Total synthesis of proposed structure of coibamide A, a highly N- and O-methylated cytotoxic marine cyclodepsipeptide

Article abstract of DOI:10.1016/j.tetlet.2014.09.047

Total synthesis of the originally proposed structure of coibamide A, a highly N- and O-methylated cytotoxic marine cyclodepsipeptide, has been accomplished by using a [(4+1)+3+3]-peptide fragment-coupling strategy and careful examination and optimization

Full text of DOI:10.1016/j.tetlet.2014.09.047

Tetrahedron Letters 55 (2014) 6109–6112
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Total synthesis of proposed structure of coibamide A, a highly N- and
O-methylated cytotoxic marine cyclodepsipeptide
Wei He a, Hai-Bo Qiu a, Yi-Jie Chen b, Jie Xi b, Zhu-Jun Yao a,b,
⇑
a State Key Laboratory of Bioorganic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 345 Lingling Road, Shanghai
200032, China
b State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Nanjing University, 22 Hankou Road, Nanjing, Jiangsu 210093, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Total synthesis of the originally proposed structure of coibamide A, a highly N- and O-methylated cyto-
toxic marine cyclodepsipeptide, has been accomplished by using a [(4+1)+3+3]-peptide fragment-cou-
pling strategy and careful examination and optimization of the multiple dense N-methylated amide-
bond formations. The synthetic sample of the proposed coibamide A could not match the natural product
in both 1H and 13C NMR spectra, but was found to exhibit low micromolar cytotoxicity against the pro-
liferation of three tested cancer cells.
Received 12 August 2014
Accepted 9 September 2014
Available online 16 September 2014
Keywords:
Coibamide A
Cyclodepsipeptide
Anticancer
Ó 2014 Elsevier Ltd. All rights reserved.
N-Methylamino acid
Total synthesis
Marine cyanobacteria are a rich source of complex secondary
metabolites, and some of them have shown striking biological
activity1–4 and have consequently become attractive synthetic tar-
gets in the quest for new leads for pharmaceutical development.5
Coibamide A (1, Fig. 1) is a cyclodepsipeptide recently isolated by
McPhail and co-workers from the marine cyanobacterium Lep-
tolyngbya sp. collected from the Coiba National Park, Panama.6 It
was found to exhibit attractive low-nanomolar inhibitory activities
against the proliferation of various cancer cells, including MDA-
MB-231 (IC50 2.8 nM), LOX IMVI (IC50 7.4 nM), HL-60 (IC50
7.4 nM), and SNB-75 (IC50 7.6 nM). The significant inhibitory
potencies and possible new mechanism of action make coibamide
A to be a promising lead in anticancer drug discovery.6–8 For the
limited sources of marine natural products, establishment of a
practical chemical synthesis of coibamide A is of extreme value
to acquire sufficient sample for further pharmaceutical develop-
ment and exploration of its structure–activity relationship.
Full assignment of natural coibamide A was carried out when its
isolation by aid of NMRs and MS studies, as well as chiral HPLC
analysis of the degraded amino acids (Fig. 1).6 Among the amino-
acid components, the absolute configuration of threonine was pre-
dicted by the MM2 method and was further confirmed with ROESY
experiments. Coibamide A comprised a 22-membered macrocycle
with a high content of N- and O-methylamino acids (eight of its
eleven amino acids are N-methylated). Such a structural feature
may help to improve the metabolic stability and enhance the
hydrophobicity,9 however, synthesis of a macrocyclic peptide
containing multiple dense N-methylated amides is often full of
challenges. Besides lower efficiency of the amide couplings with
N-methylamino acids, common problems encountered in the syn-
thesis also include the side reactions of forming diketopiperazine,
epimerization, and the lability of N-alkylated peptides to acids.10,11
Despite many reports have been addressed the development of
proper reagents for the coupling of sterically hindered
N-methylamino acids,12 a satisfactory solution to resolve all the
above problems has not yet been found.
