Total Synthesis of Microsclerodermin E

Article abstract of DOI:10.1002/anie.200352423

Succumbing to total synthesis: Microsclerodermin E, a 23-membered cyclic hexapeptide possessing antifungal activity, has been prepared for the first time. Key features in the synthesis include the construction of four unusual amino acid units, the assembl

Full text of DOI:10.1002/anie.200352423

Communications
shown to display several unique features, including a 23-
membered cyclic hexapeptide core, featuring a very complex
b-amino acid and three other unnatural amino acid residues.
Preliminary studies have indicated that all these compounds
possess intriguing antifungal activities and some of them show
potent antitumor properties,[2,3] but further evaluation of their
biological activities has been hampered by their relative
scarcity. The biological activities and the structural complex-
ity exhibited by the microsclerodermins have attracted much
attention within the community of synthetic chemists. How-
ever, although some synthetic efforts towards these com-
pounds have been reported,[4] no member of this family has
been synthesized yet. Herein we wish to describe our total
synthesis of microsclerodermin E. Notable elements of the
synthesis are the concise assembly of its b-amino acid
fragment, the assembly of the pyrrolidinone fragment with
little racemization, and the effective macrocyclization at the
Gly-GABOB site.
The retrosynthetic blueprint that guided this new cam-
paign is defined in Figure 1. The macrocyclic core of the target
molecule is disconnected into two fragments, the tetrapeptide
Hexapeptide Total Synthesis
Total Synthesis of Microsclerodermin E**
Jidong Zhu and Dawei Ma*
Lithistid sponges are well-known among marine organisms
for their ability to produce a diverse array of biologically
active metabolites.[1] From several species of the lithistid
sponge Microscleroderma sp. and Theonella sp., Faulkner and
co-workers have isolated a series of cyclic peptides named
microsclerodermins A–I.[2] Structurally, each of them was
Figure 1. Retrosynthetic analysis of microsclerodermin E. PG=protect-
ing group.
portion and the dipeptide unit. Further disconnections of
these two parts reveals that we need to employ Gly and N-
MeGly as the starting materials and develop some efficient
methods for assembling four unusual amino acid units, which
include 3-amino-10-(p-ethoxyphenyl)-2,4,5-trihydroxydeca-
7,9-dienoic acid (AETD), 4-amino-3-hydroxybutanoic acid
(GABOB), (R)-Trp-2-carboxylic acid, and a pyrrolidinone
fragment.
Our approach towards the AETD fragment is shown in
Scheme 1 (see scheme legends for abbreviations). Treatment
of lithiated methyl phenyl sulfone in THF at À788C with
oxirane 2 (derived from d-gluconolactone as shown) in the
presence of BF3·Et2O[5] afforded the desired g-hydroxy
[*] Prof. Dr. D. Ma, J. Zhu
State Key Laboratory of Bioorganic and Natural Products Chemistry
Shanghai Institute of Organic Chemistry
Chinese Academy of Sciences
354 Fenglin Lu, Shanghai 200032 (China)
Fax: (+86)21-6416-6128
E-mail: madw@pub.sioc.ac.cn
[**] The authors are grateful to the Chinese Academy of Sciences, the
National Natural Science Foundation of China, and the Science and
Technology Commission of the Municipality of Shanghai (grant
02 JC14032) for their financial support. We thank Professor T. Shioiri
for his encouragement.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5348
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200352423
Angew. Chem. Int. Ed. 2003, 42, 5348 –5351

Angewandte
Chemie
Scheme 2. Reagents and conditions: a) TsCl, Py, 60%; b) NaN3, DMF,
808C, 75%; c) TBSCl, imidazole, 100%; d) aq. NaOH, MeOH, then
HOSu, EDCI, 100%; e) 8b, EtOAc, reflux, 96%; f) Dess–Martin oxida-
tion then NaClO2, resorcinol, NaH2PO4, 74%; g) HOSu, EDCI, 85%.
