Macrocyclization studies and total synthesis of cyclomarin C, an anti-inflammatory marine cyclopeptide

Article abstract of DOI:10.1016/j.tet.2005.03.058

The studies on macrocyclization at possible sites toward the total synthesis of cyclomarin C are described. The results showed that both Trp and Phe derivatives involved in the target could not be the terminals of the final linear peptide precursors. Addi

Full text of DOI:10.1016/j.tet.2005.03.058

Tetrahedron 61 (2005) 4931–4938
Macrocyclization studies and total synthesis of cyclomarin C, an
anti-inflammatory marine cyclopeptide
*
Shi-Jun Wen, Tai-Shan Hu and Zhu-Jun Yao
State Key Laboratory of Bioorganic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, 354 Fenglin Road, Shanghai 200032, China
Received 24 January 2005; revised 8 March 2005; accepted 10 March 2005
Available online 12 April 2005
Abstract—The studies on macrocyclization at possible sites toward the total synthesis of cyclomarin C are described. The results showed
that both Trp and Phe derivatives involved in the target could not be the terminals of the final linear peptide precursors. Additionally,
preparation of corresponding dipeptides with an N-methyl amide bond is not favorable in the synthesis of linear precursors. Site d was finally
proved a proper site for the cyclopeptide formation, and the corresponding head-to-tail macrocyclization was achieved under mild conditions
and gave repeatable and satisfactory yields.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
the unique structural complexicities. In the past several
years, many efforts and achievements were reported about
the elaboration of key amino acids,2 while the first total
synthesis of cyclomarin C was demonstrated by us very
recently.3 Due to its limited availability (1 mg, 3% of
cyclomarins from nature), the biological activities of
cyclomarin C could not be studied intensively. Therefore,
the synthetic efforts toward cyclomarin C will definitely
illustrate two sides of importance. In chemistry side,
because of the structural similarity between cyclomarins,
the methodologies and strategies developed during its total
synthesis will benefit those efforts toward other members of
cyclomarins. The other income will be on the biology side,
because the total synthesis of cyclomarin C will be able to
offer more quantity of sample to meet the requirements of
biological studies, as well as possible derivatives. In our
recent preliminary communication,3 we reported the results
on the total synthesis of cyclomarin C, including methods to
Cyclomarins A–C (1–3), three novel bioactive metabolites,
were recently isolated and characterized from a marine
bacterium collected in the vicinity of San Diego by Clardy
et al.1 The major constituent (95% of the total cyclomarin
mixture), cyclomarin A displays significant anti-inflamma-
tory activities both in vitro and in vivo assays, as well as
anti-cancer activity. In the phorbol ester (PMA)-induced
mouse ear edema assay, this compound shows 92%
inhibition of edema at the standard testing dose of 50 mg/
ear. In the same assay, cyclomarin A also shows promising
in vivo activity (45% reduction in edema at 30 mg/kg ip
administration), indicating it may be a potential drug
candidate. Structurally, all three natural products are cyclic
heptapeptides, and contain four noncoded amino acids in
each. Their structures show remarkable resemblance to each
other (Fig. 1). Cyclomarin A and C each contain three
common amino acids (alanine, valine, and N-methylleucine),
two uncommon amino acids (b-methoxyphenylalanine, N-
methyl-5-hydroxyleucine), and two novel amino acids
(2-amino-3,5-dimethylhex-4-enoic acids for both, N0-pre-
nyl-b-hydroxytryptophan for cyclomarin C, and N0-(1,1-
dimethyl-2,3- expoxypropyl)-b-hydroxytryptophan for
cyclomarin A).
Cyclomarins have attracted much attention within the
synthetic community due to their potent bioactivities and
Keywords: Macrocyclization; Dipeptides; Cyclomarin C.
*
Corresponding author. Tel.: C86 21 5492 5133; fax: C86 21 6416 6128;
e-mail: yaoz@mail.sioc.ac.cn
Figure 1. The chemical structures of cyclomarins A–C.
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.03.058

4932
S.-J. Wen et al. / Tetrahedron 61 (2005) 4931–4938
elaborate those noncoded amino acids,4 convergent assem-
bly of linear peptide, and final successful macrocyclization
at the optimized site. In this paper, we would like to address
the details of our efforts to search a proper site for the final
macrocyclization, which have played a key role in the total
synthesis of cyclomarin C (3).
coupling of noncoded amino acid derivative (amine) 12
and dipeptide (acid) 10,5 using EDCI-HOBt-based con-
ditions (Scheme 1). Similarly, the other tripeptide 18 was
synthesized starting from 14 via two successive coupling
reactions. The active ester of N-Boc-Ala-OH generated in
situ was treated with b-methoxyphenylalanine derivative
(amine) 15 to afford dipeptide 16 (88%), whose N-Boc
protecting group could be removed by 10% TFA in
dichloromethane quantitatively. Another noncoded amino
acid derivative (acid) 17 was activated by EDCI and HOBt
and then allowed to react with the amine derived from
dipeptide 16, giving the desired tripeptide 18 in 80% yield.
As mentioned, seven components (single amino acid
derivatives) appear in cyclomarin C were synthesized by
various fashions, and modified as the proper derivatives.
