Abstract
A synthesis of 2-epi-amphidinolide E (1) has been accomplished via an unexpected and highly diastereoselective C(2) stereochemical inversion during the modified Yamaguchi esterification of alcohol 4b and Fe(CO)3-complexed dienoic acid 7.
The amphidinolides are a family of structurally diverse macrolides isolated from the dinoflagellate Amphidinium sp, many of which display impressive anti-tumor activity.1 As a consequence, the amphidinolides have attracted considerable interest as targets for synthesis and biological evaluation. Total syntheses of amphidinolides A,2 J,3 K,4 P,5 T,6 W,7 X8 and Y9 have been reported.
Amphidinolide E10 (2) is a 19-membered macrolactone featuring an embedded cis-tetrahydrofuran. While this structural motif is common within the amphidinolide family, the C(1)-C(6) α-chiral, β,γ,δ,ε-dienoate moiety of amphidinolide E is not found in any of the other amphidinolides. Amphidinolide E is cytotoxic to murine lymphoma L1210 and human epidermoid carcinoma KB cells, IC50 = 2.0 and 10.0 μg/mL, respectively.1 Lee has reported the total synthesis of amphidinolide E.11 Furthermore, Gurjar12 and Marshall13 have published studies toward amphidinolide E (2).
We recently reported a total synthesis of amphidinolide E (2) via the ring closing metathesis of polyene 3a (Figure 1).14 Installation of the C(1)-C(6) α-chiral, unconjugated dienoate via direct esterification of hindered alcohol 4a (and closely related anologs) even in the presence of large excesses (10-20 equiv) of acid 5 proved to be extremely challenging. Only trace amounts of the corresponding ester were isolated under a variety of conditions. Furthermore, acid 5 was typically recovered as the fully conjugated, diene migrated carboxylic acid species.14
Figure 1.

Strategy for the synthesis of amphidinolide E (1)
Ultimately, we found that use of (CO)3Fe-complexed acid 6,14, 15 a “diene protected” analog of 5, resulted in an efficient esterification of 4b under the modified Yamaguchi conditions16 (reaction pathway A, Scheme 1). Subsequent oxidative decomplexation afforded polyene 3a, which was smoothly elaborated to amphidinolide E.
Scheme 1.

Divergent Reaction Pathways for Acids 6 and 7
Surprisingly, when acid 7, the (CO)3Fe-diene diastereomer of 6, was used in the same esterification-decomplexation sequence, polyene 3a was not observed (reaction pathway B, Scheme 1). Instead, a product (3b) with a very similar 1H NMR spectrum to polyene 3a was isolated. It was subsequently determined that this product was the C(2) stereochemically inverted isomer of 3a, namely polyene 3b.17
The stereochemical assignment of C(2) of 3b was made on the basis of the data summarized in Table 1. Compounds 6, 7, 8, and 9 were transformed into diene 10, whose optical rotations were compared with material independently synthesized from commercially available methyl (S)-(+)-3-hydroxy-2-methylpropionate.18 Esters 8 and 9 are the immediate precursors of 3a and 3b, respectively. The C(2) stereochemistry of acids 6 and 7 was verified to be 2S before being subjected to the esterification reaction (entries 1 & 2, Table 1). Furthermore, the 2S stereochemistry was confirmed for amphidinolide E precursor 8 (prepared via the esterification reaction of 4b using acid 6, entry 3, Table 1).14 On the other hand, the data in entry 4 leave no doubt that inversion at C2 occurs when acid 7 is used in the esterification reaction of 4b (entry 4, Table 1).
Table 1.
