Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Sep 1.
Published in final edited form as: Angew Chem Int Ed Engl. 2014 Jul 15;53(36):9632–9636. doi: 10.1002/anie.201405455

Synthesis of (+)-Discodermolide by Catalytic Stereoselective Borylation Reactions**

Zhiyong Yu 1, Robert J Ely 1, James P Morken 1,
PMCID: PMC4171733  NIHMSID: NIHMS626771  PMID: 25045037

Abstract

The marine natural product (+)-discodermolide was first isolated in 1990 and, to this day, remains a compelling synthesis target. Not only does the compound possess fascinating biological activity, but it also presents an opportunity to test current methods for chemical synthesis and provides a forum for the inspiration of new reaction development. In this manuscript, we present a synthesis of discodermolide that employs a previously undisclosed stereoselective catalytic diene hydroboration and also establishes a strategy for chiral enolate alkylation. In addition, this synthesis of discodermolide provides the first examples of diene 1,4-diboration and borylative diene-aldehyde couplings in complex molecule synthesis.

Keywords: discodermolide, total synthesis, asymmetric catalysis, hydroboration, diboration


Discodermolide is a marine natural product that is a highly potent microtubule stabilizing agent.1 Notably, it exhibits activity against paclitaxel-resistant cell lines and has been found to possess synergism with paclitaxel in human carcinoma cells.2,3 Because of these promising features and because isolation4 from its natural source, Discodermia dissoluta, provides only 7.0 mg from 434 grams of sponge, discodermolide has been the subject of many total synthesis efforts.5 Most notably, a group at Novartis prepared 64 g of synthetic material to support clinical trials.5 While pulmonary toxicity was noted among two patients and prompted abandonment of discodermolide as a clinical candidate,1 subsequent studies on metabolism of the drug found near-complete oxidation after one hour in the presence of human liver microsomes.7 Currently, it is unknown whether the pulmonary toxicity alluded to above arises from discodermolide itself or from metabolic byproducts.

A short route to discodermolide (Figure 1 for structure) that might enable new analog synthesis8 should be founded upon enantioselective construction of key chiral building blocks from readily available feedstocks. In considering specific structural demands presented by the assembly of discodermolide, construction of the Z-trisubstituted alkene at C13-C14 has proven to be a most challenging task. Indeed, the manufacturing process for clinical trials of discodermolide employed a reaction sequence that occurs in 30% yield for construction of the C13-C14 region.6 To enable a rapid synthesis of discodermolide and new analogs, it is most critical to construct this element concisely and in an efficient fashion. To accomplish this objective, we considered two key innovations: first, we considered use of Ni-catalyzed 1,4-diene hydroboration9 to establish a Z trisubstituted alkene bearing an allylic functional group (eq. 1). Use of this reaction in the construction of discodermolide requires developing a process that results in stereoselective H-atom addition to the prochiral carbon at C12. Second, to connect C16 with an electrophile derived from the trisubstituted alkene, we considered an enolate alkylation (eq. 2). A central requirement for this technology was the design of a transformation that allows stereoselective α-alkylation of an enolate with control by an α' stereocenter. Most important in the development of this process is that it provides the product in high stereoselection and yield, without requiring an excess of precious building blocks.

Figure 1.

Figure 1

Diastereoselective diene hydroboration (eq. 1) and stereoselective alkylation of enolates (eq. 2) are advanced to facilitate the synthesis of (+)-discodermolide.

Previous studies in our laboratory established that Ni-catalyzed 1,4-hydroboration of 1,3-dienes provides Z-alkenes with excellent control over the olefin geometry (>20:1 Z:E).9,10 Application of this process to the construction of discodermolide requires the reaction of prochiral diene substrates such as in A (Figure 2). A critical question thus arises as to the influence of neighboring stereocenters on the diastereoselectivity of the reduction process (A → B) and to probe these features, we examined several chiral dienol derivatives. Reaction of an unprotected alcohol at 0 °C for 3 h with pinacolborane and 2.5% each of Ni(cod)2 and PCy3, followed by oxidative workup furnished the hydroboration product 1 in moderate yield, moderate diastereoselection and excellent olefin Z stereoselectivity. While the alcohol-derived substrate reacted well, a significant improvement in both yield and stereoselection was observed upon incorporation of silicon protecting groups. As depicted in Table 1, use of a TES protecting group furnished the reaction product 2 in not only excellent yield, but also in enhanced selectivity relative to the unprotected substrate (12:1 vs. 6:1 dr). Use of larger protecting groups served to enhance selectivity such that with the TBDPS-protected substrate (data not shown), the product was obtained as a single stereoisomer according to 1H NMR analysis. Substrates with other substituents and protecting groups also participated and the stereoselectivity trends appear to follow the model depicted by C (inset, Figure 2). We reason that the Ni complex associates with the diene in a manner that positions the metal complex antiperiplanar with respect to the adjacent oxygen atom. This orientation allows the π system of the diene to mix with the C-O σ* and enhance backbonding between the metal and the alkene.11 A conformation such as C, wherein the carbinol hydrogen atom is directed towards the metal complex and the carbinol substituent directed away, serves to minimize steric interactions with the catalyst and leads to a stereocontrolled reaction.

