Prins Reaction of Homoallenyl Alcohols: Access to Substituted Pyrans in the Halichondrin Series

Article abstract of DOI:10.1021/acs.orglett.7b02934

Prins reaction of homoallenyl alcohols with aldehyde dimethylacetals in the presence of methoxyacetic acid directly affords tetrasubstituted pyrans relevant to halichondrins with complete control of the C27 stereogenic center. Regioselective Tsuji reducti

Full text of DOI:10.1021/acs.orglett.7b02934

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX-XXX
Prins Reaction of Homoallenyl Alcohols: Access to Substituted
Pyrans in the Halichondrin Series
Hyeong-wook Choi,* Francis G. Fang, Hui Fang, Dae-Shik Kim, Steven R. Mathieu, and Robert T. Yu
Integrated Chemistry Engine, Eisai AiM Institute, 4 Corporate Drive, Andover, Massachusetts 01810, United States
S
* Supporting Information
ABSTRACT: Prins reaction of homoallenyl alcohols with aldehyde dimethylacetals in the presence of methoxyacetic acid
directly affords tetrasubstituted pyrans relevant to halichondrins with complete control of the C27 stereogenic center.
Regioselective Tsuji reduction of the resultant allylic acetates stereoselectively establishes the C25 stereogenic center and C26
exocyclic olefin. Building upon these findings, we achieved concise access to the halichondrin C14−C38 and eribulin C14−C35
fragments.
Eribulin (2) is a highly potent tubulin binding agent approved
alichondrins are polyether macrolides with potent in vitro
Hand in vivo antitumor activity isolated from the marine for the treatment of breast cancer and soft tissue sarcoma in more
sponge Halichondria okadia by Uemura, Hirata, and co-workers.1
than 50 countries. Among the common structural features of
eribulin and halichondrins, the C23−C27 pyran moiety
containing three stereogenic centers and one exocyclic olefin is
notable as the central ring in the carbon-linked trisaccharide
building units (i.e., C14−C35 and C14−C38 for eribulin and
halichondrins, respectively). The manufacturing route for
eribulin features iterative Nozaki−Hiyama−Kishi/Williamson
ether syntheses to assemble C14−C35. While the conciseness
and convergence of the synthetic route was notable, the need to
establish each ring in two separate steps and employ metals in the
C−C bond forming step suggested opportunities for improve-
ment (Scheme 1).5
The first total synthesis of halichondrins was reported by the
Kishi group in 1992.2,3 Biological evaluation of synthetic
intermediates revealed that the antitumor activity resides in the
macrocyclic portion of halichondrin B, which has led to the
discovery and development of eribulin (Figure 1).4
We were intrigued with the possibility that the central C23−
C27 tetrasubstituted pyran could be assembled in a single
sequence utilizing an overall “reductive” Prins reaction6 of a
homoallenyl alcohol (Scheme 2). Attractive features of such an
approach included the potential to derive all of the stereo-
chemical information from a single stereogenic center. Among
the issues to be addressed were (1) viability of a homoallenyl
alcohol as a Prins reaction substrate7 and (2) regio- and
stereoselective capture of the putative allylic carbocation
intermediate with a hydride source. In this paper, we report
the successful realization of these goals and a concise assembly of
halichondrin C14−C38 and eribulin C14−C35 fragments.
Received: September 19, 2017
Figure 1. Halichondrin B and Eribulin.
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b02934
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
reduction of the presumed allylic carbocation intermediate.
