Total synthesis of the chlorinated marine natural product dysamide B

Article abstract of DOI:10.1021/ol062279f

Two approaches to the synthesis of (2S,4S)-5,5-dichloroleucine are compared, and the parent amino acid was used in the first total synthesis of the polychlorinaled marine natural product, dysamide B. A key step was the lead tetraacetate-mediated decarboxylation of an α,α-dichloro acid in the presence of 1,4-cyclohexadiene to generate the dichloromethyl group.

Full text of DOI:10.1021/ol062279f

ORGANIC
LETTERS
2006
Vol. 8, No. 23
5401-5404
Total Synthesis of the Chlorinated
Marine Natural Product Dysamide B
Amanda C. Durow, G. Cliona Long, Susan J. O’Connell, and Christine L. Willis*
School of Chemistry, UniVersity of Bristol, Cantock’s Close, Bristol BS8 1TS, U.K.
chris.willis@bristol.ac.uk
Received September 15, 2006
ABSTRACT
Two approaches to the synthesis of (2S,4S)-5,5-dichloroleucine are compared, and the parent amino acid was used in the first total synthesis
of the polychlorinated marine natural product, dysamide B. A key step was the lead tetraacetate-mediated decarboxylation of an
acid in the presence of 1,4-cyclohexadiene to generate the dichloromethyl group.
r,r-dichloro
Biological halogenation occurs on a wide range of organic
scaffolds including polyketides, terpenes, and non-ribosomal
peptides, and more than 4500 halogenated natural products
have been reported.1 In the majority of cases, the halogens
are incorporated into positions in accord with biohalogenation
occurring via electrophilic species, and indeed, haloperoxi-
dases which catalyze such reactions have been widely
studied.2 However, an intriguing family of chlorinated marine
natural products exist in which the halogens are located at
aliphatic positions remote from activating groups and for
which the mechanism of halogenation is a topic of current
interest.3 For example, it has been shown that during the
biosynthesis of barbamide, isolated from the cyanobacterium
Lyngbya majuscula, chlorination of the pro-R methyl group
of leucine occurs giving dichloroleucine 1 which in turn is
converted to trichloroleucine 2 and hence to barbamide
(Scheme 1).4
The isolation and characterization of the gene cluster
encoding barbamide biosynthesis has been reported,5 and
recently, a mechanism for the reaction was proposed.6
As well as L. majuscula, extracts of the marine sponge-
cyanobacteria symbionts, Lamellodysidea-Oscillatoria (for-
merly Dysidea-Oscillatoria), have proven to be a prolific
source of unusual polychlorinated compounds. In 1993,
symmetrical polychlorinated diketopiperazines (dysamides
A and B; Figure 1) were isolated from extracts of Lamell-
odysidea fragilis (formerly Dysidea fragilis) and their
structures were determined by spectroscopic methods.7 The
absolute configuration of dysamide A (4),7 a dimer of
N-methyl (2S,4S)-trichloroleucine, was determined by X-ray
crystallography and used as a reference in the structure
elucidation of other isolated diketopiperazines. Although not
Scheme 1. Biosynthesis of Barbamide in L. majuscula
(1) Gribble, G. W. J. Chem. Educ. 2004, 81, 1441.
(2) Vaillancourt, F.; Yeh, E.; Vosburg, D. A.; Garneau-Tsodikova, S.;
Walsh, C. T. Chem. ReV. 2006, 106, 3364.
(3) Murphy, C. Nat. Prod. Rep. 2006, 23, 147.
(4) (a) Sitachitta, N.; Rossi, J.; Roberts, M. A.; Gerwick, W. H.; Fletcher,
M. D.; Willis, C. L. J. Am. Chem. Soc. 1998, 120, 7131. (b) Sitachitta, N.;
Marquez, B. L.; Williamson, R. T.; Rossi, J.; Roberts, M. A.; Gerwick,
W.H.; Nguyen, V-A.; Willis, C. L. Tetrahedron 2000, 55, 9103.
(5) Chang, Z.; Flatt, P.; Gerwick, W. H.; Nguyen, V-A.; Willis, C. L.;
Sherman, D. H. Gene 2002, 296, 235.
(6) (a) Galonic´, D. P.; Vaillancourt, F.; Walsh, C. T. J. Am. Chem. Soc.
2006, 128, 3900. (b) Flatt, P. M.; O’Connell, S. J.; McPhail, K. L.; Zeller,
G.; Willis, C. L.; Sherman, D. H.; Gerwick, W. H. J. Nat. Prod. 2006, 69,
938.
10.1021/ol062279f CCC: $33.50
© 2006 American Chemical Society
Published on Web 10/18/2006

