Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2006 Aug 11.
Published in final edited form as: Tetrahedron Lett. 2005 Jul 18;46(29):4911–4914. doi: 10.1016/j.tetlet.2005.04.146

Stereoselective synthesis of the C1–C13 segment of dolabelide B

Gary E Keck 1,*, Mark D McLaws 1,†
PMCID: PMC1540402  NIHMSID: NIHMS10546  PMID: 16909163

Abstract

The efficient construction of the C1–C13 segment of dolabelide B is described. A key element of the synthesis entails BITIP catalyzed asymmetric methallylation to establish the C7 stereocenter, which was then used to direct the stereoselective installation of the C9 and C11 centers through Evans reduction and 1,5-anti aldol condensation, respectively.

Keywords: Dolabelide, Asymmetric synthesis, Allylation, Hydroformylation

The marine natural product dolabelide B is a member of a small family of macrolides isolated from the sea hare Dolabella auricularia, which has been shown to exhibit cytotoxicity against HeLaS3 cell lines with an IC50 value of 1.3 μg/mL.1 The biological activity of dolabelide B and its limited availability from natural sources have prompted a number of recent synthetic studies directed toward this structure.2-4 Herein, we report a concise stereoselective synthesis of the C1–C13 segment of dolabelide B.

We envisioned the dolabelides as potentially being derived from the union of two subunits, 1 and 2 (Scheme 1), arising from bond disconnection at the macrolactone and about the C14–C15 olefin. Disconnection about the C14–C15 double bond to give Suzuki coupling precursors5 can be viewed taking place on either side of the olefin, and both approaches were investigated during the course of our study. However, as the synthesis progressed, continued evaluation of our approach lead us to ultimately choose disconnection across the C13–C14 bond (disconnection a) as the most prudent course.

Scheme 1.

Scheme 1

Open in a new tab

Retrosynthetic analysis of dolabelide B.

Our past experience in synthesizing skipped triol structures like that found in subunit 1 suggested that the desired stereochemical relationship could be constructed from aldehydes 3 and 5 through a sequence of asymmetric allylation using a 2-substitued allylstannane reagent, aldol reaction with 1,5-induction, and 1,3-reduction.6 A process by which the initial hydroxyl group is introduced with a high degree of stereoselection is vital, given that this stereocenter will be used to control the stereochemistry of the remaining stereocenters of the triol segment. We have found the catalytic asymmetric allylation (CAA) reaction using the catalyst prepared from BINOL and titanium isopropoxide (BITIP) to be ideally suited for establishing remote secondary hydroxyl groups in a highly stereoselective fashion.7-9 We were hopeful, therefore, that the CAA reaction would fulfill this requirement in the present instance.

Our approach commenced with the synthesis of aldehyde 5 as shown in Scheme 2. The stereochemical relationship of the triad found in this segment lends itself well to synthesis via crotylstannylation chemistry previously developed in our laboratories.10 Thus, treatment of aldehyde 6 with TiCl4 and (Z)-crotyl tributyltin provided 7 in good yield and diastereoselectivity. After protection of the hydroxyl group, the backbone was extended by one carbon through a series of transformations involving oxidative cleavage of the terminal olefin followed by two-carbon homologation and net reduction11 to arrive at aldehyde 5. Although this sequence was successful at producing the desired aldehyde in an overall yield of 30% from 6, the overall process was far more cumbersome and lengthy than we desired. Indeed, we recognized that overall this six step sequence was equivalent to a reductive carbonylation of olefin 7. Therefore, a more direct approach involving hydroformylation was investigated in order to streamline the production of intermediate 5. Hydroformylation of terminal olefins is a powerful reaction type; however, it most commonly results in mixtures of linear and branched regioisomers. A thorough search of the literature suggested that, in this instance, hydroformylation of olefin 7 might be expected to proceed with some degree of regioselectivity to favor the linear product due to the steric environment defined by the methyl substituent at C4.12 Indeed, in the event, the reaction proved exceptionally regioselective giving a linear/branched ratio of 96:4 in favor of the desired product, which was isolated by chromatography in 82% yield (Scheme 3). This three-step sequence (crotyl addition, silyation, and hydroformylation) was easily scaled to furnish multigram quantities of aldehyde 5 from compound 6 in an overall yield of 54% and as a single isomer.

