Total solid-phase synthesis of the azathiocoraline class of symmetric bicyclic peptides

Article abstract of DOI:10.1002/chem.200600815

Thiocoraline is a potent antitumor agent isolated from the marine organism Micromonospora sp. This symmetric bicyclic depsipeptide binds the minor groove of DNA. Here we report two solid-phase strategies for the syntheses of azathiocoraline and its analogues. The thioester linkage was replaced by an amide bond to improve the compound's pharmacokinetic properties. The first strategy is based on a convergent (4+4) approach, whilst the second is a stepwise synthesis, cyclizations in both approaches occurring on the solid support. These two strategies were designed to overcome problems caused by the presence of consecutive noncommercial N-methyl amino acids, to avoid epimerization during cyclization and/or fragment condensation, and to form the disulfide bridge under solid-phase conditions. The heterocyclic moiety was added in the last step of the synthesis to assist the preparation of libraries of new compounds with potential therapeutic applications.

Full text of DOI:10.1002/chem.200600815

FULL PAPER
DOI: 10.1002/chem.200600815
Total Solid-Phase Synthesis of the Azathiocoraline Class of Symmetric
Bicyclic Peptides
Nfflria Bayó-Puxan,[a] Ariadna Fernµndez,[a] Judit Tulla-Puche,[a] Estela Riego,[a]
Carmen Cuevas,[b] Mercedes lvarez,[a, c] and Fernando Albericio*[a, d]
Abstract: Thiocoraline is a potent anti-
tumor agent isolated from the marine
organism Micromonospora sp. This
symmetric bicyclic depsipeptide binds
the minor groove of DNA. Here we
report two solid-phase strategies for
the syntheses of azathiocoraline and its
analogues. The thioester linkage was
replaced by an amide bond to improve
the compoundꢀs pharmacokinetic prop-
erties. The first strategy is based on a
convergent (4+4) approach, whilst the
second is a stepwise synthesis, cycliza-
tions in both approaches occurring on
the solid support. These two strategies
were designed to overcome problems
caused by the presence of consecutive
noncommercial N-methyl amino acids,
to avoid epimerization during cycliza-
tion and/or fragment condensation, and
to form the disulfide bridge under
solid-phase conditions. The heterocyclic
moiety was added in the last step of
the synthesis to assist the preparation
of libraries of new compounds with po-
tential therapeutic applications.
Keywords: bisintercalators
· cou-
pling reagents · cyclic peptides ·
natural products · on-resin cycliza-
tion · peptides
Introduction
N
Thiocoraline is a potent new antitumor agent that has been
isolated from the marine organism Micromonospora sp.[1] It
exhibits a complex depsipeptide structure that shows biolog-
ical activity of interest, and the research efforts of many sci-
entific groups and pharmaceutical companies have conse-
quently focused on this peptide in order to improve knowl-
edge about its mode of action and to study its potential ap-
plication in cancer therapy. Thiocoraline shares several
common motifs with a family of antitumor peptide antibio-
[a] N. Bayó-Puxan, A. Fernµndez, Dr. J. Tulla-Puche, Dr. E. Riego,
Prof. Dr. M. lvarez, Prof. Dr. F. Albericio
Barcelona Biomedical Research Institute, University of Barcelona
Barcelona Science Park(PCB), 08028 Barcelona (Spain)
Fax : (+34)934-037-126
E-mail: albericio@pcb.ub.es
[b] Dr. C. Cuevas
PharmaMar, S.A., 28770-Colmenar Viejo (Spain)
[c] Prof. Dr. M. lvarez
Department of Pharmacology and Therapeutic Chemistry
University of Barcelona, 08028 Barcelona (Spain)
[d] Prof. Dr. F. Albericio
Department of Organic Chemistry, University of Barcelona
08028 Barcelona (Spain)
Fax : (+34)934-037-126
Chem. Eur. J. 2006, 12, 9001 – 9009
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9001

which may, depending on the coupling method used, require
protection. To the best of our knowledge, this study is the
first example of a solid-phase synthesis of this kind of mole-
cule.[6,7,8] The establishment of a robust solid-phase synthetic
strategy should facilitate the development of drug discovery
programmes based on these natural products.
The strategy makes provision for the following issues: in-
corporation of a heterocyclic moiety in the last step to facili-
tate efficient synthesis of analogues with distinct chromo-
phores, and a cyclization step with optimum amide bond for-
mation, taking advantage of the symmetry to simplify the
synthetic strategy and avoiding DKP formation favored by
the presence of consecutive N-methyl amino acids.
Scheme 1. Structures of thiocoraline and azathiocoraline.
The synthesis of these kinds of peptides represents a chal-
lenge because, in addition to being bicyclic (disulfide and
lactam), they contain consecutive N-methyl amino acids and
the quinaldic acid moiety. Consecutive N-methyl amino
acids can undergo internal diketopiperazine (DKP) forma-
tion[5] and, furthermore, NMe-Cys(Me) easily undergoes an
elimination side-reaction to produce didehydroalanine. Fi-
nally, the quinaldic acid moiety contains a hydroxy group,
Results and Discussion
Protection Scheme and synthesis of N-methyl cysteines: For
the azathiocoraline analogues (Scheme 2), d-2,3-diamino-
propionic acid (Dap) was used instead of d-Cys. The syn-
Scheme 2. Solid-phase strategy developed for the preparation of azathiocoraline by a (4+4) approach: a) Boc-d-Dap(Fmoc)-OH, DIEA, CH2Cl2;
G
b) MeOH; c) piperidine/DMF (1:4), piperidine/DBU/toluene/DMF (1:1:4:14); d) Fmoc-AA-OH/HATU/DIEA, DMF; e) piperidine/DMF (1:4); f) TFA/
CH2Cl2 (1:99); g) PyOAP/DIEA, CH2Cl2; h) I2, DMF; i) EDC·HCl/HOAt/DIEA, CH2Cl2 (1 mm); j) TFA/H2O (19:1); k) 3-Hydroxyquinaldic acid/
EDC·HCl/HOSu/DIEA, CH2Cl2.
