Total synthesis of natural myriaporones

Article abstract of DOI:10.1002/anie.200353313

The relative and absolute configuration of the cytotoxic myriaporones. (e.g. 4) was assigned by their total synthesis from aldehyde 1. Key steps included aldol reactions with chiral oxazolidinone 2 and with Weinreb amide 3. TBS = tert-butyldimethylsilyl,

Full text of DOI:10.1002/anie.200353313

Communications
Natural Products Synthesis
Total Synthesis of Natural Myriaporones
Marta PØrez, Carlos del Pozo, Fernando Reyes,
Alberto Rodríguez, AndrØs Francesch,
Antonio M. Echavarren,* and Carmen Cuevas*
Myriaporones 1–4 (1–4, respectively) are a new class of
cytotoxic marine polyketide-derived compounds that exhibit
significant cytotoxic activity against L1210 cells. These com-
pounds 1–4 were isolated by Rinehart and co-workers in 1995
from the bryozoan Myriapora truncata.[1] The most active
constituents, 3 and 4, were isolated as a mixture of cyclic and
open-chain isomers. The myriaporones are structurally
related to the C10–C23 region of the macrocycles tedanolide
(5)[2] and deoxytedanolide (6),[3,4] although the configuration
at C5 of 1 and 2 and at C5 and C6 of 3 and 4 were not
unequivocally determined. Additionally, the absolute config-
uration of the myriaporones was unknown. The functionality
that flanks the C7 carbonyl group confers significant lability
to the myriaporones, which makes these densely functional-
ized structures challenging synthetic targets.
Taylor et al. described progress toward the preparation of
myriaporone 1 (1) by a strategy based on an homoallenylbo-
ration and a nitrile oxide cycloaddition.[5] Furthermore,
Yonemitsu and co-workers reported the synthesis of
advanced intermediates from d-glucose.[6] On the other
[*] Prof. Dr. A. M. Echavarren
Departamento de Química Orgµnica
Universidad Autónoma de Madrid
Cantoblanco, 28049 Madrid (Spain)
Fax : (+)34-91-497-3966
E-mail: anton.echavarren@uam.es
Dr. M. PØrez, Dr. C. del Pozo, Dr. F. Reyes, Dr. A. Rodríguez,
Dr. A. Francesch, Dr. C. Cuevas
PharmaMar, S.A.
28770 Colmenar Viejo, Madrid (Spain)
Fax : (+)34-91-846-6001
E-mail: ccuevas@pharmamar.com
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
1724
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200353313
Angew. Chem. Int. Ed. 2004, 43, 1724 –1727

