Efficient total synthesis of dysidenin, dysidin, and barbamide

Article abstract of DOI:10.1002/asia.201100338

The total syntheses of three trichloroleucine-derived marine natural products-dysidenin, dysidin, and barbamide-capitalized on a practical and highly diastereoselective ruthenium catalyzed radical chloroalkylation of N-acyloxazolidinones as the key step. Copyright

Full text of DOI:10.1002/asia.201100338

COMMUNICATION
DOI: 10.1002/asia.201100338
Efficient Total Synthesis of Dysidenin, Dysidin, and Barbamide
Elizabeth A. Ilardi and Armen Zakarian*[a]
Marine sponges of the genus dysidea and the marine cya-
nobacterium Lyngba majuscula are the source of a diverse
subgroup of halogenated natural products in which the halo-
gens are located at aliphatic positions.[1] For the majority of
chlorinated natural products, the halogen atoms occupy po-
sitions that are suggestive of electrophilic chlorination as the
biosynthetic origin of halogenation.[2] In contrast, the tri-
chloromethyl group in a number of secondary metabolites is
unlikely to be formed by an electrophilic process. The bio-
genesis of this halogenation pattern has incited notable in-
terest, and recent efforts by the groups of Gerwick and
Walsh unveiled a fascinating mechanism that involves direct
chlorination of the unactivated pro-(R) methyl group of leu-
cine carried out by nonheme iron(II)-dependent halogenases
that require oxygen, chloride, and a-ketoglutarate for their
activity.[3]
Dysidenin (1),[4] dysidin (2),[5] and barbamide (3)[6] are
characteristic members of a family of natural products that
contain trichloromethyl groups. Despite a number of meth-
ods being reported for the introduction of the trichloro-
methyl group, practicality and stereocontrol remain ongoing
challenges. We have recently reported an experimentally
simple, highly efficient, diastereoselective method for the
direct, ruthenium-catalyzed radical chloroalkylation reaction
based on the valence tautomerism in titanium enolates
(Scheme 1).[7] This method provides a convenient approach
Scheme 1. Ruthenium-catalyzed radical chloroalkylation.
for the stereoselective introduction of the di- and trichloro-
methyl groups found in chloroleucine-derived natural prod-
ucts; herein, we describe an application of the method for a
concise total synthesis of the three members of the group:
dysidenin, dysidin, and barbamide.
Dysidenin (1) is a modified hexachlorinated tripeptide
isolated from the sponge Dysidea herbacea and character-
ized by Kazlauskas et al. at the Roche Research Institute
(Dee Why, New South Wales, Australia).[4] Its structure in-
cludes two chlorinated carboxylate residues and a 2-(1-ami-
noethyl)thiazole fragment. No total synthesis of dysidenin
has been reported previously. Our synthesis strategy is cen-
tered on the ruthenium-catalyzed radical trichloromethyla-
tion reaction highlighting compound 5 as a common starting
point for chlorinated intermediates 8 and 9 (Scheme 2).
After the peptide coupling and N-methylation, the synthesis
is completed by amide bond-formation between carboxylic
acid 6 and l-alanine-derived thiazole 7.
[a] E. A. Ilardi, Prof. Dr. A. Zakarian
Department of Chemistry and Biochemistry
University of California, Santa Barbara
Santa Barbara, California 93106-9510 (USA)
Fax : (+1)805-893-4120
E-mail: zakarian@chem.ucsb.edu
The synthesis of thiazole 7 is based on diastereoselective
addition of Grignard reagents to chiral sulfinimines
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100338.
2260
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2260 – 2263