Figure 2.

Figure 2

Ni-catalyzed hydroboration of chiral dienols results in chiral trisubstituted alkenes. Reactions were conducted at [substrate] = 0.25 M and oxidized with 30% H2O2 and 3 M NaOH. Yields refer to isolated yield of purified material and values are an average of two experiments. TES=triethylsilyl, TBS=tert-butyldimethylsilyl.

To establish the C15–C16 linkage in discodermolide, we considered alkylation of a ketone enolate with an electrophile derived from the above-described diene hydroboration. Enolate alkylations that establish this connection in previous syntheses of discodermolide have employed Z-enolates that engage in chelation with the β-oxygen at C13; these reactions proceed in modest selectivity (6:1 d.r.) and require an excess of electrophile (2 equivalents).5e Important precedents involving anti aldol reactions of α-chiral E enolates12 not withstanding, the use of A(1,3) strain to control the α'-alkylation of non-chelated α-chiral ketone enolates remains undeveloped.13 In line with established strategies for acyclic conformational analysis,14 it was considered that the E enolate derived from D (Figure 3), in the absence of intervening chelation effects between neighboring functional groups and the metal enolate might favor conformer F such that alkylation occurs preferentially from the Re face. In a preliminary experiment aimed at addressing this issue, alkylation to give 7 was accomplished by subjecting the corresponding ethyl ketone starting material to deprotonation with lithium tetramethylpiperidide in the presence of lithium bromide, conditions known15 to generate the E-enolate even from hindered ketones. Treatment with 1.5 equivalents of allyl iodide at −78 °C delivered 7 in excellent yield and in good selectivity. Of vital importance for eventual scale-up of complex fragment couplings, the reaction yield was excellent even when a 1:1 stoichiometry of enolate:electrophile was employed. Selective construction of 8 demonstrates that this alkylation strategy applies to syn aldol products, and selective formation of product 9 indicates simple hydrocarbon-derived nucleophiles can exhibit useful levels of selectivity.

Figure 3.

Figure 3

Stereoselective alkylation of α'-chiral E-enolates. Minimization of allylic (1,3) strain in the E-enolate derived from D, establishes a conformation wherein the RL at the stereocenter impedes approach of a reacting electrophile and results in a stereoselective alkylation.

With central technology in place, construction of discodermolide commenced. To prepare fragment 14, a reaction sequence involving catalytic enantioselective 1,4-diboration16 of trans-pentadiene followed by in situ homologation17 and oxidation furnished 1,6-diol 11 in excellent yield and selectivity (Scheme 1). While monoactivation of diol 11 was non-selective, conversion of 11 to diene 12 was readily accomplished by bis(tosylation), selective elimination with a hindered basic alkoxide (potassium tert-butoxide), and detosylation. Importantly, this simple three-step sequence could be accomplished in excellent yield. Oxidation of 12 by the Dess-Martin periodinane, was followed by use of the recently-developed18 Roush reductive aldol reaction; subsequent alkylation and TMS protection furnished 14.

Scheme 1.

Scheme 1

Synthesis of 14 from pentadiene in 8 steps is enabled by catalytic enantioselective tandem diene diboration/homologation/oxidation and reductive aldol methods. DMP=1,1,1-Triacetoxy-1,1-dihydro-1,2-benziodoxol-3(1H)-one, TMS=trimethylsilyl.