MeO
NH
Me
N
O
O
NMe
O
O
O
HN
O
OMe
Me
O
Me
O
O
N
N
O
N
N
Me
Me2N
N
Me
O
Me
O
O
MeO
Coibamide A (1)
Figure 1. Proposed structure of coibamide A.
⇑
Corresponding author. Tel./fax: +86 21 54925123.
E-mail address: yaoz@sioc.ac.cn (Z.-J. Yao).
http://dx.doi.org/10.1016/j.tetlet.2014.09.047
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.

6110
W. He et al. / Tetrahedron Letters 55 (2014) 6109–6112
Me
BocN
1
O
CO2Bn
Coibamide A (
)
1) Et2NH, CH3CN
1) 5 M HCl in dioxane
CO2Bn
O
N
2) HATU, DIPEA, DMF
2) HATU, DIPEA, DMF
OBn
Me
CO2H
N(Me)Fmoc
MeO
MeO
Site B
NH
CO2H
10
8
(74%)
N(Me)Boc
9
11
Me
N
N(Me)Boc
Site A
O
O
NMe
OBn
OBn
Me
O
O
O
O
CO2Bn
O
1) 5 M HCl in dioxane
Me
Me
O
O
BocN
N
O
N
N
BocN
Me
+
N
NMe
2) HATU, DIPEA, DMF
CO2H
O
Me
Me
N
N
CO2H
HN
O
O
O
O
Me
OMe
Me
MeO
O
O
Site C
N
Me
N
BnO2C
N(Me)Boc
N
N
Me
14
12
(40%)
(44%)
13
Me
Boc
O
2
O
OH
Me
BocN
O
1) 20% Pd(OH)2/C, H2
EtOAc
O
CO2Allyl
Me
EDCI, DMAP, DCM
CO2H
N
3
N
N
MeO
2) NaHCO3, DMF
Br
Me
Me
OMe
O
O
Site C
N(Me)Fmoc
O
H
N
5
Me
N
6 (80%)
HO2C
N
H
NHFmoc
O
MeN
O
N(Me)Fmoc
CO2Allyl
NMe
O
O
O
O
CO2H
O
O
O
O
Me
Me
N
N
7
N
Boc
N
FmocHN
O
N
Fmoc
Me
Me
O
5
O
OMe
Me
N
Boc
MeN
OH
CO2Allyl
N
15
(71%)
Scheme 1. Synthesis of tetrapeptide 6.
Boc
O
O
CO2Allyl
N
Me
O
Me
Me
MeN
N
N
N
O
Me
Me
O
O
4
6
Figure 2. Retrosynthetic analysis of coibamide A.
H
1) 5 M HCl in dioxane
BnO2C
N
CO2Bn
NHBoc
NMe
Boc
2) EDCI, HOBt
DIPEA, DCM
O
MeO
Our retrosynthetic analysis of coibamide A is shown in Figure 2.
At first, the side chain is disconnected from the macrocycle at the
Ser(OMe)-N-MeLeu junction, affording fragment 2 and macrocycle
3 having a residue of N-MeLeu, which would reduce the difficulty
of future segment-coupling with 2. Further disconnection of the
macrocycle 3 renders significant strategic importance and may dic-
tate the success of the synthesis. According to the common guide-
lines of peptide synthesis, a proper cyclization position should
avoid those sterically encumbered by the N-methylated amide
bond.10 Esterification is firstly excluded from the potential cycliza-
tion protocols, because unfavorable kinetics of lactone-formation
based cyclization (at site A) may imply a major risk of racemiza-
tion. For larger distribution of Z-amide conformer in the N-methyl-
ated peptides (compared to the non-N-methylated peptides),
cyclization at N-MeLeu-Tyr(OMe) junction (site B) would be also
unfavorable because of a high risk of 2,5-diketopiperizine forma-
tion resulting in the degradation of the linear peptide. Therefore,
a potentially suitable macrocyclization position is determined at
the N-MeIIe-Ala amide bond (site C), affording the linear precursor
4. Precursor 4 can be further broken into three fragments: Fmoc-N-
Me-Ala-OH (5), tetrapeptide 6, and tripeptide 7.
