Total Synthesis of the 7,10-Epimer of the Proposed Structure of Amphidinolide N, Part II: Synthesis of C17-C29 Subunit and Completion of the Synthesis

Article abstract of DOI:10.1002/chem.201504675

The total synthesis of 7,10-epimer of the proposed structure of amphidinolide N was accomplished. The requisite chiral C17-C29 subunit was assembled stereoselectively via Keck allylation, Shi epoxidation, diastereoselective 1,3-reduction, and a later oxidative synthesis of the THF framework. The C1-C13 and C17-C29 subunits were successfully coupled using a Enders RAMP "linchpin" as the C14-C16 three carbon unit, thereby controlling the chirality at C14 and C16. The labile allyl epoxy moiety was successfully constructed by Grieco-Nishizawa olefination at a final stage of the synthesis.

Full text of DOI:10.1002/chem.201504675

DOI: 10.1002/chem.201504675
Communication
&
Natural Products
Total Synthesis of the 7,10-Epimer of the Proposed Structure of
Amphidinolide N, Part II: Synthesis of C17–C29 Subunit and
Completion of the Synthesis
Koji Ochiai,[a] Sankar Kuppusamy,[a] Yusuke Yasui,[a] Kenji Harada,[a] Nishant R. Gupta,[a]
Yohei Takahashi,[b] Takaaki Kubota,[b, c] Jun’ichi Kobayashi,[b] and Yujiro Hayashi*[a, d]
Ketone 2 would be obtained by alkynylation of 3 with 6. The
known aldehyde 5[3] would be converted into chiral aldehyde
3 by Keck asymmetric allylation.[4] On the other hand, the ter-
Abstract: The total synthesis of 7,10-epimer of the pro-
posed structure of amphidinolide N was accomplished.
minal alkyne 6 would be derived from 10 via epoxide opening
followed by intramolecular oxidative cyclization. The chiral ep-
oxide 10 would be synthesized by Shi asymmetric epoxida-
tion[5] of the corresponding trans-enyne, which would be readi-
ly prepared from commercially available alkyne 11.[6]
Our synthetic route to the chiral aldehyde 3 is illustrated in
Scheme 2. Keck asymmetric allylation[4] of the known aldehyde
5[3] with allylstannane proceeded in the presence of 10 mol%
of (S)-BINOL and Ti(OiPr)4. This afforded the chiral homoallyl al-
cohol 12 in excellent yield with excellent enantioselectivity.
The hydroxy group of 12 was protected by TBS to give 13. Di-
hydroxylation of double bond followed by oxidative cleavage
of the resulting diol afforded aldehyde 3 in good yield.
Synthesis of the terminal alkyne 6 began by converting the
commercially available 11 to trans-vinyl iodide 14 according to
the literature procedure[6] (Scheme 3). Sonogashira coupling[7]
of vinyl iodide 14 with trimethylsilylacetylene gave enyne 15
in good yield. The enantioselective epoxidation of 15 was ac-
complished by the method of Shi using organocatalyst 16 to
give the chiral epoxide 10 in moderate yield and good enatio-
selectivity, which was determined in the next step.[5] The chiral
epoxide 10 was converted into alcohol 17 via regioselective
epoxide opening with 3,4-dimethoxybenzyl alcohol as a nucleo-
phile in the presence of CSA as catalyst. The desired 17 could
not be obtained when Lewis acids such as BF3·Et2O, Ti(OiPr)4,
and Sc(OTf)3 were employed instead of CSA.[8] Removal of the
TMS group of 17 afforded 18. Oxidative acetal formation was
accomplished by the treatment of 3,4-dimethoxybenzyl ether
18 with DDQ under anhydrous conditions to afford acetal 6 in
moderate yield.
The requisite chiral C17–C29 subunit was assembled ste-
reoselectively via Keck allylation, Shi epoxidation, diaste-
reoselective 1,3-reduction, and a later oxidative synthesis
of the THF framework. The C1–C13 and C17–C29 subunits
were successfully coupled using a Enders RAMP “linchpin”
as the C14–C16 three carbon unit, thereby controlling the
chirality at C14 and C16. The labile allyl epoxy moiety was
successfully constructed by Grieco–Nishizawa olefination
at a final stage of the synthesis.
In the preceding paper in this issue,[1] we described that 7,10-
epimer of the proposed structure of amphidinolide N (4) could
be disconnected into three subunits 7, 8 and 9 (Scheme 1),
and further described the synthesis of the C1–C13 subunit 7
via a highly enantioselective and diastereoselective route. In
this paper, we report the synthesis of the C17–C29 subunit 9
and a RAMP-Enders-based coupling of the three units to com-
plete the total synthesis of the macrolide 4.