Stereochemical Correlations Implicating Inversion at C(2) During the Esterification Reaction Leading to 3b (via 9) 
| entry | starting material | steps | overall yield | product [(+)-or(-)-10] |
|---|---|---|---|---|
| 1 | ![]() |
A | 84% | (+)-(R)-10 |
| 2 | ![]() |
A | 68% | (+)-(R)-10 |
| 3 | ![]() |
B | 56% | (+)-(R)-10 |
| 4 | ![]() |
C | 61% | (-)-(S)-10 |
Steps A: 1) BH3·DMS, THF; 2) TBSCl, imidazole, CH2Cl2; 3) CAN, acetone, 0 °C. Steps B: 1) DIBAL, MePh, -78 °C; 2) TBSCl, imidazole, CH2Cl2; 3) CAN, acetone, 0 °C
Closure of the 19-membered macrocycle in 11 was accomplished in 60% yield via ring closing metathesis with 20 mol % of Grubbs’ first generation catalyst (Scheme 2). In addition, an inseparable mixture of enyne metathesis products was also isolated in 15% yield. Use of Grubbs’ second generation19 or the Grubbs-Hoveyda19 catalysts resulted in only trace amounts of macrocyle 11 and significant decomposition of polyene 3b. Stannylalumination-protonolysis20 of alkyne 11 afforded vinylstannane 12 (51% yield). Treatment of vinylstannane 12 with N-iodosuccinimide yielded vinyl iodide 13 (93% yield). Acidic hydrolysis of the triethylsilylether and acetonide protecting groups in 13 afforded an inseparable 10:1 mixture of the desired C(18) and undesired C(17) lactone regioisomers. This is in contrast with our synthesis of amphidinolide E, in which case the analogous deprotection step provided only the 19-membered lactone.14 Stille21 cross coupling of the crude mixture of vinyl iodides with vinylstannane 1412 followed by HPLC purification afforded 2-epi-amphidinolide E in 34% yield from 13. Biological data for 2-epi-amphidinolide E will be reported in due course.
Scheme 2.

Completion of Synthesis of 2-epi-Amphidinolide E (1)
The striking diastereoselectivity (dr = 20:1) for the formation of the C(2)-epimers of 8 and 9 prompted further studies to discern the role, if any, that the alcohol plays in influencing the diastereoselectivity of these esterification reactions. Table 2 summarizes the diastereoselectivites and yields obtained for the esterification reactions of three different alcohols with acids 6 and 7. Esterification of the sterically unhindered, primary alcohol 3-phenylpropanol afforded a 1:1 C(2)-diastereomeric mixture with both acids 6 and 7 (entries 1 & 2, Table 2).22 In contrast, esterification of the hindered primary alcohol, 2,2-dimethylpropanol, with acid 6 yielded a single isomer 1723, but a 1:1 mixture of ent-17 and 18 was obtained when acid 7 was used (entries 3 & 4).22 Interestingly, coupling of 3-pentanol and acid 7 afforded a 5:1 mixure favoring 20 (entry 6), whereas 1924 was obtained exclusively from the coupling of 6 and 3-pentanol (entry 5). Ester 20 is the enantiomer of 19, and can be obtained only if inversion of C(2) of 7 occurs during the esterification reaction.
Table 2.
Esterification Reactions of Acids 6 and 7 with Primary and Secondary Alcohols
| entry | alcohol | acid | product | dr | % yield |
|---|---|---|---|---|---|
| 1 | 6 | ![]() |
1:1 | 99% | |
| 2 | 7 | ![]() |
1:1 | 99% | |
| 3 | ![]() |
6 | ![]() |
>20:1 | 79% |
| 4 | 7 | ![]() |
1:1 | 69% | |
| 5 | ![]() |
6 | ![]() |
>20:1 | 99% |
| 6 | 7 | ![]() |
5:1 | 99% |
Esterification conditions: alcohol (1 equiv), acid (1.3 equiv), 2,4,6-trichlorobenzoyl chloride, Et3N, DMAP, THF, 0 to 25 °C.
