TETRAHEDRON LETTERS Tetrahedron Letters 43 (2002) 185–187
Pergamon
Synthesis of C13C22 fragment of the marine sponge polyketide callystatin A† Luiz C. Dias* and Paulo R. R. Meira Instituto de Quı´mica, Universidade Estadual de Campinas/UNICAMP CP 6154, 13083 -970, Campinas, SP, Brazil Received 1 November 2001; accepted 6 November 2001
Abstract—We wish to report here our initial efforts towards the total synthesis of callystatin A, describing the synthesis of the C13C22 fragment. This asymmetric approach uses an efficient syn-aldol reaction, and a diastereoselective epoxidation of an allylic alcohol with m-CPBA followed by epoxide opening with Me2CuLi. © 2002 Elsevier Science Ltd. All rights reserved.
In 1997, Kobayashi and co-workers reported the isolation of callystatin A (1), a potent antitumor polyketide from the marine sponge Callyspongia truncata (Scheme 1).1 The relative as well as the absolute stereochemistry of callystatin A was established by a combination of spectroscopic methods and chemical synthesis.2
O
13 10
Me
OH
16
18
Me
Me
Me
12
Me
6
Wittig coupling
Me
Callystatin A (1) epoxide
O
O opening
1
O
12
Br
8
Me
PBu3
EtO
C1-C12 (2)
PMB OTBS
O
Me
+
Me
Me
Me
22
Me
C13-C22 (3) Horner-Emmons
OTBS
O 18 1
OiPr
As the natural supply is extremely restricted, and attracted by its potent cytotoxicity, we initiated a project directed towards its total synthesis. An efficient and flexible synthesis is essential to provide further material for more extensive biological studies, along with access to novel analogs.3,4 The approach described here to the C13C22 fragment might give access to callystatin A and additional derivatives with potential relevance to biological studies.4,5
18 13
Me
6
22
Me
Callystatin A (1), a polyketide with a terminal a,bunsaturated lactone and two diene systems shows remarkably high activity (IC50=0.01 ng/mL) against KB tumor cells.
HO
Me
15
Me
4
O
Me
22
OH
Our disconnection, summarized in Scheme 1, involved cleavage of the C12C13 bond to give fragments C1C12 (2) and C13C22 (3).6 Unsaturated ester 3 is viewed as arising from an allylic alcohol 4 by selective epoxidation followed by epoxide opening with Me2CuLi. The allylic alcohol 4 may be further dissected in a straightforward manner to give Weinreb amide 5. Of the available options, we speculate that the desired syn–syn stereocenters in 5 might be established through a boron enolate mediated aldol reaction.
MeO
N 5 Me
Me
18
Me
Me
22
Evans' aldol
Scheme 1. * Corresponding author. Fax: +55-19-3788-3023; e-mail: ldias@ iqm.unicamp.br † This paper is dedicated to the Brazilian Chemical Society (SBQ).
Synthesis of fragment C13C22 began with the known acyloxazolidinone 6, which was most conveniently prepared by acylation of the corresponding (S)-oxazolidinone (Scheme 2).7 Asymmetric aldol addition of the boron enolate derived from oxazolidinone 6 with 2-(S)methylbutanal 7 gave the aldol adduct 8 as a crystalline solid (mp 88°C) in 89% yield and >95:5 diastereoselectivity (Scheme 2).8 Exchange of the oxazolidinone auxiliary in the all syn-aldol 8 with N,O-dimethylhydroxyl-
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 0 - 4 0 3 9 ( 0 1 ) 0 2 1 0 4 - 9
L. C. Dias, P. R. R. Meira / Tetrahedron Letters 43 (2002) 185–187
186 O
O
O 18
O
N
Me
1. n-Bu2BOTf CH2Cl2, Et3N, -78°C
6
O
O
Me 20
MeONHMe.HCl 2. TBSOTf, 2,6-lutidine o
CH2Cl2, 25 C
8
18
MeON Me
Me
N Me
Me
9
Me
THF, 0oC (90%, 2 steps)
>95:5
Me
OH
Me Me
18
22 2. (EtO)2POCH2CO2Me
Me 9
OTBS 18
OTBS
10
Me
diastereoselection > 95:5
O MeO
22
1. DIBALH, CH2Cl2 MeO 15 0 oC
OTBS
Me Me
Bn
H Me 7 Me 22 89%
1. Me3Al, THF, 00C
O
18
N
O
Bn
O
OH
22
DIBALH 15 CH2Cl2
OTBS 18
-23 oC, 90%
83% (over 2 steps)
m-CPBA CH2Cl2
4
Me
Me
19
11
22
Me
18
HO
Me 0 oC, 96%
19
OTBS
O
15
22
Me
Me
22
Me
diastereoselection >95:5
PMP
Scheme 2.