to give 7 in an excellent 92% yield with complete stereo-
control. Following conversion of 7 to the diBoc derivative
8, the ester was cleaved via a π-allyl complex using Pd-
(PPh3)4/NaBH4.12 Reduction of acid 9 with a borane di-
methylsulfide complex gave alcohol 10 which was oxidized
under Swern conditions13 to give the required aldehyde 11
in 59% overall yield from 7.
The first approach to the synthesis of (2S,4S)-dichloro-
leucine relied upon a decarboxylation of R,R-dichloro acid
16 as the pivotal step using cyclohexadiene as the hydride
source to access the required dichloromethyl group (Scheme
3). The substrate was readily prepared by a straightforward
Figure 1. Structure of dysamides A and B.
proven, it is reasonable to propose that the chlorinated
leucines 1 and 2 are biosynthetic precursors to diketopip-
erazines 3 and 4, respectively. Further members of this family
have been isolated,8 but to date, the synthesis of none of
these natural products has been reported. Herein, we describe
two approaches for the preparation of (2S,4S)-dichloroleucine
1 as well as the first total synthesis of dysamide B 3.
Because dysamide B is a symmetrical dimer based on
N-methyl (2S,4S)-dichloroleucine, our studies began with the
synthesis of the parent amino acid. Two approaches were
explored via a common intermediate, aldehyde 11. We have
previously reported the synthesis of the analogous methyl
ester9 in which the asymmetric center at C-4 was generated
via a directed alkylation of the lithium enolate of the
N-protected glutamate10 giving the (2S,4S)-diastereomer with
excellent stereocontrol. However, we had found that the
selective reduction of the terminal ester to the required
aldehyde was capricious and good yields could only be
obtained reliably when fresh DIBALH was employed. To
overcome the problem, we now favor the use of selective
reduction of a terminal acid rather than reduction of the ester
with DIBALH as shown in Scheme 2. First, the known
Scheme 3. Synthesis of (2S,4S)-5,5-Dichloroleucine 1
series of transformations beginning with an initial Wittig
reaction under salt-free conditions to give alkene 12 in 95%
yield. Hydroboration of 12 followed by Swern oxidation of
the resultant primary alcohol 13 cleanly gave aldehyde 14.
R,R-Dichlorination of 14 was achieved using tert-butylamine
and N-chlorosuccinimide,14 and optimum yields were ob-
tained when the reaction was conducted in the presence of
molecular sieves giving dichloro aldehyde 15. Williard and
de Laszlo15 have reported the use of KMnO4 to oxidize an
R,R-dichloro aldehyde, but in the case of oxidation of 15,
we found that sodium chlorite with hydrogen peroxide proved
more reliable.
Scheme 2. Synthesis of Aldehyde 11
The penultimate step in the synthesis of dichloroleucine
was decarboxylation of acid 16. Halodecarboxylation of
carboxylic acids has been achieved under a variety of
conditions including the classical Hunsdiecker reaction
(treatment of the silver salt of the acid with bromine)16 or
the Kochi reaction using Pb(OAc)4 and lithium chloride.17
Barton and co-workers18 described the formation of nor-
alkanes from carboxylic acids under radical conditions by
irradiation of the N-hydroxythione derivative with tri-n-
glutamic acid derivative 511 was temporarily protected as
an allyl ester 6 by treatment with allyl bromide and cesium
carbonate prior to the required alkylation which proceeded
(7) Su, J. -Y.; Zhong, Y. -L.; Zheng, L. -M.; Wei, S.; Wong, Q. -W.;
Mak, T. C. W; Zhou, Z. -Y. J. Nat. Prod. 1993, 56, 637.
(8) (a) Fu, X.; Ferreira, M. L. G.; Schmitz, F. J.; Kelly-Borges, M. J. J.
Nat. Prod. 1998, 61, 1226. (b) Dumdei, E. J.; Simpson, J. S.; Garson, M.
J.; Byriel, K. A.; Kennard, C. H. L. Aust. J. Chem. 1997, 50, 139.
(9) Gerwick, W. H.; Leslie, P.; Long, G. C.; Marques, B. L.; Willis, C.
L. Tetrahedron Lett. 2003, 44, 285.
(12) Gigg, R.; Gigg, J. J. Chem. Soc. 1968, 15, 1903.
(13) Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651.
(14) Verhe, R.; DeKimpe, N.; DeBuyck, L.; Schamp, N. Synthesis 1975,
455.
(15) de Laszlo, S. E.; Williard, P. G. J. Am. Chem. Soc. 1985, 107, 199.
(16) Borodin, A. Liebigs Ann. Chem. 1861, 119, 121.
(17) Kochi, J.; Bacha, J.; Bethea, T. J. Am. Chem. Soc. 1967, 89, 6538.
(18) Barton, D. H.; Crick, D.; Motherwell, W. B. J. Chem. Soc., Chem.
Commun. 1983, 939
(10) Hannesian, S.; Margarita, R. Tetrahedron Lett. 1998, 39, 5887.
(11) Tomasz, J. Acta Chim. Acad. Sci. Hung. 1971, 70, 225.
5402
Org. Lett., Vol. 8, No. 23, 2006

