptides,
without purification of intermediates. The convergent
synthesis described above shows that arginine side-chain
protection is not required and that in this case the simple
precipitation of intermediates provides products of suf-
ficient purity to carry out subsequent reactions efficiently.
The tripeptides were coupled at the Pro-Cha junction to
minimize racemization via the oxazolone pathway. Ad-
equate quantities of the final product (64 g) could be
purified by rp-HPLC to >97% purity in an overall yield
of 12.5% from the commercially available amino acids.
(30) Benoiton, N. L.; Lee, Y. C.; Steinaur, R.; Chen, F. M. Int. J .
Pept. Protein Res. 1992, 40, 559-66.
(31) Steinauer, R.; Chen, F. M.; Benoiton, N. L. Int. J . Pept. Protein
Res. 1989, 34, 295-8.
4468 J . Org. Chem., Vol. 68, No. 11, 2003

Solution-Phase Synthesis of Ac-Phe-[Orn-Pro-D-Cha-Trp-Arg]
Ethyl chloroformate (26 mL, 272 mmol) was added, and
stirring at -15 °C was continued for 30 min. The solution
prepared above containing H-Orn(Z)-Pro-OMe (273 mmol) in
DCM was added, and the temperature was maintained at 0
°C for 15 min then stirred at room temperature for a further
4 h. The precipitate of N-methylmorpholine hydrochloride was
filtered off and washed with THF, and the combined fractions
were evaporated. The oil residue was redissolved in ether/DCM
3:1 (1.2 L), washed with 2 M HC1 (2 × 300 mL), brine (300
mL), saturated NaHCO3 (300 mL), brine (300 mL), and dried
over MgSO4. The solvent was evaporated giving the protected
tripeptide as a clear, colorless oil (171 g, >100%): HRMS
625.3253 MH+, calcd for C33H45N4O8+ 625.3232; 1H NMR (500
MHz, DMSO-d6) multiple minor conformations were observed,
data refers to the major conformer: 8.03 (d, J ) 8.0 Hz, 1H),
7.39-7.12 (m, 10H), 7.27 (1H), 6.89 (d, J ) 8.7 Hz), 5.01 (s,
2H), 4.53 (m, 1H), 4.29 (dd, J ) 8.6, 5.2 Hz, 1H), 4.18 (m, 1H),
3.65 (m, 1H), 3.60 (s, 3H), 3.54 (m, 1H), 3.09-2.97 (m, 2H),
2.94 (dd, J ) 13.8, 4.0 Hz, 1H), 2.70 (dd, J ) 13.8, 10.5 Hz,
1H), 2.17 (m, 1H), 1.89 (m, 1H), 1.82 (m, 2H), 1.69 (m, 1H),
1.58-1.41 (m, 3H), 1.28 (s, 9H). Resonances at δ 4.23 4.03,
3.93, 3.38, 1.18, and 1.09 correspond to N-ethoxycarbonyl-Pro-
OMe, (separated later). Analytical rpHPLC tR ) 20.9 min.
This work will be of value for the synthesis of other cyclic
peptidomimetic drugs that will be discovered in the
future.
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All amino acid derivatives were
purchased from Novabiochem and used as received. DMF,
TFA, and diisopropylethylamine (DIPEA) were “peptide grade”
obtained from Auspep Pty Ltd., Australia. THF was HPLC
grade (less than 0.01% H2O). Other chemicals were obtained
from Aldrich and used as received. 1H NMR spectra (500 MHz,
δ) were recorded at 298 K in DMSO-d6 and were referenced
to the residual solvent peak δ 2.50 ppm. Assignments and
chemical shifts for overlapped signals were determined from
TOCSY and DQ-COSY spectra in each case. Standard analyti-
cal HPLC conditions were used for all compounds: 30% B to
100% B linear gradient over 20 min followed by a further 10
min at 100% B where solvent B was 90% MeCN, 10% H2O +
0.1% TFA. Solvent A was H2O + 0.1% TFA. Column was
Phenomenex Luna C18, 5 µm, 100 Å, 250 × 4.6 mm, flow rate
1 mL/min. Mass spectra were obtained using electrospray
ionization (ESI-MS) from solutions in 70% MeCN 30% H2O +
0.1% formic acid.