CO2H
16
MeO
17 (90%)
N(Me)Boc
13
O
H
10% Pd/C, H2
EtOAc
1) 5 M HCl in dioxane
BnO2C
N
NHFmoc
N
Me
2) BEP, DIPEA, DCM
CO2H
O
NHFmoc
MeO
19
(63%)
18
O
H
HO2C
N
NHFmoc
N
Me
O
MeO
7 (100%)
Scheme 2. Synthesis of tripeptide acid 7.
ester 15. Coupling with a single N-methylamino acid is also
advantageous in reducing the extent of epimerization during this
ester-bond formation.
The total synthesis began with the preparation of tetrapeptide 6
(Scheme 1). In order to overcome the possible epimerization of the
active ester of N-methylated amino acid component (via oxazolone
formation) during the couplings, stepwise coupling protocol was
applied in the synthesis.13 Coupling of N-Boc-N-Me-Ser(OMe) (9)
with the N-Fomc deblocked product derived from 8 yielded dipep-
tide 10. Removal of the N-Boc group of 10 with 5 M HCl in dioxane
followed by condensation with the active ester of N-Boc-N-Me-
Thr(OBzl) (11) afforded tripeptide12. Using a similar operation, tet-
rapeptide 14 was obtained after coupling with acid 13. Considering
the orthogonality of the protecting groups, the benzyl ester of 14
was altered with an allyl ester by hydrogenolysis of 14 and subse-
quent selective O-allylation with allyl bromide in the presence of
NaHCO3 in DMF. In order to reduce the steric hindrance of seg-
ments coupling, the resulting segment 6 was firstly coupled with
the single N-methylamino acid 5, providing the corresponding
Tripeptide fragment 7 was synthesized in a fashion from the
C-terminal to the N-terminal, beginning with the tyrosine deriva-
tive 16 (Scheme 2). Use of an N-Boc derivative 13 resulted in higher
yield and minimized DKP formation in the second step of coupling.
The benzyl ester protecting group of tripeptide 19 was then
removed by hydrogenolysis, affording tripeptide acid 7 in quantita-
tive yield.
To improve the coupling efficiency, we decided to synthesize
the side-chain fragment 26 (a derivative of 2, Fig. 2) from its
N-terminal to C-terminal (Scheme 3). We reasoned that the pres-
ence of ester bond could be exploited to decrease the extent of oxa-
zolone formation, which might otherwise lead to epimerization.
The chiral
a-hydroxy acid 21 was prepared readily from L
-Val.14
Esterification of alcohol 21 with acid 20 was carried out in the
presence of EDCI and a catalytic amount of DMAP, affording ester
22 in quantitative yield. Removal of the O-allyl group of 22 with

W. He et al. / Tetrahedron Letters 55 (2014) 6109–6112
6111
with DEPBT and DIPEA in THF at À20 °C (entry 5).17 Though
increasing the reaction temperature (entries 5–7) could improve
the yield, the extent of epimerization was also increased. Using
the optimized conditions, we were able to synthesize the linear
precursor 4 in 43% isolated yield over two steps with 99% HPLC
purity.
CO2Allyl
O
EDCI, DMAP
DCM
NH(Me) 24
O
CO2R
CO2H
NHCbz
20
CbzHN
BEP, DIPEA, DCM
O
CO2Allyl
22 (R = allyl, 100%)
Pd(PPh3)4
OH
21
23
(R = H, 100%)
DMBA, THF
After removing O-allyl and N-Fmoc groups from the linear pre-
cursor 4, final cyclization was carried out in the presence of EDCI,
HOAt, and DIPEA in DCM/DMF (10:1), affording 28% yield of prod-
uct 3 and a dimerized by-product (ꢀ29% yield) (Scheme 4). Use of
other condensation reagents18 and reducing the reaction concen-
tration could not reduce the ratio of dimer and improve the results.