Scheme 2.

Scheme 2

Open in a new tab

Reagents and conditions: (a) (Z)-crotyltributyltin, TiCl4, −90 °C, 2.5 h, 80% (15:1 dr); (b) TBSOTf, Et3N, CH2Cl2, −30 to 0 °C, 1 h, 83%; (c) O3, CH2Cl2, −78 °C, 15 min; PPh3, rt, 2 h; (d) (C6H5)3P = CHCO2C2H5, CH2Cl2, rt, 18 h; (e) SmI2, MeOH, DMA, 0 °C, 15 min, 55% (three steps); (f) DIBAL, THF, −78 °C, 4 h; (g) (COCl)2, DMSO, Et3N, −78 °C, 4 h, 83% (two steps).

Scheme 3.

Scheme 3

Open in a new tab

Hydroformylation of 7.

The synthesis of the triol portion of subunit 1 can be approached from either of the two directions in which the stereochemistry of the hydroxyl bearing carbon at either the C7 or C11 stereocenters is set through asymmetric methallylation. While an approach that first establishes the C11 hydroxyl group would be considered more convergent, experimentation revealed that the subsequent aldol reaction between methyl ketone 10 and aldehyde 5 required a three-fold excess of 5 to achieve acceptable conversion (Scheme 4). Therefore, a more linear approach was taken to minimize consumption of such an advanced intermediate, as shown in Scheme 5. BITIP catalyzed methallylation of 5 gave alcohol 12 in 96% yield and with 94:6 diastereoselectivity. After protection of the resulting hydroxyl group as the PMB ether, the methylene group was oxidatively cleaved to give the aldol precursor 13. Aldol condensation of ketone 13 with aldehyde 3b using conditions developed by Paterson et al.13 afforded the desired 1,5 anti product with a pleasing 90:10 diastereomeric ratio, but in a disappointing 30% yield. The meager yield was attributed to decomposition of aldehyde 3b during the reaction, and a significant improvement was achieved by using the more robust aldehyde 14. Likewise, condensation with acrolein was successful and afforded the desired 1,5-diol in 83% yield and as a single isomer (Scheme 6). The resulting alcohol 17 was protected as the silyl ether and the PMB group was removed with DDQ to unmask the directing group for anti reduction. Reduction of the β-hydroxy ketone with SmI2/methanol14 occurred essentially quantitatively, however the diastereomeric ratio was only 2:1 in favor of the desired anti product. We have found,15-17 as have others,18,19 that the syn/anti ratio obtained from β-directed reduction of ketones with SmI2 can be greatly influenced by a number of factors, including the nature of the neighboring functionality, solvent, additives, and proton source. In many cases, the diastereoselectivity can be increased, or even switched, by attenuating one or more of these factors. Unfortunately, all attempts to improve the syn/anti selectivity of this particular SmI2 mediated reduction failed. Improved selectivity was ultimately achieved using tetramethylammonium triacetoxyborohydride20 to give the diol as a 26:1 (anti/syn) mixture that was separable by chromatography, thereby affording the desired isomer in 85% yield. Completion of the synthesis of this segment was accomplished uneventfully by acetylation of the resulting diol, providing 19 with all stereocenters in place, in a total of 11 linear steps with an overall yield of 24% from compound 5.21,22 Future events leading to subunit coupling will require the removal of the benzyl protecting group found at C1, and oxidation of the alcohol to the acid. Hydroboration of the olefin at C12–C13 with 9-BBN is expected to provide the appropriate Suzuki coupling partner for ring closure.

Scheme 4.

Scheme 4

Open in a new tab

C7–C8 bond construction via 1,5-anti aldol condensation.

Scheme 5.

Scheme 5

Open in a new tab

Reagents and conditions: (a) (R)-(+)-1,1′-bi-2-naphthol, Ti(OiPr)4,4 Å MS, CH2Cl2, −20 °C, 5 days, 96% (94:6 dr); (b) NaH, p-methoxybenzyl bromide, KI, THF, 0 °C, 6 h, 91%; (c) OsO4, NMO, t-BuOH/THF/H2O, rt, 1.5 h; NaIO4, rt, 2 h, 89%; (d) (C6H11)2BCl, Et3N, Et2O, −78 to −20 °C, 22 h.

