Synthesis of the C9-C21 fragment of debromoaplysiatoxin and oscillatoxins A and D

Article abstract of DOI:10.1016/j.tetlet.2006.05.187

A highly stereoselective synthesis of the C9-C21 fragment of debromoaplysiatoxin and oscillatoxins A and D has been devised. This new approach relies on the cross coupling of titanium enolates from N-acyl-1,3-thiazolidine-2-thiones and dialkyl acetals and

Full text of DOI:10.1016/j.tetlet.2006.05.187

Tetrahedron Letters 47 (2006) 5819–5823
Synthesis of the C9–C21 fragment of debromoaplysiatoxin and
oscillatoxins A and D
*
*
´
`
`
Annabel Cosp, Enric Llacer, Pedro Romea and Felix Urpı
`
´ ´
Departament de Quı´mica Organica, Universitat de Barcelona, Martı i Franques 1-11, 08028 Barcelona, Catalonia, Spain
Received 6 April 2006; revised 21 April 2006; accepted 17 May 2006
Abstract—A highly stereoselective synthesis of the C9–C21 fragment of debromoaplysiatoxin and oscillatoxins A and D has
been devised. This new approach relies on the cross coupling of titanium enolates from N-acyl-1,3-thiazolidine-2-thiones and dialkyl
acetals and the selective hydrogenolysis of O-benzyl protecting groups.
Ó 2006 Elsevier Ltd. All rights reserved.
Debromoaplysiatoxin (1) is a bicyclic diolide isolated
from the sea hare Stylocheilus longicauda1 as well as the
blue-green algae belonging to Oscillatoriaceae class along
with several structurally related bioactive metabolites as
oscillatoxin A (2), oscillatoxin D (3), and 30-methyloscil-
latoxin D (4) shown in Scheme 1.2–4 Debromoaplysia-
toxin is a highly inflammatory agent responsible for
severe contact dermatitis that sometimes affects sea swim-
mers (swimmers itch),2–4 but interest on it and oscillatoxin
A mainly stems from their remarkable character as tumor
promoters that operate on protein kinase C, which has
been used to better understand carcinogenic processes.4,5
Conversely, oscillatoxin D and 30-methyloscillatoxin D
are nontoxic metabolites with antileukemic activity.3b
In this context, we disclose herein our studies toward the
synthesis of the benzyl-protected C9–C21 fragment of
the debromoaplysiatoxin (1) and oscillatoxin A (2)
based on the cross-coupling reactions between titanium
enolates from N-acyl-l,3-thiazolidine-2-thiones 5 and 6
(see Scheme 2) and dialkyl acetals previously developed
in our group,10–12 and the selective cleavage of benzyl
ethers.13 Furthermore, the configuration of the new
stereocenters has been confirmed by correlation to an
advanced intermediate for the synthesis of oscillatoxin
D, taking advantage that C9–C21 fragment is common
to all of these metabolites.14
According to these ideas, we envisaged the retrosyn-
thetic analysis for the benzyl-protected C9–C21 frag-
ment summarized in Scheme 2, which relies on two
crucial steps: (i) the stereoselective C10–C11 bond for-
mation by means of the addition of the titanium enolate
from N-propanoyl auxiliary 5 to dibenzyl acetal 7; and
(ii) the stereoselective formation of C15 methyl ether
through addition of the titanium enolate from N-acetyl
auxiliary 6 to dimethyl acetal 8.
Synthetic efforts on this family of metabolites have cul-
minated in several total syntheses6–8 and a plethora of
reports on advanced intermediates.9 Most of these ap-
proaches depend on the construction of an acyclic pre-
cursor, whose careful manipulation should eventually
trigger the formation of the spiro or the macrocyclic sys-
tems. Success of such strategy depends to a great extent
on the proper choice of C20-phenol protecting group,
since it must be removed in a late step of the synthesis
under very mild conditions. Not surprisingly, total syn-
theses reported so far have made use of benzyl like pro-
tecting groups that can be cleaved by hydrogenolysis.
The synthesis began with the preparation of dimethyl
acetal 8 in multigram scale (>5 g, 80% yield) from com-
mercially available 3-hydroxybenzaldehyde following
standard procedures.15 With acetal 8 in hand, we faced
the stereoselective construction of methyl ether at C15.
As expected,10b cross coupling reaction between titanium
enolate from N-acetyl-4-isopropyl-1,3-thiazolidine-2-thi-
one (6) and dimethyl acetal 8 in 10 mmol scale afforded a
mixture (dr 87:13 by HPLC) that was purified by column
chromatography. Eventually, the diastereomerically
pure adduct 9 was isolated in 82% yield (see Scheme 3).