9002
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 9001 – 9009

Solid-Phase Synthesis of Azathiocoralines
FULL PAPER
thetic approaches follow
a
fluorenylmethoxycarbonyl
ments, which indicated a 95–97% yield per cycle. Once the
tetrapeptide had been assembled, 2/3 of the protected pep-
tide resin was cleaved with TFA/CH2Cl2 (1:99) and, without
prior purification and with use of (7-azabenzotriazol-1-
yloxy)-tris(pyrrolidino)phosphonium hexafluorophosphate
(PyAOP) and DIEA as coupling reagents, it was coupled to
the peptide resin, from which the Fmoc group had previous-
ly been removed.[21,22] After removal of the Fmoc group, the
disulfide bridge was formed under solid-phase conditions
with I2 (5 equiv, 2.5 equiv”Acm) in DMF for 10 min (Fig-
ure 1a, b), by a method developed in our laboratory.[23] This
on-resin oxidation greatly facilitates the removal of the
excess of reagent and soluble side-products by simple filtra-
tion and extensive washing with CHCl3 to remove I2. The
cyclic peptide was cleaved from the resin and, without fur-
ther purification, the second cyclization was attempted at
1 mm concentration in CH2Cl2 (Figure 1c). Several coupling
reagents, such as N,N’-diisopropylcarbodiimide (DIPCDI)/
(Fmoc)/tert-butoxycarbonyl (Boc) strategy with use of a
chlorotrityl chloride resin (CTC), which minimizes the for-
mation of DKPs and allows cleavage of protected peptides
under very mild acid conditions.[9] The acetamidomethyl
(Acm) group, an orthogonal protecting group to the Fmoc
and Boc groups, was chosen to maskthe thiol group of the
Cys during peptide elongation and to facilitate the direct
formation of a disulfide bridge. H-NMe-Cys-OH, the precur-
sor of the two Cys-based amino acids, was synthesized by
the reduction of cyclic l-thiaproline with sodium and ammo-
nium as previously described.[6,10] H-NMe-Cys-OH was S-
methylated with MeI in a NaHCO3 solution[11] and the Acm
group was introduced by treatment with N-(hydroxymethy-
l)acetamide in a water solution with the addition of tri-
fluoroacetic acid (TFA), catalyzed with trifluoromethanesul-
fonic acid.[10,12] The Fmoc group was introduced by treat-
ment with the most reactive reagent, Fmoc-Cl.[12]
In our hands, direct methylation of Boc-Cys
OH did not workproperly.
A
hydroxybenzotriazole
(HOBt)/DIEA,
DIPCDI/HOAt/
[13]
DIEA, PyOAP/DIEA, 1-ethyl-3-(3’-dimethylamino-propyl)-
carbodiimide hydrochloride (EDC·HCl)/7-aza-hydroxyben-
zotriazole (HOAt)/DIEA, were tested. The use of the
water-soluble carbodiimide EDC·HCl and HOAt achieved a
cleaner crude product and had the additional advantage that
the reagents were removed in the aqueous workup treat-
ment. DIEA was added to neutralize the trifluoroacetate
salt and to avoid unwanted trifluoroacetylation, as in the
synthesis of Kahalalide F.[24] After removal of the Boc group
with TFA/H2O (19:1), the 3-hydroxyquinaldic acid compo-
nent was incorporated by use of EDC·HCl/hydroxysuccini-
mide (HOSu)/DIEA (Figure 2d). The use of more reactive
additives such as HOAt or HOBt in place of HOSu resulted
in more complex crude products because of over-acylation
of the hydroxy function of the heterocyclic moiety.[25] The
final product was purified (3.5% overall nonoptimized
yield) by preparative TLC with CH2Cl2/MeOH (95:5) as elu-
ents (Figure 1e). The product was characterized by HPLC
and HRMS.[26]
under solid-phase conditions with use of the 2-nitrobenzene-
sulfonyl protecting group, as described by Miller and Scan-
lan, provoked b-elimination, mainly at the NMe-Cys(Me)
residue,.[14]
The chromophore group, 3-hydroxyquinaldic acid, was
synthesized by a new efficient strategy developed from com-
mercially available 3-aminoquinoline.[15]
Convergent (4+4) approach: The first solid-phase strategy
used for the preparation of azathiocoraline followed a con-
vergent (4+4) approach (Scheme 2). Because of the sym-
metric character of the peptide, both fragments were pre-
pared from only one synthesis. Although macrolactamiza-
tion can be performed in any position, the most favorable
option appeared to be between the a-amino function of
Gly—as the Gly was not hindered—and the carboxylic acid
function of d-Dap. The a-amino function of d-Dap, the car-
boxylic component, was thus protected in the form of a car-
bamate, thereby preventing the formation of oxazolone and
minimizing epimerization.[16] In addition, the first amino
acid (d-Dap) grows through the side chain, thereby avoiding
DKP formation.
Solid-phase bicyclization approach: The second strategy
(Scheme 3), used to synthesize [NMe-Leu4, NMe-
Leu8]azathiocoraline, an analogue of azathiocoraline in
which NMe-Cys(Me) is substituted by NMe-Leu, started
with the attachment of d-Dap onto the CTC resin through
the b-amino group. This allowed the formation of the two
cycles under solid-phase conditions, taking advantage of the
pseudo-dilution phenomenon[27] associated with the solid-
phase and allowing easy removal of the coupling reagents
used for lactamization by simple filtration and washing. An
additional advantage of this strategy is that the bicyclic pep-
tide obtained after cleavage from the resin is unsymmetrical
with respect to the a-amino function of d-Dap. The a-amino
group that was anchored to the resin was therefore unpro-
tected and hence ready to be acylated, while the other was
protected by the Boc group, which can be removed either
before acylation or after the first acylation has been per-
formed. In the first case, both a-amino groups are acylated
Fragment synthesis thus began with limited anchoring of
Boc-d-Dap(Fmoc)-OH on the CTC resin (0.7 mmol of pro-
G
tected amino acid per g of resin, the functionalization being
1.5 mmolgÀ1).[17] The stepwise synthesis was carried out
through the b-amino function of d-Dap.[18] The first Fmoc
group was removed with piperidine/DMF (1:4) and addi-
tional
treatment
with
piperidine/DBU/DMF/toluene
(5:5:70:20). These conditions facilitated the complete re-
moval of this highly unreactive carbamate.[19] Couplings
were performed with 1-[bis(dimethylamino)methylene]-1H-
1,2,3-triazolo[4,5-b]pyridinium hexafluorophosphate 3-oxide
G
(HATU) and N,N-diisopropylethylamine (DIEA), which as-
sured complete acylation of the N-methylamino acids. The
coupling reaction was tested by the De Clercq test[20] and
the removal of Fmoc groups was followed by UV measure-
Chem. Eur. J. 2006, 12, 9001 – 9009
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9003

F. Albericio et al.
Figure 1. HPLC chromatograms of: a) linear octapeptide, b) after formation of disulfide bridge on solid support, c) after macrolactamization, d) crude
azathiocoraline, e) purified azathiocoraline. Reversed-phase C18 columns were used for the analysis with elution by linear gradient (over 15 min) of
0.036% TFA in CH3CN and 0.045% TFA in H2O from 1:9 to 7:3 in a–d and from 5:5 to 9:1 in e.
with the same intercalating moiety, while in the second strat-
egy two distinct moieties can be incorporated, thus introduc-
ing a new source of diversity.[28]
This scheme requires the orthogonal protection of the car-
boxylic function of d-Dap, which was achieved with the aid
Conclusion
Here we report two optimized solid-phase strategies for the
preparation of azathiocoralines, both of which overcome the
problems caused by the presence of consecutive noncom-
mercial N-methyl amino acids. These two strategies were de-
signed for addition of the heterocyclic moiety in the last
step of the synthesis to favor the preparation of libraries, to
avoid epimerization during cyclization and/or fragment con-
densation, and to form the disulfide bridge under solid-
phase conditions, thereby facilitating the removal of the
excess of I2. Whilst the first strategy takes advantages of
symmetry to provide a rapid synthesis, the second provides
an unsymmetrical compound that may be of interest for ex-
ploration of the relationship between symmetry and activity
while allowing most steps under solid-phase conditions. A
cornerstone of the two approaches is the concourse of the
distinct coupling reagents. Under solid-phase conditions,
aminium/uronium and phosphonium salts are used, thereby
facilitating the removal of excess and side-products by filtra-
tion and washing. Whilst HATU is used for the stepwise
of an allyl ester. Limited attachment of H-d-Dap(Fmoc)-
A
Oallyl,[29] to the CTC resin through the a-amino group was
also performed, and a stepwise elongation of the peptide
chain was carried out with HATU/DIEA (Figure 2a). Re-
moval of the allyl ester was accomplished with [Pd(PPh3)4]
A
(0.1 equiv) in the presence of PhSiH3 (10 equiv) as scaveng-
er.[30] The formation of the disulfide bridge was achieved as
indicated above. Cyclization on the solid support was incom-
plete after use of DIPCDI/HOBt, as shown by the ninhydrin
test, but was efficiently accomplished with DIPCDI/HOAt.