With those fragments in hand, the following major task is
how to link them together, affording the corresponding
linear heptapeptide precursors for the final macrocycliza-
tion. In order to enhance the synthetic efficiency, two key
points were initially considered in our retrosynthetic
analysis for this stage of total synthesis. First, a convergent
peptide-segment assembly strategy rather than a stepwise
one should be more preferable. Second, because the
noncoded amino acid, N0-prenyltryptophan, was highly
sensitive to various acidic conditions, it would be favorable
to incorporate this amino acid at the last stage of the
synthesis of linear heptapeptide. Figure 2 illustrates all of
our trials for a proper macrocyclization site and the
corresponding linear precursors.
Coupling between the above two tripetides was executed as
Scheme 2. At first, methyl ester 18 and N-Boc derivative 13
were treated with LiOH in water–THF and 10% TFA in
dichloromethane, respectively, affording the corresponding
acid and amine. The amide bond formation between these
two intermediates was achieved under EDCI and HOBt
conditions, and the linear hexapeptide 20 was obtained in
63% yield. After removal of N-Boc functionality of 20, the
resulting amine reacted with tryptophan derivative 22 in the
presence of EDCI and HOAt6 to give heptapeptide 4 in
reasonable yield. Treatment of heptapeptide 4 with 10%
piperidine in dichloromethane followed by LiOH in THF–
water afforded macrocyclization precursor 23 with both
amino and carboxylic acid functionalities. It is noteworthy
that the hydrolysis of methyl ester was a low-yield reaction
and generated many undesired by-products. To make
matters worse, macrocyclization of 23 turned out to be
troublesome. A variety of condensation conditions were
2. Synthesis and discussion
Disconnection of the amide bond at site a afforded the first
route (see Fig. 2), in which Trp derivative 22 was
incorporated at the last step of the linear precursor 4, and
a [3C3C1] peptide-segment-assembly strategy was
adopted. The first tripeptide 13 was prepared by the
Figure 2. Outline of tested macrocyclization sites and their corresponding
linear precursors.
Scheme 1. Synthesis of two tripeptides, 13 and 18.

S.-J. Wen et al. / Tetrahedron 61 (2005) 4931–4938
4933
Scheme 3. Synthetic efforts via route b and its [4C3] peptide assembly
strategy.
methane. Linkage of segments 26 with 27 was achieved
through amide bond formation in the presence of EDCI and
HOAt (60% yield) to give heptapeptide 6, whose N-Fmoc
group was removed smoothly (78%) using 10% piperidine
in dichloromethane. Subsequent saponification of methyl
ester gave 41% yield of precursor 28. The lower yield of
ester hydrolysis may be partially due to the elimination of
C-terminal residue, b-methoxylphenylalanine methyl ester.
To our disappointment, all efforts towards macrocyclization
of precursor 28 again failed. Similar to route a, the last step
resulted in complex mixtures under a variety of conditions.
These results indicate b-methoxyphenylalanine residue is
not a suitable C-terminal for the cyclization precursor.
Scheme 2. Coupling between tripeptides 13 and 18, and trials for the
macrocyclization.
examined for this cylcization, but none led to satisfactory
results. For example, in the presence of BOP and
diisopropylethylamine in DMF or dichloromethane, only
trace of macrocycle 24 (w2–4%) was formed as detected by
mass spectrum and the linear precursor was decomposed
during the reaction (Scheme 2). Based on this failure and
our experience on the preparation of 22, we realized that not
only the free form of 22 (nonprotected form) is unstable, the
linear peptides with it as the terminal moiety, like 23, were
not stable either. So, in the alternative macrocyclization
precursors, the unstable component 22 should be placed in
the middle region of the final precursor (with two amide
bond in both C and N terminals of this amino acid, as it is in
natural product) for macrocyclization.
The third approach to cyclomarin C (3) through bond
disconnection at site c (Fig. 2) also encountered problems at
the early stage of total synthesis, and failed to prepare the
designed heptapeptide (Scheme 4). Dipeptide 29 with an N-
methyl amide functionality was found to easily cyclize
intramolecularly7 when it was treated with 10% piperidine
in dichloromethane, giving 2,5-piperazinedione derivative
31 (67%). An explanation for this result is the confor-
mational equilibrium (trans and cis) of N-methyl amide bond,
in which its cis conformation favors the observed cyclization.
Because of the severe side reactions in preparing the
dipeptides, further investigations were not explored.
With the above considerations in mind, the second route (b)
was designed and heptapeptide 6 was envisioned to be the
cyclization precursor (Fig. 2), in which the b-hydroxyl Trp
derivative 22 was at the middle of the molecule to avoid the
involvement in final cyclization reaction. Synthesis of the
linear peptide 6 adopted a [4C3] peptide assembly strategy
(Scheme 3). Tripeptide 13 was saponified with LiOH in
water, and then treated with 10% TFA in dichloromethane
and FmocOSu in the presence of Na2CO3 sequentially to
give peptide acid 27 quantitatively. The tetrapeptide
segment (amine) 26 was prepared by coupling of tripeptide
18 derived amine and b-hydroxyl Trp derivative 22 (44%),
and N-Fmoc removal by 10% piperidine in dichloro-
Analyzing the above unsatisfactory results, we concluded
that the following situations should be avoided during the
total synthesis: (1) the Trp and Phe derivatives involved
should not be the terminal of the final linear peptide
precursors; (2) preparations of dipeptides with an N-methyl
amide bond is not favorable (see also the conversion of 9 to
32, Scheme 5). With those principles, only limited sites
could be chosen for the macrocylization. Path d was the final
and successful one we tried to elaborate the whole molecule.