We hypothesize that ketene intermediates may be involved in these esterification reactions (Figure 2). Ketenes are known to be generated from active ester intermediates in esterifications of carboxylic acids with acidic alpha protons.24 Subsequent addition of the alcohol to ketene 22 and reformation of the C(2) stereocenter may occur diastereoselectively via protonation anti to the (CO)3Fe unit in the lowest energy conformation 23. Alternatively, DMAP could add to the ketene intermediate.25 Subsequent diastereoselective protonation of the enolate 24 by ROH would provide the acyl pyridinium salt 25.
Figure 2.

Proposed Esterification Pathway
The rate of ketene formation could well be different for the two epimeric Fe(CO)3 complexed dienoic acids 6 and 7. In addition, unhindered alcohols (e.g., 3-phenylpropanol) could react with the active ester intermediates generated from 6 and 7 at rates competitive with ketene formation. Consequently, high diastereoselectivity is realized in these esterification reactions only with hindered, less reactive secondary alcohols (e.g., 4b and 3-pentanol) for which the rate of direct esterification from the initially formed active ester intermediate is substantially slower than the rates of ketene formation from 6 and 7.
An alternative mechanism is also plausible based on the work of Donaldson.15 Donaldson has demonstrated that the methyl ester corresponding to 7 rapidly epimerizes under basic conditions, with an equilibrium ratio of approximately 1.4 : 1 at C(2). Donaldson also demonstrated that the diastereomeric methyl ester 21 (R = Me) undergoes hydrolysis faster than the methyl ester of 715 Accordingly, it is conceivable that a rapid C(2) epimerization of the active esters generated from 6 and 7 followed by dynamic kinetic resolution of this mixture could also play a role in some of the reactions reported herein. However, the fact that both 6 and 7 display C(2) epimerization without significant asymmetric induction in esterification with simple alcohols (e.g., Table 2, entries 1, 2, and 4), and the fact that 6 and 7 display strikingly different behavior in esterication reactions with neopentanol (entries 6 and 7), leads us to favor the ketene reaction pathway in competition with direct esterification of the active ester intermediates (but without dynamic resoluation) to rationalize the stereoselectivity observed these reactions, and especially in those cases that proceed with excellent asymmetric induction (e.g., 8 and 9, and entries 3, 5, and 6 of Table 2).
In summary, we have synthesized 2-epi-amphidinolide E via an unexpected and highly diastereoselective C(2) stereochemical inversion that occurs during the modified Yamaguchi esterification reaction of 4b and 7. A mechanistic rationale is presented that implicates diastereoselective protonation of enol (23) or enolate (24) intermediates derived from addition of the alcohol to the ketene intermediate 22.
Supplementary Material
Acknowledgment
This work was supported by the National Institutes of Health (GM 38436).
References
- 1(a).For reviews on the amphidinolides:Kobayashi J, Ishibashi M. Chem. Rev. 1993;93:1753.Chakraborty TK, Das S. Curr. Med. Chem.: Anti-Cancer Agents. 2001;1:131. doi: 10.2174/1568011013354660.Kobayashi J, Shimbo K, Kubota T, Tsuda M. Pure Appl. Chem. 2003;75:337.Kobayashi J, Tsuda M. Nat. Prod. Rep. 2004;21:77. doi: 10.1039/b310427n.