Ac-P h e-Or n (Z)-P r o-OMe 10. The crude oily product from
the above procedure (171 g, containing Boc-Phe-Orn(Z)-Pro-
OMe 5, 273 mmol) was treated with neat TFA (300 mL). CO2
and heat were evolved. After a homogeneous solution had been
obtained, the TFA was evaporated on a rotary evaporator at
40 °C/20 mbar; however, 120 g of TFA was retained and could
not be evaporated. The residue was dissolved in DCM (1.2 L),
cooled to 0 °C, and carefully neutralized with KOH (59 g, in
water/ice 500 mL) and finally with 10% K2CO3 solution. The
DCM layer was washed with 10% K2CO3 (200 mL), and the
aqueous layers were back-extracted with DCM (200 mL) (free
amine 9 HRMS 525.2716 MH+, calcd for C28H37N4O6+ 525.2708).
Acetic anhydride (27 mL, 286 mmol) was added, and the
solution was stirred at rt for 1 h. Mass spectrometry showed
that complete acetylation had occurred. The solution was
washed with saturated NaHCO3 (200 mL), water (200 mL),
and 1 M HCl (200 mL), dried over MgSO4, and evaporated to
give a viscous colorless clear oil (155 g >100%): HRMS
567.2814 MH+, calcd for C30H39N4O7+ 567.2813; 1H NMR (500
MHz, DMSO-d6) 8.16 (d, J ) 7.7 Hz, 1H), 8.03 (d, J ) 8.4 Hz,
1H), 7.39-7.14 (m, 10H), 7.28 (1H), 5.02 (s, 2H), 4.53 (m, 1H),
4.48 (m, 1H), 4.29 (dd, J ) 8.6, 5.3 Hz), 3.65 (m, 1H), 3.60 (s,
3H), 3.52 (m, 1H), 3.02 (m, 2H), 2.96 (dd, J ) 13.9, 4.4 Hz,
1H), 2.70 (dd, J ) 13.9, 10.0 Hz, 1H), 2.16 (m, 1H), 1.96-1.76
(m, 3H), 1.75 (s, 3H), 1.69 (m, 1H), 1.58-1.38 m). Resonances
at δ 4.23 4.03, 3.93, 3.38, 1.18 and 1.09 correspond to
N-ethoxycarbonyl-Pro-OMe (separated later). Analytical
rpHPLC tR ) 15.9 min.
Boc-Or n (Z)-P r o-OMe, 6. A solution of Boc-L-ornithine(Z)-
OH (100 g, 273 mmol) in dry THF (1 L) and N-methylmor-
pholine (36 mL, 327 mmol, 1.2 equiv) was stirred under argon
and cooled to -15 °C. Ethyl chloroformate (32 mL, 335 mmol,
1.2 equiv) was then added while the temperature was main-
tained at -15 °C. Stirring was continued for 30 min, and then
N-methylmorpholine (50 mL, 455 mmol, 1.6 equiv) was added
followed by a solution of H-pro-OMe‚HCl (63 g, 380 mmol, 1.4
equiv) in DMF (100 mL). The mixture was allowed to warm
to rt with stirring for a further 6 h. The precipitate of
N-methylmorpholine hydrochloride was filtered off and washed
with THF and the combined THF solution evaporated to
dryness. The oil residue was redissolved in ether/DCM 3:1 (1.2
L), washed with 2 M HCl (2 × 300 mL), brine (300 mL),
saturated NaHCO3 (300 mL), and brine (300 mL), and dried
over MgSO4. The solvent was evaporated giving the protected
dipeptide as a clear, colorless oil (131 g > 100%). The product
contains some N-ethoxycarbonyl-Pro-OMe but this is easily
removed at a later stage. This procedure gives the best yield
based on the ornithine derivative which is the most expensive
ingredient: HRMS 478.2556 MH+ calcd for C24H36N3O7
+
1
478.2548; H NMR (500 MHz, DMSO-d6) 7.39-7.28 (m, 5H),
7.24 (t, J ) 5.5 Hz, 1H), 6.92 (d, J ) 8.0 Hz, 1H), 5.01 (s, 2H),
4.31 (dd, J ) 8.6, 5.0 Hz, 1H), 4.15 (m, 1H), 3.66 (m, 1H), 3.58
(s, 3H), 3.52 (m, 1H), 3.06-2.93 (m, 2H), 2.17 (m, 1H), 1.90
(m, 2H), 1,80 (m, 1H), 1.59 (m, 1H), 1.64-1.40 m, 3H), 1.36
(s, 9H). Resonances at δ 4.23 4.03, 3.93, 3.38, 1.18, and 1.09
correspond to N-ethoxycarbonyl-Pro-OMe, (separated later).