The partially protected bicyclic peptide was cleaved as in
the case of the previous peptide (Figure 2b). Finally, the
Boc group was removed, the 3-hydroxyquinaldic acid was
incorporated, and the target was purified as above (Fig-
ure 2c, d). The product was characterized by HPLC and
HRMS.[31]
9004
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 9001 – 9009

Solid-Phase Synthesis of Azathiocoralines
FULL PAPER
Scheme 3. Solid-phase strategy developed for the preparation of azathiocoraline analogues by a stepwise approach. a) H-d-Dap
CH2Cl2; b) MeOH; c) piperidine/DMF (1:4), piperidine/DBU/toluene/DMF (1:1:4:14); d) Fmoc-AA-OH/HATU/DIEA, DMF; e) piperidine/DMF (1:4);
f) Boc-d-Dap(Fmoc)-OH/HATU/DIEA, DMF; g) [Pd(PPh3)4], PhSiH3, CH2Cl2; h) I2, DMF; i) DIPCDI/HOAt, DMF; j) TFA/CH2Cl2 (1:99); k) TFA/
(Fmoc)-OAllyl, DIEA,
A
ACHTREUNG
H2O (19:1); l) 3-hydroxyquinaldic acid/EDC·HCl/HOSu/DIEA, CH2Cl2.
Figure 2. HPLC chromatograms of: a) linear octapeptide, b) after formation of the bicyclic peptide on solid support, c) after the coupling of the 3-hy-
droxyquinaldic acid, d) purified [NMe-Leu4, NMe-Leu8]azathiocoraline. Reversed-phase C18 columns were used for the analysis with elution by linear
gradient (over 15 min) of 0.036% TFA in CH3CN and 0.045% TFA in H2O from 0:1 to 1:0 in a and b, 5:5 to 10:0 in c, and from 5:5 to 9:1 in d.
Chem. Eur. J. 2006, 12, 9001 – 9009
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9005

F. Albericio et al.
ical shifts (d) are expressed in parts per million downfield from tetrame-
thylsilane. Coupling constants are expressed in Hertz.
elongation of the peptidic chain to assure complete acyla-
tion of N-methylamino acids, PyAOP is preferred for frag-
ment condensation in the first (4+4) approach because it
does not cause capping of the amino function in slow cou-
plings. For solid-phase cyclization, HOAt gave better results
than HOBt as an additive to DIPCDI. The cyclization and
incorporation of 3-hydroxyquinaldic acid was performed in
solution with EDC·HCl to allow easy removal of carbodii-
mide derivatives during the workup. While HOAt is the best
additive for cyclization, the less reactive HOSu is preferable
for the final acylation because it prevents over-incorporation
of the carboxylic acid. These state-of-the-art strategies are
valid for a broad range of thiocoraline analogues and should
contribute to the discovery of new compounds with thera-
peutic applications.
HCl·HNMe-Cys-OH:[10] (R)-(À)-Thiazolidine-4-carboxylic acid (8 g,
60.2 mmol) and sodium were added sequentially in small portions to
liquid ammonia (200 mL), thereby preserving the excess of sodium. Am-
monium chloride was then added. The solvent was removed under re-
duced pressure, and the product was dissolved in the minimum possible
amount of water and was acidified with hydrochloric acid until pH 2. The
water was removed under reduced pressure and the product was extract-
ed with methanol to afford the title compound (6.9 g, 40.1 mmol, 66%)
1
as a white solid. H NMR (D2O, 400 MHz): d = 3.93 (m, 1H; CHa), 3.03
(m, 2H; CH2b), 2.65 ppm (s, 3H; NMe); 13C NMR (D2O, 100 MHz): d =
169.9 (CO Cys), 60.7 (CHa), 32.0 (CH2b), 22.9 ppm (CH3, NMe); ES(+):
m/z: calcd for C4H9NO2S 135.0; found: 136.0 [M+H]+.
Fmoc-NMe-Cys(Me)-OH: HCl·HNMe-Cys-OH[10] (4.1 g, 23.8 mmol) was
dissolved in H2O/THF (1:1, 170 mL) and the mixture was cooled (48C).
Iodomethane (2.1 mL, 33.3 mmol) was dissolved in THF (68 mL), and
NaHCO3 (5.3 g, 71.4 mmol) in H2O (68 mL). Both solutions were sequen-
tially added dropwise to the amino acid solution. After the addition was
complete, the mixture was stirred for 4 h at 258C, the product was acidi-
fied to pH 5, and the solvent was removed under reduced pressure. The
product was then dissolved in H2O/dioxane (1:1, 80 mL) and the solution
was cooled to 48C. Fmoc-Cl (10 g, 38.8 mmol) was dissolved in dioxane
(8 mL) and the amino acid solution was added. The mixture was stirred
for 2 h at 48C and for 3 days more at 258C, the pH being maintained at
9–10. Dioxane was removed and the aqueous solution was washed with
TBME (3”50 mL), the aqueous layer was acidified with HCl until pH 2,
and the product was extracted with EtOAc (3”50 mL). The solvent was
removed under reduced pressure to give the product (1.23 g, 71% purity
Experimental Section
Materials and equipment: Protected amino acid derivatives, HOBt,
PyAOP and Fmoc-Cl were purchased from Applied Biosystems (Fra-
mingham, MA), Bachem (Bubendorf, Switzerland), Albatross (Montreal,
Canada) and NovaBiochem (Läufelfingen, Switzerland). CTC resin was a
gift from Rohm & Haas (Philadelphia, PA). DIEA, DIPCDI, piperidine,
TFA, ammonia, iodomethane, allyl bromide, 2-quinoxalinecarboxylic acid
and quinaldic acid were from Aldrich (Milwaukee, WI), and EDC· HCl
was a gift from Luxembourg Industries (Tel Aviv, Israel). DMF, CH2Cl2,
acetonitrile (HPLC grade), methanol (HPLC grade), dioxane, Et2O,
TBME and EtOAc were from SDS (Peypin, France). (R)-(À)-Thiazoli-
dine-4-carboxylic acid, trifluoromethanesulfonic acid, N-(hydroxymethy-
l)acetamide, and N-hydroxysuccinimide were obtained from Fluka
(Buchs, Switzerland). All commercial reagents and solvents were used as
received, with the exception of DMF and CH2Cl2, which were degassed
with nitrogen to remove volatile contaminants (DMF) and stored over
activated 4 Š molecular sieves (Merck, Darmstadt, Germany) (DMF) or
CaCl2 (CH2Cl2). Et2O was stored over Na.