EDCI=1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride,
Su=succinimide.
was protected with TBSCl to give 10. After 10 was hydrolyzed
with NaOH, the resultant acid was converted into the
corresponding activated ester,[11] which when coupled with
the amine 8b, to provide amide 11a in 96% yield. The
hydroxy group in 11a was oxidated stepwise to the corre-
sponding acid by reaction with the Dess–Martin reagent[12]
and sodium chlorite, followed by esterification with HOSu
and EDCI[13] to give dipeptide 11b in 63% yield over three
steps; this intermediate was ready for connection with the
tetrapeptide part.
For the synthesis of the pyrrolidinone part, the starting d-
aspartic acid derivative 12 was converted into the urethane N-
carboxyanhydride 14 (Scheme 3),[14] and then treated with the
lithium enolate of trimethylsilylethyl (TMSE) acetate to give
b-keto ester 15. This operation is essential for obtaining
enantiopure 15 (> 98% ee determined by HPLC) because
attempts to use other activated esters prepared with N,N’-
carbonyldiimidazole,[4c] N-hydroxysuccinimide, and penta-
fluorophenol led to partial or full racemization of 15. After
hydrogenolysis of the benzyl group over Pd/C, the resultant
acid was converted into its mixed anhydride, which was
Scheme 1. Reagents and conditions: a) 2,2-dimethoxypropane,
acetone, TsOH, 80%; b) NaBH4/MeOH, 100%; c) TsCl, Py, 85%;
d) K2CO3, MeOH, 89%; e) PhSO2CH2Li, BF3·Et2O, À788C, 84%; f) 1n
HCl, THF then 2,2-dimethoxypropane, acetone, TsOH, 67%; g) MsCl,
DMAP, Et3N, 97%; h) NaN3, DMF, 1208C, 83%; i) Pd/C, H2 then
(CF3CO)2O, Py, 62%; j) nBuLi, THF, À788C then 6; k) PhCOCl, DMAP,
Et3N; l) Na(Hg), MeOH, 77% for 3 steps; m) PPTS, MeOH, 458C,
87%; n) TBSCl, imidazole, 92%; o) MOMCl, iPr2NEt, 100%; p) TBAF,
THF, 100%; q) NaBH4, EtOH, 100%. DMAP=4-dimethylaminopyri-
dine, DMF=dimethylformamide, MOM=methoxymethyl, Ms=metha-
nesulfonyl, PPTS=pyridinium p-toluenesulfonate, Py=pyridine,
TBAF=tetrabutylammonium fluoride, TBS=tert-butyldimethylsilyl,
Ts =toluenesulfonyl.
sulfone 3. Subsequent exposure of 3 to 1n HCl and
regioselective reprotection of the resultant alcohol with 2,2-
dimethoxypropane produced 4a in 67% yield. This approach
relied on the preferential formation of the acetonides with
less steric hindrance. Mesylation of the hydroxy function of
4a, followed by azidation of the resultant mesylate 4b
furnished azide 5a, which was hydrogenated over Pd/C, and
then treated with trifluoroacetic anhydride to provide amide
5b. Julia reaction[6] of 5b with a,b-unsaturated aldehyde 6
afforded the desired diene 7, together with its E/Z isomer in a
ratio of 5:1. After the terminal acetonide in 7 was selectively
removed with the assistance of PPTS,[7] the two liberated
hydroxy groups were protected selectively with TBSCl and
MOMCl, and the TBS ether was then cleaved with TBAF to
deliver amide 8a. Removal of the CF3CO group in 8a by
[8]
reaction with NaBH4 led to the free amine 8b, which was
Scheme 3. Reagents and conditions: a) Allyl bromide, KHCO3, DMF,
97%; b) (Boc)2O, DMAP, 100%; c) [RhCl(PPh3)3], 90% EtOH, 81%;
d) SOCl2, Py, DMF, 90%; e) LiCH2CO2TMSE, THF, À788C, 80%; f) Pd/
C, H2, 100%; g) iBuOCOCl, NMM; h) aq NH3; i) MsCl, Et3N, 81% for
3 steps; j) TBAF, TsOH then C6F5OH, EDCI, 73%. Boc=tert-butoxycar-
bonyl, NMM=N-methylmorpholine, TMSE=trimethylsilylethyl.
required for the coupling with the GABOB unit.