In this route, cyclomarin C was disconnected at site d and a
[4C3] peptide assembly strategy was employed (Fig. 2). In
order to avoid the complexity caused by saponifications at

4934
S.-J. Wen et al. / Tetrahedron 61 (2005) 4931–4938
Scheme 4. 2,5-Piperazinedione derivative formation at the early stage of
route c.
C-terminals, route d adapted allyl ester as the acid
protecting form, because it could be easily removed by
Pd(0) or Rh(I)-catalyzed isomerization protocols8–9 in mild
conditions (Scheme 5). Dipeptide 33 derived from N-Boc
Val-OH and N-Me Leu(Oallyl) was treated with 10% TFA,
giving the dipeptide TFA salt 34. The N-Boc amino acid 15a
was then allowed to couple with 34 in the presence of EDCI
and HOBt to afford tripeptide 35 (62%). Preparation of the
other segment, tetrapeptide 40 was started from the coupling
of N-Boc amino acid 17 and Ala(Oallyl) using EDCI and
HOBt. The resultant dipeptide 36 was first treated with 10%
TFA and then coupled with N-Fmoc amino acid 22 using
BOPCl as the carboxyl activation reagent to afford
tripeptide 38 (70%). The fourth amino acid derivative 11a
was finally incorporated into amino-liberated tripeptide 39,
affording tetrapeptide 40 (72%). Parallel removals of N-Boc
protecting group of 35 (10% TFA in dichloromethane) and
O-allyl ester of 40 (Pd(PPh3)4, PhNHMe, THF)10 afforded
the both precursors essential for the followed coupling to
generate the desired heptapeptide 41 (Scheme 6).
Scheme 5. Synthesis of tetrapeptide 40 and tripeptide 35.
up to 60–70 8C. Spatially hindrance enhancement after
macrocycle formation may be a reason for the difficulty to
remove O-TBS ether of 42. Therefore, we decided to
remove O-TBS ether before the cyclization step, that is, at
the linear stage (Scheme 7). The subtle change of the
reaction sequence proved very successful. Removal of caps
at C- and N-terminals of 41 followed by TBS-deprotection
with TBAF gave a satisfactory yield of 43 (98%).
Macrocylclization was again accomplished under similar
conditions and afforded macrocycle 44 (63%). Finally,
treatment of 44 with K2CO3 in methanol at rt afforded
cyclomarin C (3, 38 mg, 80% yield), and reproducibility of
this reaction was perfectly high.
This coupling reaction was performed very well under the
conditions of EDCI and HOAt.11 It is noteworthy here that
replacement of HOAt with HOBt resulted in much lower
efficiency. Upon stable heptapeptide 41, both the N- and C-
terminal protections were similarly removed by the above
methods, Pd(0)-catalyzed deprotection of allyl ester and
10% piperidine for N-Fmoc removal. Under the presence of
PyBOP12 and diisopropylethylamine, the resulting precur-
sor was cyclized smoothly by head to tail in dichloro-
methane at a diluted concertration (1.4 mM), affording
macrocycle 24 in 55% yield. Deprotection of benzoate at
one of the side chains of 24 was achieved by K2CO3 in
methanol at rt, giving 42 in 62% yield. Unfortunately, the
followed removal of O-TBS ether existing in 42 using
TBAF failed. No reactions were observed at rt, while
decomposition happened when the temperature was raised
The synthetic sample of 3 purified by silica gel chromato-
graphy showed high purity (O98%, by HPLC analysis using
same conditions described in Ref. 1). However, significant
difference was observed in comparison of specific optical
rotation of synthetic sample ([a]2D0ZK72.8 (c 0.75,
CHCl3)) with that reported for natural product
([a]2D0ZK19.7 (c 1.0, CHCl3)). Because of unavailability

S.-J. Wen et al. / Tetrahedron 61 (2005) 4931–4938
4935
3. Conclusion
In summary, our efforts to search for the suitable
macrocylization sites and the first total synthesis of
cyclomarin C were described in details. These studies
showed that the Trp and Phe derivatives involved in the
target could not be the terminals of the final linear peptide
precursors, and preparations of the corresponding dipeptides
with an N-methyl amide bond is not favorable. Site d finally
proved to be a proper site for macrocyclization, and the
head-to-tail macrocyle formation was realized under the
dilute conditions at ambient temperature in repeatable and
satisfactory yields. The first total synthesis of cyclomarin C
was thus achieved based on these explorations. All these
will be helpful for the synthesis of other members of
cyclomarins. Further applications of the developed methods
and strategies to cyclomarins A and B, as well as the
evaluation of activity contribution caused by each amino
acid component in cyclomarins are under investigation in
our laboratory.
4. Experimental
4.1. General
Physical data of compounds in rout d are available in
supporting information of Ref. 3.
Scheme 6. Macrocyclization of linear heptapeptide.
4.1.1. Dipeptide 9. To a solution of N-Boc, N-methyl-L-
Leu-OMe (3.8 g, 14.7 mmol) in dichloromethane (60 mL)
cooled with ice-water bath was added dropwise TFA
(9 mL). After stirred at room temperature for about 4.5 h,
the reaction mixture was concentrated under reduced
pressure, and the residue was dissolved in ethyl acetate
(30 mL) and basified to pH 9 with sat. aq. NaHCO3 solution
at 0 8C. The aqueous phase was separated and extracted with
ethyl acetate (20 mL!3). The organic layers were dried
with anhydrous Na2SO4 and concentrated to give compound
8 (2.42 g) as a yellow oil, which was used in the next step
without further purification.
of natural product sample for a CD spectrum, peak-to-peak
comparisons4 of both 1H NMR and 13C NMR spectra were
adapted. These results show the synthetic sample is
1
exactly same as the natural one, while the reported H
NMR spectrum of natural product (in supporting
information of Ref. 1) shows that it contains small
amount of impurities.