Mindful of Tsuji’s work,9 we decided that an allylic acetate could
serve as a viable substrate for the allylic reduction and was thus
the initial target for the Prins reaction. When allene 7 and
aldehyde 6b were subjected to the Prins reaction in the presence
of trifluoroacetic acid,10a the desired product 8a was obtained in
37% yield along with the diene 9 as a side product.11 Encouraged
by the initial success, efforts to improve the overall reaction
profile focused on (1) evaluation of more nucleophilic carboxylic
acids to more efficienly capture the tertiary allylic cation and (2)
inclusion of a strong Lewis acid to allow the reaction to be
conducted at lower temperature. Thus, the Prins reaction was
attempted in the presence of acetic acid and various Lewis
acids.10 After screening the reaction conditions, we identified
BF3−OEt2 as the preferred Lewis acid; 15 equiv of acetic acid
provided the optimum yield of the allylic ester, and the
temperature range of −40 to −20 °C gave the best reaction
profile. Finally, even though both dimethyl acetal and the
corresponding aldehyde could serve as the substrate for the
allene-Prins reactions, dimethyl acetal gave slightly higher
chemical yield with additional convenience in handling due to
its stability. Under the optimized conditions, a mixture of
homoallenyl alcohol 7 and dimethyl acetal 6a in methylene
chloride was treated with 15 equiv of acetic acid and 3 equiv of
BF3−OEt2 at −40 to −20 °C. Typically, the reaction was
completed in 1 h, and the desired product 8b was obtained in
75% yield after column chromatographic purification.
Scheme 1. Synthesis of Eribulin C14−C35 Fragment
(Previous Works and This Work)
Scheme 2. Ovearall “Reductive” Prins Reaction to
Tetrahydropyrans
With the desired allylic acetates in hand, Tsuji reduction was
attempted to convert the allylic esters to the corresponding exo-
olefins (Scheme 3).9 Even though the allylic acetate 8b did not
show any reaction progress under the reported conditions,12 the
corresponding trifluoroacetate 8a showed a smooth conversion
to the desired product 10 under the same reaction conditions.
Apparently, trifluoroacetate has better leaving group ability to act
as a more reactive substrate for the allylic reduction. However,
because trifluoroacetic acid was a poor nucleophile for the Prins
reaction, it seemed that a balance between the nucleophilicity of
the carboxylic acid for the Prins reaction and the leaving group
ability of the carboxylic acid for the Tsuji reduction would need
to be struck to achieve an optimal overall reductive Prins
sequence. After various carboxylic acids with different pKa were
screened, methoxyacetic acid was found to be optimal for the
two-step procedure: 75% for the allene-Prins and 88% for the
Tsuji reduction to give 10 as a single isomer by NMR. Several
chiral and achiral palladium ligands for the allylic reduction were
also investigated before concluding that triphenylphosphine is
the optimal ligand. Regarding solvents, a faster reaction rate was
observed in DMF even at lower reaction temperature (40 vs 60
°C). However, migration of the terminal olefin in 10 was
observed as a side reaction in DMF, although the regio- and
stereoselectivity was similar to that obtained in refluxing THF.
Considering the presence of extra exo-olefin functionality at C19
in the halichondrins C14−C26 fragment, THF was chosen as the
reaction solvent to avoid potential complications.
Homoallenyl alcohol 7 served as a model system to test the
viability of the Prins reaction (Scheme 3).8 At the outset, it was
decided to decouple the cyclization sequence from the allylic
Scheme 3. Allene-Prins Reaction and Regio- and
Stereoselective Tsuji Reduction
A stereochemical rationale for the observed selectivity in the
allene-Prins reaction and Tsuji reduction sequence is depicted in
Scheme 4. The allene-Prins reaction would generate the allylic
tertiary carbocation B via the oxonium A, in which both C23 and
C27 substituents reside in pseudoequatorial positions to
selectively establish the C27 stereogenic center. It turned out
the methyl substitution at the allene is critical for the success of
the allene-Prins reactions.13 It is conceivable that the methyl
group stabilizes the resulting carbocations B to facilitate the
B
DOI: 10.1021/acs.orglett.7b02934
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Explanation of Regio- and Stereoselectivity
Scheme 5. Synthesis of Eribulin C20−C35 and C14−C35
Fragments via Allene-Prins Reaction
The allene-Prins approach was also applied to the synthesis of
the halichondrin C14−C38 fragment (Scheme 6). The allene-
Scheme 6. Synthesis of the Halichondrin C14−C38 Fragment
allene-Prins reaction. The resulting tertiary carbocation could be
captured as a primary or tertiary ester due to its allylic nature.