Analytical rpHPLC tR ) 18.8 min.
Ac-P h e-Or n (Boc)-P r o-OMe, 12. The crude oily product
from the above procedure (155 g, containing Ac-Phe-Orn(Z)-
Pro-OMe 7 273 mmol) was dissolved in THF (650 mL) and 2
M HCl (100 mL) and hydrogenated over 10% Pd on carbon (1
g) at 35 psi and room temperature for 3 h. The catalyst was
filtered off, and water (1 L) and ether (500 mL) were added to
the filtrate and shaken. The aqueous layer was washed again
with ether (300 mL) to complete the removal of neutral
impurities such as N-ethoxycarbonyl-Pro-OMe. The aqueous/
THF solution containing Ac-Phe-Orn-Pro-OMe 11 (273 mmol)
[HRMS 433.2456, calcd for C22H33N4O5+ 433.2446] was basified
with solid K2CO3 (30 g), a solution of di-tert-butyl dicarbonate
(60 g, 275 mmol) in THF (200 mL) was added, and the solution
was stirred vigorously for 1 h. The solution was extracted with
ether/DCM 2:1 (3 × 500 mL), and the combined extracts were
washed with 1 M HCl (300 mL), brine (300 mL), saturated
NaHCO3 (300 mL), and brine (300 mL) and dried over MgSO4.
Removal of solvent gave Ac-Phe-Orn(Boc)-Pro-OMe 9 as a
colorless viscous gum 150 g >100%: HRMS 533.2974 MH+,
H-Or n (Z)-P r o-OMe, 7. The crude oily product from the
above procedure (131 g containing 273 mmol of Boc-Orn(Z)-
Pro-OMe) was treated with neat trifluoroacetic acid (250 mL)
with swirling. CO2 was evolved, and the mixture became
warm. No attempt was made to cool the mixture as it does
not mix well if cold. After about 15 min, the TFA was
evaporated on a rotary evaporator at 40 °C/20 mbar. The
residue was dissolved in DCM (1.2 L), cooled to 0 °C, and
carefully neutralized with KOH (59 g, in water/ice 500 mL)
and finally with 10% K2CO3 solution. The DCM layer was
washed with 10% K2CO3 (200 mL), and the aqueous layers
were back-extracted with DCM (200 mL). The combined DCM
layers were dried over MgSO4 and concentrated to about 500
mL. The solution was kept cool and used immediately for the
next step to avoid possible diketopiperazine formation: HRMS
378.2026 MH+ calcd for C19H28N3O5 378.2024.
+
Boc-P h e-Or n (Z)-P r o-OMe, 8. A solution of Boc-Phe-OH
(75 g, 283 mmol) and N-methylmorpholine (32 mL, 291 mmol)
in THF (1 L) was stirred under argon and cooled to -15 °C.
+
1
calcd for C27H41N4O7 533.2970; H NMR (500 MHz, DMSO-
d6) 8.14 (d, J ) 7.8 Hz, 1H), 8.02 (d, J ) 8.4 Hz, 1H), 7.28-
J . Org. Chem, Vol. 68, No. 11, 2003 4469

Reid et al.
7.15 (m, 5H), 6.80 (t, J ) 5.6 Hz, 1H), 4.52 (m, 1H), 4.47 (m,
1H), 4.29 (dd, J ) 8.5, 5.0 Hz), 3.65 (m, 1H), 3.61 (s, 3H), 3.54
(m, 1H), 3.00-2.84 (m, 2H), 2.95 (dd, J ) 13.9, 4.4 Hz, 1H),
2.69 (dd, J ) 13.9, 10.0 Hz, 1H), 2.16 (m, 1H), 1.97-1.76 (m,
3H), 1.74 (s, 3H), 1.65 (m, 1H), 1.54-1.36 m), 1.38 (s, 9H).
Analytical rpHPLC tR ) 15.2 min.
solution of H-Arg-OEt‚2HCl (22.5 g, MW ) 274, 82.11 mmol)
and N-methylmorpholine (18 mL, 2 equiv) in DMF (75 mL).