The GABOB unit 10 was constructed from methyl (R)-
3,4-dihydroxybutanoate 9[9] (Scheme 2). Selective tosyla-
tion[10] of the primary hydroxy group, followed by azidation
of the resultant monotosylate afforded a hydroxy azide, which
Angew. Chem. Int. Ed. 2003, 42, 5348 –5351
www.angewandte.org
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5349

Communications
treated with aqueous ammonia to give a hydroxypyrrolidi-
was hydrolyzed to the acid function under alkali conditions
and then esterified to afford the (R)-Trp-2-carboxylic acid
derivative 20.
none. Subjecting this intermediate to elimination through
mesylation provided pyrrolidinone 16 in 81% overall yield.
When the TMSE group was removed by treatment with
TBAF, the liberated acid was found to be racemized due to
the basic conditions directed by TBAF. The racemization
could be hampered in the presence of TsOH,[15] although the
best R/S ratio was 4:1. The free acid was converted to its PFP-
activated ester[16] 17 by treatment with C6F5OH and EDCI in
80% yield.
Construction of the (R)-Trp-2-carboxylic acid part com-
menced with introduction of a cyano group to the indole ring
of 18, which was derived from d-tryptophan according to
Danishefsky's protocol[17] (Scheme 4). The cyano group of 19
With the four unnatural amino acid fragments in hand, we
set out on the next step—coupling these units to complete the
synthesis of the macrocyclic core. The reaction was initially
carried out on compound 20. After catalytic removal of the
benzyl groups over Pd(OH)2, the free amine was coupled with
Boc-N-MeGly-OH promoted by EDCI–HOBt[18] to provide a
dipeptide (93%). Regioselective hydrolysis of the aliphatic
carboxylic ester with LiOH, followed by coupling with H-Gly-
OTMSE produced tripeptide 21. The Boc group was removed
by treatment with TFA, and without purification the free
amine was coupled with the PFP-activated ester 17 to afford
tetrapeptide 22 as a inseparable mixture (4:1), which resulted
from the partial racemization of 17. Removal of the Boc
group in 22 followed by coupling with the dipeptide 11b gave
the linear precursor 23 in 80% yield. The TMSE group of 23
was removed with TBAF and TsOH to give the free
carboxylic acid. Although attempts to reduce the azide
group to the amine failed when 23 was treated with Ph3P,
nBu3P, and SnCl2,[19] less sterically hindered Me3P[20] proved to
be suitable for this transformation and furnished the reduced
product in good yield.
The stage was now set for the crucial macrocyclization. To
our delight, although EDCI–HOAt was not effective for this
conversion, treatment of the above amino acid with DPPA[21]
for 14 days afforded lactam 24 in 40% yield (three steps from
23), together with its separable isomer (9% yield, resulting
from the partial racemization during assembly of the pyrro-
lidinone fragment). Noteworthy is that much lower yields
were observed when the reaction time was shortened to two
to three days with either DPPA or HATU[22] as the coupling
reagents. Removal of methyl ester was achieved by treatment
with LiOH to give the free acid, and the next task was the
removal of the MOM and acetonide group. This operation
proved far from trivial, mainly because both the substrate and
the product are unstable to strong acidic conditions. After
some experimentation, we finally observed that the removal
was accomplished with Amberlyst-15[23] to give 1[24] in 61%
yield (two steps). The analytical data of the synthetic
compound were identical with those reported,[2b] thereby
confirming the structure of microsclerodermin E proposed by
Faulkner and co-workers.