Keywords:
Debromoaplysiatoxin;
Oscillatoxin;
Stereoselective
synthesis.
*
Corresponding authors. Tel.: +34 93 402 12 47/403 91 06; fax: +34 93
339 78 78 (F.U.); e-mail addresses: pedro.romea@ub.edu; felix.urpi@
ub.edu
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.05.187

5820
A. Cosp et al. / Tetrahedron Letters 47 (2006) 5819–5823
O
O
O
O
OBn
OMe
O
O
OMe
R
OH
O
OMe
9
X
9
15
21
O
O
O
R
15
21
9
O
15
21
O
OBn
Benzyl–Protected C9–C21 Fragment
O
OH
OH
OH
1 R=Me Debromoaplysiatoxin
2 R=H Oscillatoxin A
3 R=H Oscillatoxin D
4 R=Me 30-Methyloscillatoxin D
Scheme 1.
O
OBn
OMe
late from tert-butyl acetate with iodide 11 provided
quantitatively ester 12, which was easily transformed
into the corresponding carboxylic acid 13. The acylation
of (S)-4-benzyl-1,3-oxazolidinone with 13 and the subse-
quent stereoselective alkylation with methyl iodide were
carried out according to the reported procedures. Anal-
ysis of the reaction mixture revealed the presence of a
11
X
10
S
O
OBn
S
N
1
i
-Pr
single diastereomer 15 (dr >97:3 by H NMR), which
BnO
BnO
OMe
X
OMe
5
was isolated in 78% yield after purification by column
chromatography. Finally, removal of the chiral auxiliary
afforded alcohol 16 in 91% yield. Unfortunately, neither
15 nor alcohol derivative 16 were crystalline solids and
we decided to explore a more expeditious route based
on Myers methodology.17 In this case, alkylation of lith-
ium enolate from (R,R)-N-propanoylpseudoephedrine
with iodide 11 provided quantitatively amide 17 that
was easily reduced to alcohol 16, identical to that ob-
tained along the former route. Therefore, enantiomeri-
cally pure alcohol 16 had been obtained from acetal 8
in 60% yield after a five-step sequence.
15
11
OBn
OBn
7
MeO
MeO
S
N
O
S
15
i
-Pr
OBn
6
8
Scheme 2.
Next, we confronted the synthesis of dibenzyl acetal
7 required for the last cross-coupling reaction (see
Scheme 4). After careful optimization, it was established
that such acetal could be easily obtained through Swern
oxidation of alcohol 16 followed by treatment of the
crude reaction mixture with BnOTMS under Noyori’s
conditions.18 Thus, the stage was now set for a double
stereodifferentiation reaction involving N-propanoyl
auxiliary 5 and acetal 7. We were pleased to confirm10a
that the Lewis acid mediated addition of the titanium
enolate from 5 to 7 proceeded in a highly stereoselective
Removal of the chiral auxiliary was achieved under very
mild conditions and the resulting alcohol 10 was con-
verted into iodide 11 in excellent yield.
Once the C15 methyl ether was installed, we focused our
attention on the construction of C12 stereocenter (see
Scheme 4). Looking for crystalline products, we initially
envisioned that it could arise from a stereoselective
alkylation of a more elaborated intermediate using
Evans methodology.16 Thus, treatment of lithium eno-
1
manner (dr >98:2 by H NMR and HPLC), being able
to isolate diastereomerically pure adduct 18 in 74%
yield. At this point, it is worth mentioning that adduct
18 should be viewed as a benzyl protected derivative of
the corresponding anti aldol intermediate, whose stereo-
chemical array is still reluctant to most of the standard
methodologies.
S
S
O
O
OMe
a
S
N
S
N
i
-Pr
i-Pr
OBn
9
6
b
Such exceptional stereocontrol can be understood con-
sidering that it is a matched case of a double asymmetric
process. Indeed, the reaction takes place through an
open transition state A that involves the addition of
the chelated titanium enolate to a putative oxonium
intermediate I (Scheme 5). Hence, the configuration of
C10 is controlled by the auxiliary (the Re face of the
enolate is blocked by the isopropyl group), whereas that
of C11 arises from the stereoelectronically favored
OMe
OMe
c
I
HO
OBn
OBn
11
10
Scheme 3. Reagents and conditions: (a) i. TiCl4, i-Pr2NEt, CH2Cl2,
À78 °C, 2 h; ii. BF3ÆOEt2, 8, 2 h; 82%. (b) NaBH4, THF, rt, 4 h; 87%.