Importantly, if H-Arg-OEt is not totally solubilized in the
basic DMF solution before being added to the mixed anhydride,
coupling will also occur at the arginine side chain giving a
substantial amount of side product! The reaction was stirred
for 2 h allowing it to come to room temperature. This solution
was subsequently poured into 10% KHSO4 (500 mL) with
vigorous stirring. A gel slowly precipitates out of solution. The
solution is stirred for a further 10 min and the gel filtered,
washed with water several times, and air-dried to give the
HSO4- salt of the dipeptide (32.64 g, 70% yield). The compound
was >93% pure by 1H NMR, ESI-MS, and rpHPLC: HRMS
517.2764 MH+, calcd for C25H37N6O6+ 517.2769; 1H NMR δ 9.64
(br s, 1H), 9.26 (br s, 2H), 8.51 (d, J ) 7.02 Hz), 8.24 (s, 1H),
8.23 (br s, 1H), 8.00 (br s, 1H), 7.96 (br m, 1H), 7.76 (d, J )
7.6 Hz, 1H), 7.34 (m, 2H), 7.04(d, J ) 8.1 Hz, 1H), 4.32 (m,
1H), 4.27 (m, 1H), 4.08 (m, 2H), 3.09 (m, 2H), 3.06 (m, 1H),
2.95 (m, 1H), 1.76 (m 1H), 1.68 (m, 1H), 1.53 (m, 2H), 1.28 (s,
9H), 1.17(t, 3H). Analytical rpHPLC tR ) 13.2 min.
H-Tr p (F or )-Ar g-OEt, 19. Boc-Trp(For)-Arg-OEt 18 was
dissolved in a mixture of 90% TFA-10% water (5 mL/gram of
dipeptide). The solution was stirred at room temperature for
15 min, the TFA was evaporated in vacuo, and the product
was precipitated with diethyl ether. The ether layer was
decanted from the solid, and the solid was washed with two
further volumes of diethyl ether to remove as much of the TFA
as possible. The solid was dried in vacuo and used directly for
the next coupling.
Ac-P h e-Or n (Boc)-P r o-OH, 2. The viscous gum from the
above procedure (150 g containing Ac-Phe-Orn(Boc)-Pro-OMe
273 mmol) was dissolved in MeOH (500 mL), and then a
solution of NaOH (12 g, 300 mmol) in water (100 mL) was
added. The solution was stirred at rt for 2 h, and the hydrolysis
was monitored periodically by mass spectrometry. The solution
was diluted with water (700 mL), washed with ether (2 × 500
mL), and then acidified to pH 3 with solid citric acid (60 g).
The mixture was extracted with ether/DCM 2:1 (3 × 500 mL),
and then the combined extracts were washed with brine (2 ×
300 mL) and dried over MgSO4. Removal of solvent in vacuo
gave Ac-Phe-Orn(Boc)-Pro-OH as a colorless glass (127 g, 90%).
The product was crushed to a dry white powder for convenient
storage. Analysis by Mass spectrometry, NMR, and HPLC
showed the product to be greater than 98% pure: HRMS
519.2847 MH+, calcd for C26H39N4O7+ 519.2813; 1H NMR (500
MHz, DMSO-d6) 12.4 (br s, 1H), 8.13 (d, J ) 8.0 Hz, 1H), 8.02
(d, J ) 8.6 Hz, 1H), 7.29-7.13 (m, 5H), 6.77 (t, J ) 5.5 Hz,
1H), 4.52 (m, 1H), 4.46 (m, 1H), 4.22 (dd, J ) 8.7, 4.6 Hz),
3.62 (m, 1H), 3.53 (m, 1H), 3.00-2.85 (m, 2H), 2.96 (dd, J )
13.8, 4.3 Hz, 1H), 2.70 (dd, J ) 13.8, 9.8 Hz, 1H), 2.13 (m,
1H), 1.95-1.78 (m, 3H), 1.74 (s, 3H), 1.66 (m, 1H), 1.54-1.38
m), 1.37 (s, 9H). Analytical rpHPLC tR ) 13.0 min. Anal. Calcd
for C26H38N4O7: C, 60.2; N, 10.8. Found: C, 58.9; N, 10.2.