butylstannane. To prepare the target dichloride, decarboxy-
lation of acid 16 was required in the presence of a hydride
source which would not lead to concomitant reduction of
the dichloride. The key to the success of this reaction was
treatment of 16 with Pb(OAc)4 in the presence of 1,4-
cyclohexadiene generating the required dichloromethyl prod-
uct 17 in 58% yield.
A more succinct approach to the required gem-dichloride
is directly from aldehyde 11. Several methods are known
for the conversion of aldehydes to dichlorides; however, we
have found, for example, that use of Ph3P/CCl4/Et3N led to
a vinyl dichloride,9 whereas under more forcing conditions,
e.g., SOCl2/DMF or PCl5, decomposition occurred. Recently,
Rodr´ıguez19 reported a valuable modification of the Takeda
conditions (involving treatment of an aldehyde with hydra-
zine monohydrate in anhydrous MeOH followed by reaction
with copper(II) chloride20) in the synthesis of dichloroleucine
which has been used by the group of Walsh6a as well as by
Gerwick Sherman, Willis, and co-workers in biosynthetic
studies on barbamide.6b Using this method, we converted
aldehyde 11 directly to the required dichloride 17; however,
we found that the reaction was capricious and yields varied
from 30 to 45% (Scheme 4). Other solvents were examined
our goal was to investigate a general approach which could
be adapted for the synthesis of a series of symmetrical and
unsymmetrical dysamides. Interestingly, in 1978 Wells and
co-workers reported the isolation of a series of secondary
metabolites from Dysidea herbacea including chlorinated
diketopiperazines which, on reduction, gave a symmetrical
diketopiperazine assembled from N-methylleucine.23 The
absolute and relative stereochemistry of the product was not
assigned.
Matsunari and co-workers have reported the only synthesis
of such a symmetrical diketopiperazine, and their approach
began with a thermal dehydration of L-leucine to give
diketopiperazine 18 (Scheme 5). N-Methylation of 18 with
Scheme 5. Synthesis of Diketopiperazines 19 and 2024
Scheme 4. Synthesis of Protected 5,5-Dichloroleucine 17
MeI and silver oxide gave the cis product 19 in 10% yield,
and with NaH and MeI, approximately a 1:2 mixture of cis-
and trans-diketopiperazines 19 and 20 was obtained which
were separated by fractional crystallization.24 Thus, 19 was
selected as our initial target to develop the synthesis of
diketopiperazines by an approach which could be adapted
for the preparation of symmetrical and unsymmetrical
products
The strategy involved coupling of two orthogonally
protected N-methyl amino acids 23 and 24 followed by
deprotection and cyclization (Scheme 6). First, N-methylation
was achieved by treatment of Cbz-L-leucine 21 with sodium
hydride and methyl iodide in either CH3CN or THF/DMF
to give the known acid 24 and methyl ester 22, respectively.25
Removal of the Cbz group in 22 with HBr/AcOH gave the
required secondary amine 23 in 88% yield. With both the
orthogonally protected amino acids 23 and 24 in hand, next
it was necessary to couple them using conditions which
would not lead to racemization. Several standard coupling
conditions were investigated (including HOAt, EDCI, NaH-
CO3 and HOBt, and DCC)26 which indeed gave the required
dipeptide 25, but the optimum yield was achieved using
BOPCl/Et3N to give 25 as a single diastereomer.27 To
complete the synthesis of the model diketopiperazine 19, the
Cbz group was removed from 25 using HBr/AcOH followed
including dichloromethane, toluene, and isobutanol, but these
gave a lower yield of the required gem-dichloride.
The final stage of the synthesis of dichloroleucine 1 was
removal of the Boc group and hydrolysis of the tert-butyl
ester. Interestingly, treatment of 17 with concentrated HCl
under reflux led to the formation of a quantity (ca. 15%) of
a byproduct tentatively identified as vinyl chloride (1H NMR
δ1.82, d, J ) 1.5 Hz, 4-CH3 and δ6.15, br. s, dCHCl; 13C
NMR δ141.2, C-4 and δ110.0, C-5), an observation also
noted by Macko et al.21 in the hydrolysis of the fungal
phytotoxin victorin C. It was possible to avoid formation of
this unwanted byproduct by carrying out the deprotection in
neat TFA giving 1 in quantitative yield after ion-exchange
chromatography. With quantities of (2S,4S)-dichloroleucine
1 in hand, next we turned to the synthesis of dysamide B
(3).
Several methods have been reported for the synthesis of
diketopiperazines,22 and although dysamide B is symmetrical,
(19) Arda´, A.; Jime´nez, C.; Rodr´ıguez, J. Tetrahedron Lett. 2004, 45,
3241.
(20) Takeda, T.; Sasaki, R.; Yamaguchi, S.; Fujiwara, T. Tetrahedron
1997, 53, 557.
(21) Macko, V.; Wolpert, T. J.; Acklin, W.; Juan, B.; Seibl, J.; Meili, J.;
Arigoni, D. Experientia 1985, 41, 1366.
(23) Kazlauskas, P.; Murphy, P. T.; Wells, R. J. Tetrahedron Lett. 1978,
49, 4945.
(24) Yoshimura, J.; Nakamura, H.; Matsanuri, K. Bull. Chem. Soc. Jpn.
1975, 48, 605.
(22) (a) Hulme, C.; Morrissette, M.; Volz, F.; Burns, C. Tetrahedron
Lett. 1998, 39, 1113. (b) Siwicka, A.; Wojtasiewicz, K.; Rosiek, B.;
Leniewski, A.; Maurin, J. K.; Czarnocki, Z. Tetrahedron: Asymmetry 2005,
16, 975. (c) Dinsmore, C. J.; Beshore, D. C. Tetrahedron 2002, 58, 3297.
(25) Cheung, S.; Benoiton, T. Can. J. Chem. 1977, 55, 916.
(26) Chen, Y.; Bilban, M.; Foster, C. A.; Foster, D. L. J. Am. Chem.
Soc. 2002, 124, 5431.
(27) Tung, R. D.; Rich, D. H. J. Am. Chem. Soc. 1985, 107, 4342.
Org. Lett., Vol. 8, No. 23, 2006
5403