and 7% yield), which was purified by MPLC. Analytical HPLC (tR
=
12.8 min, conditions 3:7 to 7:3, 99% purity); 1H NMR (CDCl3,
400 MHz): d = 9.1 (brs, 1H; COOH), 7.75 (m, 2H; 2”CHarom Fmoc),
7.60 (m, 2H; 2”CHarom Fmoc), 7.55 (m, 2H; 2”CHarom Fmoc), 7.34 (m,
2H; 2”CHarom Fmoc), 4.87 & 4.67 (1:2) (m, 1H; CHa), 4.46 (m, 2H; CH2
Fmoc), 4.29 & 4.22 (1:2) (m, 1H; CH Fmoc), 3.11 & 2.82 (1:2) (m, 2H;
b
CH2 ), 2.96 & 2.89 (1:2) (s, 3H; NMe), 2.12 & 1.88 ppm (1:2) (s, 3H;
SMe); 13C NMR (CDCl3, 100 MHz): d = 175.0 (CO Cys), 157.5 (CO
Fmoc), 143.9 (2”Carom Fmoc), 141.6 (COarom Fmoc), 128.0 (2”CHarom
Fmoc), 127.4 (2”CHarom Fmoc), 125.3 (2”CHarom Fmoc), 120.3 (2”
CHarom Fmoc), 68.0 (CH2 Fmoc), 58.6
(CHa), 47.3 (CH Fmoc), 33.4
U
(CH2b), 32.1 (CH3, NMe), 15.7 ppm (CH3, SMe); MALDI-TOF MS
(DHB): calcd for C20H21NO4S: 371.1; found: 372.0 [M+H]+, 393.9
[M+Na]+, 409.9 [M+K]+.
Solution reactions were performed in round-bottomed flasks. Organic
solvent extracts were dried over anhydrous MgSO4, followed by solvent
removal under reduced pressure at temperatures below 408C.
Fmoc-NMe-Cys
(Acm)-OH: HCl·HNMe-Cys-OH[10] (4.1 g, 23.8 mmol)
C
Solid-phase syntheses were performed in polypropylene syringes
(2.5 mL) fitted with polyethylene porous discs. Solvents and soluble re-
agents were removed by suction. The Fmoc group was removed with pi-
peridine/DMF (1:4, 1”1 min, 3”5 min, 1”10 min). Washings between
deprotection, coupling and final deprotection steps were performed with
DMF (5”1 min) and CH2Cl2 (5”1 min) with use of 5 mL solvent per
g resin for each wash. Peptide synthesis transformations and washes were
performed at 258C.
was dissolved in H2O (12.3 mL) and the solution was purged with Ar. N-
(Hydroxymethyl)acetamide (3.54 g, 39.7 mmol) was added, and the mix-
ture was cooled to 48C under Ar. A solution of trifluoroacetic acid in tri-
fluoromethanesulfonic acid (95:5, 41 mL) was added, the mixture was
stirred for 16 h at 258C, the product was dissolved in H2O (2% Na2CO3)/
dioxane (1:1, 80 mL), and the solution was then cooled to 48C. Fmoc-Cl
(10 g, 38.8 mmol) was dissolved in dioxane and added to the amino acid
solution, and the mixture was stirred for 2 h at 48C and for 3 days more
at 258C with the pH at 9–10. The dioxane was removed, the aqueous sol-
ution was washed with TBME (3”50 mL), the aqueous layer was acidi-
fied with HCl until pH 2, and the product was extracted with EtOAc (3”
50 mL). The solvent was removed under pressure and coevaporated with
Et2O to give a white solid (3.0 g, 71% purity, 33% yield). The product
was purified by MPLC. Analytical HPLC (tR = 8.8 min, conditions 3:7 to
7:3, 99% purity); 1H NMR (CDCl3, 400 MHz): d = 7.76 (m, 2H; 2”
CHarom Fmoc), 7.58 (m, 2H; 2”CHarom Fmoc), 7.39 (m, 2H; 2”CHarom
Fmoc), 7.29 (m, 2H; 2”CHarom Fmoc), 7.01 & 6.55 (1:3.7) (m, 1H; NH
Acm), 4.95 & 4.73 (m, 1H; CHa), 4.55, 4.25 (m, 4H; CH2 Fmoc, CH2
Acm), 4.21 (m, 1H; CH Fmoc), 3.25 & 2.9 (m, 2H; CH2b), 2.91 & 2.90
HPLC columns (Symmetry C18 reversed-phase column, 5.0 mm”
4.6 mm”150 mm) were obtained from Waters (Ireland). Analytical
HPLC was performed on a Waters instrument containing two solvent de-
livery pumps (Waters 1525), an automatic injector (Waters 717 autosam-
pler), a dual wavelength detector (Waters 2487) and a system controller
(Breeze V3.20), and on an Agilent 1100 instrument incorporating two
solvent delivery pumps (G1311A), an automatic injector (G1329A) and
a DAD (G1315B). UV detection was at 215 or 220 nm, and linear gradi-
ents of CH3CN (+0.036% TFA) to H2O (+0.045% TFA) were run at a
flow rate of 1.0 mLminÀ1 for 15 min.
IR spectra were obtained by using a Nicolet 510 FT-IR spectrophotome-
ter. MALDI-TOF and ES(+)-MS analyses of peptide samples were per-
formed on an Applied Biosystems VoyagerDE RP, by using ACH matrix,
(s, 3H; NMe), 2.02
&
1.99 ppm (s, 3H, Acm); 13C NMR (CDCl3,
100 MHz): d = 172.6 (CO Cys), 157.9 (CO Fmoc), 156.0 (CO Acm),
144.0 (2”Carom Fmoc), 141.6 (COarom Fmoc), 128.0 (2”CHarom Fmoc),
127.4 (2”CHarom Fmoc), 125.3 (2”CHarom Fmoc), 120.2 (2”CHarom
Fmoc), 68.5 (CH2 Fmoc), 58.8 (CHa), 47.4 (CH Fmoc), 42.1 (CH2 Acm),
in
a Waters Micromass ZQ spectrometer, and in an Agilent Ion
Trap 1100 Series LC/MSDTrap. 1H NMR (400 Hz) and 13C NMR
(100 MHz) spectroscopy was performed on a Varian Mercury 400. Chem-
9006
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 9001 – 9009

Solid-Phase Synthesis of Azathiocoralines
FULL PAPER
32.0 (CH2b), 31.3 (CH3, NMe), 23.1 ppm (CH3, Acm); MALDI-TOF MS
(DHB): m/z: calcd for C22H24N2O5S: 428.1; found: 429.2 [M+H]+, 451.2
[M+Na]+, 467.2 [M+K]+.
purity); MALDI-TOF MS (DHB): m/z: calcd for C37H50N6O10S2: 802.3;
found: 825.5 [M+Na]+, 841.5 [M+K]+.