To a solution of N-Boc-L-valine (3.25 g, 15 mmol) in
dichloromethane (60 mL) at 0 8C were added successively
HOBt (2.18 g, 16 mmol), EDCI (3.13 g, 16 mmol) and
diisopropylethylamine (2.85 mL, 16 mmol). The mixture
was stirred for 20 min, and then a solution of compound 8
(14.7 mmol) in dichloromethane (60 mL) was added. The
resulting mixture was stirred at room temperature until
starting material disappeared, and then quenched with aq.
NH4Cl solution at 0 8C. The aqueous phase was extracted
with dichloromethane (60 mL!3). The organic phases
were combined, washed with sat. aq. NH4Cl solution
(50 mL!3) and brine (50 mL) and dried with anhydrous
Na2SO4. After filtration, the filtrate was concentrated and
the residue was purified by silica gel chromatography (PE/
EA, 5/1, v/v) to give compound 9 (4.42 g, 84%) as an oil.5
[a]2D0ZK26.8 (c 1.05, CHCl3). IR: nmax 1014, 1176, 1270,
1367, 1496, 1648, 1709, 1745, 2961, 3327 cmK1. 1H NMR
(300 MHz, CDCl3): d 0.85–1.02 (m, 12H), 1.43 (s, 9H),
1.67–1.77 (m, 3H), 2.00 (m, 1H), 3.02 (s, 3H), 3.70 (s, 3H),
4.43 (dd, 1H, JZ6.9, 9.3 Hz), 5.22 (d, 1H, JZ8.7 Hz), 5.38
Scheme 7. Completion of total synthesis of cyclomarin C.

4936
S.-J. Wen et al. / Tetrahedron 61 (2005) 4931–4938
(dd, 1H, JZ6.9, 9.0 Hz) ppm. EI-MS (m/z): 302 (MCK56),
285 (MCK100).
was removed under reduced pressure and the residue was
purified by silica gel chromatography (PE/EA, 2/1, v/v) to
give compound 18 (500 mg, 80%) as white solid. IR: nmax
1
714, 1142, 1274, 1509, 1721, 1752, 2977, 3322 cmK1. H
4.1.2. Tripeptide 13. To a solution of compound 9 (700 mg,
1.96 mmol) in THF/H2O (10 mL, v/v 2:1) at 0 8C was added
LiOH$H2O (168 mg, 4 mmol). After 3 h, 4 N HCl was
added to adjust the pH to 2–3. The mixture was extracted
three times with ethyl acetate. The organic phases were
dried with anhydrous Na2SO4 and then concentrated to give
the crude product 10.5 The crude acid was dissolved in
dichloromethane (5 mL) and cooled with ice-water bath. To
it were added sequentially diisopropylethylamine (0.16 mL,
0.95 mmol), HOBt (83 mg, 0.61 mmol) and EDCI (119 mg,
0.61 mmol). After 15 min, a solution of 12 (0.47 mmol) in
dichloromethane (2 mL) was added dropwise. The resulting
mixture was stirred at 0 8C for 2 h, then allowed to warm to
room temperature and stirred for additional 10 h, and then
quenched with aq. NH4Cl solution at 0 8C. The aqueous
phase was extracted with dichloromethane. The organic
phases were combined, washed with sat. aq. NH4Cl solution
and brine (50 mL) and dried with anhydrous Na2SO4. After
concentration, the residue was purified by chromatography
(PE/EA, 5/1, v/v) to give compound 13 (208 mg, 89%) as an
oil. [a]2D0ZK56.7 (c 1.1, CHCl3). IR: nmax 1015, 1174,
1367, 1518, 1630, 1685, 1743, 2933, 2964, 3315 cmK1. 1H
NMR (300 MHz, CDCl3): d 0.82–1.02 (m, 15H), 1.43 (s,
9H), 1.52–1.70 (m, 3H), 1.65 (s, 6H), 1.92–2.05 (m, 1H),
2.75–2.85 (m, 1H), 3.01 (s, 3H), 3.71 (s, 3H), 4.39–4.45 (m,
2H), 4.86 (d, 1H, JZ9.6 Hz), 5.15–5.22 (m, 2H), 6.34 (d,
1H, JZ8.1 Hz) ppm. ESI-MS (m/z): 498.3 (MHC).
NMR (300 MHz, CDCl3): d 1.11 (d, 3H, JZ6.6 Hz), 1.20
(d, 3H, JZ7.2 Hz), 1.47 (s, 9H), 1.70–2.18 (m, 3H), 2.77 (s,
3H), 3.26 (s, 3H), 3.75 (s, 3H), 4.20–4.30 (m, 2H), 4.40–
4.60 (m, 2H), 4.70–4.82 (m, 2H), 6.37–6.74 (m, 2H), 7.20–
7.40 (m, 5H), 7.40–7.50 (m, 2H), 7.54–7.60 (m, 1H), 8.04
(d, JZ7.8 Hz, m, 2H) ppm. ESI-MS (m/z): 628.3 (MHC).