Scheme 6. Synthesis of Diketopiperazines 3 and 19
by cyclization of amino ester 26 with AcOH/butanol/NMM.28
This synthetic route gave diketopiperazine 19 as a single
diastereomer in six steps from L-leucine and 26% overall
yield.
(c 1.6 in MeOH) and [R]D +20 (c 1.6 in EtOH) which is in
accord with the value for the natural product reported in the
original paper.7
In conclusion, two approaches to the synthesis of (2S,4S)-
dichloroleucine 1 have been described which have led to the
synthesis of a series of novel protected amino acids 12-16
as well as a new method for the decarboxylation of an R,R-
dichloro acid to a dichloride using Pb(OAc)4 and 1,4-
cyclohexadiene as the hydride source. The first total synthesis
of dysamide B 3 has been reported confirming the structure
and absolute configuration of the marine natural product.
The same approach was used for the synthesis of dysamide
B 3 from (2S,4S)-dichloroleucine 1 giving the target com-
pound as a white crystalline solid (Scheme 6). The synthetic
material had the same melting point (mp 147-149 °C) as
that reported for the natural product, and their spectral data
correlated well. There are two papers which describe the
isolation of dysamide B 3 from different sponges, namely,
Lamellodysidea (Dysidea) fragilis and Lamellodysidea (Dys-
idea) chlorea.29 However, although the spectral data for 3
are in good agreement, the optical rotation values are very
different: [R]D +13.7 (c 0.117 in MeOH)7 and [R]D -31 (c
0.8 in EtOH).8a Hence, the optical rotation of the synthetic
material was recorded in both solvents giving [R]D +11.4
Acknowledgment. We are grateful to the BBSRC
(A.C.D.) and EPSRC (G.C.L. and S.J.O.) for funding this
work.
Supporting Information Available: Preparation and
characterization of the compounds described in the paper.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(28) Suzuki, K.; Sasaki, Y.; Endo, N.; Mihara, Y. Chem. Pharm. Bull.
1981, 29, 233.
(29) It is now believed that the chlorinated natural products may indeed
be metabolites of the symbiont cyanobacteria (Oscillatoria) rather then from
the sponge itself.
OL062279F
5404
Org. Lett., Vol. 8, No. 23, 2006