Nonpurified lyophilized product was added to the unprotected peptide
resin fraction with PyOAP (22.4 mg, 43 mmol) and DIEA (22 mL,
129 mmol) in DMF. The mixture was stirred overnight at 258C. After a
positive De Clercq test without filtration, we added more PyOAP (22 mg,
43 mmol) and DIEA (22 mL, 129 mmol) to the mixture for 4 h, followed
by more PyOAP (11 mg, 22 mmol) and DIEA (11 mL, 65 mmol) for a fur-
ther 2 h until the De Clercq test was negative. We removed the Fmoc
group as described above and measured fragment coupling yields, which
were 81%. An aliquot was cleaved from the resin for analysis by HPLC-
MS. Analytical HPLC (tR = 8.3 min, conditions 1:9 to 7:3, 91% purity);
MALDI-TOF MS (DHB): m/z: calcd for C44H78N12O15S4: 1142.5; found:
1143.4 [M+H]+, 1165.4 [M+Na]+.
H-d-Dap
ACHTREUNG
pension of Boc-d-Dap
AHCTREUNG
(5 mL). The mixture was stirred at 258C until dissolution was complete,
and allyl bromide (1 mL, 11.6 mmol) was then added. The resulting mix-
ture was heated at reflux for 1 h, cooled, filtered, diluted with EtOAc
(5 mL) and finally washed with aqueous NaHCO3 (5%, 3”5 mL), HCl
(1n, 3”5 mL) and NaCl (sat, 3”5 mL). The organic layer was then con-
centrated to give a crude oil, which was purified by crystallization from
Et2O and dioxane to afford Boc-d-Dap(Fmoc)-Oallyl (128 mg,
R
0.27 mmol, 58%) as a white, amorphous powder. The Boc group was re-
moved by treatment with a solution of TFA/H2O (95:5). The mixture was
stirred for 1.5 h at 258C, the solvent was removed in vacuo, and the re-
maining acid was eliminated by co-evaporations with Et2O to afford the
title compound (97 mg, 0.26 mmol, 98%) as a white solid. 1H NMR
(CDCl3, 400 MHz): d = 7.70 (m, 2H; 2”CHarom Fmoc), 7.55 (m, 2H; 2”
CHarom Fmoc), 7.30 (m, 2H; 2”CHarom Fmoc), 7.21 (m, 2H; 2”CHarom
Fmoc), 5.88 (m, 1H; CH allyl), 5.28 (dm, J = 17.2 Hz, 1H; =CHH’
allyl), 5.18 (dm, J = 10.6 Hz, 1H; =CHH’ allyl), 4.68 (m, 1H; CHa), 4.59
(m, 1H; CH Fmoc), 4.32, 4.10 (m, 2H; CH2 allyl or CH2 Fmoc),
3.56 ppm (m, 2H; CH2b); 13C NMR (CDCl3, 100 MHz): d = 167.6 (CO
Dap), 157.6 (CO Fmoc), 143.9 (2”Carom Fmoc), 141.4 (COarom Fmoc),
130.0 (CH allyl), 128.1 (2”CHarom Fmoc), 127.4 (2”CHarom Fmoc), 125.2
(2”CHarom Fmoc), 120.1 (2”CHarom Fmoc), 67.6 (OCH2 allyl), 67.4 (CH2
{[Boc-d-Dap(&1)-Gly-MeCys(Acm)-MeCys(Me)&2][H-Gly-MeCys-
A
A
proach: CTC resin (300 mg, 1.6 mmolgÀ1) was placed in a 10 mL polypro-
pylene syringe fitted with a polyethylene filter disc. The resin was then
washed with DCM (5”1 min), and a solution of Boc-d-Dap(Fmoc)-OH
G
(90 mg, 0.7 mmol) and DIEA (244 mL, 4.7 equiv) in CH2Cl2 (2.0 mL) was
then added. After 5 min, more DIEA (122 mL, 2.3 equiv) was added and
the mixture was stirred for 55 min at 258C. The reaction was terminated
by the addition of MeOH (240 mL) and the mixture was stirred for a fur-
ther 10 min. The Boc-d-Dap(Fmoc)-O-CTC resin was subjected to the
R
following washings/treatments: filtration, CH2Cl2 (5”30 s), DMF (5”
30 s), piperidine/DMF (1:4, 1”1 min, 3”5 min, 1”10 min), piperidine/
DBU/toluene/DMF (5:5:20:70, 2”5 min). Loading, calculated by measur-
b
Fmoc), 53.2 (CH Fmoc), 47.0 (CHa), 41.0 ppm (CH2 allyl); ES(+): m/z:
calcd for C21H22N2O4: 366.2; found: 367.1 [M+H]+, 389.1 [M+Na]+.
ing absorbance at 290 nm, was 0.64 mmolgÀ1
.
{[Boc-d-Dap(&1)-Gly-MeCys(Acm)-MeCys(Me)&2][H-Gly-MeCys-
U
The elongation of the peptide was achieved by sequential addition of
Fmoc-AA-OH with HATU and DIEA (1:1:2) as coupling reagents in
DMF in pre-activation mode. The mixture was stirred for 35 min and
after filtration the corresponding colorimetric test indicated completion
of the coupling. Next, the peptide resin was washed with DMF (5”30 s)
and treated with a solution of piperidine/DMF (1:4) to remove the Fmoc
group, and the coupling plus Fmoc removal was then measured, as shown
in Table 2.
A
proach: CTC resin (150 mg, 1.6 mmolgÀ1) was placed in a 10 mL polypro-
pylene syringe fitted with a polyethylene filter disc. The resin was then
washed with DCM (5”1 min) and a solution of Boc-d-Dap(Fmoc)-OH
E
(45 mg, 0.21 mmol) and DIEA (123 mL, 1.33 equiv) in CH2Cl2 (2.5 mL)
was then added. After 5 min, more DIEA (60 mL, 0.66 equiv) was added
and the mixture was stirred for 55 min at 258C. The reaction was termi-
nated by addition of MeOH (120 mL) and the mixture was stirred for a
An aliquot was cleaved for analysis by HPLC-MS. This technique
showed two close peaks with the same mass. The global linear synthesis
yield was 78%. Analytical HPLC (tR = 8.3, 8.4 min, conditions 1:9 to
7:3, 83% purity); ES(+): m/z: calcd for C44H78N12O15S4: 1142.5; found:
1143.9 [M+H]+.
{[Boc-d-Dap&1-Gly-MeCys(&2)-MeCys(Me)&3][H-Gly-MeCys(&2)-Me-
Cys(Me)&1][Boc-d-Dap(&3)-O-CTC resin]}—disulfide bond formation
on solid support: A solution of I2 (122 mg, 5 equiv, 2.5 equiv”Acm) in
DMF was added to the peptide resin and the mixture was stirred for
10 min at 258C. After filtration, the resin was washed repeatedly with
DMF (5”30 s), CH2Cl2 (10”30 s) and CHCl3 (5”30 s). HPLC-MS analy-
sis of an aliquot cleaved from the resin indicated that the disulfide bond
formation had been completed. The peptide resin was cleaved with a sol-
ution of TFA/H2O (1:99, 5”30 s), and filtrates were collected over H2O
(18 mL, 60 mL H2O per gram resin). The combined filtrates were evapo-
further 10 min. The Boc-d-Dap(Fmoc)-O-CTC resin was subjected to the
R
following washings/treatments: filtration, CH2Cl2 (5”30 s), DMF (5”
30 s), piperidine/DMF (1:4, 1”1 min, 3”5 min, 1”10 min), piperidine/
DBU/toluene/DMF (5:5:20:70, 2”5 min). Loading, calculated by measur-
ing absorbance at 290 nm, was 0.61 mmolgÀ1
.