Construction of fragment 22 was accomplished in four synthesis steps from chiral aldehyde 16 (Scheme 2). In designing this route, it was considered that access to 16 by hydroformylation of protected allyl alcohol derivatives would be an ideal strategy. Hydroformylation of this substrate class has been addressed by Landis using the bisdiazaPhos ligand19 and similar reactions have been employed by Burke20 and Leighton21 for the construction of polypropionate arrays. Unfortunately, the bisdiazaphos ligands are not readily available. Expecting that related electron-rich bidentate diphosphines might provide sufficient selectivity, alternative ligands were examine in the hydroformylation of TBS-protected allyl alcohol 15. The electronic and steric properties of the commercially available DuPhos22 family of ligands appeared ideally suited to the task and it was found that phenyl-substituted bis(phospholano)ethane (PhBPE, 17) was most effective, delivering chiral aldehyde 16 in outstanding levels of enantioselection and yield, and with moderate-but-useful levels of regiocontrol.23 Importantly, very low catalyst loadings are effective and the unpurified reaction product is sufficiently free from contaminants that it may be used directly in subsequent operations. To complete construction of 22, aldehyde 16 was subjected to Walsh chelation-controlled addition24 of vinyl zinc 18, which is readily available from hydroboration of 19 followed by boron-zinc exchange. Ni-catalyzed hydroboration/oxidation of 21 followed by iodinolysis furnished 22 in excellent yield and diastereoselection.

Scheme 2.

Scheme 2

Synthesis of 22 in 5 steps from TBS-protected allyl alcohol 15. TBSOTf=tert-butyldimethylsilyl trifluoromethane sulfonate, HBpin=pinacolborane, ONf=nonafluorobutane sulfonyl.

Lastly, construction of fragment 29 employed a sequence involving catalytic diastereoselective borylative diene-aldehyde coupling (Scheme 3).25 To begin this sequence, aldehyde 24 was prepared by Rh/PhBPE catalyzed hydroformylation of the cyclic o-xylyl acetal26 of acrolein (23). Likely due to the influence of a second electronegative heteroatom at the allylic carbon, this hydroformylation occurred with outstanding branch:linear selectivity (b:l = 13:1) and, like the example in Scheme 2, with excellent enantiocontrol. While the subsequent borylative diene-aldehyde coupling reaction has precedent in our laboratory using P(SiMe3)3 as a ligand, it was found that PBn3 is equally effective and allows for high levels of stereoinduction from aldehyde 24 in conversion to 25. Iridium-catalyzed hydroboration/oxidation27 followed by neutral hydrogenolytic deprotection of the o-xylyl acetal, oxidation, and esterfication furnished 26. Primary alcohol 26 was converted to fragment 29 by oxidation, chlorination and carbonyl substitution employing a phosphoryl anion derived from 28.

Scheme 3.

Scheme 3

Synthesis of 29 from acrolein cyclic o-xylyl acetal 23. B2(pin)2=bis(pinacolato)diboron, pTsOH=para-toluene sulfonic acid, dppm=bis(diphenylphosphino)methane, cod=1,5-cyclooctadiene.

In line with the enolate alkylation reactions described above, the E-enolate derived from 14 (Scheme 4) was treated with 1.0 equivalent of allyl iodide 22 and this furnished 30 in excellent yield and stereoselectivity. After replacement of the TMS protecting group in 30 with a requisite primary carbamate, the primary TBS protecting group was removed and the product alcohol oxidized to give aldehyde 31. The reaction sequence for completion of the target molecule was influenced by expertise developed in the course of reported prior syntheses.5d,28 Thus 31 was subjected to Still-Gennari olefination with 29 to give 32.29 Final conversion of 32 to discodermolide was accomplished by the three-step reduction and deprotective lactonization sequence depicted in Scheme 4. 1H and 13C NMR spectra, optical rotations and mass spectral data were identical to those reported for the natural product.

Scheme 4.

Scheme 4

Stereoselective enolate alkylation to give 30 and the conversion to discodermolide. LiTMP=lithium 2,2,6,6-tetramethylpiperamide, PPTS=pyridinium para-toluene sulfonate, p-TsOH=para-toluene sulfonic acid,

Overall, the synthesis of discodermolide was accomplished in a total of 36 steps with a longest linear sequence of 17 steps (13% yield) from commercially available materials. While it is anticipated that the synthesis strategy described herein will offer access to new discodermolide analogs that may address biological limitations of the natural product itself, it is also anticipated that the synthetic methods developed to address this structure may have value in other ventures.

Supplementary Material

Supporting Information

Footnotes

**

This work was supported by grants from the National Institutes of Health (GM 059417 & GM 064451). Z.Y. and R.J.E. were supported by LaMattina Fellowships.

Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/anie.201xxxxxx.