First of all, the synthesis of C17–C29 subunit 9 will be de-
scribed. The retrosynthesis is shown in Scheme 1. The THF ring
of 9 would be obtained via intramolecular oxidative cyclization
of 1, triggered by DDQ. The anti 1,3-diol of 1 would be formed
by diastereoselective anti-reduction of b-hydroxy ketone 2.[2]
[a] Dr. K. Ochiai, Dr. S. Kuppusamy, Y. Yasui, K. Harada, Dr. N. R. Gupta,
Prof. Dr. Y. Hayashi
Before July, 2012: Department of Industrial Chemistry
Faculty of Engineering, Tokyo University of Science, Kagurazaka
Shinjuku-ku, Tokyo 162-8601 (Japan)
Synthesis of the intermediate 1 is shown in Scheme 4. First,
we attempted to construct the C21 stereocenter via reagent-
controlled alkynylation as reported by Shibasaki.[9] However,
the method was found ineffective between 6 and 3, which
possess several functional groups, and resulted in recovery of
the starting materials. Therefore, we decided to construct the
C21 stereogenic center under substrate control. Alkynylation of
3 with lithium acetylide of 6 afforded 19 as a diastereomeric
mixture in good yield. TBS deprotection of secondary alcohol
followed by MnO2 oxidation gave b-hydroxy ketone 21. Ketone
21 was reduced to 2 by alkyne hydrogenation and carbonyl
diastereoselective hydride addition with Me4NBH(OAc)3.[2] This
[b] Dr. Y. Takahashi, Prof. Dr. T. Kubota, Prof. Dr. J. Kobayashi
Graduate School of Pharmaceutical Sciences, Hokkaido University
Sapporo 060-0812 (Japan)
[c] Prof. Dr. T. Kubota
Present address: Showa Pharmaceutical University
3-3165 Higashi-Tamagawagakuen, Machida, Tokyo 194-8543 (Japan)
[d] Prof. Dr. Y. Hayashi
Present address: Department of Chemistry, Graduate School of Science
Tohoku University, Aoba-ku, Sendai 980-8578 (Japan)
Fax: (+81)22-795-6566
E-mail: yhayashi@m.tohoku.ac.jp
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201504675.
Chem. Eur. J. 2016, 22, 3287 – 3291
3287
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Scheme 1. Retrosynthesis of 7,10-epi-amphidinolide N (4) and the C17–C29 subunit 9, Cbz=benzyloxycarbonyl, PMB=p-methoxybenzyl, TES=triethylsilyl,
TBS=tert-butyldimethylsilyl, Bn=benzyl, DMP=3,4-dimethoxyphenyl, TMS=trimethylsilyl.
Scheme 2. Synthesis of aldehyde 3. a) allyltributyltin, Ti(OiPr)4, (S)-BINOL,
MS4Š/DCM, À208C, quant. (99% ee); b) TBSCl, imdiazole/DMF, RT, 96%;
c) OsO4, NMO/dioxane, H2O, RT then NaIO4, RT, 90%. Bn=benzyl,
BINOL=1,1’-bi-2-naphthol, MS=molecular sieves, TBS=tert-butyldimethyl-
silyl, NMO=N-methyl morpholine N-oxide.
Scheme 3. Synthesis of terminal alkyne 6. a) TMSCꢀCH, Pd(Ph3P)4, Et2NH/
sequence and order of reduction gave the desired anti-diol
DMF, RT, 80%; b) 16, Oxone, K2CO3/DMM, MeCN, 08C, 65%; c) 3,4-dimethoxy-
1 as a single isomer in excellent yield; otherwise, hydride addi-
benzyl alcohol, CSA/DCM, 59% (87% ee); d) K2CO3/MeOH, RT, 75%; e) DQQ,
tion of 21 with Me4NBH(OAc)3 gave almost no diastereoselec-
tivity.
MS4Š/DCM, 08C, 61%. TMS=trimethylsilyl, DMM=dimethoxymethane,
CSA=camphorsulfonic acid, DMB=3,4-dimethoxybenzyl, DDQ=2,3-di-
chloro-5,6-dicyano-p-benzoquinone, MS=molecular sieves, DMP=3,4-dime-
thoxyphenyl.
Next, we studied the oxidative THF ring formation using
DDQ from 1 (Scheme 5). To the best of our knowledge, an oxi-
dative THF ring formation of this type using DDQ has not
been reported.[10] At first, the reaction of 1 with DDQ in the
presence of molecular sieves resulted in almost full recovery of
the starting material. Therefore, we investigated the effect of
additives in the expectation to promote the oxidation, and it
was found that the selection and amount of additive were
both crucial for successful THF ring formation: for example,
a low yield of 22 was observed in the presence of Brønsted
acid such as PPTS and CSA. Although an equimolar amount of
BF3·Et2O was found moderately successful, affording the prod-
uct in moderate yield (59%), optimal results were obtained
with a catalytic amount of BF3·Et2O (15 mol%). Namely, the de-
sired THF derivative 22 was obtained as a single isomer in
good yield without formation of the undesired THP ring. Hy-
Chem. Eur. J. 2016, 22, 3287 – 3291
www.chemeurj.org
3288
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Scheme 6. Synthesis of the C17-C29 subunit 9. a) TBSOTf, 2,6-lutidine/DCM,
RT, 96%; b) NaOMe/MeOH, reflux, 91%; c) PMBOC(=NH)CCl3, TfOH/Et2O, RT;
d) H2, Raney Ni (W-2)/EtOH, RT, 54% (2 steps); e) I2, Ph3P, imidazole/benzene,
RT, 94%. Bn=benzyl, DMP=3,4-dimethoxyphenyl, PMB=p-methoxybenzyl.
without affecting the PMB ether by hydrogenation with Raney
nickel to give the primary alcohol 26 in moderate yield.[13] Fi-
nally, iodination of primary alcohol with PPh3 and I2 gave the
desired C17–C29 subunit 9.