(c) I2, Ph3P, imidazole, CH2Cl2, rt, 1.5 h; 97%.

A. Cosp et al. / Tetrahedron Letters 47 (2006) 5819–5823
5821
O
O
OMe
O
OMe
a
c
e
RO
O
N
R
Bn
OMe
OMe
OBn
OBn
12 R=
t
-Bu
14 R=H
b
d
I
HO
13 R=H
15 R=Me
OBn
OBn
O
OMe
11
16
f
g
N
OH
h
Cln
Ti
OBn
17
S
O
S
N
S
O
OBn
OMe
BnO
BnO
OMe
O
OMe
i-Pr
i
j
S
N
i
H
-Pr
OBn
OBn
OBn
18
7
17
Scheme 4. Reagents and conditions: (a) t-BuOAc, LDA, THF–DMPU, À78 °C, 2 h; 100%. (b) TMSOTf, 2,6-lutidine, CH2Cl2, 0 °C, 2 h; 91%.
(c) i. t-BuOCOCl, Et3N, THF, 0 °C, 2 h; ii. (S)-Benzyl-1,3-oxazolidin-2-one, BuLi, THF, À78 °C, 30 min; 93%. (d) i. NaHMDS, THF, À78 °C, 2 h;
ii. MeI, À78 °C, 3 h; 78%. (e) LiBH4, MeOH, THF, À20 °C, 17 h; 91%. (f) (R,R)-N-Propanoylpseudoephedrine, LDA, LiCl, THF, 0 °C, 5 h.
(g) LiH2NÆBH3, THF, rt, 2 h; 80% over two steps. (h) (COCl)2, DMSO, Et3N, CH2Cl2, À78 °C; 96%. i. BnOTMS, TMSOTf cat., CH2Cl2, À50 °C;
78%. (j) SnCl4, CH2Cl2, À20 °C, 2 h; 74%.
OBn
OBn
OBn
OMe
OMe
OMe
antiperiplanar approach of the enolate to conformation
B preferred by the Felkin model.19–21
HO
a
19
20
Finally, removal of the chiral auxiliary was carried out
under very mild conditions to afford enantiomerically
pure alcohol 19, ester 20, and Weinreb amide 21 (see
Scheme 6) as potential representatives of the C9–C21
fragment in high yields.
OBn
OBn
OBn
O
O
b
18
MeO
Although the configuration of adduct 18 was rooted on
well-established asymmetric methodologies16,17 and the
chemistry developed in our own group,10 it was obvi-
ously desirable to prove it unambiguously. Additionally,
we were interested in challenging the synthetic potential-
ity of our approach and go ahead in the synthesis of the
metabolites described in Scheme 1. For these reasons,
we decided to transform amide 21 into a more elabo-
rated intermediate 22 (see Scheme 7) in route to oscilla-
toxin D already reported in the literature.3b Keeping in
mind both goals, we adjusted the protecting group strat-
egy. Selective hydrogenolysis conditions disclosed in the
c
N
MeO
21
Scheme 6. Reagents and conditions: (a) NaBH4, THF, rt, 5 h; 65%. (b)
MeOH, DMAP cat., rt, 24 h; 95%. (c) MeNH(OMe)ÆHCl, Et3N,
DMAP cat., CH2Cl2, rt, 16 h; 90%.
preceding communication yielded dihydroxyamide 23
without affecting the integrity of C15 methyl ether.
Then, it was subsequently treated with TBSOTf/2,6-luti-
dine and the ensuing silicon protected Weinreb amide 24
was smoothly reduced with DIBALH. Submission of the
resulting aldehyde 25 to Still–Gennari olefination condi-
tions22 afforded the a,b-unsaturated ester 26 (91% yield,
Z/E ratio >95:5), which was finally converted into the
desired lactone 22 in 58% yield. Eventually, analysis of
BnO
H
OMe
Ar
Cln
Ti
I
S
O
OBn
Me
H
R
H
Me
O
R'
S
N
H
H
1
the H NMR of lactone 2223 supported the configura-
Nu
Bn
i-Pr
A
B
tion of the C9–C21 fragment.
S
O
N
OBn
OMe
Ar
In summary, we have disclosed a highly stereoselective
synthesis of the C9–C21 fragment of debromoaplysia-
toxin and oscillatoxins A and D based on cross-coupling
reactions of chiral titanium enolates and acetals.
Furthermore, the preparation of a more advanced
S
i-Pr
18
Scheme 5.

5822
A. Cosp et al. / Tetrahedron Letters 47 (2006) 5819–5823
7. For a total synthesis of 3-deoxydebromoaplysiatoxin, see:
O
OR
OMe
c
Toshima, H.; Suzuki, T.; Nishiyama, S.; Yamamura, S.