Boc-D-Ch a -Tr p (F or )-Ar g-OEt, 20. Boc-Trp(For)-Arg-OEt
18 (42.00 g, 68.44 mmol) was deprotected as described above.
The deprotected solid was dissolved in DMF (70 mL) and to
the solution was added N-methylmorpholine (15 mL, 2 equiv).
Importantly, if H-Trp(For)-Arg-OEt is not totally solubilized
in the basic DMF solution before before being added to the
mixed anhydride, coupling will also occur at the arginine side
chain giving a substantial amount of side product! In a
separate flask, Boc-D-Cha-OH (18.18 g, 67.33 mmol) was
dissolved in DMF (70 mL) and to it was added N-methylmor-
pholine (9 mL, 1.2 equiv). The solution was cooled to -10 °C,
and ethyl chloroformate (6.43 mL, 67.9 mmol) was added. The
reaction mixture was stirred at -10 °C for a further 15 min,
and to it was added the solution of H-Trp(For)-Arg-OEt in 1
portion. The reaction mixture was stirred for a further 2 h,
after which time 10% KHSO4 (1 L) was added. The precipitate
was filtered off, washed several times with water, and air-
dried to give the tripeptide (40.66 g, 80% yield). The product
was >90% pure by 1H NMR, ESI-MS, and rpHPLC: HRMS
670.3935 MH+, calcd for C34H52N7O7+ 670.3923; 1H NMR δ 9.65
(br s, 1H), 9.23 (br s, 2H), 8.47 (m, 1H), 8.21 (m, 1H), 8.16 (br
s, 1H), 8.00 (br s, 1H), 7.74 (d, J ) 7.57 Hz, 1H), 7.57 (br s,
1H), 7.33 (m, 2H), 6.76 (m, 1H), 4.67 (br s, 1H), 4.26 (br s,
1H), 4.07 (m, 2H), 3.92 (m, 1H), 3.11 (m, 1H), 3.09 (m, 2H),
2.93 (m, 1H), 1.78 (m, 1H), 1.66 (m, 1H), 1.55 (m, 4H), 1.50
(m, 2H), 1.40 (m, 1H), 1.32 (s, 9H), 1.16 (t, J ) 7 Hz), 1.12-
0.91 (m, 6H), 0.681 (m, 2H); MS 670.39 MH+. Analytical
Boc-D-Cycloh exyla la n in e, 15. H-D-phenylalanine-OH (20
g, MW ) 165, 121 mmol) was dissolved in a 1:1 mixture of
deionized water/TFA (80 mL), and to the hydrogenator vessel
was added PtO2 (800 mg, 4% w/w). The vessel was heated to
60 °C at 50 psi for 4 h. The solution was decanted and filtered
from the catalyst, and the cyclohexylalanine was precipitated
out of solution by the addition of concd HCl until no more
precipitation was observed. The solid was filtered off, washed
three times with acetone, and air-dried. This procedure gave
D-cyclohexylalanine as the HCl salt (20 g, 80% yield). The
catalyst could be reused without significant reduction in
reactivity for at least 5 further 20 g batches of H-D-phenyla-
lanine-OH before discarding. The use of TFA is essential for
quick hydrogenation times. If acetic acid is used the pheny-
lalanine is not as soluble in the solution and was found to clog
the hydrogenator lines. Total reaction time in acetic acid took
2 days in one experiment.
D-Cyclohexylalanine‚HCl (36.5 g, 176 mmol) was dissolved
in a 1:1 solution of water/THF (600 mL). Potassium carbonate
(48.7 g, 352 mmol) was added, and the solution was cooled to
0 °C and Boc carbonate (46 g, 1.2 equiv, 211 mmol) was added
over 15 min, adjusting the pH to 10-11 as the addition
proceeded by adding more potassium carbonate. If the pH falls
below ∼6 the unprotected amino acid precipitates out of
solution as the zwitterion. When all the Boc carbonate was
added, the addition of a further 100 mL of water gave a
homogeneous solution. This was stirred overnight at room
temperature, and the THF was removed from the basic
solution by rotary evaporation. The basic aqueous layer was
extracted with ethyl acetate (2 × 100 mL) to remove unreacted
Boc carbonate. The water layer was acidified to pH ) 2-3 by
the addition of citric acid and extracted again with ethyl
acetate (3 × 150 mL), and the combined organic layers were
dried and evaporated. This procedure yields Boc-D-cyclohexy-
rpHPLC tR
68H104N14O18S: C, 56.8; N, 13.6; S, 2.2. Found: C, 57.1; N,
13.6; S, 2.1.