Angewandte
Chemie
elimination reactions, oxidation to the keto group at C7 was
postponed until the final steps of the synthesis.
Aldehyde 10 was prepared from 12, readily available in
nine steps from methyl (S)-(+)-3-hydroxy-2-methylpropio-
nate by the procedure described by Roush and Lane.[7]
Protection of the primary alcohol of 12 as the TBS ether
and oxidative cleavage of the olefin furnished aldehyde 10
(Scheme 2). Reaction of the boron enolate of chiral crotonate
hand, several syntheses of the C10–C23 region of 5 and 6 have
been described[4,7–9] as part of synthetic efforts directed
towards the total synthesis of the macrolides. However, a
total synthesis of any member of the myriaporone family has
not yet been described. Herein we report the total synthesis of
myriaporones 1, 3–4, which allowed the confirmation of their
structure and the determination of the absolute configuration.
Our retrosynthetic approach to the myriaporones is
outlined in Scheme 1, and was based on the assumption that
the configurations of C5 and C6 were identical to those of
tedanolide (5). We planned to introduce the ethyl side chain
and the Z alkene from Weinreb amide 7, itself available by an
aldol reaction of 8 with acetamide 9. For the key stereo-
Scheme 2. a) TBSCl, imidazole, CH2Cl2, 238C, 90%; b) O3, CH2Cl2; Ph3P, ꢀ788C,
70%; c) (R)-11, nBu2BOTf, Et3N, CH2Cl2, ꢀ788C; 10, CH2Cl2, ꢀ308C, 90%;
d) LiBH4, THF/H2O, 0!238C; H2O2, Et2O, 08C, 95%; e) TBSCl, imidazole,
CH2Cl2, 238C, 92%; f) O3, CH2Cl2; Ph3P, ꢀ788C, 82%. Tf=trifluoromethane-
sulfonyl.
ꢀ
selective C C formation, we relied on an Evans aldol
reaction[10] between aldehyde 10 and chiral oxazolidinone
11.[11] To avoid epimerization, as well as retro-aldol and
imide (R)-11[11] with 10 furnished aldol 13 with excellent
stereoselectivity (> 99:1).[10] Although the secondary alcohol
generated in the aldol reaction will ulti-
mately be oxidized to a ketone, we found that
the aldol product with the opposite config-
uration at C7 (myriaporone numbering)
rearranged quantitatively under acidic con-
ditions to the corresponding tetrahydrofuran
derivative through an intramolecular epox-
ide-opening reaction. Reduction of the acyl
[12]
oxazolidinone with LiBH4 gave 14, whose
primary alcohol group was protected as a
TBS ether. Oxidative cleavage of the termi-
nal double bond provided 8, the central core
of the myriaporones, in good overall yield
(41% from 12, six steps).
Although pure aldehyde 8 can be stored
at ꢀ 308C for several months, an intramo-
lecular epoxide-opening reaction[13] occurs
quantitatively under acidic conditions at
room temperature to give tetrahydrofuran
15 (Scheme 3). A similar cyclization of 14,
followed by the formation of the p-methoxy-
benzylidene derivative, gave tetrahydrofuran
Scheme 1. Retrosynthetic analysis for the synthesis of myriaporone 4 (4). TBS=tert-butyl-
dimethylsilyl, PMB=p-methoxybenzyl.
Angew. Chem. Int. Ed. 2004, 43, 1724 –1727
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1725

Communications
1
16. Analysis of H–1H coupling constants
and NOESY correlations of 16 allowed the
assignment of the configurations of C8–C12