Anal. Calcd for C33H45N3O9: C, 63.14; H, 7.23; N, 6.69.
Found: C, 63.08; H, 7.24; N, 6.55.
4.1.5. Hexapeptide 20. Acid 19 (0.30 mmol) derived from
ester 18 by saponification was dissolved in dichloromethane
(3 mL) and cooled with ice-water bath. To it were added
sequentially diisopropylethylamine (52 mL, 0.30 mmol),
HOBt (49 mg, 0.36 mmol) and EDCI (70 mg, 0.36 mmol).
After 15 min, a solution of an amine (0.30 mmol, prepared
from 13 by treatment with TFA) in dichloromethane (3 mL)
was added dropwise. The resulting mixture was stirred at
0 8C for 2 h, then at room temperature overnight, and then
quenched with aq. NH4Cl solution at 0 8C. The aqueous
phase was extracted with dichloromethane. The combined
organic phase was washed with sat. aq. NH4Cl solution and
brine, and dried with anhydrous Na2SO4. After concen-
tration, the residue was purified by chromatography (PE/
EA, 1/1, v/v) to give compound 20 (186 mg, 63%) as a solid.
[a]2D0ZK75.9 (c 0.7, CHCl3). IR: nmax 713, 1100, 1274,
1522, 1627, 1655, 1689, 1723, 1745, 2933, 2966, 3310 cmK1
.
1H NMR (300 MHz, CDCl3): d 0.90–1.00 (m, 18H), 1.13 (d,
3H, JZ6.3 Hz), 1.50 (s, 9H), 1.58–1.78 (m, 4H), 1.66 (s,
6H), 1.88–2.16 (m, 3H), 2.80 (s, 3H), 2.78–2.82 (m, 1H),
2.99 (s, 3H), 3.31 (s, 3H), 3.70 (s, 3H), 4.26 (d, 2H, JZ
4.5 Hz), 4.08–4.30 (m, 2H), 4.60–4.92 (m, 6H), 5.18–5.25
(m, 1H), 6.34 (d, 1H, JZ9.3 Hz), 6.70–6.79 (m, 2H), 7.13–
7.38 (m, 5H), 7.42–7.50 (m, 2H), 7.55–7.60 (m, 1H), 8.08
(d, 2H, JZ7.8 Hz) ppm. HR-MS (ESI, m/z) for C53H80N6-
O12NaC: 1015.5726, found: 1015.5719.
4.1.3. Dipeptide 16. To a mixture of compound 15
(3.2 mmol) and N-Boc-L-Ala-OH (725 mg, 3.8 mmol) in
dichloromethane (15 mL) at 0 8C were added sequentially
diisopropylethylamine (0.75 mL, 4.18 mmol), HOBt
(569 mg, 4.18 mmol) and EDCI (818 mg, 4.18 mmol).
The reaction mixture was stirred at the same temperature
for 2 h, additional 1 h at room temperature, and then
quenched with aq. NH4Cl solution at 0 8C. The aqueous
phase was extracted with dichloromethane (10 mL!3). The
combined organic phase was washed with sat. aq. NH4Cl
solution (7 mL!3) and brine (7 mL), and dried with
anhydrous Na2SO4. After filtration, the filtrate was con-
centrated and the residue was purified by silica gel
chromatography (PE/EA, 4/1, v/v) to give compound 16
(1.07 g, 88%) as an oil. [a]2D0ZK52.2 (c 1.4, CHCl3). IR:
nmax 716, 1098, 1178, 1555, 1676, 1749, 2933, 3271,
4.1.6. Heptapeptide 4. To a solution of acid 22 (94 mg,
102 mmol) in dichloromethane (2 mL) were added sequen-
tially diisopropylethylamine (36 mL, 0.20 mmol), HOBt
(28 mg, 0.20 mmol) and EDCI (40 mg, 0.20 mmol). After
15 min, a solution of compound 21 (70 mmol, prepared from
20 by treatment with TFA) in dichloromethane (2 mL) was
added dropwise. The resulting mixture was stirred at 0 8C
for 2 h, then at room temperature overnight, and then
quenched with aq. NH4Cl solution at 0 8C. The aqueous
phase was extracted with dichloromethane. The combined
organic phases were washed with sat. aq. NH4Cl solution
and brine, and dried with anhydrous Na2SO4. After
concentration, the residue was purified by chromatography
(CH2Cl2/MeOH, 100/1, v/v) to give compound 4 (62 mg,
59%) as a solid. IR: nmax 742, 1099, 1211, 1453, 1529, 1631,
1
3364 cmK1. H NMR (300 MHz, CDCl3): d 1.25 (m, 3H),
1.45 (s, 9H), 3.28 (s, 3H), 3.76 (s, 3H), 4.13 (m, 1H), 4.73–
4.85 (m, 3H), 6.70 (d, 1H, JZ8.7 Hz), 7.25–7.37 (m,
5H) ppm. ESI-MS (m/z): 403.2 (MCNaC).