However, carboxylic acids would prefer to attack at the primary
carbon rather than the sterically congested tertiary carbon to give
8 as the exclusive product.
Upon treatment with a palladium catalyst, allylic ester 8 would
form a π-allyl palladium species (Scheme 4). As proposed by
Tsuji,9 the palladium species would prefer to reside on the
primary carbon to deliver the hydride to the tertiary carbon to
give the desired exo-olefin. The bulky palladium species would
approach from the opposite face of two substituents at C23 and
C27 and deliver the hydride to the same face via a cyclic
intermediate C to set the C25 chiral center as the desired fashion.
Consequently, the desired compound 10 was obtained as the
only observable product by 1H NMR.
Prins reaction of the halichondrin C27−C38 dimethyl acetal 20
with allene 16 gave the desired product in 57% yield under the
standard conditions. Subsequent Tsuji reduction also proceeded
smoothly to give the halichondrin C14−C38 fragment 21 in 63%
yield.
The optimized allene-Prins reaction and Tsuji reduction
sequence were applied to the synthesis of eribulin C20−C35 and
C14−C35 fragments (Scheme 5). The allene-Prins reaction of
dimethyl acetal 6a with allene 11 gave allylic ester 12, which was
subjected to Tsuji reduction to furnish 13 in 74% yield in two
steps as a single isomer by 1H NMR. Chemoselective
debenzoylation was accomplished with magnesium methoxide.14
The resulting diol was protected as a bis-TBS ether to give the
eribulin C20−C35 fragment 14.
The allene 16 required for the preparation of the eribulin
C14−C35 fragment was prepared from the corresponding vinyl
triflate.5b,15 The allene-Prins reaction of 16 with dimethyl acetal
15 proceeded smoothly to give 17 along with 7% of the
corresponding diene. Tsuji reduction of 17 provided 18 in 70%
yield in two steps. Protecting group manipulation of 18 as in the
C20−C35 fragment gave the C14−C35 fragment 19 in 58% in
two steps. The stereoselectivity at C25 was determined to be
48:1 by HPLC analysis.16
In summary, an allene-Prins reaction/Tsuji reduction
sequence of homoallenyl alcohols was demonstrated in the
context of the halichondrin C23−C27 pyran. Notable features
are the highly stereoselective establishment of C25 and C27
chiral centers from a single stereogenic center (C23). The
described methods were successfully applied for concise
syntheses of halichondrin C14−C38 and eribulin C14−C35
fragments. Additional applications of the Prins reaction in the
halichondrin series are under investigation and will be reported
in due course.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b02934.
Experimental procedures and characterization data for all
new compounds (PDF)
C
DOI: 10.1021/acs.orglett.7b02934
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(6) Reviews on Prins reaction: (a) Crane, E. A.; Scheidt, K. A. Angew.
Chem., Int. Ed. 2010, 49, 8316. (b) Pastor, I. M.; Yus, M. Curr. Org. Chem.
2007, 11, 925. (c) Adams, D. R.; Bhatnagar, S. P. Synthesis 1977, 1977,
661.
(7) For the Prins-type reaction with homoallenyl alcohols, see:
(a) Cho, Y. S.; Karupaiyan, K.; Kang, H. J.; Cha, J. H.; Pae, A. N.; Koh, H.
Y.; Chang, M. H. Chem. Commun. 2003, 2346. (b) Kang, H. J.; Kim, S.
H.; Pae, A. N.; Koh, H. Y.; Chang, M. H.; Choi, K. I.; Han, S.-Y.; Cho, Y.
S. Synlett 2004, 2004, 2545.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: hyeongwook_choi@eisai.com.
ORCID
Hyeong-wook Choi: 0000-0002-5923-4878
Notes
The authors declare no competing financial interest.
(8) For the preparation of 6 and 7, see Supporting Information.
(9) (a) Tsuji, J.; Minami, I.; Shimizu, I. Synthesis 1986, 1986, 623.