In summary, we have developed a new, efficient strategy
for the enantiospecific synthesis of microsclerodermins
exemplified by the first total synthesis of microscleroder-
min E (1% overall yield for 26 linear steps). The methods
expressed here for the synthesizing AETD and the pyrroli-
dinone parts should be applicable for installation of the other
polyhydroxy-b-amino acid residues and pyrrolidinone units in
the microsclerodermins. We are currently extending these
methods to the total synthesis of microsclerodermins A–D
and their analogues for in-depth biological evaluation.
Results will be reported in due course.
Scheme 4. Reagents and conditions: a) tBuOCl, Et3N, À788C then
TMSCN, BF3·Et2O, 79%; b) aq KOH, MeOH then MeI, KHCO3, DMF,
71%; c) Pd(OH)2, H2, MeOH, 70%; d) BocN(Me)CH2CO2H, EDCI,
HOBt, DIEA, 96%; e) aq LiOH, MeOH then NH2CH2CO2TMSE, EDCI,
HOBt, DIEA, 88%; f) TFA then 17, NaHCO3, DMF, 458C; 81%; g) TFA
then 11b, NaHCO3, MeCN, 80%; h) TBAF, THF; i) Me3P, THF;
j) DPPA, DMF, 40% for 3 steps; k) LiOH, MeOH, THF, then
Amberlyst-15, MeOH, H2O, 61%. DIEA=diisopropylethylamine,
DPPA=diphenylphosphoryl azide, HOBt=1-hydroxybenzotriazole,
TFA= trifluoroacetic acid.
Received: July 21, 2003 [Z52423]
5350
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2003, 42, 5348 –5351

Angewandte
Chemie
Keywords: amino acids · Julia reaction · macrocycles · peptides ·
.
total synthesis
[1] C. A. Bewley, D. J. Faulkner, Angew. Chem. 1998, 110, 2280;
Angew. Chem. Int. Ed. 1998, 37, 2162.
[2] a) C. A. Bewley, C. Debitus, D. J. Faulkner, J. Am. Chem. Soc.
1994, 116, 7631; b) E. W. Schmidt, D. J. Faulkner, Tetrahedron
1998, 54, 3043; c) A. Qureshi, P. L. Colin, D. J. Faulkner,
Tetrahedron 2000, 56, 3679.
[3] A. E. Wright, S. A. Pomponi, R. E. Longley, R. A. Isbrucker,
US 6384187, 2002 [Chem. Abstr. 2002, 136, 319381]
[4] a) S. Sasaki, Y. Hamada, T. Shioiri, Tetrahedron Lett. 1997, 38,
3013; b) S. Sasaki, Y. Hamada, T. Shioiri, Tetrahedron Lett. 1999,
40, 3187; c) S. Sasaki, Y. Hamada, T. Shioiri, Synlett 1999, 4, 453.
[5] S. Marczak, J. Wicha, Synth. Commun. 1990, 20, 1511.
[6] S. L. Schreiber, H. V. Meyers, J. Am. Chem. Soc. 1988, 110, 5198.
[7] R. Van Rijsbergen, R. M. J. O. Anteunis, A. De Bruyn, J.
Carbohydr. Chem. 1983, 2, 395.
[8] Z. H. Kudzin, P. Lyzwa, J. Luczak, G. Andrijewski, Synthesis
1997, 44.
[9] S. Saito, T. Hasegawa, M. Inaba, R. Nishida, T. Fujii, S. Nomiz, T.
Moriwake, Chem. Lett. 1984, 1389.
[10] F. D. Bellamy, M. Bondoux, P. Dodey, Tetrahedron Lett. 1990, 31,
7323.
[11] E. Wunsch, E. Drees, Chem. Ber. 1966, 99, 110.
[12] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155.
[13] J. C. Sheehan, P. A. Cruickshank, G. L. Boshart, J. Org. Chem.