4.1.4. Tripeptide 18. To a solution of 17 (438 mg,
1.2 mmol) in dichloromethane (5 mL) at 0 8C were added
successively HOBt (177 mg, 16 mmol), EDCI (254 mg,
1.3 mmol), and diisopropylethylamine (0.80 mL, 4.4 mmol)
and then a solution of dipeptide 16 derived amine (1 mmol)
in dichloromethane (5 mL). After stirred at 0 8C for 2 h and
then room temperature 2 h, the reaction mixture was
quenched with aq. NH4Cl solution at 0 8C. The aqueous
phase was extracted with dichloromethane. The combined
organic phase was washed with sat. aq. NH4Cl solution and
brine, and then dried with anhydrous Na2SO4. The solvent
1
1686, 1722, 2928, 2960, 3310 cmK1. H NMR (300 MHz,
CDCl3): d K0.24 (m, 3H), 0.15 (s, 3H), 0.78–1.00 (m, 27H),
1.15 (d, 3H, JZ6.9 Hz), 1.52–1.74 (m, 16H), 1.90–2.10 (m,
3H), 2.64 (s, 3H), 2.80 (m, 1H), 2.91 (s, 3H), 3.10 (s, 3H),
3.70 (s, 3H), 4.22–4.47 (m, 5H), 4.50–4.92 (m, 8H), 5.00–
5.35 (m, 5H), 5.80–6.12 (m, 2H), 6.38 (d, 1H, JZ8.7 Hz),
6.62 (d, 1H, JZ7.2 Hz), 7.00–7.66 (m, 18H), 7.72–7.82 (m,
2H), 7.90 (d, 1H, JZ9.0 Hz), 8.00 (t, 2H, JZ7.5 Hz), 8.13

S.-J. Wen et al. / Tetrahedron 61 (2005) 4931–4938
4937
(m, 1H) ppm. HR-MS (ESI, m/z) for C85H114N8O14SiNaC:
1521.8116, found: 1521.8107.
(6 mL, 34 mmol), HOAt (4 mg, 29 mmol) and EDCI (6 mg,
31 mmol). After 15 min, a solution of compound 26
(24 mmol, prepared from 25 by treatment with piperidine)
in dichloromethane (1 mL) was added dropwise. The
resulting mixture was stirred at 0 8C for 2 h, then at room
temperature for 5 h, and then quenched with aq. NH4Cl
solution at 0 8C. The aqueous phase was extracted with
dichloromethane. The combined organic phases were
washed with sat. aq. NH4Cl solution and brine, and dried
with anhydrous Na2SO4. After concentration, the residue
was purified by chromatography (PE/EA, 2/1, v/v) to give
4.1.7. Compound 24. To a solution of 4 (100 mg, 67 mmol)
in dichloromethane (5 mL) was added piperidine (0.5 mL)
at 0 8C. The mixture was stirred for 2 h and then
concentrated. The residue was purified by flash chromato-
graphy (PE/EA, 2:1 then CH2Cl2/MeOH, 20/1) to give the
de-Fmoc product (66 mg, 77%). The product obtained
above was dissolved in THF/H2O (8 mL, 3:1) and cooled
with ice-water bath. To it was added LiOH$H2O (9 mg,
0.21 mmol), and the resulting mixture was stirred for 9 h,
neutralized with 1 N HCl to pH 7 and then extracted with
ethyl acetate. The combined organic phases were dried with
anhydrous Na2SO4, filtered and concentrated. The residue
was purified by chromatography (CH2Cl2/MeOH, 100/1,
40/1 and then 10/1) to give 23 (16 mg, 25%) as a solid. 23
(15 mg, 12 mmol) was dissolved in dichloromethane
(10 mL), and was added slowly via syringe pump to a
mixture of BOP (16 mg, 36 mmol) and diisopropylethyla-
mine (7 mL, 40 mmol) in dichloromethane (10 mL) over
12 h. After complete addition, the reaction mixture was
stirred for additional 4 h and then quenched with aq. NH4Cl
solution at 0 8C. The aqueous phase was extracted with
dichloromethane. The combined organic phases were
washed with sat. aq. NH4Cl solution and brine, and dried
with anhydrous Na2SO4. After concentration, the residue
was purified by chromatography (PE/EA 4/1, 2/1, 1/1) to
give compound 24 (4 mg). ESIMS (m/z): 1263.7 (MHC).
1
compound 6 (23 mg, 60%) as a solid. H NMR (300 MHz,
d6-DMSO): d K0.38 (s, 3H), K0.07 (s, 3H), 0.64 (s, 9H),
0.89–0.98 (m, 15H), 1.12 (d, 3H, JZ7.5 Hz), 1.24 (m, 3H),
1.41 (s, 3H), 1.52 (s, 3H), 1.60 (s, 3H), 1.63 (s, 3H), 1.40–
1.70 (m, 4H), 1.92–2.10 (m, 3H), 2.80 (m, 1H), 2.93 (s, 3H),
3.10 (s, 3H), 3.34 (s, 3H), 3.50 (s, 3H), 3.85–4.34 (m, 8H),
4.45 (dd, 2H, JZ4.5, 9.0 Hz), 4.66 (d, 1H, JZ4.8 Hz),
4.82–5.10 (m, 6H), 5.38–5.50 (m, 1H), 6.10 (dd, 1H, JZ
11.1, 18.6 Hz), 6.90–7.04 (m, 3H), 7.20–7.54 (m, 10H),
7.54–7.80 (m, 7H), 7.80–8.00 (m, 6H), 8.40–8.52 (m,
1H) ppm. ESI-MS (m/z): 1520.8 (MCNHC4 ), 1521.7 (MC
NaC).
4.1.10. Dipeptide 29. The title compound was prepared
according to the procedure similar to compound 6:
[a]2D0ZK38.2 (c 1.2, CHCl3). IR: nmax 741, 838, 1071,
1274, 1452, 1721, 2929, 2955, 3310 cmK1 1H NMR
.