Me2CuLi OH OH OTBS THF, 0 oC 16 18 15
amine generated the Weinreb amide, whose purification was facilitated by isolation of the recyclable oxazolidinone chiral auxiliary (92%) by efficient crystallization from the reaction mixture.9 Subsequent protection of the alcohol functionality as its TBS ether cleanly provided the Weinreb amide 9, in 83% yield (over two steps). This amide was smoothly reduced to the aldehyde on treatment with diisobutylaluminum hydride at 0°C (Scheme 3). This unpurified aldehyde was directly subjected to a Horner–Emmons homologation with the required phosphonate reagent to give a,b-unsaturated ester 10 in 90% isolated yield over the two-step sequence.10 Reduction of 10 with 2.2 equiv. of diisobutylaluminum hydride at −23°C gave allylic alcohol 4 (90% yield). Epoxidation of allylic alcohol 4 with m-CPBA proceeded with complete stereoselectivity from the opposite side of the C19 tert-butyldimethylsilyl group to yield the anti-epoxy alcohol 11 as a single product in high purity (96% yield, >95:5).6,11 It is noteworthy that the diastereoselectivity associated with this epoxidation was exceptional and compared in both yield and selectivity to related transformations described earlier by Isobe et al. and later by Miyashita et al.11 Epoxide opening proceeded smoothly with high regioselectivity after treatment of the epoxy alcohol 11 with Me2CuLi to give diol 12 in 90% yield and >95:5 selectivity, possessing the anti-anti–syn-syn stereochemistry for the five contiguous stereocenters.12 Although the hydroxyl function at C17 in diol 12 should be converted to a ketone carbonyl later in the synthesis, this epoxidation/epoxide opening sequence enabled control of the C16 stereocenter. Formation of pmethoxybenzylidene acetal 13 was accomplished by treatment of the corresponding diol with p-methoxybenzaldehyde dimethyl acetal and a catalytic amount of PPTS (96% yield). The use of CSA instead of PPTS in this reaction gave the secondary alcohol with loss of the TBS protecting group. The stereochemical outcome of the epoxide opening was determined by coupling constant analysis in the 1H NMR spectra of benzylidene acetal 13. The large vicinal coupling constants between Ha–Hb (10.1 Hz) and Hb–Hd (11.2 Hz), together with the small observed value between Hb–Hc (4.8 Hz) unambiguously established the proposed 1,2-anti relative stereochemistry between C16 and C17.
90%
Me
Me
17
OMe
Me
12
OTBS 18
Me
19
Me
Me
96% Hb Hc
16
Me
O
15
Me MeO 22
Me
diastereoselection >95:5
O
PPTS
17
22
13
OMe 15 O
O
R
JHa-Hb = 10.1 Hz JHb-Hd = 11.2 Hz JHb-Hc = 4.8 Hz
PMP
17 Hd
H
Ha
13
Scheme 3.
The stereochemistry of the secondary alcohols at C17 and C19 was determined on the basis of the 13C NMR analysis of the corresponding 1,3-diol acetonide 15 (Scheme 4). Treatment of diol 12 with TBAF in THF gave triol 14 in 98% yield. A sequence of selective protection of the primary alcohol functionality in 14 followed by treatment with 2,2-dimethoxypropane and catalytic amounts of PPTS gave acetonide 15 in 85% yield for the two steps. 13C NMR observed resonances at 23.7, 25.3, and 100.2 are characteristic of an anti acetonide.13 Selective DIBALH-mediated acetal cleavage in 13 gave the desired p-methoxybenzyl ether alcohol 16 in 94% yield with complete regioselectivity (Scheme 5).14 Dess– Martin periodinane oxidation15 of 16 under standard conditions followed by Wittig10 type coupling with carboethoxyethylidene–triphenylphosphorane in refluxing CH2Cl2 gave a,b-unsaturated ester 3 (E:Z>95:5) in 90% overall yield for the two-step sequence, corresponding to the C13C22 fragment of callystatin A. OH
OH
OTBS
15
17
19
Me
Me
TBAF, THF Me
Me
22
OH
OH
OH
15
17
19
98%
Me
12 25.3
1. TBDPSCl Me
Me
DMF, 25 oC TBDPSO
O
O
15
17
19
imidazole
PPTS,
0 oC
85% (2 steps)
Scheme 4.
Me Me
14 23.7
2. Me2C(OMe)2
Me
Me
Me
100.2 Me Me
15
22
22
L. C. Dias, P. R. R. Meira / Tetrahedron Letters 43 (2002) 185–187 PMP O
O
15
OTBS 18
Me
DIBALH CH2Cl2
Me
Me 22
Me
0 oC, 94%
OH
PMB O OTBS 18
15
Me
13
Me 16
NOESY
OEt H
O
1.Dess-Martin CH2Cl2, rt
Me 2. O Me
22
EtO
3.
PPh3
Me CH2Cl2, ∆ (90%, 2 steps)
4.
PMB OTBS 18
O
13
Me
15
Me
Me
Me
22
Me
E:Z > 95:5 C13-C22 fragment (3)
5.
Scheme 5.
The E-geometry for ester 3 was confirmed by the illustrated NOESY interaction between vinylic hydrogen at C15 and hydrogens of the ethyl group. Ester 3 is a very useful intermediate to be used in the synthesis of callystatin A, by the same strategies already used by others4.