Scheme 6.

Scheme 6

Open in a new tab

Reagents and conditions: (a) acrolein, (C6H11)2BCl, Et3N, Et2O, −78 to −20 °C, 22 h, 83% (single isomer); (b) TBSOTF, 2,6-lutidine, THF, 0 °C, 1 h, 96%; (c) DDQ, 10:1 CH2Cl2/pH 7 buffer, 0 °C, 40 min, 89%; (d) Me4NBH(OAc)3, 1:1 acetonitrile/AcOH, −15 °C, 5 h, 85%; (e) Ac2O, DMAP, pyridine, rt, 18 h, 94%.

Acknowledgements

Financial assistance provided by the National Institutes of Health (through Grant GM-28961) and by Pfizer, Inc., is gratefully acknowledged.

References and notes

  • 1.Ojika M, Nagoya T, Yamada K. Tetrahedron Lett. 1995;36:7491–7494. [Google Scholar]
  • 2.Grimaud L, de Mesmay R, Prunet J. Org. Lett. 2002;4:419–421. doi: 10.1021/ol017122w. [DOI] [PubMed] [Google Scholar]
  • 3.Desroy N, Le Roux R, Phansavath P, Chiummiento L, Bonini C, Genêt J-P. Tetrahedron Lett. 2003;44:1763–1766. [Google Scholar]
  • 4.Schmidt DR, Park PK, Leighton JL. Org. Lett. 2003;5:3535–3537. doi: 10.1021/ol035431b. [DOI] [PubMed] [Google Scholar]
  • 5.Miyaura N, Suzuki A. Chem. Rev. 1995;95:2457–2483. [Google Scholar]
  • 6.Keck GE, Wager CA, Wager TT, Savin KA, Covel JA, McLaws MD, Krishnamurthy D, Cee VJ. Angew. Chem., Int. Ed. 2001;40:231–234. [PubMed] [Google Scholar]
  • 7.Keck GE, Tarbet KH, Geraci LS. J. Am. Chem. Soc. 1993;115:8467–8468. [Google Scholar]
  • 8.Keck GE, Krishnamurthy D. Syn. Prep. 1998:12–18. [Google Scholar]
  • 9.Keck GE, Krishnamurthy D, Grier MC. J. Org. Chem. 1993;58:6871–6874. [Google Scholar]
  • 10.Keck GE, Savin KA, Cressman ENK, Abbott DE. J. Org. Chem. 1994;59:7889–7896. [Google Scholar]
  • 11.Inanaga J, Sakai S, Handa Y, Yamaguchi M, Yokoyama Y. Chem. Lett. 1991:2117–2118. [Google Scholar]
  • 12.Ojima I, Tsai C-Y, Tzamarioudaki M, Bonafoux D. Org. React. 2000;56:1–354. [Google Scholar]
  • 13.Paterson I, Gibson KR, Oballa RM. Tetrahedron Lett. 1996;37:8585–8588. [Google Scholar]
  • 14.Keck GE, Wager CA, Sell T, Wager TT. J. Org. Chem. 1999;64:2172–2173. [Google Scholar]
  • 15.Keck GE, Wager CA. Org. Lett. 2000;2:2307–2309. doi: 10.1021/ol006072c. [DOI] [PubMed] [Google Scholar]
  • 16.Keck GE, Knutson CE, Wiles SA. Org. Lett. 2001;3:707–710. doi: 10.1021/ol015531m. [DOI] [PubMed] [Google Scholar]
  • 17.Keck GE, Truong AP. Org. Lett. 2002;4:3131–3134. doi: 10.1021/ol026456y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Chopade PR, Davis TA, Prasad E, Flowers RA., II Org. Lett. 2004;6:2685–2688. doi: 10.1021/ol049129u. [DOI] [PubMed] [Google Scholar]