(300 MHz, CDCl3): d K0.27 (s, 3H), 0.04 (s, 3H), 0.84 (s,
9H), 0.98 (d, 3H, JZ6.6 Hz), 1.60–1.70 (m, 1H), 1.70 (d,
6H, JZ3.9 Hz), 1.80–1.92 (m, 1H), 1.98–2.10 (m, 1H), 2.71
(s, 3H), 3.46 (s, 3H), 4.04–4.38 (m, 5H), 5.10 (d, 1H, JZ
17.1 Hz), 5.14–5.25 (m, 4H), 5.79 (d, 1H, JZ8.1 Hz), 6.15
(dd, 1H, JZ10.2, 17.1 Hz), 7.02–7.14 (m, 2H), 7.22–7.62
(m, 11H), 7.71–7.82 (m, 3H), 8.01 (d, 2H, JZ6.9 Hz) ppm.
ESI-MS (m/z): 908.4 (MCNaC), 754.3 (MCK131). Anal.
Calcd for C52H63N3O8Si: C, 70.48; H, 7.17; N, 4.74. Found:
C, 70.62; H, 7.42; N, 4.49.
4.1.8. Heptapeptide 6. To a solution of compound 13
(30 mg, 60 mmol) in THF/H2O (2.5 mL, v/v 4:1) at 0 8C was
added LiOH$H2O (5 mg, 120 mmol). After 4 h, 4 N HCl
was added to adjust the pH to 3–4. The mixture was
extracted three times with ethyl acetate. The organic phases
were combined, dried with anhydrous Na2SO4, and then
concentrated to give the crude product. The crude acid
(29 mg) was dissolved in dichloromethane (2 mL) and
cooled with ice-water bath. To it was added TFA (0.5 mL),
and the resulting mixture was stirred for 2 h, and then
concentrated. The residue was dissolved in THF (2 mL) and
neutralized with sat. aq. NaHCO3 at 0 8C, then solid sodium
carbonate (13 mg, 120 mmol) and Fmoc-Osu (24 mg,
63 mmol) were added. After 3 h, 0.5 N HCl was added to
adjust the pH to 2–3. The aqueous phase was extracted with
ethyl acetate (10 mL!3). The organic phases were
combined, and dried with anhydrous Na2SO4. After
concentration, the residue was purified by chromatography
(PE/EA, 4/1, then CH2Cl2/MeOH, 20/1) to give dipeptide-
acid 27 (38 mg, 100%).
4.1.11. Compound 31. To a solution of 29 (22 mg,
25 mmol) in dichloromethane (2 mL) at 0 8C was added
dropwise piperidine (0.2 mL). The mixture was stirred for
1 h, and then concentrated. The residue was purified by
chromatography (PE/EA, 3/1 then CH2Cl2/MeOH, 50/1) to
give compound 31 (11 mg, 67%) as a solid. 1H NMR
(300 MHz, CDCl3): d K0.16 (s, 3H), K0.06 (s, 3H), 0.89 (s,
9H), 1.10 (d, 3H, JZ6.6 Hz), 1.62–1.70 (m, 1H), 1.72 (s,
3H), 1.76 (s, 3H), 2.30–2.42 (m, 1H), 2.58–2.69 (m, 1H),
3.02 (s, 3H), 3.88 (d, 1H, JZ8.7 Hz), 4.12–4.20 (m, 2H),
4.33 (d, 1H, JZ10.5 Hz), 5.13 (d, 1H, JZ17.4 Hz), 5.23 (d,
1H, JZ10.8 Hz), 5.64 (s, 1H), 5.76 (d, 1H, JZ1.8 Hz), 6.12
(dd, 1H, JZ10.8, 17.4 Hz), 7.06–7.19 (m, 2H), 7.30 (s, 1H),
7.40–7.63 (m, 5H), 8.05 (d, 2H, JZ8.4 Hz) ppm. ESI-MS
(m/z): 500.3 (MCK131).
4.1.9. Compound 25. The title compound was prepared
according to the procedure similar to compound 4: 1H NMR
(300 MHz, CDCl3): d K0.31 to K0.23 (m, 3H), 0.14 (d, 3H,
JZ5.1 Hz), 0.81 (s, 9H), 0.85 (d, 3H, JZ6.6 Hz), 1.10 (d,
3H, JZ7.2 Hz), 1.50–1.80 (m, 9H), 2.63 (d, 2H, JZ4.8 Hz),
2.80 (d, 1H, JZ3.0 Hz), 3.10 (s, 2H), 3.21 (s, 1H), 3.60 (s,
2H), 3.70 (s, 1H), 4.10–4.38 (m, 6H), 4.60–4.76 (m, 5H),
5.00–5.32 (m, 2H), 5.81 (d, 1H, JZ4.5 Hz), 5.98–6.18 (m,
1H), 6.65–6.78 (m, 2H), 7.00–8.04 (m, 18H) ppm.