1961, 26, 2525.
[14] M. Paris, J. Fehrentz, A. Heitz, A. Loffet, J. Martinez,
Tetrahedron Lett. 1996, 37, 8489.
[15] J. P. Ferezou, M. Julia, Y. Li, L. W. Liu, A. Pancrazi, Bull. Soc.
Chim. Fr. 1995, 132, 428.
[16] U. Schmidt, M. Kroner, H. Griesser, Synthesis 1989, 832.
[17] J. M. Schkeryantz, J. G. Woo, P. Siliphaivanh, K. M. Depew, S. J.
Danishefsky, J. Am. Chem. Soc. 1999, 121, 11969.
[18] W. Kꢀnig, R. Geiger, Chem. Ber. 1970, 103, 788.
[19] E. F. V. Scriven, K. Turnbull, Chem. Rev. 1988, 88, 297.
[20] B. M. Trost, S. R. Pulley, Tetrahedron Lett. 1995, 36, 8737.
[21] a) D. L. Boger, P. Yohannes, Tetrahedron Lett. 1989, 30, 5061;
b) T. Sugiura, Y. Hamada, T. Shioiri, Tetrahedron Lett. 1987, 28,
2251; c) M. P. Forst, G. P. Concepcion, G. B. Caraan, C. M.
Ireland, J. Org. Chem. 1992, 57, 6671.
[22] L. A. Carpino, A. El-Faham, C. A. Minor, F. Albericio, J. Chem.
Soc. Chem. Commun. 1994, 201.
[23] B. M. Trost, J. D. Chisholm, S. T. Wrobleski, M. Jung, J. Am.
Chem. Soc. 2002, 124, 12420.
[24] [a]2D5 = À22.7 (c = 0.4, 1:1 MeOH/0.1n NH4HCO3 (aq); lit.[2b]
:
1
[a]2D5 = À24 (c = 0.4, 1:1 MeOH/0.1n NH4HCO3 (aq)); H NMR
(300 MHz, [D7]DMF with trace of TFA): d = 11.58 (s, 1H), 10.60
(s, 1H), 8.88 (d, J = 4.8 Hz, 1H), 8.62 (d, J = 8.7 Hz, 1H), 8.49 (t,
J = 6.2 Hz, 1H), 7.72 (d, J = 7.8 Hz, 1H), 7.59 (br, 1H), 7.54 (d,
J = 8.1 Hz, 1H), 7.40 (d, J = 8.7 Hz, 2H), 7.35 (m, 1H), 7.31 (t,
J = 7.8 Hz, 1H), 7.16 (t, J = 7.8 Hz, 1H), 6.91 (d, J = 8.7 Hz, 2H),
6.73 (dd, J = 15.9, 10.8 Hz, 1H), 6.45 (d, J = 15.9 Hz, 1H), 6.24
(dd, J = 15.0, 10.8 Hz, 1H), 5.83 (m, 1H), 5.45 (s, 1H), 5.42 (m,
1H), 4.68 (s, 1H), 4.64 (d, J = 15.9 Hz, 1H), 4.33 (m, 1H), 4.27
(m, 1H), 4.05 (q, J = 5.7 Hz, 2H), 3.95 (br, 1H), 3.73 (m, 1H),
3.68 (m, 2H), 3.61 (m, 2H), 3.47 (d, J = 15.9 Hz, 1H), 3.45 (m,
2H), 3.07 (s, 3H), 2.91 (m, 2H), 2.59 (dd, J = 18.9, 4.8 Hz, 1H),
2.48 (d, J = 13.5 Hz, 1H), 2.39 (m, 2H), 2.21 (m, 1H), 1.35 ppm
(t, J = 5.7 Hz, 3H); negative ESI-MS m/z 929 (MÀH)À.
Angew. Chem. Int. Ed. 2003, 42, 5348 –5351
www.angewandte.org
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5351