  • 19.Davis TA, Chopade PR, Himersson G, Flowers RA., II Org. Lett. 2005;7:119–122. doi: 10.1021/ol047835p. [DOI] [PubMed] [Google Scholar]
  • 20.Evans DA, Champman KT, Carreira EM. J. Am. Chem. Soc. 1988;110:3560–3578. [Google Scholar]
  • 21.Dale JA, Mosher HS. J. Am. Chem. Soc. 1973;95:512–519. The absolute stereochemistry of the triol segment was confirmed by a NMR spectral analysis of the C7 Mosher esters and the C7–C9 and C9–C11 acetonides. See: [Google Scholar]; Rychnovsky SD, Rogers B, Yang G. J. Org. Chem. 1993;58:3511–3515. [Google Scholar]
  • 22. All new compounds were characterized by IR, 1H NMR, 13C NMR, and mass data. Spectroscopic and physical characterization data of selected compounds: Compound 5: Rf = 0.40 (15% EtOAc/hexanes); [α]D20 +10.2 (c 0.95, CHCl3); 300 MHz 1H NMR (CDCl3) δ 9.74 (t, J = 1.8 Hz, 1H), 7.35–7.26 (m, 5H), 4.49 (ABq, JAB = 11.9 Hz, Δv = 14.4 Hz, 2H), 3.59–3.46 (m, 2H), 3.30 (dd, J = 8.9, 7.4 Hz, 1H), 2.55–2.30 (m, 2H), 2.04–1.89 (m, 1H), 1.78–1.42 (m, 3H), 0.97 (d, J = 6.9 Hz, 3H), 0.89 (s, 9H), 0.87 (d, J = 6.9 Hz, 3H), 0.04 (s, 6H); 75 MHz 13C NMR (CDCl3) δ 202.6, 138.7, 128.3, 127.5, 127.4, 76.9, 73.0, 72.9, 42.3, 37.9, 35.9, 26.8, 26.1, 18.4, 15.0, 14.1, −3.8, −4.1; IR (neat) 2931, 1726, 1458, 1254, 1097, 1057, 774, 698 cm−1. Anal. Calcd for C22H38O3Si requires C, 69.79; H, 10.12. Found: C, 69.52; H, 10.27. Compound 7: Rf = 0.60 (35% EtOAc/hexanes); [α]D20 −0.8 (c 1.09, CHCl3); 300 MHz 1H NMR (CDCl3) δ 7.37–7.26 (m, 5H), 5.91–5.79 (m, 1H), 5.06–5.03 (m, 1H), 5.00 (dd, J = 1.1, 1.1 Hz, 1H), 4.51, (s, 2H), 3.66 (dd, J = 9.1, 4.3 Hz, 1H), 3.49 (dd, J = 9.1, 6.2 Hz, 1H), 3.40 (ddd, J = 6.9, 4.8, 4.8 Hz, 1H), 3.23–3.21 (m, 1H), 2.37–2.27 (m, 1H), 2.00–1.85 (m, 1H), 1.04 (d, J = 6.7 Hz, 3H), 0.96 (d, J = 7.0, 3H); 75 MHz 13C NMR (CDCl3) δ 142.2, 137.7, 128.4, 127.7, 127.6, 114.0, 78.8, 74.7, 73.5, 40.9, 35.6, 14.4, 13.2; IR (neat) 3487 (broad), 2965, 1455, 1092, 737, 698 cm−1. Anal. Calcd for C15H22O2 requires C, 76.88; H, 9.46. Found: C, 76.66; H, 9.56. Compound 12: Rf = 0.26 (5% EtOAc/hexanes); [α]D20 +8.0 (c 1.28, CHCl3); 300 MHz 1H NMR (CDCl3) δ 7.35–7.25 (m, 5H), 4.90–4.80 (m, 2H), 4.40 (ABq, JAB = 11.9 Hz, Δv = 9.9 Hz, 2H), 3.72–3.62 (m, 1H), 3.56 (dd, J = 8.9, 4.5 Hz, 1H), 3.51 (dd, J = 5.9, 2.1 Hz, 1H), 3.29 (dd, J = 8.2, 8.2 Hz, 1H), 2.24–1.91 (m, 