(b) Mandai, T.; Matsumoto, T.; Kawada, M.; Tsuji, J. Tetrahedron 1994,
50, 475. (c) Mandai, T.; Kaihara, Y.; Tsuji, J. J. Org. Chem. 1994, 59,
5847. (d) Ma, D.; Wu, W.; Deng, P. Tetrahedron Lett. 2001, 42, 6929.
(e) Chau, A.; Paquin, J.-F.; Lautens, M. J. Org. Chem. 2006, 71, 1924.
(f) Cheng, H.-Y.; Sun, C.-S.; Hou, D.-R. J. Org. Chem. 2007, 72, 2674.
(10) (a) Bahnck, K. B.; Rychnovsky, S. D. J. Am. Chem. Soc. 2008, 130,
13177. (b) Woo, S. K.; Kwon, M. S.; Lee, E. Angew. Chem., Int. Ed. 2008,
47, 3242.
(11) Primary allylic esters 8 were obtained as the allene-Prins products
(rather than the corresponding tertiary esters) presumably due to the
preferred nucleophilic attack on the primary carbon rather than the
sterically congested tertiary carbon. Structure of 8d was unambiguously
determined by 2D NMR experiments.
(12) Allylic reduction of 8b showed no reaction or very low conversion
under the tested reaction conditions with a combination of palladium
source (Pd(OAc)2, Pd2(dba)3, or Pd(PPh3)4), ligands (PPh3, PBu3, or
(t-Bu)2P(biphenyl)), and solvents (THF, DMF, or CH3CN).
(13) Allene substrates missing the 3-methyl substituent (i.e., nona-
1,2,8-trien-5-ol or 4-methylnona-1,2,8-trien-5-ol) did not give any
desired products under the same reaction conditions.
(14) (a) Xu, Y.-C.; Lebeau, E.; Walker, C. Tetrahedron Lett. 1994, 35,
6207. (b) Xu, Y.-C.; Bizuneh, A.; Walker, C. Tetrahedron Lett. 1996, 37,
455.
ACKNOWLEDGMENTS
■
We thank our colleagues Dr. A. Endo, C.-A. Lemelin, and J.
Cutter for the supply of compound 16, and Dr. C.E. Chase for
valuable feedback and proofreading.
DEDICATION
■
Dedicated to Prof. Yoshito Kishi at Harvard University on the
occasion of his 80th birthday.
REFERENCES
■
(1) (a) Uemura, D.; Takahashi, K.; Yamamoto, T.; Katayama, C.;
Tanaka, J.; Okumura, Y.; Hirata, Y. J. Am. Chem. Soc. 1985, 107, 4796.
(b) Hirata, Y.; Uemura, D. Pure Appl. Chem. 1986, 58, 701.
(2) For the synthesis of halichondrins, see: (a) Aicher, T. D.; Buszek, K.
R.; Fang, F. G.; Forsyth, C. J.; Jung, S. H.; Kishi, Y.; Matelich, M. C.;
Scola, P. M.; Spero, D. M.; Yoon, S. K. J. Am. Chem. Soc. 1992, 114, 3162.
(b) Jackson, K. L.; Henderson, J. A.; Motoyoshi, H.; Phillips, A. J. Angew.
Chem., Int. Ed. 2009, 48, 2436. (c) Dong, C.-G.; Henderson, J. A.;
Kaburagi, Y.; Sasaki, T.; Kim, D.-S.; Kim, J. T.; Urabe, D.; Guo, H.; Kishi,
Y. J. Am. Chem. Soc. 2009, 131, 15642. (d) Yamamoto, A.; Ueda, A.;
́
Bremond, P.; Tiseni, P. S.; Kishi, Y. J. Am. Chem. Soc. 2012, 134, 893.
(15) For the preparation of 16, see Supporting Information.
(16) HPLC analysis was conducted using C18 column, and the ratio
was determined based on comparison of retention times with those of
the authentic samples. The minor isomer was not detectible by 1H NMR
analysis (400 MHz).