cleavage of the oxazolidinone
was carried out with lithium
aluminum hydride to avoid the
mono-dechlorination routinely
observed with sodium borohy-
dride.[10] Nitrile 13 was derived
from intermediate alcohol 12
by trifluoromethanesulfonation
followed by substitution with
cyanide (71% yield). The nitrile
was advanced in a divergent
manner to (2S,4S)-5,5,5-trichlor-
oleucine (8) and N-acyloxy suc-
cinimide 15. Reduction to the
aldehyde followed by condensa-
Scheme 2. Retroynthesis of 1.
tion with (S)-tert-butanesulfina-
mide afforded 14. A highly se-
lective, scandium-triflate-cata-
lyzed Strecker reaction followed by hydrolysis with HCl pro-
vided 8 (Scheme 4).[11]
Intermediate 15, which was activated for peptide coupling,
was accessed in three steps involving reduction of the nitrile
to the aldehyde, oxidation to the acid, and condensation
with N-hydroxysuccinimide mediated by DCC.[12]
Completion of the total synthesis of dysidenin required
two amidation reactions and an N-methylation of an amide
group. This process was achieved in a direct fashion as illus-
trated in Scheme 5. Thus, N-acylation of amino acid 8 with
imide 15, N-methylation (NaH, MeI, THF), and a final
amide bond-formation with 7 using EDC and 1-hydroxy-7-
azabenzotriazole (HOAt) in tetrahydrofuran afforded the
natural product in 58% overall yield. The presence of
HOAt was found to be essential to suppress epimerization
Scheme 3. Preparation of the amino-thiazole.
(Scheme 3).[8] We have found
that the stereochemical out-
come is opposite to what is
commonly observed in this type
of reaction. This result can be
rationalized as a chelation-con-
trolled addition through an
open transition structure (i), as
was proposed for the addition
of Grignard reagents to related
N-tert-butanesulfinyl 2-pyridi-
nealdimines.[9] Hydrolysis of the
addition product with 4m HCl
in 1,4-dioxane/diethyl ether af-
fords the desired thiazole cou-
pling partner in 83% over three
steps (d.r. 94:6).
The unified enantioselective
synthesis of the chlorinated
subunits began with compound
5, obtained efficiently in multi-
gram quantities by the key
ruthenium-catalyzed trichloro-
methylation process. Reductive Scheme 4. Synthesis of building blocks 8 and 15. PPTS=pyridinium p-toluenesulfonate, DCC=N,N’-dicyclo-
hexylcarbodiimide.
Chem. Asian J. 2011, 6, 2260 – 2263
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2261