The elongation of the peptide was achieved by sequential addition of
Fmoc-AA-OH with HATU and DIEA (1:1:2) as coupling reagents in
DMF in pre-activation mode. The mixture was stirred for 35 min and
after filtration the corresponding colorimetric test indicated the comple-
tion of the coupling. Next, the peptide resin was washed with DMF (5”
30 s) and treated with a solution of piperidine/DMF (1:4) to remove the
Fmoc group, and the coupling plus Fmoc removal was then measured, as
shown in Table 1.
The peptide resin was split into fractions (1/3 and 2/3). The former was
treated with a piperidine/DMF (1:4) solution as described above, whilst
the latter was cleaved from the resin with a TFA/H2O solution (1:99, 5”
30 s), and filtrates were collected over H2O (3 mL, 60 mL H2O per gram
resin). The combined filtrates were concentrated to dryness under re-
duced pressure to provide the protected tetrapeptide (34.4 mg, 43 mmol,
78% yield). Analytical HPLC (tR = 8.7 min, conditions 3:7 to 10:0, 88%
Table 2. Report on the synthesis of the octapeptide resin synthesized by
carboxylic acid anchoring.[a]
Fmoc-AA-OH
Amount
[mg (equiv)]
Test
Yield
[%]
A
ACHTREUNG
Fmoc-NMe-Cys(Me)-OH
214, (3)
247, (3)
285, (5)
285, (5)
246, (3)
214, (3)
247, (3)
285, (5)
ninhydrin (À)
De Clercq (À)
De Clercq (+)
De Clercq (À)
ninhydrin (À)
ninhydrin (À)
De Clercq (À)
De Clercq (À)
95
97
Table 1. Report on the synthesis of the tetrapeptide resin.[a]
Fmoc-NMe-Cys(Acm)-OH
A
Fmoc-Gly-OH
97
Fmoc-AA-OH
Amount
[mg (equiv)]
Test
Yield
[%]
A
ACHTREUNG
Boc-d-Dap
A
97
98
97
94
Fmoc-NMe-Cys(Me)-OH
102, (3)
118, (3)
136, (5)
ninhydrin (À)
De Clercq (À)
De Clercq (À)
90
96
Fmoc-NMe-Cys(Me)-OH
Fmoc-NMe-Cys
A
Fmoc-NMe-Cys(Acm)-OH
H
Fmoc-Gly-OH
Fmoc-Gly-OH
[a] In all cases the coupling reactions were performed in DMF for
35 min.
[a] In all cases the coupling reactions were performed in DMF for
35 min.
Chem. Eur. J. 2006, 12, 9001 – 9009
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9007

F. Albericio et al.
rated to dryness under reduced pressure to provide the cyclic peptide
(86 mg, 57% yield). Analytical HPLC (tR = 9.4 min, conditions 1:9 to
7:3, 82% purity); MALDI-TOF MS (DHB): m/z: calcd for
C38H66N10O13S4: 998.4; found: 999.3 [M+H]+, 1021.3 [M+Na]+, 1037.3
[M+K]+.
{[Boc-d-Dap(&1)-Gly-MeCys(&2)-MeCys(Me)&3][Boc-d-Dap(&3)-Gly-
MeCys(&2)-MeCys(Me)&1]}: Cyclic peptide (85 mg, 85 mmol) dissolved
in CH2Cl2 (85 mL, 1 mm) was added to a solution of HOAt (53 mg,
340 mmol) in the minimum amount possible of DMF. DIEA (30 mL,
170 mmol) and later EDC·HCl (65 mg, 340 mmol) were added and the
mixture was stirred for 35 min at 258C. The solvent was removed in
vacuo, a solution of CH2Cl2/Et2O (1:1, 6 mL) was added, and the organic
layer was washed with aqueous NaHCO3 (5%, 2”4 mL), HCl (1n, 2”
4 mL) and NaCl (sat, 2”4 mL). The solvent was removed under reduced
pressure. Analytical HPLC (tR = 9.4 min, conditions 1:9 to 7:3, 88%
purity); ES(+): m/z: calcd for C38H64N10O12S4: 980.4; found: 982 [M+H]+.
Table 3. Report on the synthesis of the octapeptide resin synthesized by
amino anchoring.[a]
Fmoc-AA-OH
Amount
[mg (equiv)]
Test
Yield [%]
A
ACHTREUNG
Fmoc-NMe-Leu-OH
24, (3)
27, (3)
37, (5)
26, (3)
24, (3)
27, (3)
27, (3)
37, (5)
ninhydrin (À)
De Clercq (À)
De Clercq (À)
ninhydrin (À)
ninhydrin (À)
De Clercq (+)
De Clercq (+)
De Clercq (À)
97
95
99
95
98
Fmoc-NMe-Cys(Acm)-OH
G
Fmoc-Gly-OH
Boc-d-Dap
A
Fmoc-NMe-Leu-OH
Fmoc-NMe-Cys
Fmoc-Gly-OH
A
99
97
[a] In all cases the coupling reactions were performed in DMF for
35 min.
and CHCl3 (5”30 s). HPLC-MS analysis of the aliquot cleaved from the
resin indicated completion of disulfide bond formation.
Azathiocoraline: The bicyclical peptide was dissolved in TFA/H2O (19:1,
3 mL) and the mixture was vigorously stirred for 1.5 h at 258C. TFA was
evaporated under reduced pressure and the residual acid was removed
by co-evaporation with Et2O. H2O was added and the product was
lyophilized. A white solid was obtained (16.7 mg). Unprotected bicycle
peptide (16.7 mg, 21.4 mmol) was dissolved in CH2Cl2 (250 mL), and
DIEA (11.1 mL) was then added. 3-Hydroxyquinoline-2-carboxylic acid
(12.3 mg, 64.2 mmol) was pre-activated with EDC·HCl (12.4 mg,
64.2 mmol) and HOSu (7.4 mg, 11 mmol) in CH2Cl2 (1 mL), and after
15 min the solution was added to the peptide solution. The solution was
stirred for 16 h at 258C and more coupling reagents were then added:
EDC·HCl (12.4 mg, 64.2 mmol) and HOSu (7.4 mg, 11 mmol) for two
more days. The crude product was purified by PTLC (SiO2, CH2Cl2/
MeOH 95:5 as eluent) to afford the final product (2.5 mg). Analytical
In a first attempt, the cycling step was performed with DIPCDI (2 equiv)
and HOBt (2 equiv) as coupling reagents in DMF for 40 min. Because
the ninhydrin test was positive, a second coupling was performed with
more reactive reagents DIPCDI (2 equiv)/HOAt (2 equiv) for 40 min.