of 8. A relative anti configuration at C6–C7
was determined from the small value of
the coupling constant (JH6,H7 = 1.2 Hz)
observed in acetonide 17, prepared from
14 (Scheme 3).
For the next stage, the aldol reaction
between aldehyde 8 and Weinreb acet-
amide 9 proceed readily in the presence of
LDA as base to give 18a and 18b[14] in good
yield but low stereoselectivity (1:2.5),
which were chromatographically separated
(Scheme 4). The use of lithium hexame-
thyldisilazide provided 18a and 18b in a 1:3
ratio, whereas the sodium or potassium
enolates of 9 furnished almost exclusively
anti-18b. The latter could be partially
recycled to 18a by oxidation of the secon-
dary alcohol with Dess–Martin periodi-
nane[15] followed by reduction of the result-
ing ketone with NaBH4 in MeOH to afford
18a and 18b in a 1:1 ratio (62% yield, two
steps).
Selective protection of the less-hin-
dered secondary alcohol of 18a with
TBSOTf and lutidine, followed by oxida-
tion of the alcohol at C7 with DMP and
oxidative removal of the PMB group gave
19 (Scheme 4). Based on literature prece-
dent,[4,5] the Z olefin was constructed by
oxidation with DMP followed by Wittig
ethylidenation of the resulting aldehyde to
give 20. Chemoselective addition of ethyl-
magnesium bromide to the Weinreb amide
of 20 afforded ethyl ketone 21[16] in sat-
isfactory yield on a gram scale. Deprotec-
tion of the silyl ethers of 21 with TBAF led
to decomposition. However, the use of
Scheme 4. a) 9 + LDA, THF, ꢀ788C, 91%, 18a,b (1:2.5); b) TBSOTf, 2,6-lutidine, CH2Cl2,
08C, 81%; c) DMP, CH2Cl2, 238C; d) DDQ, THF/H2O, 238C, 51%, two steps; e) DMP,
CH2Cl2, 238C; f) Ph3PCH2CH3Br, KtBuO, toluene, ꢀ788C, 60%, two steps; g) EtMgBr, THF,
238C, 68%; h) TBAF, HOAc, DMF, 238C, 35%; i) TBAF, HOAc, THF, 238C, 80%. LDA=
lithium diisopropylamide; DMP=Dess–Martin periodinane; DDQ=2,3-dichloro-5,6-
dicyano-1,4-benzoquinone; TBAF=tetra-n-butylammonium fluoride.
TBAF and HOAc in THF led to the selective removal of
three of the protecting groups to give a mixture of 22 and 23 in
good yield. Global deprotection of 21 with TBAF and HOAc
in DMF gave myriaporones 3,4 (1:1). The myriaporones 3,4
in CD3OD were identical in all respects (1H and 13C NMR,
HRMS, HPLC, and TLC) to an authentic sample, which
allowed the assignment of their relative configurations as
those shown. The cytotoxic activity of synthetic and natural
myriaporones 3–4 was found to be the same. The myriapor-
ones 3–4 undergo partial dehydration on silica gel. Accord-
ingly, acetylation (Ac2O, Et3N, CHCl3) of the crude dehy-
dration products obtained after column chromatography gave
1 (30% yield over two steps).
The C5 epimers 24–25 of 3–4 were also prepared from 18b
by the same reaction sequence.[17] Similarly, the aldol reaction
of 10 and the boron enolate of imide (S)-11 led to the series
with the non-natural configuration at C6 or C5–C6
Scheme 3. a) Acid-catalyzed cyclization of epoxyaldehyde 8. b) Derivatives 16
and 17 prepared from alcohol 14 for the determination of the relative
configurations. The arrows indicate NOE correlations.
1726
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 1724 –1727