- 2(a).Lam HW, Pattenden G. Angew. Chem., Int. Ed. Engl. 2002;41:508. doi: 10.1002/1521-3773(20020201)41:3<508::aid-anie508>3.0.co;2-7. [DOI] [PubMed] [Google Scholar]; (b) Maleczka RE, Jr., Terrell LR, Geng F, Ward JS., III Org. Lett. 2002;4:2841. doi: 10.1021/ol0262284. [DOI] [PubMed] [Google Scholar]; (c) Trost BM, Chisholm JD, Wrobleski SJ, Jung M. J. Am. Chem. Soc. 2002;124:12420. doi: 10.1021/ja027883+. [DOI] [PubMed] [Google Scholar]; (d) Trost BM, Harrington PE. J. Am. Chem. Soc. 2004;126:5028. doi: 10.1021/ja049292k. [DOI] [PubMed] [Google Scholar]; (e) Trost BM, Wrobleski ST, Chisholm JD, Harrington PE, Jung M. J. Am. Chem. Soc. 2005;127:13589. doi: 10.1021/ja0533646. [DOI] [PubMed] [Google Scholar]; (f) Trost BM, Harrington PE, Chisholm JD, Wrobleski ST. J. Am. Chem. Soc. 2005;127:13598. doi: 10.1021/ja053365y. [DOI] [PubMed] [Google Scholar]
- 3.Williams DR, Kissel WS. J. Am. Chem. Soc. 1998;120:11198. [Google Scholar]
- 4.Williams DR, Meyer KG. J. Am. Chem. Soc. 2001;123:765. doi: 10.1021/ja005644l. [DOI] [PubMed] [Google Scholar]
- 5(a).Williams DR, Myers BJ, Mi L. Org. Lett. 2000;2:945. doi: 10.1021/ol0000197. [DOI] [PubMed] [Google Scholar]; (b) Trost BM, Papillon JPN. J. Am. Chem. Soc. 2004;126:13618. doi: 10.1021/ja045449x. [DOI] [PubMed] [Google Scholar]; (c) Trost BM, Papillon JPN, Nussbaumer T. J. Am. Chem. Soc. 2005;127:17921. doi: 10.1021/ja055967n. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6(a).Fürstner A, Aissa C, Riveiros R, Ragot J. Angew. Chem., Int. Ed. Engl. 2002;41:4763. doi: 10.1002/anie.200290042. [DOI] [PubMed] [Google Scholar]; (b) Aiessa C, Riveiros R, Ragot J, Fürstner A. J. Am. Chem. Soc. 2003;125:15512. doi: 10.1021/ja038216z. [DOI] [PubMed] [Google Scholar]; (c) Ghosh AK, Liu C. J. Am. Chem. Soc. 2003;125:2374. doi: 10.1021/ja021385j. [DOI] [PubMed] [Google Scholar]; (d) Ghosh AK, Liu C. Strategies Tactics Org. Synth. 2004;5:255. [Google Scholar]; (e) Colby EA, O’Brien KC, Jamison TF. J. Am. Chem. Soc. 2004;126:998. doi: 10.1021/ja039716v. [DOI] [PubMed] [Google Scholar]; (f) Colby EA, O’Brien KC, Jamison TF. J. Am. Chem. Soc. 2005;127:4297. doi: 10.1021/ja042733f. [DOI] [PMC free article] [PubMed] [Google Scholar]; (g) O’Brien KC, Colby EA, Jamison TF. Tetrahedron. 2005;61:6243. [Google Scholar]; (h) Deng L-S, Huang X-P, Zhao G. J. Org. Chem. 2006;71:4625. doi: 10.1021/jo0605086. [DOI] [PubMed] [Google Scholar]
- 7(a).Ghosh AK, Gong G. J. Am. Chem. Soc. 2004;126:3704. doi: 10.1021/ja049754u. [DOI] [PubMed] [Google Scholar]; (b) Ghosh AK, Gong G. J. Org. Chem. 2006;71:1085. doi: 10.1021/jo052181z. [DOI] [PubMed] [Google Scholar]
- 8.Lepage O, Kattnig E, Fürstner A. J. Am. Chem. Soc. 2004;126:15970. doi: 10.1021/ja044130+. [DOI] [PubMed] [Google Scholar]
- 9.Fürstner A, Kattnig E, Lepage O. J. Am. Chem. Soc. 2006;128:9194. doi: 10.1021/ja061918e. [DOI] [PubMed] [Google Scholar]