The ninhydrin test was clearer and a second coupling with same reagents
was required to complete the cyclization. Analytical HPLC (tR
=
8.3 min, conditions 0:1 to 1:0, 31% purity); MALDI-TOF MS (DHB): m/
z: calcd for C37H64N10O10S2: 872.4; found: 873.7 [M+H]+, 895.6 [M+Na]+
, 911.5 [M+K]+.
[NMe-Leu4, NMe-Leu8]azathiocoraline: The peptide was cleaved from
the resin as described above. After removal of the solvent, a solution of
TFA/H2O (95:5) was added and the mixture was stirred for 1.5 h at 258C.
The solvent was removed under reduced pressure, H2O was then added,
and the product was lyophilized. The 3-hydroxyquinoline-2-carboxylic
acid was coupled as described in the synthesis of azathiocoraline. The
HPLC (tR
C48H58N12O12S4: 1122.3180; found: 1123.3253 [M+H]+.
{[Boc-d-Dap(&1)-Gly-MeCys(Acm)-MeLeu&2][H-Gly-MeCys
MeLeu&1][CTC
CTC resin
= 7.1 min, conditions 5:5 to 9:1); HRMS: m/z calcd for
A
ACHTREUNG
resin-d-Dap(&2)-Oallyl]}:
product was characterized by HPLC-MS. Analytical HPLC (tR
=
1.6 mmolgÀ1) was placed in a 10 mL polypropylene syringe fitted with a
polyethylene filter disc. The resin was then washed with DCM (5”
9.4 min, conditions 5:5 to 9:1); HRMS: m/z calcd for C52H66N12O12S2:
1114.4365; found: 1115.4437 [M+H]+.
1 min), and a solution of H-d-Dap(Fmoc)-Oallyl (12.8 mg, 35 mmol) and
A
DIEA (40 mL, 6.67 equiv) in CH2Cl2 (0.5 mL) was then added. After
5 min, more DIEA (20 mL, 3.33 equiv) was added and the mixture was
stirred for 55 min at 258C. The reaction was terminated by addition of
MeOH (40 mL) and the mixture was stirred for a further 10 min. The
Acknowledgements
CTC resin-NH-d-Dap(Fmoc)-Oallyl resin was subjected to the following
A
This workwas partially supported by funds from CICYT (BIO2002–
2301, BQU2003–00089, and PETRI 95–0658-OP), Generalitat de Catalu-
nya (Grup Consolidat and Centre de Refer›ncia en Biotecnología) and
PharmaMar. The help of Dr. Jan Spengler in the hexafluoroacetone pro-
tection (Ref. [13]) is gratefully acknowledged.
washings/treatments: filtration, CH2Cl2 (5”30 s), DMF (5”30 s), piperi-
dine/DMF (1:4, 1”1 min, 3”5 min, 1”10 min), piperidine/DBU/toluene/
DMF (5:5:20:70, 2”5 min). Loading, calculated by measuring absorbance
at 290 nm, was 0.43 mmol per gram. Elongation of the peptide was ach-
ieved by sequential addition of Fmoc-AA-OH with HATU, and DIEA as
coupling reagent in DMF was added to the peptide resin. The mixture
was stirred for 35 min and after filtration the corresponding colorimetric
test indicated completion of the coupling. In this case, the resin was
washed with DMF (5”30 s) and treated with a solution of piperidine/
DMF (1:4, 1”1 min, 3”5 min, 1”10 min) to remove the Fmoc group.
Absorbance was determined in each case to obtain the yield of the cou-
pling as shown in Table 3.
[1] F. Romero, F. Espliego, J. Perez Baz, T. Garcia de Quesada, D.
Gravalos, F. De la Calle, J. L. Fernandez-Puentes, J. Antibiot. 1997,
50, 734–737.
[2] a) M. Waring, J. Path. Biol. 1992, 1022–1034; b) H. Otsuka, J. Shoji,
Tetrahedron 1967, 23, 1535–1542.
[3] This peptide can be considered to be made up of two antiparallel
symmetric chains with the amino terminals capped by the chromo-
phore groups.
[4] E. Erba, D. Bergamaschi, S. Ronzoni, M. Faretta, S. Taverna, M.
Bonfanti, C. V. Catapano, G. Faircloth, J. Jimeno, M. DꢀIncalci, Brit-
ish J. Cancer 1999, 80, 971–980.
[5] M. Teixidó, F. Albericio, E. Giralt, J. Pept. Res. 2005, 57, 153–166.
[6] Some elegant solution syntheses of thiocoraline and related mole-
cules have recently been published: a) D. L. Boger, S. Ichikawa, J.
Am. Chem. Soc. 2000, 122, 2956–2957; b) D. L. Boger, J. K. Lee, J.
Org. Chem. 2000, 65, 5996–6000; c) M. Shin, K. Inouye, N. Higuchi,
Y. Kyogoku, Bull. Chem. Soc. Jpn. 1984, 57, 2211–2215; d) D. L.
Boger, J. H. Chen, K. W. Saionz, J. Am. Chem. Soc. 1996, 118, 1629–
1644; e) D. L. Boger, M. W. Ledeboer, M. Kume, M. Searcey, Q. Jin,
An aliquot was cleaved from the resin and analyzed as described above.
We achieved an 81% yield of linear peptide and 79% purity of the crude
product. Analytical HPLC (tR
= 7.3 min, conditions 0:1 to 1:0, 79%
purity); MALDI-TOF MS (DHB): m/z: calcd for C46H82N12O13S2: 1074.6;
found: 1075.7 [M+H]+, 1097.7 [M+Na]+, 1113.7 [M+K]+.
{[Boc-d-Dap(&1)-Gly-MeCys(&2)-MeLeu&3][CTCresin-d-Dap(&3)-Gly-
MeCys(&2)-MeLeu&1]}: Allyl ester was removed with three treatments
with [Pd(PPH3)4] (6 mg, 0.1 equiv) and PhSiH3 (62 mL, 10 equiv) dis-
G
solved in CH2Cl2 for 15 min under Ar. The remaining Pd was removed
with three treatments of diethyldithiocarbamate sodium in DMF (0.2m)
for 15 min. A solution of I2 (81 mg, 5 equiv, 2.5 equiv”Acm) in DMF was
added and the solution was stirred for 10 min. After filtration, the pep-
tide resin was washed repeatedly with DMF (5”30 s), CH2Cl2 (10”30 s)
9008
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 9001 – 9009

Solid-Phase Synthesis of Azathiocoralines
FULL PAPER
J. Am. Chem. Soc. 1999, 121, 11375–11383; f) D. L. Boger, M. W.
Ledeboer, M. Kume, Q. Jin, Angew. Chem. 1999, 111, 2533–2536;
Angew. Chem. Int. Ed. 1999, 38, 2424–2426; ; g) Y. B. Kim, H. K.
Kim, J. Y. Park, S. K. Kim, Bioorg. Med. Chem. Lett, 2004, 14, 541–
544; h) K. B. Lorenz, U. Diederichsen, J. Org. Chem. 2004, 69,
3917–3927.
[18] This strategy also prevents the formation of DKPs. This peptide is
prone to undergoing this side-reaction because of the presence of N-
methyl amino acids and Gly.