Tetrahedron Lett. 1989, 30, 6725–6728.
N
MeO
8. For a total synthesis of oscillatoxin D, see: Toshima, H.;
Goto, T.; Ichihara, A. Tetrahedron Lett. 1995, 36, 3373–
3374.
21 R=Bn
23 R=H
a
b
OR
d
O
OTBS
OMe
9. (a) Walkup, R. D.; Cunningham, R. T. Tetrahedron Lett.
1987, 28, 4019–4022; (b) Ireland, R. E.; Thaisrivongs, S.;
Dussault, P. H. J. Org. Chem. 1988, 110, 5768–5779; (c)
Walkup, R. D.; Kane, R. R.; Boatman, P. D., Jr.;
Cunningham, R. T. Tetrahedron Lett. 1990, 31, 7587–
7590; (d) Walkup, R. D.; Boatman, P. D., Jr.; Kane, R.
R.; Cunningham, R. T. Tetrahedron Lett. 1991, 32, 3937–
3940; (e) Stolze, D. A.; Perron-Sierra, F.; Heeg, M. J.;
Albizati, K. F. Tetrahedron Lett. 1991, 32, 4081–4084; (f)
Walkup, R. D.; Kahl, J. D.; Kane, R. R. J. Org. Chem.
1998, 63, 9113–9116; (g) Toshima, H.; Ichihara, A. Biosci.
Biotechnol. Biochem. 1998, 62, 599–602.
24 R=TBS
H
OTBS
OMe
OMe
25
OTBS
O
O
26
OTBS
e
O
OMe
´
10. (a) Cosp, A.; Romea, P.; Talavera, P.; Urpı, F.; Vilarrasa,
J.; Font-Bardia, M.; Solans, X. Org. Lett. 2001, 3, 615–
22
OH
´
617; (b) Cosp, A.; Romea, P.; Urpı, F.; Vilarrasa, J.
Tetrahedron Lett. 2001, 42, 4629–4631; (c) Cosp, A.;
Scheme 7. Reagents and conditions: (a) i. H2, 10% Pd/C, toluene, rt,
3 h; ii. H2, 5% Rh/Al2O3, MeOH, rt, 1 h; 95%. (b) TBSOTf, 2,6-
lutidine, CH2Cl2, rt, 13 h; 61%. (c) DIBALH, THF, À78 °C, 3 h; 88%.
(d) (CF3CH2O)2POCH2CO2Me, K2CO3, 18-crown-6, toluene, 0 °C,
4 h; 91%. (e) HC1, MeOH, rt, 24 h; 58%.
´
´
Larrosa, I.; Vilasıs, I.; Romea, P.; Urpı, F.; Vilarrasa, J.
Synlett 2003, 1109–1112.
11. Chiral 1,3-thiazolidine-2-thiones were first used in highly
diastereoselective aldol processes by Nagao and Fujita: (a)
Nagao, Y.; Hagiwara, Y.; Kumagai, T.; Ochiai, M.; Inoue,
T.; Hashimoto, K.; Fujita, E. J. Org. Chem. 1986, 51,
2391–2393; For a recent review on the application of 1,3-
thiazolidine-2-thiones in asymmetric synthesis, see: (b)
intermediate toward oscillatoxin D proves the poten-
tiality of our approach to tackle the synthesis of the
abovementioned metabolites.
´
Velazquez, F.; Olivo, H. F. Curr. Org. Chem. 2002, 6, 1–38.
12. For relevant applications of chiral 1,3-thiazolidine-2-
thiones in stereoselective aldol reactions, see: (a) Hsiao,
C.-N.; Liu, L.; Miller, M. J. J. Org. Chem. 1987, 52, 2201–
Acknowledgments
´
´
´
2206; (b) Gonzalez, A.; Aiguade, J.; Urpı, F.; Vilarrasa, J.
Tetrahedron Lett. 1996, 37, 8949–8952; (c) Crimmins, M.
T.; King, B. W.; Tabet, E. A.; Chaudhary, K. J. Org.
Chem. 2001, 66, 894–902; (d) Evans, D. A.; Downey, C.
W.; Shaw, J. T.; Tedrow, J. S. Org. Lett. 2002, 4, 1127–
1130; (e) Zhang, Y.; Phillips, A. J.; Sammakia, T. Org.
Lett. 2004, 6, 23–25.
Financial support from the Ministerio de Ciencia y Tec-
´
nologıa and Fondos FEDER (Grant BQU2002-01514),
and from the Generalitat de Catalunya (2001SGR00051
and 2005SGR00584), and a doctorate studentship (Gen-
eralitat de Catalunya) to A.C. are acknowledged.