The N-Boc protecting group was then removed from the macrocy-
cle 3, and the resulting free amine was immediately coupled to
acid 26 using DEPBT as the coupling reagent, affording the whole
skeleton 27.19 Finally, hydrogenolysis of the N-Cbz group of 27 fol-
lowed by in situ reductive amination with formaldehyde afforded
the proposed structure of coibamide A (1).
O
O
O
Pd(PPh3)4
DMBA, THF
O
O
O
HN
Cbz
N
CO2Allyl
HN
Cbz
N
CO2H
Me
(94%)
Scheme 3. Synthesis of side-chain acid 26.
Me
O
O
26
(99%)
25
Table 1
Screening conditions for the coupling of 7 and 15
In order to confirm the stereochemical configuration of cyclic
peptide 3, we decided to re-synthesize the same macrocycle using
different synthetic sequences. The Tyr(OMe)-N-MeAla amide
bond was chosen as the alternative cyclization site, and the N-
MeIIe-Ala junction was applied for the coupling of segments. To
our delight, the same cyclic peptide 3 was finally afforded in a
O
N(Me)Fmoc
O
O
Boc
MeN
O
O
CO2Allyl
Me
N
N
N
Me
Me
O
15
1) Et2NH, MeCN
low yield (15%) along with epi-3 (46%, epimerized at the a-posi-
tion of Tyr) through the new route (Scheme 5). Generation of a
7
2) , DEPBT
DIPEA, THF, -20 °C
O
O
H
N
H
N
Me
N
Me
N
O
NMe
NMe
O
O
O
O
O
O
O
O
+
O
O
NH
Me
FmocHN
O
FmocHN
O
N
O
O
Me
N
Me
N
O
O
NMe
CO2Allyl
CO2Allyl
1) Pd(PPh3)4, DMBA, THF
2) Et2NH, CH3CN
1) TFA/DCM (1:10)
Et3SiH, 0 o
Me
N
Me
N
N
N
O
N
N
O
N
Me
Me
C
Me
Me
Boc
Boc
O
O
O
4
3) EDCI, HOAt, DIPEA,
DCM/DMF (10:1)
O
O
26
2) , DEPBT, DIPEA
HN
O
O
THF, rt
Me
O
N
N
Me
N
epi-4
4
(43%, 99% HPLC purity)
Me
Me
O
Boc
Entry Conditions
4/epi-4
O
Combined yielda (%) HPLC ratiob
3
(28%)
O
1
2
3
4
5
6
7
8
9
10
PyBOP, NMM, DCM, rt
BOP, NMM, DCM, rt
59
67
28
62
43
75
89
13
75
61:39
61:39
95:5
66:34
99:1
BEP, DIPEA, DCM, À15 °C to rt
HATU, DIPEA, DCM, 0 °C to rt
DEPBT, DIPEA, THF, À20 °C
DEPBT, DIPEA, THF, 0 °C
DEPBT, DIPEA, THF, rt
Bop-Cl, DIPEA, DCM, rt
EDCI, HOAt, DIPEA, DCM, rt
EDCI, HOAt, DIPEA, DCM, À20 °C 53
NH
O
Me
N
O
NMe
93:7
O
O
H2, aq. HCHO
87:13
81:19
63:37
75:25
O
O
CbzHN
O
cat. 20% Pd(OH)2/C
MeOH
HN
O
O
O
O
O
Me
O
N
Me
N
N
N
Me
N
a
Me
O
Isolated yields over 2 steps.
HPLC conditions: C18 column, k = 220 nm, isocratic 90% CH3CN/H2O at a flow
Me
b
O
rate of 1.0 mL/min.
27 (50%)
O
O
NH
1 mol % Pd(PPh3)4 provided quantitative yield of acid 23, which
was further coupled with the freshly prepared amine 24 with
BEP (2-bromo-1-ethylpyridinium tetrafluoroborate)15 and DIPEA
in DCM, affording 25 in 94% yield. Deprotection of the O-allyl group
of 25 gave the required acid 26.