Angewandte
Chemie
(Scheme 5, 26–29).[17,18] The NMR spectra of compounds 24–
29 were clearly distinct from those of the natural myriapor-
ones 3–4.
[5] a) R. E. Taylor, B. R. Hearn, J. P. Ciavarri, Org. Lett. 2002, 4,
2953 – 2955; b) R. E. Taylor, J. P. Ciavarri, B. R. Hearn, Tetrahe-
dron Lett. 1998, 39, 9361 – 9364.
[6] a) B.-Z. Zheng, M. Yamauchi, H. Dei, O. Yonemitsu, Chem.
Pharm. Bull. 2000, 48, 1761 – 1765; b) B.-Z. Zheng, M. Yamau-
chi, H. Dei, S. Kusaka, K. Matsui, O. Yonemitsu, Tetrahedron
Lett. 2000, 41, 6441 – 6444; d) T. Matsushima, M. Mori, B.-Z.
Zheng, H. Maeda, N. Nakajima, J.-I. Uenishi, O. Yonemitsu,
Chem. Pharm. Bull. 1999, 47, 308 – 321; e) T. Matsushima, M.
Mori, N. Nakajima, H. Maeda, J.-I. Uenishi, O. Yonemitsu,
Chem. Pharm. Bull. 1998, 46, 1335 – 1336.
[7] W. R. Roush, G. C. Lane, Org. Lett. 1999, 1, 95 – 98.
[8] a) T. Matsushima, B.-Z. Zheng, H. Maeda, N. Nakajima, J.
Uenishi, O. Yonemitsu, Synlett 1999, 780 – 782; b) B.-Z. Zheng,
H. Maeda, M. Mori, S.-I. Kusaka, O. Yonemitsu, T. Matsushima,
N. Nakajima, J.-I. Uenishi, Chem. Pharm. Bull. 1999, 47, 1288 –
1296.
[9] T.-P. Loh, L.-C. Feng, Tetrahedron Lett. 2001, 42, 3223 – 3226.
[10] D. A. Evans, E. B. Sjogren, J. Bartroli, R. L. Dow, Tetrahedron
Lett. 1986, 27, 4957 – 4960.
[11] D. A. Evans, K. T. Chapman, J. Bisaha, J. Am. Chem. Soc. 1984,
106, 4261 – 4263.
[12] T. D. Penning, S. W. Djuric, R. A. Haack, V. J. Kalish, J. M.
Miyashiro, B. W. Rowell, S. S. Yu, Synth. Commun. 1990, 20,
307 – 312.
[13] See reference [9] for a similar transformation.
[14] The configuration of the aldol products was assigned by
conversion of the 1,3-diols into the corresponding acetonides.
The acetonide of syn-1,3-diol 18b gave rise to a signal for the
axial methyl group at d = 20.2 ppm and for the equatorial methyl
group at d = 30.0 ppm in the 13C NMR spectrum, whereas the
acetonide of anti-1,3-diol 18a showed the methyl groups at d =
24.4 and 25.2 ppm. See: S. D. Rychnovsky, B. Rogers, G. Yang, J.
Org. Chem. 1993, 58, 3511 – 3515.
[15] a) D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155 – 4256;
b) D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277 –
7287.
[16] S. Nahm, S. M. Weinreb, Tetrahedron Lett. 1981, 22, 3815 – 3818.
[17] Reaction conditions and yields are given in the Supporting
Information.
Scheme 5. The C5 epimers 24 and 25 of 4 and 3, respectively, as well
the series with the non-natural configuration at C6 or C5–C6, 26–29
(see the Supporting Information for details on the synthesis).
In summary, the total synthesis of the myriaporones 1, 3,
and 4 (1, 3–4) was based on two consecutive aldol reactions.[19]
The ethyl ketone was introduced in a straightforward manner
by the addition of an ethyl Grignard reagent to the highly
functionalized Weinreb amide 20 bearing keto and epoxide
functions. The simplicity of this approach is promising for the
preparation of new derivatives from this key intermediate.
Additionally, this route allows the synthesis of a variety of
stereoisomers of 1, 3–4 and other derivatives for further
biological testing. This work allowed the unambiguous
determination of the relative and absolute configurations of
these cytotoxic compounds to be those shown in formulae 1,
3–4.
[18] These series with the non-natural configuration at C6, could also
be synthesized from 30, which was prepared by oxidation of 14.
Further elaboration, as shown in Scheme 4, included the
cleavage of the vinyl group by ozonolysis, followed by reduction
in situ with NaBH4 to form the alcohol.
Received: November 13, 2003 [Z53313]
Keywords: aldol reaction · antitumor agents ·
.
diastereoselectivity · natural products · total synthesis
[19] For an alternative total synthesis of myriaporones 1, 3, and 4 see
following Communication in this issue: K. N. Fleming, R. E.
Taylor, Angew. Chem. 2004, 116, 1760–1762; Angew. Chem. Int.
Ed. 2004, 43, 1728–1730.
[1] a) K. L. Rinehart, K. Tachibana, J. Nat. Prod. 1995, 58, 344 – 358;
b) K. L. Rinehart, J. F. Cheng, J.-S. Lee, US Patent 5,514,708,
1996 [Chem. Abstr. 1996, 125, 5896].
[2] a) F. J. Schmitz, S. P. Gunasekera, G. Yalamanchili, M. B.
Hossain, D. Van der Helm, J. Am. Chem. Soc. 1984, 106, 7251 –
7252; b) T. Matsushima, K. Horita, N. Nakajima, O. Yonemitsu,
Tetrahedron Lett. 1996, 37, 385 – 388.
[3] N. Fusetani, T. Sugawara, S. Matsunaga, H. Hirota, J. Org. Chem.
1991, 56, 4971 – 4974.
[4] for the total synthesis of deoxytedanolide (6), see: A. B. Smith,
C. M. Adams, S. A. Barbosa, A. P. Degnan, J. Am. Chem. Soc.
2003, 125, 350 – 351.
Angew. Chem. Int. Ed. 2004, 43, 1724 –1727
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1727