13. See the preceding communication. Tetrahedron Lett. 2006,
47, see doi:10.1016/j.tetlet.2006.05.109.
References and notes
14. Walkup et al. first recognized that a general approach to
all of the members of this family of metabolites could take
advantage of the common fragment C9–C21. See Ref. 9f.
15. Landais, Y.; Robin, J.-P. Tetrahedron 1992, 48, 7185–7196.
16. Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem.
Soc. 1982, 104, 1737–1739.
17. Myers, A. G.; Yang, B. H.; Chen, H.; McKinstry, L.;
Kopecky, D. J.; Gleason, J. L. J. Am. Chem. Soc. 1997,
119, 6496–6511.
18. Tsunoda, T.; Suzuki, M.; Noyori, R. Tetrahedron Lett.
1980, 21, 1357–1358.
19. For a previous analysis of double asymmetric reactions
involving oxonium cations, see Ref. 10c.
20. For a close and inspiring example of Lewis acid mediated
aldol reactions involving boron enolates from N-propa-
noyl-1,3-oxazolidin-2-ones, see: Walker, M. A.; Heath-
cock, C. H. J. Org. Chem. 1991, 56, 5747–5750.
21. For transition states of related Lewis acid mediated aldol
addition, see: Mahrwald, R. Chem. Rev. 1999, 99, 1095–
1120.
1. Kato, Y.; Scheuer, P. J. J. Am. Chem. Soc. 1974, 96, 2245–
2246.
2. (a) Mynderse, J. S.; Moore, R. E.; Kashiwagi, M.; Norton,
T. R. Science 1977, 196, 538–540; (b) Mitchell, S. S.;
Faulkner, D. J.; Rubins, K.; Bushman, F. D. J. Nat. Prod.
2000, 63, 279–282.
3. Complete structures including absolute stereochemistries
have been established by Moore et al. through spectro-
scopic and crystallographic studies. See: (a) Moore, R. E.;
Blackman, A. J.; Cheuk, C. E.; Mynderse, J. S.; Matsu-
moto, G. K.; Clardy, J.; Woodard, R. W.; Graig, J. C. J.
Org. Chem. 1984, 49, 2484–2489; (b) Entzeroth, M.;
Blackman, A. J.; Mynderse, J. S.; Moore, R. E. J. Org.
Chem. 1985, 50, 1255–1259.
4. For a comprehensive review on natural products from
marine blue-green algae, see: Burja, A. M.; Banaigs, B.;
Abou-Mansour, E.; Burgess, J. G.; Wright, P. C. Tetra-
hedron 2001, 57, 9347–9377.
22. Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405–
4408.
5. (a) Kishi, Y.; Rando, R. R. Acc. Chem. Res. 1998, 31, 163–
172; (b) Knust, H.; Hoffmann, R. W. Helv. Chim. Acta
2003, 86, 1871–1892.
6. For a total synthesis of debromoaplysiatoxin, see: Park,
P.-u.; Broka, C. A.; Johnson, B. F.; Kishi, Y. J. Am.
Chem. Soc. 1987, 109, 6205–6207.
23. Selected spectroscopic data for 22: [a]D +8.8 (c 0.25,
CHCl3); IR (film): 3338, 1702 cmÀ1 1H NMR (CDCl3,
;
400 MHz): d 7.21 (1H, t, J = 7.7 Hz, ArH), 6.85–6.80 (2H,
m, ArH), 6.79–6.75 (1H, m, ArH), 6.64 (1H, dd, J = 9.7,

A. Cosp et al. / Tetrahedron Letters 47 (2006) 5819–5823
5823
2.1 Hz, H9), 5.94 (1H, dd, J = 9.7, 2.6 Hz, H8), 5.25 (1H, br
s, OH), 4.05–4.00 (1H, m, H15), 4.02 (1H, dd, J = 10.7, 2.7
Hz, H11), 3.21 (3H, s, OCH3), 2.65–2.58 (1H, m, H10),
1.84–1.72 (1H, m, H12), 1.70–1.58 (2H, m, H14), 1.50–1.48
(2H, m, H13), 1.06 (3H, d, J = 7.3 Hz, Me10), 0.95 (3H, d,
J = 6.7 Hz, Me12); 13C NMR (CDCl3, 75.4 MHz): d 164.9,
156.0, 152.6, 143.7, 129.6, 120.0, 119.5, 114.8, 113.3, 85.5,
83.7, 56.6, 35.2, 33.7, 30.8, 29.5, 15.8, 13.4.