) 17.0 min. Anal. Calcd for sulfate salt
C
Ac-P h e-Or n (Boc)-P r o-D-Ch a -Tr p -Ar g-OH , 22. Boc-D-
Cha-Trp(For)-Arg-OEt‚HSO4 21 (20.14 g, MW ) 766, 26.29
mmol) was deprotected according to the same procedure used
for H-Trp(For)-Arg-OEt. The solid, after ether precipitation,
was dissolved in DMF (50 mL), and to it were added Ac-Phe-
Orn(Boc)-Pro-OH (13.07 g, 25.28 mmol), DIPEA (4 equiv, 17.2
mL), and after complete dissolution, BOP (11.18 g, 25.28
mmol). The reaction mixture was stirred overnight, to it was
added 10% KHSO4 solution (500 mL), and the aqueous layer
was extracted with butan-1-ol/ethyl acetate 1:2 (3 × 100 mL).
The combined butan-1-ol/ethyl acetate layers were extracted
with 10% KHSO4 (3 × 100 mL), saturated NaHCO3, and water
(5 × 100 mL) and evaporated to dryness. The resultant oil was
1
lalanine (48 g, 100%) as a colorless oil which was pure by H
NMR and ESI-MS: MS 272.19 MH+.
Boc-Tr p (F or )-Ar g-OEt, 18. Boc-Trp(For)-OH (25 g, MW
) 332, 75.2 mmol) was dissolved in peptide grade DMF (75
mL), and to it was added N-methylmorpholine (18 mL, 2
equiv). The solution was cooled to -10 °C, ethyl chloroformate
(7.12 mL, 75.2 mmol) was added, and the solution was stirred
for a further 10 min. To this mixed anhydride was added a
4470 J . Org. Chem., Vol. 68, No. 11, 2003

Solution-Phase Synthesis of Ac-Phe-[Orn-Pro-D-Cha-Trp-Arg]
TABLE 2. 1H NMR (500 MHz, DMSO-d 6, δ) for 1
dissolved in a 1:1 mixture of water/ethanol, to it was added
NaOH (2.0 g, 50 mmol), and the solution was stirred for 1 h.
The solution was poured into 10% KHSO4 (1 L), and the
resultant precipitate was extracted with butan-1-ol/ethyl
acetate 1:2 (3 × 100 mL). The combined butanol/ethyl acetate
layers were washed with water several times, and the solution
was evaporated in vacuo. Trituration of the oil with ether
caused the compound to precipitate as a creamy solid which
was filtered off and dried in an oven at 50 °C (20 g, 72%). The
3
residue NH J NH-CHR HR
Hâ
Hγ
others
AcPhe 8.11
8.4
7.0
4.50 2.69, 2.96
4.55 1.44, 1.64 1.22 2.76, 3.38,
7.04 (NHꢀ)
1.74 (Me), 7.23
Orn
7.96
Pro
dCha
Trp
4.57 1.63, 1.99 1.63 3.41, 3.62
4.01 1.15, 1.26 0.93 0.62, 0.69
8.17
8.41
4.6
7.0
4.23 2.96, 3.26
7.15, 7.39,
10.86 (NH)
1
product was pure (>94%) by H NMR, ESI-MS, and rpHPLC:
Arg
7.82
8.2
4.12 1.63, 1.87 1.50 3.11, 7.59 (NHꢀ)
MS 1014 MH+, 507 M2H2+; HRMS 1014.5765 MH+, calcd for
C52H76N11O10 1014.5771; 1H NMR (500 MHz, DMSO-d6) δ
+
10.76 (s, 1H), 8.27 (d, J ) 7.10 Hz, 1H), 8.13 (d, J ) 7.10 Hz,
1H), 8.04 (d, J ) 7.95 Hz, 1H), 8.03 (d, J ) 7.74 Hz, 1H), 7.82
(d, J ) 8.81 Hz, 1H), 7.62 (d, J ) 7.52 Hz, 1H), 7.52 (br s, 1H),
7.30 (d, J ) 7.74 Hz, 1H), 7.27-7.14 (m, 5H), 7.11 (d, J ) 1.93
Hz, 1H), 7.08 (t, J ) 7.52 Hz, 1H), 6.95 (t, J ) 7.52 Hz, 1H),
6.73 (br s, 1H), 4.56 (m, 1H), 4.52 (m, 1H), 4.45 (m, 1H), 4.31
(m, 1H), 4.22 (m, 1H), 3.56 (m, 2H), 3.19-3.06 (m, 3H), 2.98-
2.91 (m, 2H), 2.89 (m, 2H), 2.70 (dd, 1H, J ) 9.67, 13.75 Hz),
1.98 (m, 1H), 1.84 (m, 1H), 1.80-1.71 (m, 2H), 1.75 (s, 3H),
1.70-1.63 (m, 3H), 1.54 (m, 2H), 1.5-1.38 (m, 7H) 1.36 (s, 9H),
1.35 (m, 1H), 1.15 (t, 2H), 1.07-0.93 (m, 4H), 0.70 (m, 2H).