6. 7. 8.
9.
The 12-step sequence from 6 to 3 proceeded in an overall yield of 42% and is amenable to a multigram scale-up. Thus, the synthesis of C13C22 fragment of callystatin A has been described. Notable features of this approach include an efficient syn aldol reaction, a diastereoselective epoxidation of an allylic alcohol followed by epoxide opening with Me2CuLi, and a diastereoselective Wittig coupling. As a result, the route to the C13C22 fragment presented here is, in principle, readily applicable for the preparation of callystatin A and additional analogs. Efforts are being undertaken to complete the total synthesis of callystatin A and these studies will be described in a full account of this work.16
10. 11.
12.
13.
Acknowledgements We are grateful to FAPESP and CNPq for financial support. We thank also Professor Carol H. Collins, from IQ-UNICAMP, for helpful suggestions about English grammar and style.
14. 15.
References 16. 1. Kobayashi, M.; Higuchi, K.; Murakami, N.; Tajima, H.; Aoki, S. Tetrahedron Lett. 1997, 38, 2859. 2. Murakami, N.; Wang, W.; Aoki, M.; Tsutsui, Y.;
187
Higuchi, K.; Aoki, S.; Kobayashi, M. Tetrahedron Lett. 1997, 38, 5533. (a) Hamamoto, T.; Seto, H.; Beppu, T. J. Antibiot. 1983, 36, 646; (b) Absolute stereochemistry and synthesis of leptomycin B: Kobayashi, M.; Wang, W.; Tsutsui, Y.; Sugimoto, M.; Murakami, N. Tetrahedron Lett. 1998, 39, 8291. For total syntheses of callystatin A, see: (a) Murakami, N.; Wang, W.; Aoki, M.; Tsutsui, Y.; Sugimoto, M.; Kobayashi, M. Tetrahedron Lett. 1998, 39, 2349; (b) Crimmins, M. T.; King, B. W. J. Am. Chem. Soc. 1998, 120, 9084; (c) Smith, A. B., III; Brandt, B. M. Org. Lett. 2001, 3, 1685; (d) Kalesse, M.; Quitschalle, M.; Khandavalli, C. P.; Saeed, A. Org. Lett. 2001, 3, 3107. For previous synthesis of C13C22 fragment of callystatin A, see: (a) Marshall, J. A.; Fitzgerald, R. N. J. Org. Chem. 1999, 64, 4477; (b) Marshall, J. A.; Schaaf, G. M. J. Org. Chem. 2001, 66, 7825. The numbering of 1 as well as of each intermediate follows that suggested in Ref. 1. Gage, J. R.; Evans, D. A. Org. Synth. 1990, 68, 83. (a) Evans, D. A.; Bartroli, J.; Shih, T. L. J. Am. Chem. Soc. 1981, 103, 2127; (b) Evans, D. A.; Taber, T. R. Tetrahedron Lett. 1980, 21, 4675. Levin, J. I.; Turos, E.; Weinreb, S. M. Synth. Commun. 1982, 12, 989. Maryanoff, B. E.; Reitz, A. B. Chem. Rev. 1989, 89, 863. (a) Isobe, M.; Kitamura, M.; Mio, S.; Goto, T. Tetrahedron Lett. 1982, 23, 221; (b) Kitamura, M.; Isobe, M.; Ichikawa, Y.; Goto, Y. J. Org. Chem. 1984, 49, 3517; (c) Maruyama, K.; Ueda, M.; Sasaki, S.; Iwata, Y.; Miyazawa, M.; Miyashita, M. Tetrahedron Lett. 1998, 39, 4517. (a) Nagaoka, H.; Kishi, Y. Tetrahedron 1981, 37, 3873; (b) Lipshutz, B. H.; Kozlowski, J.; Wilhelm, R. S. J. Am. Chem. Soc. 1982, 104, 2305; (c) Kigoshi, H.; Suenaga, K.; Mutou, T.; Ishigaki, T.; Atsumi, T.; Ishiwata, H.; Sakakura, A.; Ogawa, T.; Ojika, M.; Yamada, K. J. Org. Chem. 1996, 61, 5326. (a) Rychnovsky, S. D.; Skalitzky, D. J. Tetrahedron Lett. 1990, 31, 945; (b) Evans, D. A.; Rieger, D. L.; Gage, J. R. Tetrahedron Lett. 1990, 31, 7099; (c) Rychnovsky, S. D.; Rogers, B. N.; Richardson, T. I. Acc. Chem. Res. 1998, 31, 9. Takano, S.; Akiyama, M.; Sato, S.; Ogasawara, K. Chem. Lett. 1983, 1593. (a) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277; (b) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155; (c) Ireland, R. E.; Liu, L. J. Org. Chem. 1993, 58, 2899. New compounds and the isolatable intermediates gave satisfactory 1H and 13C NMR, IR, HRMS and analytical data. Yields refer to chromatographically and spectroscopically homogeneous materials.