A. Zakarian et al.
COMMUNICATION
85% overall yield. An E-selective O-methylation was best
achieved upon enolization of 18 with sodium hydride in
HMPA as the solvent followed by treatment with freshly
distilled dimethyl sulfate. Under these conditions, the de-
sired product was produced in 75% yield as a single stereo-
isomer. Saponification of the ester with barium hydroxide in
refluxing methanol afforded carboxylic acid 19 in 80%.[15]
In preparation for the final amide coupling, the carboxylic
acid was converted into the corresponding chloroanhydride
upon treatment with oxalyl chloride and catalytic N,N-dime-
thylformamide in dichloromethane. Tetramic acid 17 was
then exposed to methylmagnesium bromide in tetrahydro-
furan to generate the soluble magnesium salt.[13b] Slow addi-
tion of the acid chloride 20 to the solution of the magnesium
amide at 08C and warming to room temperature yielded the
desired natural product in 72% (Scheme 7).
Scheme 5. Dysidenin: endgame. EDC=3-(ethyliminomethyleneamino)-
N,N-dimethyl-propan-1-amine, THF=tetrahydrofuran.
at the C5 position. In this manner, the first synthesis of dysi-
denin was completed in 10 direct steps from 5.
Dysidin was the second total synthesis target in this study.
Two total syntheses of dysidin in its racemic form have been
described previously.[13] Our plan calls for a direct amide for-
mation between the tetramic acid and the carboxylic acid
subunits as the last operation in the synthesis Scheme 6).
Barbamide (3) was isolated
from the lipid extract of L. ma-
juscula collected in CuraÅao
and was employed as a proto-
typic natural product for the
elucidation of the biosynthetic
origin of natural products con-
taining
trichloromethyl
groups.[3,6] Bioassays have indi-
cated that barbamide possesses
Scheme 6. Synthesis of tetramic acid. DCC=N,N’-dicyclohexylcarbodiimide, TFA=trifluoroacetic acid.
Accordingly, the preparation of the known O-methyltetr-
amic acid 17 was accomplished based on a procedure de-
scribed in the literature.[14] Condensation of Meldrumꢁs acid
with N-Boc-l-valine was carried out in the presence dicyclo-
hexylcarbodiimide and 4-dimethylaminopyridine (DMAP)
in dichloromethane, followed by thermolysis in methanol to
afford pyrrolidinone derivative 16. Upon methylation under
Mitsunobu conditions and removal of the tert-butoxycarbon-
yl-protecting group, tetramic acid 17 was furnished in 75%
yield over the four steps.
The preparation of the chlorinated carboxylic acid frag-
ment of dysidin began with (3S)-4,4,4-trichloro-3-methylbu-
tanoic acid 9, developed previously for the synthesis of dysi-
denin (Scheme 7). Generation of the acyl chloride by the
action of thionyl chloride, coupling with Meldrumꢁs acid,
and heating in methanol afforded methyl b-ketoester 18 in
anti-molluscidal activity (LC50 10.0 mgmLÀ1); however, the
full extent of its biological properties remains unknown. In
2001, Gerwick and co-workers reported the first total syn-
thesis of barbamide (3) from a resolution of the racemic
acid 9.[16]
To avoid potential epimerization at the C7 position that
could occur upon oxygen methylation under basic condi-
tions, (S)-N-methyldolaphenine and the (E)-b-methoxyacryl-
ic acid units were synthesized separately and combined in
the final stage by N-methylamide coupling, following a simi-
lar outline as in previous successful efforts with dysidenin
and dysidin.
The addition of a solution of benzylmagnesium chloride
in tetrahydrofuran to sulfinimine 21 in dichloromethane af-
forded thiazole 22 in 79% yield (Scheme 8). Subsequent
methylation with sodium hydride and iodomethane followed
by hydrolysis yielded (S)-N-
methyldolaphenine
(22).[17]
Upon amide coupling between
carboxylic acid 19 and N-meth-
ylamine 22 using EDCI in tetra-
hydrofuran, the total synthesis
of barbamide was completed in
11 steps and 21.8% overall
yield from 5.
In conclusion, this study fur-
ther expands the scope of appli-
cations of the direct ruthenium-
catalyzed radical haloalkylation
of N-acyloxazolidinones via ti-
tanium enolates for the efficient
Scheme 7. Synthesis of dysidin (2). DMF=N,N-dimethylformamide, Py=pyridine, HMPA=hexamethylphos-
phoramide.
2262
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2260 – 2263