- 10(a).Kobayashi J, Ishibashi M, Murayama T, Takamatsu M, Iwamura M, Ohizumi Y, Sasaki T. J. Org. Chem. 1990;55:3421. [Google Scholar]; (b) Kubota T, Tsuda M, Kobayashi J. i. J. Org. Chem. 2002;67:1651. doi: 10.1021/jo016326n. [DOI] [PubMed] [Google Scholar]
- 11.Kim CH, An HJ, Shin WK, Yu W, Woo SK, Jung SK, Lee E. Angew. Chem Int. Ed. 2006;45 doi: 10.1002/anie.200603363. ASAP. [DOI] [PubMed] [Google Scholar]
- 12.Gurjar MK, Mohapatra S, Phalgune UD, Puranik VG, Mohapatra DK. Tetrahedron Lett. 2004;45:7899. [Google Scholar]
- 13.Marshall JA, Schaaf G, Nolting A. Org. Lett. 2005;7:5331. doi: 10.1021/ol0523493. [DOI] [PubMed] [Google Scholar]
- 14.Va P, Roush WR. J. Am. Chem. Soc. 2006;128:ASAP. doi: 10.1021/ja066663j. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15(a).Donaldson WA, Craig R, Spanton S. Tetrahedron Lett. 1992;33:3967. [Google Scholar]; (b) Wasicak JT, Craig RA, Henry R, Dasgupta B, Li H, Donaldson WA. Tetrahedron. 1997;53:4185. [Google Scholar]
- 16.Hikota M, Sakurai Y, Horita K, Yonemitsu O. Tetrahedron Lett. 1990;31:6367. [Google Scholar]
- 17.Chronologically, polyene 3b was synthesized prior to 3a and was elaborated to 2-epi-amphidinolide E. We initially suspected that the stereochemistry of natural amphidinolide E may have been misassigned. It ultimately became apparent that a stereochemical inversion had occurred at C(2) in the esterification of 4b and 7 after we repeated Kobayashi’s stereochemical assignments for amphidinolide E by using advanced intermediates such as 11 as correlation compounds. Details of these stereochemical assignments will be reported in a full paper.
- 18.Diene 10 was independently synthesized in 5 steps from methyl (S)-(+)-3-hydroxy-2-methylpropionate, see Supporting Information.
- 19(a).For reviews of olefin metathesis:Gradillas A, Perez-Castells J. Angew. Chem., Int. Ed. 2006;45:6086. doi: 10.1002/anie.200600641.Nicolaou KC, Bulger PG, Sarlah D. Angew. Chem., Int. Ed. 2005;44:4490. doi: 10.1002/anie.200500369.Fürstner A. Angew. Chem., Int. Ed. 2000;39:3012.
- 20.Sharma S, Oehlschlager AC. J. Org. Chem. 1989;54:5064. [Google Scholar]
- 21(a).Stille JK, Groh BL. J. Am. Chem. Soc. 1987;109:813. [Google Scholar]; (b) Han X, Stoltz BM, Corey EJ. J. Am. Chem. Soc. 1999;121:7600. [Google Scholar]
- 22.A 1:1 mixture of C(2)-epimers was also obtained in the esterification of 3-phenylpropanol and acid 6 using EDCI·MeI and DMAP in CH2Cl2. The reaction does not proceed without DMAP.
- 23.The C(2) stereochemistry of 17 and 19 was confirmed via a three step transformation to (+)-(R)-10: 1) DIBAL, MePh, -78 °C; 2) TBSCl, imidazole, CH2Cl2; 3) CAN, acetone, 0 °C.
- 24(a).Nahmany M, Melman A. Org. Lett. 2001;3:3733. doi: 10.1021/ol0166855. [DOI] [PubMed] [Google Scholar]; (b) Shelkov R, Nahmany M, Melman A. J. Org. Chem. 2002;67:8975. doi: 10.1021/jo0263824. [DOI] [PubMed] [Google Scholar]
- 25(a).Hodous BL, Ruble JC, Fu GC. J. Am. Chem. Soc. 1999;121:2637. [Google Scholar]; (b) Wiskur SL, Fu GC. J. Am. Chem. Soc. 2005;127:6176. doi: 10.1021/ja0506152. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.