Analytical rpHPLC tR ) 15.8 min.
The fully protected hexapeptide before de-esterification was
hygroscopic and difficult to handle. It was found that the
partially deprotected product dries to a nonhygroscopic powder.
The reaction was repeated several times giving product in 80%
yield.
water (1 L) and TFA (10 mL) with stirring and warming at
approximately 40 °C for 15 min. The cloudy solution was
applied to a column of reversed-phase C18 silica gel (Fluka
cat. 60757) of depth 10 cm contained in a sintered glass filter
funnel of diameter 10 cm, and light vacuum was applied. The
column was eluted with a further 500 mL of 50% MeCN/50%
water, and then the combined eluent was partially evaporated
on a rotary evaporator until the product began to precipitate.
A small volume of MeCN was added sufficient to redissolve
the precipitate, and then the solution was applied in 50 mL
aliquots to a preparative HPLC column (Vydac C18, 10-15
µm, 300 Å, 50 × 250 mm) and eluted with 38% MeCN/61.9%
water/0.1% TFA at 70 mL/min with UV detection at 280 nm.
The peak containing the cyclic peptide (retention time 18-23
min) was fractionated into six separate vessels that were
analyzed for diastereomer content by analytical HPLC (ana-
lytical HPLC conditions: column: Vydac peptide & protein
C18, 300 Å 5 µm, 4.6 × 250 mm, flow rate 1 mL/min. 70%
A/30% B to 55% A/45% B over 30 min where buffer A is water
+ 0.1% TFA, buffer B is 90% MeCN/10% water + 0.1% TFA.
Retention times: linear peptide 16.2 min, cyclic product 1
26.5 min, minor diastereomer 28.1 min). Pure fractions were
combined and lyophilized giving a white powder (33 g): MS
Ac-P h e-Or n -P r o-Ch a -Tr p -Ar g-OH, 24. Ac-Phe-Orn(Boc)-
Pro-D-Cha-Trp-Arg-OH (100 g, 90 mmol) was added to a
solution of concd HCl/dioxane (200 mL) and stirred at room
temperature for 1 h. The solution was cooled to 0 °C, and to it
was slowly added a 4 M NaOH solution (∼70 mL) until the
pH of the reaction mixture was ∼7. The resultant neutral
solution was extracted with butanol/ethyl acetate 1:2 (3 × 300
mL), and the combined layers were washed with water (2 ×
50 mL). Evaporation of the solvent and trituration with ether
gave the fully unprotected linear peptide as an off-white
powder (91 g, 95%). The product is greater than 93% pure by
1H NMR, ESI-MS, and rpHPLC: MS 914.53 MH+ 457.77
M2H2+; HRMS 914.5269 MH+, calcd for C47H68N11O8+ 914.5247;
1H NMR (500 MHz, DMSO-d6) δ 10.77 (s, 1H), 8.36 (d, J )
7.6 Hz, 1H), 8.26 (d, J ) 8.06 Hz, 1H), 8.04 (d, J ) 8.17 Hz,
1H), 8.01 (d, J ) 8.28 Hz, 1H), 7.90 (d, J ) 8.50 Hz, 1H), 7.66
(br s, 2H), 7.63 (d, J ) 8.07 Hz, 1H), 7.54 (t, J ) 5.56 Hz, 1H),
7.31 (d, J ) 7.96 Hz, 1H), 7.27-7.15 (m, 5H), 7.13 (d, J ) 1.85
Hz, 1H), 7.04 (t, J ) 7.30 Hz, 1H), 6.96 (t, J ) 7.52 Hz, 1H),
4.57 (m, 1H), 4.50 (m, 2H), 4.30 (m, 2H), 4.22 (m, 1H), 3.54
(m, 2H), 3.18-3.07 (m, 3H), 2.96-2.87 (m, 2H), 2.77-2.67 (m,
3H), 2.02 (m, 1H), 1.90-1.62 (m, 7H), 1.75 (s, 3H), 1.61-1.50
(m, 8H), 1.39 (d, 1H), 1.15 (t, 2H), 1.08-0.93 (m, 4H), 0.70 (m,
2H). Analytical rpHPLC tR ) 11.7 min, peak at rt ) 13.5 is
tert-butylated material.