With segments 7 and 15 in hand, we continued to synthesize
the key linear precursor 4 for the macrocyclization. Surprisingly,
low efficiency was observed in the coupling between acid 7 and
the amine derived from 15, and the degree of epimerization was
also increased. After screening of various coupling conditions
(Table 1),16 the two segments was finally jointed upon treatment
Me
N
O
O
NMe
O
N
O
HN
O
O
O
O
Me
O
O
N
Me
N
O
O
N
Me
Me2N
N
Me
O
Me
O
1 (65%)
Scheme 4. Total synthesis of the proposed structure of coibamide A (1).

6112
W. He et al. / Tetrahedron Letters 55 (2014) 6109–6112
Acknowledgments
O
N(Me)Fmoc
1) Pd(PPh3)4, DMBA, THF
O
O
This work was financially supported by the Ministry of Science
and Technology of the People’s Republic of China (2010CB833200,
2013AA092903) and the National Natural Science Foundation of
China (21032002).
O
O
CO2Allyl
2) 28, DEPBT, DIPEA, THF, -20 °C
Me
N
Me
N
Boc
N
N
Me
Me
O
H
O
RHN
N
CO2Bn
N
Me
15
O
O
19
(R = Fmoc)
Et2NH,
CH3CN
Supplementary data
28 (R = H)
O
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2014.09.
047.
BnO2C
Me
NH
Fmoc
O
N
1) 10% Pd/C,
EtOAc
O
NMe
O
O
2) Et2NH,
CH3CN
O
epi-3
(46%)
References and notes
3
+
HN
3) EDCI, HOAt, (15%)
DIPEA,
DCM:DMF
(10:1)
O
Me
O
N
N
Me
1. Simmons, T. L.; Coates, R. C.; Clark, B. R.; Engene, N.; Gonzalez, D.; Esquenazi,
E.; Dorrestein, P. C.; Gerwick, W. H. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 4587–
4589.
2. Bull, A. T.; Stach, J. E. M. Trends Microbiol. 2007, 15, 481–499.
3. Luesch, H.; Harrigan, G. G.; Goetz, G.; Horgen, F. D. Curr. Med. Chem. 2002, 9,
1791–1806.
N
Me
N
Me
Boc
O
O
29
(57%)
4. Nunnery, J. K.; Mevers, E.; Gerwick, W. H. Curr. Opin. Biotechnol. 2010, 21, 787–
793.
Scheme 5. Alternative route to the cyclized product 3.
5. Rawat, D. S.; Joshi, M. C.; Joshi, P.; Atheaya, H. Anti-Cancer Agents Med. Chem.
2006, 6, 33–40.
6. Medina, R. A.; Goeger, D. E.; Hills, P.; Mooberry, S. L.; Huang, N.; Romero, L. I.;
Ortega-Barría, E.; Gerwick, W. H.; McPhail, K. L. J. Am. Chem. Soc. 2008, 130,
6324–6325.
7. McPhail, K. L.; Medina, R. A.; Gerwick, W. H.; Goeger, D. E.; Capeon, T. L. US
08034780, 2011.
8. Hau, A. M.; Greenwood, J. A.; Löhr, C. V.; Serrill, J. D.; Proteau, P. J.; Ganley, I. G.;
McPhail, K. L.; Ishmael, J. E. PLOS ONE 2013, 8, 1–16.
9. Chatterjee, J.; Gilon, C.; Hoffman, A.; Kessler, H. Acc. Chem. Res. 2008, 41, 1331–
1342.
10. Humphrey, J. M.; Chamberlin, A. R. Chem. Rev. 1997, 97, 2243–2266.
11. Teixidó, M.; Albericio, F.; Giralt, E. J. Peptide Res. 2005, 65, 153–166.
12. For reviews on coupling reagents, see: (a) El-Faham, A.; Albericio, F. Chem. Rev.
2011, 111, 6557–6602; (b) Valeur, E.; Bradley, M. Chem. Soc. Rev. 2009, 38, 606–
631; (c) Prasad, K. V. S. R. G.; Bharathi, K.; Haseena, B. B. Int. J. Pharm. Sci. Rev.
Res. 2011, 8, 108–119.