(e) Ueda, A.; Yamamoto, A.; Kato, D.; Kishi, Y. J. Am. Chem. Soc. 2014,
136, 5171. (f) Kishi, Y.; Fang, F. G.; Forsyth, C. J.; Scola, P. M.; Yoon, S.
K. U.S. Patent 5338866; Int. Patent WO93/17690.
(3) Review of synthetic work on halichondrins: Jackson, K. L.;
Henderson, J. A.; Phillips, A. J. Chem. Rev. 2009, 109, 3044 and refs cited
therein.
(4) For eribulin reference, see: (a) Zheng, W.; Seletsky, B. M.; Palme,
M. H.; Lydon, P. J.; Singer, L. A.; Chase, C. E.; Lemelin, C. A.; Shen, Y.;
Davis, H.; Tremblay, L.; Towle, M. J.; Salvato, K. A.; Wels, B. F.; Aalfs, K.
K.; Kishi, Y.; Littlefield, B. A.; Yu, M. Bioorg. Med. Chem. Lett. 2004, 14,
5551. (b) Littlefield, B. A.; Palme, M. H.; Seletsky, B. M.; Towle, M. J.;
Yu, M. J.; Zheng, W. U.S. Patents 6214865 and 6365759; Int. Patent
WO99/65894. (c) Vahdat, L. T.; Pruitt, B.; Fabian, C. J.; Rivera, R. R.;
Smith, D. A.; Tan-Chiu, E.; Wright, J.; Tan, A. R.; DaCosta, N. A.;
Chuang, E.; Smith, J.; O’Shaughnessy, J.; Shuster, D. E.; Meneses, N. L.;
Chandrawansa, K.; Fang, F.; Cole, P. E.; Ashworth, S.; Blum, J. L. J. Clin.
Oncol. 2009, 27, 2954.
(5) For the synthesis of eribulin, see: (a) Chase, C. E.; Fang, F. G.;
Lewis, B. M.; Schnaderbeck, M. J.; Wilkie, G. D.; Zhu, X. Synlett 2013,
24, 323−326. (b) Austad, B. C.; Benayoud, F.; Calkins, T. L.; Campagna,
S.; Chase, C. E.; Choi, H.-w.; Christ, W.; Costanzo, R.; Cutter, J.; Endo,
A.; Fang, F. G.; Hu, Y.; Lewis, B. M.; Lewis, M. D.; McKenna, S.; Noland,
T. A.; Orr, J. D.; Pesant, M.; Schnaderbeck, M. J.; Wilkie, G. D.; Abe, T.;
Asai, N.; Asai, Y.; Kayano, A.; Kimoto, Y.; Komatsu, Y.; Kubota, M.;
Kuroda, H.; Mizuno, M.; Nakamura, T.; Omae, T.; Ozeki, N.; Suzuki, T.;
Takigawa, T.; Watanabe, T.; Yoshizawa, K. Synlett 2013, 24, 327.
(c) Austad, B. C.; Calkins, T. L.; Chase, C. E.; Fang, F. G.; Horstmann,
T. E.; Hu, Y.; Lewis, B. M.; Niu, X.; Noland, T. A.; Orr, J. D.;
Schnaderbeck, M. J.; Zhang, H.; Asakawa, N.; Asai, N.; Chiba, H.;
Hasebe, T.; Hoshino, Y.; Ishizuka, H.; Kajima, T.; Kayano, A.; Komatsu,
Y.; Kubota, M.; Kuroda, H.; Miyazawa, M.; Tagami, K.; Watanabe, T.
Synlett 2013, 24, 333. (d) Belanger, F.; Chase, C. E.; Endo, A.; Fang, F.
G.; Li, J.; Mathieu, S. R.; Wilcoxen, A. Z.; Zhang, H. Angew. Chem., Int.
Ed. 2015, 54, 5108. (e) Lee, J. H.; Li, Z.; Osawa, A.; Kishi, Y. J. Am. Chem.
Soc. 2016, 138, 16248.
D
DOI: 10.1021/acs.orglett.7b02934
Org. Lett. XXXX, XXX, XXX−XXX