References

  • 1.Longley RE. Natural Products and Cancer Drug Discovery. New York: Springer; 2013. Discodermolide: past, present, and future; pp. 39–58. [Google Scholar]
  • 2.Kowalski RJ, Giannakakou P, Gunasekera SP, Longley RE, Day BW, Hamel E. Mol. Pharmacol. 1997;52:613–622. [PubMed] [Google Scholar]
  • 3.Martello LA, McDaid HM, Regl DL, Yang CH, Meng D, Pettus TRR, Kaufman MD, Arimoto H, Danishefsky SJ, Smith AB, Horwitz SB. Clin. Cancer Res. 2000;6:1978–1987. [PubMed] [Google Scholar]
  • 4.Gunasekera SP, Gunasekera M, Longley RE. J. Org. Chem. 1990;55:4912–4915. [Google Scholar]
  • 5.For completed total syntheses of discodermolide, see; Nerenberg JB, Hung DT, Somers PK, Schreiber SL. J. Am. Chem. Soc. 1993;115:12621–12622. Hung DT, Nerenberg JB, Schreiber SL. Chem. Biol. 1994;1:67–71. doi: 10.1016/1074-5521(94)90042-6. Smith AB, Qiu Y, Jones DR, Kobayashi K. J. Am. Chem. Soc. 1995;117:12011–12012. Hung DT, Nerenberg JB, Schreiber SL. J. Am. Chem. Soc. 1996;118:11054–11080. Harried SS, Yang G, Strawn MA, Myles DC. J. Org. Chem. 1997;62:6098–6099. Marshall JA, Johns BA. J. Org. Chem. 1998;63:7885–7892. doi: 10.1021/jo971900+. Smith AB, Kaufman MD, Beauchamp TJ, LaMarche MJ, Arimoto H. Org. Lett. 1999;1:1823–1826. doi: 10.1021/ol9910870. Halstead DP. Ph.D. Thesis. Cambridge, MA: Harvard University; 1999. Paterson I, Florence GJ, Gerlach K, Scott J. Angew. Chem., Int. Ed. 2000;39:377–380. doi: 10.1002/(sici)1521-3773(20000117)39:2<377::aid-anie377>3.0.co;2-e. Angew. Chem. 2000;112:385–388. Harried SS, Lee CP, Yang G, Lee TIH, Myles DC. J. Org. Chem. 2003;68:6646–6660. doi: 10.1021/jo034521r. Smith AB, Freeze BS, Brouard I, Hirose T. Org. Lett. 2003;5:4405–4408. doi: 10.1021/ol035697i. Paterson I, Delgado O, Florence GJ, Lyothier I, Scott JP, Sereinig N. Org. Lett. 2003;5:35–38. doi: 10.1021/ol0270780. Mickel SJ, Niederer D, Daeffler R, Osmani A, Kuesters E, Schmid E, Schaer K, Gamboni R, Chen W, Loeser E, Kinder FR, Konigsberger K, Prasad K, Ramsey TM, Repic O, Wang RM, Florence G, Lyothier I, Paterson I. Org. Process Res. Dev. 2004;8:122–130. Arefolov A, Panek JS. J. Am. Chem. Soc. 2005;127:5596–5603. doi: 10.1021/ja043168j. Smith AB, Freeze BS, Xian M, Hirose T. Org. Lett. 2005;7:1825–1828. doi: 10.1021/ol050455z. Paterson I, Lyothier I. Org. Lett. 2004;6:4933–4936. doi: 10.1021/ol0478842. de Lemos E, Porée FH, Commerçon A, Betzer JF, Pancrazi A, Ardisson J. Angew. Chem. Int. Ed. 2007;46:1917–1921. doi: 10.1002/anie.200604629. Angew. Chem. 2007;119:1949–1953. For a discussion of partial syntheses, see reference 5n.
  • 6.Mickel SJ. Pure Appl. Chem. 2007;79:685–700. [Google Scholar]
  • 7.Fan Y, Schreiber EM, Day BW. J. Nat. Prod. 2009;72:1748–1754. doi: 10.1021/np900245k. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.For an overview of total synthesis and analogs, see: Smith AB, III, Freeze BS. Tetrahedron. 2008;64:261–298. doi: 10.1016/j.tet.2007.10.039.
  • 9.Ely RJ, Morken JP. J. Am. Chem. Soc. 2010;132:2534–2535. doi: 10.1021/ja910750b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.For other catalytic hydroboration of dienes, see: Zaidlewicz M, Meller J. Tetrahedron Lett. 1997;38:7279–7282. Satoh M, Nomoto Y, Miyaura N, Suzuki A A. Tetrahedron Lett. 1989;30:3789–3792. Matsumoto Y, Hayashi T. Tetrahedron Lett. 1991;32:3387–3390. Wu JY, Moreau B, Ritter T. J. Am. Chem. Soc. 2009;131:12915–12917. doi: 10.1021/ja9048493. Sasaki Y, Zhong C, Sawamura M, Ito H. J. Am. Chem. Soc. 2010;132:1226–1227. doi: 10.1021/ja909640b. Semba K, Shinomiya M, Fujihara T, Terao J, Tsuji Y. Chem. Eur. J. 2013;19:7125–7132. doi: 10.1002/chem.201300443.