Ac-P h e-[Or n -P r o-D-Ch a -Tr p -Ar g], 1 (TF A Sa lt). A solu-
tion of the fully deprotected hexapeptide Ac-Phe-Orn-Pro-Cha-
Trp-Arg-OH (100 g, 105 mmol) in DMF (1 L) and diisopropy-
lethylamine (100 mL, 570 mmol, 5.5 equiv) was stirred at rt
until homogeneous and then cooled to -10 °C. BOP reagent
(solid 50 g, 113 mmol, 1.08 equiv) was added, and the solution
was stirred at -10 to -5 °C for 2 h. The DMF was evaporated,
and the residue was dissolved in 1-butanol/EtOAc 3:1 (1 L)
and washed with brine (300 mL), 2 M HCl (300 mL), and water
(2 × 300 mL). The solvent was evaporated in vacuo, and the
residue was triturated with ether giving a pale cream solid
98 g. The solid product was filtered off, washed on the filter
with ether, and dried under high vacuum. The crude product
dries to a nonhygroscopic powder. Analysis of the crude cyclic
peptide by analytical HPLC shows the desired product together
with a minor diastereomer in the ratio 96:4. The cyclization
has been done on two batches of 100 g with similar results.
The crude product (98 g) was dissolved in 50% MeCN/50%
896.52 MH+, 448.76 M2H2+; HRMS 896.5105 MH+, calcd
+
for
C
26H39N4O7 896.5141. Anal. Calcd for TFA salt
C49H66F3N11O9: C, 58.3; N, 15.3. Found: C, 56.8; N, 15.5.
Analytical rpHPLC tR ) 14.3 min. See Table 2 for NMR data
(Table 2).
Ac-P h e-[Or n -P r o-D-Ch a -Tr p -Ar g], 1 (Aceta te Sa lt). The
crude cyclized product (98 g) was dissolved in 50% MeCN/50%
water (1 L) and glacial acetic acid (10 mL) and applied to a 10
× 10 cm precolumn of reversed-phase C18 silica gel as
described above for the TFA salt. The solution obtained was
purified by preparative HPLC using a solvent consisting of
38% MeCN/62% water, sodium acetate 6 g/L, adjusted to pH
6 with glacial acetic acid at 70 mL/min. Fractions containing
the pure cyclic peptide were combined and partially evaporated
to remove as much MeCN as possible without causing pre-
cipitation of the product. The solution (500 mL aliquots) was
applied to a preparative HPLC column (Vydac C18 300 Å, 50
× 250 mm) previously equilibrated with 1% AcOH in water
and eluted with 1% AcOH in water at 70 mL/min for 30 min
to complete the de-salting process. The product was quickly
removed from the column by elution with 50% MeCN/49%
water/1% AcOH, and the solution was lyophilized giving the
acetate salt as a white powder (31 g): MS 896.52 MH+, 448.76
M2H2+. Analytical rpHPLC tR )14.3min.
Ack n ow led gm en t. We thank Dr Martin J . Sto-
ermer for help in assigning the NMR spectra for 23. We
also thank Promics Pty Ltd. and NHMRC for financial
support.
Su p p or tin g In for m a tion Ava ila ble: Reproductions of 1H
NMR spectra and HPLC chromatograms of 1 and intermediate
peptides. This material is available free of charge via the
Internet at http://pubs.acs.org.
J O034228R
J . Org. Chem, Vol. 68, No. 11, 2003 4471