3H), 1.76 (s, 3H), 1.72 (br s, 1H), 1.62–1.37 (m, 4H), 1.24–1.10 (m, 1H), 0.98 (d, J = 6.9 Hz, 3H), 0.89–0.87 (m, 12H), 0.05 (s, 3H), 0.03 (s, 3H); 75 MHz 13C NMR (CDCl3) δ 142.8, 138.8, 128.3, 127.5, 127.4, 113.5, 77.5, 73.1, 72.9, 68.9, 46.1, 37.9, 36.5, 35.4, 30.7, 26.1, 22.4, 18.4, 15.2, 14.3, −3.7, −4.1; IR (neat) 3456 (broad), 2931, 1456, 1254, 1097, 836, 773 cm−1. Anal. Calcd for C26H46O3Si requires C, 71.83; H, 10.67. Found: C, 72.10; H, 10.65. Compound 17: Rf = 0.38 (25% EtOAc/hexanes); [α]D20 +22.5 (c 1.63, CHCl3); 300 MHz 1H NMR (CDCl3) δ 7.34–7.19 (m, 7H), 6.85 (d, J = 8.5 Hz, 2H), 5.82 (ddd, J = 16.0, 10.5, 5.5 Hz, 1H), 5.25 (app dt, J = 17.2, 1.4 Hz, 1H), 5.12 (app dt, J = 10.5, 1.4 Hz, 1H), 4.57–4.35 (m, 5H), 3.90–3.84 (m, 1H), 3.78 (s, 3H), 3.53 (dd, J = 9.16, 4.7 Hz, 1H), 3.48 (dd, J = 6.1, 2.7 Hz, 1H), 3.28, (dd, J = 8.7, 7.6 Hz, 1H), 3.10 (br d, J = 3.8 Hz, 1H), 2.76–2.42 (m, 4H), 1.99–1.88 (m, 1H), 1.60–1.35 (m, 4H), 1.25–1.08 (m, 1H), 0.96 (d, J = 6.9 Hz, 3H), 0.88 (s, 9H), 0.85 (d, J = 6.9 Hz, 3H), 0.02 (s, 6H); 75 MHz 13C NMR (CDCl3) δ 210.2, 159.2, 138.9, 138.7, 130.3, 129.4, 128.3, 127.5, 127.4, 114.9, 113.7, 77.2, 75.5, 73.0, 72.9, 71.3, 68.5, 55.2, 50.1, 48.4, 37.8, 36.6, 32.4, 29.9, 26.1, 18.4, 15.3, 14.3, −3.7, −4.1; IR (neat) 3453 (broad), 2931, 1710, 1514, 1250, 1093, 1038, 836, 757 cm−1. Anal. Calcd for C36H56O6Si requires C, 70.55; H, 9.21. Found: C, 70.36; H, 9.28. Compound 19:Rf = 0.64 (25% EtOAc/hexanes); [α]D20 +7.9 (c 0.1.54, CHCl3); 300 MHz 1H NMR (CDCl3) δ 7.34–7.25 (m, 5H), 5.82 (ddd, J = 16.9, 10.4, 6.1 Hz, 1H), 5.17 (app. dt, J = 17.1, 1.4 Hz, 1H), 5.08–4.99 (m, 2H), 4.94–4.86 (m, 1H), 4.48 (ABq, JAB = 12.1 Hz, Δv = 9.1 Hz, 2H), 4.13 (ddd, J = 6.1, 6.1, 6.1 Hz, 1H), 3.52 (dd, J = 9.0, 4.6 Hz, 1H), 3.47 (dd, J = 6.4, 2.6 Hz, 1H), 3.28 (dd, J = 9.0, 7.6 Hz, 1H), 2.00 (s, 3H), 1.99 (s, 3H), 1.95–1.32 (m, 9H), 1.20–1.08 (m, 1H), 0.95 (d, J = 7.0 Hz, 3H), 0.90 (s, 9H), 0.87 (s, 9H), 0.83 (d, J = 6.7 Hz, 3H), 0.06 (s, 3H), 0.03 (s, 3H), 0.02 (s, 3H), 0.01 (s, 3H); 75 MHz 13C NMR (CDCl3) δ 170.6, 170.3, 140.5, 138.7, 128.3, 127.4, 114.4, 77.1, 73.1, 73.0, 70.7, 70.1, 67.3, 43.2, 38.8, 37.9, 36.3, 33.0, 30.1, 26.1, 25.8, 21.1, 18.4, 18.2, 15.2, 14.0, −3.7, −4.1, −4.4, −4.9; IR (neat) 2931, 1741, 1470, 1363, 1249, 1095, 1027, 837, 775 cm−. Anal. Calcd for C38H68O7Si2 requires C, 65.85; H, 9.89. Found: C, 65.80; H, 9.92.

RESOURCES