[19] S. A. Kates, N. A. SolØ, M. Beyermann, G. Barany, F. Albericio,
Pept. Res. 1996, 9, 106–113.
[20] In our hands, this test showed a superior performance to the ninhy-
drin and chloranil tests for the detection of primary and secondary
amines. a) A. Madder, N. Farcy, N. G. C. Hosten, H. De Muynck,
P. J. De Clercq, J. Barry, A. P. Davis, Eur. J. Org. Chem. 1999, 2787–
2791; b) J. Vµzquez, G. Qushair, F. Albericio, Colorimetric Tests for
Solid Phase Synthesis. In Methods in Enzymology, Vol. 369 (Ed.: G.
Morales), Academic Press, San Diego, 2003, 21–35.
[21] Phosphonium derivatives are more convenient than uronium re-
agents such as HATU for rather slow couplings, since the latter can
terminate the peptide chain through a guanylidation reaction. F. Al-
bericio, J. M. Bofill, A. El-Faham, S. A. Kates, J. Org. Chem. 1998,
63, 9678–9683.
[22] Alternatively, the elongation of the full sequence can be performed
stepwise by the same synthetic protocol (Scheme 1). In this case,
yields of both strategies were very similar.
[23] a) C. García-Echeverría, F. Albericio, M. Pons, G. Barany, E. Giralt,
Tetrahedron Lett. 1989, 30, 2441–2444; b) D. Andreu, F. Albericio,
N. A. SolØ, M. C. Munson, M. Ferrer, G. Barany, “Formation of di-
sulfide bonds in synthetic peptides and proteins”, in Methods in Mo-
lecular Biology, Vol. 35 : Peptide Synthesis Protocols (Eds.: M. W.
Pennington, B. M. Dunn), Humana Press, Totowa, N. J., 1994, 91–
169.
[7] During the preparation of this manuscript the solid-phase synthesis
of TANDEM, a cyclic depsipeptide with some features in common
with thiocoraline, but lacking the N-methyl amino acids and the free
hydroxy group of the heterocycle, has been published: J. P. Malkin-
son, M.K. Anim, M. Zloh, M. Searcey, A. J. Hampshire, K. R. Fox,
J. Org. Chem. 2005, 70, 7654–7661.
[8] For a short report on the solid-phase synthesis of complex peptides,
see: M.A. Ciufolini, in Strategies and Tactics in Organic Synthesis,
Vol. 6 (Ed.: M. Harmata), Elsevier, Amsterdam, NL, 2005, pp. 25–
26.
[9] K. Barlos, D. Gatos, W. Schäfer, Angew. Chem. 1991, 103, 571–572;
Angew. Chem. Int. Ed. Engl. 1991, 30, 590–593.
[10] J-F. Liu, X-X. Tang, B. Jiang, Synthesis 2002, 1499–1501.
[11] Use of MeBr resulted in a S-permethylation followed by a b-elimi-
nation reaction.
[12] F. Albericio, A. Grandas, A. Porta, E. Pedroso, E. Giralt, Synthesis
1987, 271.
[13] Methylation with MeI under basic conditions was achieved only on
a small scale (<100 mg), whilst on larger scales b-elimination of
Cys(SMe) was observed, as well as N-methylation of the amide
G
bond of the Acm protecting group. Alternatively, treatment of these
derivatives with paraformaldehyde and p-toluenesulfonic acid or
with CH2I2 under basic conditions did not result in the formation of
the corresponding oxazolone, reduction of which could have afford-
ed the target compounds. Finally, hexafluoroacetone was used to
protect both the amino and the carboxyl group, but the subsequent
N-methylation did not work.
[24] J. C. JimØnez, ꢂ. López-Maciꢃ, C. Gracia, S. Varon, M. Carrascal, C.
Cuevas, M. Royo, E. Giralt, F. Albericio, unpublished results.
[25] During the coupling with EDC·HCl/HOAt, incorporation of three
units of 3-hydroxyquinaldic acid was detected.
[26] Analytical HPLC (tR = 7.1 min, 81% purity, conditions 50% ACN
(0.036% TFA), 50% H2O (0.045% TFA) to 90% ACN (0.036%
TFA), 10% H2O (0.045% TFA) in 15 min); HRMS: m/z calcd for
C48H58N12O12S4: 1122.3180; found: 1123.3253 [M+H]+.
[27] F. Albericio, R. P. Hammer, C. Garcia-Echeverria, M. A. Molins,
J. L. Chang, M. C. Munson, M. Pons, E. Giralt, G. Barany, Int. J.
Pept. Protein Res. 1991, 37, 402–413.
[28] These thiocoraline analogues with two different bisintercalation
chromophore groups should allow the study of the effects on biolog-
ical activity both of broken symmetry and of the presence of two
distinct chromophore groups.
[14] a) S. C. Miller, T. S. Scanlan, J. Am. Chem. Soc. 1998, 120, 2690–
2691; b) S. C. Miller, T. S. Scanlan, J. Am. Chem. Soc. 1997, 119,
2301–2302.
[15] Commercial 3-aminoquinoline was transformed into 3-hydroxyqui-
noline by the Bucherer reaction (93%) and protected as a MOM
ether (67%). Then, 1,2-addition of methyllithium to the quinoline
ring (88%), followed by oxidation of the methyl group to a carbox-
ylic acid in a two-step procedure afforded the target compound in
an overall yield of 37%: E. Riego, N. Bayó, C. Cuevas, F. Albericio,
M. lvarez, Tetrahedron 2005, 61, 1407–1411. In our hands, this
method has given clearly better results than the previous method
described by D. L. Boger, J. Chen, J. Org. Chem. 1995, 60, 7369–
7371.
[29] This was prepared from commercially available Boc-d-Dap(Fmoc)-
R
OH.
[30] P. Gómez-Martínez, M. Dessolin, F. GuibØ, F. Albericio, J. Chem.
Soc. Perkin Trans. 1 1999, 2871–2874.
[31] Analytical HPLC (tR
= 9.3 min; conditions 50% ACN (0.036%
TFA), 50% H2O (0.045% TFA) to 90% ACN (0.036% TFA), 10%
H2O (0.045% TFA) in 15 min; HRMS: m/z calcd for
C52H66N12O12S2: 1114.4365; found: 1115.4437 [M+H]+.
[16] P. Lloyd-Williams, F. Albericio, E. Giralt, ChemicalApproaches to
the Synthesis of Peptides and Proteins, CRC, Boca Raton (FL,
USA), 1997.
[32] For the nomenclature used see: J. Spengler, J. C. JimØnez, E. Giralt,
F. Albericio. Nomenclature for Cyclic and Branched Homo- and
Heterodetic Peptides, J. Peptide Res. 2005, 65, 550–555.
[17] The use of Cl-Trt-Cl-resins of higher loadings (>0.5 mmolgÀ1) for
the preparation of hydrophobic peptides usually results in complex
crude products: C. Chiva, M. Vilaseca, E. Giralt, F. Albericio, J.
Pept. Sci. 1999, 5, 131—140. Unreacted reactive sites were capped
with MeOH/DIEA.
Received: June 9, 2006
Published online: September 5, 2006
Chem. Eur. J. 2006, 12, 9001 – 9009
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9009