Total Synthesis of Dysidenin, Dysidin, and Barbamide
2009, 131, 7570–7572; d) S. L.
Neidleman, J. Geigert, Biohaloge-
nation: Principles, Basic Roles
and Application, Halsted, New
York, 1986.
[4] a) R. Kazlauskas, R. O. Lidgard,
R. J. Wells, W. Vetter, Tetrahe-
dron Lett. 1977, 18, 3183–3186;
b) J. E. Biskupiak, C. M. Ireland,
Tetrahedron Lett. 1984, 25, 2935–
2936.
[5] W. Hofheinz, W. E. Oberhansli,
Helv. Chim. Acta 1977, 60, 660–
669.
[6] J. Orjala, W. H. Gerwick, J. Nat.
Prod. 1996, 59, 427–430.
[7] a) S. Beaumont, E. A. Ilardi,
Scheme 8. Synthesis of barbamide (3).
L. R. Monroe, A. Zakarian, J.
Am. Chem. Soc. 2010, 132, 1482–
1483; b) Z. Gu, A. Zakarian, Angew. Chem. 2010, 122, 9896–9899;
Angew. Chem. Int. Ed. 2010, 49, 9702–9705.
[8] a) G. Liu, D. A. Cogan, J. A. Ellman, J. Am. Chem. Soc. 1997, 119,
9913–9914; b) M. T. Robak, M. A. Herbage, J. A. Ellman, Chem.
Rev. 2010, 110, 3600–3740.
[9] S. D. Kuduk, R. M. DiPardo, R. K. Chang, C. Ng, M. G. Bock, Tetra-
hedron Lett. 2004, 45, 6641–6643.
[10] Up to 30% of 2-dichloromethyl-1-propanol is isolated with NaBH4.
[11] S. Mabic, A. A. Cordi, Tetrahedron 2001, 57, 8861–8866.
[12] G. W. Anderson, J. E. Zimmerman, F. M. Callahan, J. Am. Chem.
Soc. 1964, 86, 1839–1842.
stereoselective total syntheses of chloroleucine-derived nat-
ural products.[18,19] The total syntheses of dysidenin (1,
10 steps, 21.5% overall yield), dysidin (2, 12 steps, 20.2%
overall yield), and barbamide (3, 11 steps, 21.8% overall
yield) have been completed following a robust modular
strategy, which allows for the synthesis of large quantities of
the natural products and their analogues.
[13] a) H. Kçhler, H. Gerlach, Helv. Chim. Acta 1984, 67, 1783–1792;
b) P. G. Williard, S. E. De Laszlo, J. Org. Chem. 1984, 49, 3489–
3493.
[14] S. Suntornchashwej, K. Suwanborirux, M. Isobe, Tetrahedron 2007,
63, 3217–3226.
Acknowledgements
Amgen and Eli Lilly are thanked for unrestricted support of this work.
Financial support was provided by the National Science Foundation
(NSF CHE-0836757). We are grateful to Dr. Hongjun Zhou (UC Santa
Barbara) for continued assistance with NMR spectroscopy.
[15] Y. Iwaki, M. Kaneko, H. Akita, Tetrahedron: Asymmetry 2009, 20,
298–304.
[16] V-A. Nguyen, C. L. Willis, W. H. Gerwick, Chem. Commun. 2001,
1934–1935.
[17] a) N. Irako, Y. Hamada, T. Shioiri, Tetrahedron 1992, 48, 7251–
7264; b) G. R. Pettit, J. Barkoczy, J. K. Srirangam, S. B. Singh, D. L.
Herald, M. D. Williams, D. Kantoci, F. Hogan, T. L. Groy, J. Org.
Chem. 1994, 59, 2935–2938.
Keywords: catalysis · ruthenium · titanium · total synthesis ·
trichloromethylation
[18] Recently, an anionic method for the addition of the CCl3 group to
imines has been reported: a) Y. Li, Y. Cao, J. Gu, W. Wang, H,
Wang, T. Zheng, Z. Sun, Eur. J. Org. Chem. 2011, 676–679.
[19] For a recent highlight, see: T. Amatov, U. Jahn, Angew. Chem. Int.
Ed. 2011, 50, 4542–4544.
[1] a) G. W. Gribble, J. Chem. Educ. 2004, 81, 1441–1449; b) C.
Wagner, M. El Omari, G. M. Kçnig, J. Nat. Prod. 2009, 72, 540–553.
[2] A. Butler, J. V. Walker, Chem. Rev. 1993, 93, 1937–1944.
[3] a) P. M. Flatt, S. J. Oꢁ Connell, K. L. McPhail, G. Zeller, C. L. Willis,
D. H. Sherman, W. Gerwick, J. Nat. Prod. 2006, 69, 938–944;
Received: April 4, 2011
Published online: June 22, 2011
´
b) D. P. Galonic, F. H. Vaillancourt, C. T. Walsh, J. Am. Chem. Soc.
2006, 128, 3900–3901; c) D. K. Bedke, G. M. Shibuya, A. Pereira,
W. H. Gerwick, T. H. Haines, C. Vanderwal, J. Am. Chem. Soc.
Chem. Asian J. 2011, 6, 2260 – 2263
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2263