  • 11.For a related orbital analysis in Rh-catalyzed hydroboration of alkenes, see: Burgess K, van der Donk WA, Jarstifer MB, Ohlmeyer MJ. J. Am. Chem. Soc. 1991;113:6139–6144.
  • 12. Paterson I, Goodman JM, Isaka M. Tetrahedron Lett. 1989;30:7121–7124. Evans DA, Ng HP, Clark JS, Rieger DL. Tetrahedron Lett. 1992;48:2127–2142. Paterson I, Wallace DJ, Velázquez SM. Tetrahedron Lett. 1994;35:9083–9086. For a review, see: Cowden CJ, Paterson I. Organic Reactions. 1997;51:1–200.
  • 13.Hoffmann RW. Chem. Rev. 1989;89:1841–1860. [Google Scholar]
  • 14.Eliel EL, Wilen SH, Mander LN. Stereochemistry of organic compounds. J. Wiley & Sons; 1994. [Google Scholar]
  • 15.Hall PL, Gilchrist JH, Collum DB. J. Am. Chem. Soc. 1991;113:9571–9574. [Google Scholar]
  • 16.Schuster CH, Li B, Morken JP. Angew. Chem. Int. Ed. 2011;50:7906–7909. doi: 10.1002/anie.201102404. [DOI] [PMC free article] [PubMed] [Google Scholar]; Angew. Chem. 2011;123:8052–8055. [Google Scholar]
  • 17.Sadhu KM, Matteson DS. Organometallics. 1985;4:1687–1689. [Google Scholar]
  • 18.Nuhant P, Allais C, Roush WR. Angew. Chem. Int. Ed. 2013;52:8703–8707. doi: 10.1002/anie.201302535. [DOI] [PMC free article] [PubMed] [Google Scholar]; Angew. Chem. 2013;125:8865–8869. [Google Scholar]
  • 19.McDonald RI, Wong GW, Neupane RP, Stahl SS, Landis CR. J. Am. Chem. Soc. 2010;132:14027–14029. doi: 10.1021/ja106674n. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Risi RM, Burke SD. Org. Lett. 2012;14:2572–2575. doi: 10.1021/ol3008765. [DOI] [PubMed] [Google Scholar]
  • 21.Ho S, Bucher C, Leighton JL. Angew. Chem. Int. Ed. 2013;52:6757–6761. doi: 10.1002/anie.201302565. [DOI] [PMC free article] [PubMed] [Google Scholar]; Angew. Chem. 2013;125:6889–6893. [Google Scholar]
  • 22.Burk MJ, Feaster JE, Nugent WA, Harlow RL. J. Am. Chem. Soc. 1993;115:10125–10138. [Google Scholar]
  • 23.For other examples of asymmetric hydroformylation with Ph(BPE), see: Axtell AT, Cobley CJ, Klosin J, Whiteker GT, Zanotti-Gerosa A, Abboud KA. Angew. Chem. Int. Ed. 2005;44:5834–5838. doi: 10.1002/anie.200501478. Angew. Chem. 2005;117:5984–5988. Zhang X, Cao B, Yu S, Zhang X. Angew. Chem. Int. Ed. 2010;49:4047–4050. doi: 10.1002/anie.201000955. Angew. Chem. Int. Ed. 2010;122:4141–4144.
  • 24.Stanton GR, Kauffman MC, Walsh PJ. Org. Lett. 2012;14:3368–3371. doi: 10.1021/ol301354w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cho HY, Morken JP. J. Am. Chem. Soc. 2010;132:7576–7577. doi: 10.1021/ja101513d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Machinaga N, Kibayashi C. Tetrahedron Lett. 1989;30:4165–4168. [Google Scholar]
  • 27.Yamamoto Y, Fijukawa R, Umemoto T, Miyaura N. Tetrahedron. 2004;60:10695–10700. [Google Scholar]
  • 28.Paterson I, Lyothier I. J. Org. Chem. 2005;70:5494–5507. doi: 10.1021/jo050481a. [DOI] [PubMed] [Google Scholar]
  • 29.Still WC, Gennari C. Tetrahedron Lett. 1983;24:4405–4408. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information

RESOURCES