13. Synthesis of 6 by sequential N?C extension afforded the racemic product, and
it was less efficient than the route by an C?N sequence.
14. Liang, B.; Portonovo, P.; Vera, M. D.; Xiao, D.; Joullié, M. M. Org. Lett. 1999, 1,
1319–1322.
common product 3 through two different routes (Schemes 4 and
5) logically excluded the possibility of epimerization during the
previous macrocyclization (Scheme 4), and also verified the ste-
reochemical structure of 3.
Unfortunately, analytical data of synthetic sample of 1 were
inconsistent with those reported for natural coibamide A.6 Signifi-
cant differences were observed in both 1H and 13C NMR spectra of
our synthetic sample from those of natural product. Based on the
results from our study, further determination and reassignment
of the full structure of coibamide A would be needed.20 Interest-
ingly, some of the synthetic samples were found to exhibit low
micromolar cytotoxicities against three cancer cell lines in our bio-
logical screenings, including the final product 1 (IC50 17.98
lM
(MDA-MB-231); 11.77
the cyclic intermediate 3 (IC50 8.30
(MCF-7); and 16.79 M (A549)).
l
M (MCF-7) and 22.80
l
M (A549)) and
15. Li, P.; Xu, J. C. Tetrahedron 2000, 56, 8119–8131.
16. Colucci, W. J.; Tung, R. D.; Petri, J. A.; Rich, D. H. J. Org. Chem. 1990, 55, 2895–
2903.
lM (MDA-MB-231); 8.35 l
M
l
In summary, a total synthesis of the literature structure of coi-
bamide A has been accomplished, for the first time, using a conver-
gent [(4+1)+3+3]-peptide fragment-coupling strategy. Formations
of multiple dense N-methylated amide-bonds of this unique mar-
ine natural cyclodepsipeptide were carefully examined and opti-
mized to avoid possible epimerization during the fragment
couplings. Reverse-sequence synthesis of the linear peptides and
application of an alternative macro-cyclization site have been pro-
ven to be two useful tools for examination of the alpha-carbon
racemization of the active esters involved in the synthesis. Though
the final product 1 could not match the natural product in both 1H
and 13C NMR spectra, the synthetic sample of the proposed coiba-
mide A and a late-stage cyclic intermediate was identified to exhi-
bit low-micromolar inhibitory activity against the proliferation of
three tested cancer cells.
17. (a) Nagam, A. C.; Radford, S. E.; Warriner, S. L. Synlett 2007, 2517–2520; (b) Wu,
X. Y.; Stockdill, J. L.; Wang, P.; Danishefsky, S. J. J. Am. Chem. Soc. 2010, 132,
4098–4100; (c) Wu, X. Y.; Stockdill, J. L.; Park, P. K.; Danishefsky, S. J. J. Am.
Chem. Soc. 2012, 134, 2378–2384.
18. Representative cyclization conditions in the syntheses of cyclic peptides, see:
(a) Sleebs, M. M.; Scanlon, D.; Karas, J.; Maharani, R.; Hughes, A. B. J. Org. Chem.
2011, 76, 6686–6693; (b) Wen, S.-J.; Yao, Z.-J. Org. Lett. 2004, 6, 2721–2724; (c)
Marcucci, E.; Tulla-Puche, J.; Albericio, F. Org. Lett. 2012, 14, 612–615; (d)
Suenaga, K.; Mutou, T.; Shibata, T.; Itoh, T.; Fujita, T.; Takada, N.; Hayamizu, K.;
Takagi, M.; Irifune, T.; Kigoshi, H.; Yamada, K. Tetrahedron 2004, 60, 8509–
8527.
19. An alternative stepwise synthesis of 27 has also been completed from the
linear precursor 4 using same condensation conditions to couple the side chain
and elongate the side chain. Because the two synthesis routes afforded the
same product 27, we concluded that epimerization did not happen during the
side-chain couplings with DEPBT and DIPEA in THF.
20. For inadequate evidences, we are unable to conclude any structural revision at
this moment referring to the NMR differences between the proposed structure
(the synthesized 1 from this work) and the natural product.