4.1.12. Cyclomarin C.3 [a]D20ZK72.8 (c 0.8, CHCl3); IR:
nmax 703, 742, 1095, 1376, 1458, 1509, 1649, 1686, 2873,
1
2962, 3310 cmK1. H NMR (500 MHz, CDCl3): d 0.62 (d,
3H, JZ6.3 Hz), 0.65 (m, 1H), 0.73 (d, 3H, JZ6.7 Hz), 0.82
(d, 3H, JZ6.6 Hz), 0.88 (d, 3H, JZ6.6 Hz), 0.93 (d, 3H,
JZ6.6 Hz), 1.05 (d, 3H, JZ6.3 Hz), 1.05 (m, 1H), 1.25 (s,
3H), 1.31 (d, 3H, JZ7.2 Hz),1.43 (m, 2H), 1.55 (s, 3H),
1.63 (m, 1H), 1.70 (s, 3H), 1.73 (s, 3H), 2.15–2.35 (m, 3H),
To a solution of acid 27 (19 mg, 31 mmol) in dichloromethane
(1 mL) were added sequentially diisopropylethylamine

4938
S.-J. Wen et al. / Tetrahedron 61 (2005) 4931–4938
2.42 (br s, 1H), 2.72 (s, 3H), 2.82 (s, 3H), 3.10–3.30 (m,
2H), 3.35 (s, 3H), 4.07 (t, 1H, JZ9.8 Hz), 4.30 (s, 1H), 4.40
(t, 1H, JZ8.7 Hz), 4.56 (br s, 1H), 4.77 (d, 1H, JZ10.0 Hz),
4.78–4.83 (m, 2H), 4.88 (t, 1H, JZ4.9 Hz), 4.82–4.93 (m,
1H), 5.07 (d, 1H, JZ5.2 Hz), 5.15 (d, 1H, JZ17.4 Hz), 5.22
(d,1H, JZ10.7 Hz), 5.31 (d, 1H, JZ3.0 Hz), 6.07 (dd, 1H,
JZ10.7, 17.4 Hz), 6.72 (d, 1H, JZ2.9 Hz), 7.05 (dd, 1H,
JZ7.2, 7.2 Hz), 7.12 (d, 1H, JZ5.3 Hz), 7.15–7.21 (m,
2H), 7.22–7.30 (m, 4H), 7.33 (s, 1H), 7.48 (d, 1H, JZ
8.3 Hz), 7.55 (d, 1H, JZ7.8 Hz), 7.95 (d, 1H, JZ7.7 Hz),
8.05 (d, 1H, JZ9.3 Hz), 8.17 (d, 1H, JZ10.3 Hz) ppm. 13C
NMR (125 MHz, CDCl3): d 17.8, 18.5, 18.9, 19.3, 20.0,
20.8, 22.4, 23.5, 25.1, 25.7, 27.9, 27.9, 29.3, 29.5, 30.9,
33.2, 33.4, 35.6, 38.9, 50.6, 52.7, 55.3, 55.9, 57.7, 58.1,
58.6, 59.3, 59.3, 66.4, 68.5, 80.0, 111.4, 113.8, 114.4, 118.7,
119.5, 121.5, 123.1, 124.8, 126.8, 127.9, 128.3, 128.6,
128.7, 134.5, 135.1, 135.8, 143.7, 168.4, 168.8, 169.6,
170.6, 171.1, 171.6, 172.6 ppm. ESI-MS (m/z): 1049.4
(MCNaC), 1009.5 (MHCKH2O).
References and notes
1. Renner, M. K.; Shen, Y. C.; Cheng, X. C.; Jensen, P. R.;
Frankmolle, W.; Kauffman, C. A.; Fenical, W.; Lobkovsky, E.;
Clardy, J. J. J. Am. Chem. Soc. 1999, 121, 11273–11276.
2. (a) Wen, S.-J.; Zhang, H.-W.; Yao, Z.-J. Tetrahedron Lett.
2002, 43, 5291–5294. (b) Sugiyama, H.; Shioiri, T.;
Yokokawa, F. Tetrahedron Lett. 2002, 43, 3489–3492.
(c) Tarver, J. E.; Joullie, M. M., Jr. J. Org. Chem. 2004,
69, 815–820.
3. Wen, S.-J.; Yao, Z.-J. Org. Lett. 2004, 6, 2721–2724.
4. See Supporting Information for Ref. 3 at http://pubs.acs.org/.
5. Hasuoka, A.; Nishikimi, Y.; Nakayama, Y.; Kamiyama, K.;
Nakao, M.; Miyagawa, K.-i.; Nishimura, O.; Fujino, M.
J. Antibiot. 2002, 55, 322–336.
6. Hoeg-Jensen, T.; Holm, A.; Sorensen, H. Synthesis 1996,
383–387.
7. Umemiya, H.; Kagechika, H.; Hashimoto, Y.; Shudo, K.
Nucleotides Nucleosides 1996, 15, 465–467.
8. Naegele, E.; Schelhaas, M.; Kuder, N.; Waldmann, H. J. Am.
Chem. Soc. 1998, 120, 6889–6902.
Acknowledgements
9. Waldmann, H.; Kunz, H. Liebigs Ann. Chem. 1983, 10,
1712–1725.
We appreciate the National Natural Science Foundation of
China (20425205 and 20321202), Ministry of Science and
Technology of China (G2000077502), Chinese Academy of
Sciences (KGCX2-SW-209) and Shanghai Municipal
Committee of Science and Technology (04DZ14901) for
financial support of this project.
10. Kiho, T.; Nakayama, M.; Yasuda, K.; Miyakoschi, S.;
Inukai, M.; Kogen, H. Bioorg. Med. Chem. Lett. 2003, 13,
2315–2318.
11. Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397.
12. Coste, J.; Le-Nguyen, D.; Castro, B. Tetrahedron Lett. 1990,
31, 205.