Next, we began to install the full carbon framework of our
target molecule 4 by using Enders RAMP hydrazone diastereo-
selective method via double alkylation of the readily pre-
pared[11,14] three carbon unit of C14–C16 (Scheme 7). The se-
quential alkylation of hydrazone 8, first with the C17–C29 sub-
unit 9 and then with the C1–C29 subunit 7, followed by re-
moval of the chiral auxiliary under mild conditions,[15] gave 29
in good yield over three steps. It is noteworthy that our novel
protective C9-hydroxyl ester group[1] was inert under these
conditions.
Scheme 4. Synthesis of 1. a) nBuLi/THF, À788C, then 3, À788C to RT, 85%;
b) TBAF/THF, RT, quant.; c) MnO2/CHCl3, RT then 508C, 75%; d) H2, Pd-C/
EtOAc, RT, 99%; e) Me4NBH(OAc)3, AcOH/MeCN, À308C, 91%. DMP=3,4-di-
methoxyphenyl, Bn=benzyl, TBS=tert-butyldimethylsilyl, TBAF=tetra-n-bu-
tylammonium fluoride.
Scheme 5. THF ring formation triggered by DDQ. Bn=benzyl, DMP=3,4-di-
methoxyphenyl, DDQ=3,4-dichloro-5,6-dicyano-p-benzoquinone, MS=mo-
lecular sieves.
droxy group at C21 reacts at C24 position via complete inver-
sion of stereochemistry without reaction at C25. Thus, we have
established the novel and efficient method for the THF ring
formation.
Synthesis of the C17–C29 subunit 9 from 22 is summarized
in Scheme 6. TBS protection of secondary alcohol gave 23. We
considered the 3,4-dimethoxybenzoyl group to be too labile
under various conditions during the total synthesis, especially
during the RAMP hydrazone diastereoselective alkylation
stage.[11] Therefore, it was converted into a PMB ether via
a two-step procedure. Thus, treatment of 23 with NaOMe fol-
lowed by PMB protection[12] of the resulting secondary alcohol
afforded 25. Selective Bn etherification of 25 was accomplished
Scheme 7. Segment coupling by RAMP hydrazone diastereoselective alkyla-
tion. a) tBuLi/THF, À788C then 9, À788C; b) tBuLi/THF, À788C then 7,
À788C; c) sat. (COOH)2 aq./Et2O, RT, 73% (3 steps). RAMP=(R)-(+)-1-amino-
2-methoxymethylpyrrolidine, TBS=tert-butyldimethylsilyl, PMB=p-methoxy-
benzyl, Cbz=benzyloxycarbonyl, TES=triethylsilyl.
Chem. Eur. J. 2016, 22, 3287 – 3291
www.chemeurj.org
3289
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Scheme 8. Synthesis of 35. a) DDQ/DCM, phosphate buffer, 08C, 67%; b) Pd-C, Et3SiH/DCM, RT; c) NaClO2, NaH2PO4, 2-methyl-2-butene/tBuOH, THF, H2O, 08C,
64% (2 steps); d) 2,4,6-trichlorobenzoyl chloride, Et3N/toluene, RT then DMAP, RT, 80%; e) HF aq./MeCN, DCM, 08C; f) TBSOTf, 2,6-lutidine/DCM, À788C, 37%
(2 steps). TBS=tert-butyldimethylsilyl, Cbz=benzyloxycarbonyl, PMB=p-methoxybenzyl, TES=triethylsilyl, DDQ=2,3-dichloro-5,6-dicyano-p-benzoquinone.
Scheme 9. Total synthesis of 4. a) DDQ, 2,6-di-tBu-pyridine/DCM, H2O, 08C, 50%; b) HCOONH4/THF, EtOH, 508C, 64%; c) o-nitrophenylseleno cyanate, nBu3P/
benzene, 758C; d) xylene, 1208C; e) TPAP, NMO, MS4Š/DCM, RT; f) NH4F/MeOH, RT, 10% (4 steps). Cbz=benzyloxycarbonyl, PMB=p-methoxybenzyl, TES =
triethylsilyl, DDQ=2,3-dichloro-5,6-dicyano-p-benzoquinone, TPAP=tetra n-propylammonium perruthenate, NMO=N-methylmorpholine N-oxide, MS=mo-
lecular sieves.
Chem. Eur. J. 2016, 22, 3287 – 3291
www.chemeurj.org
3290
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Synthesis of 35 is illustrated in Scheme 8. To our delight,
when 29 was treated with 1 equiv of DDQ, selective deprotec-
tion of the PMB ether of the secondary alcohol at C25 was re-
alized, without removal of the PMB ether of the primary alco-
hol, to afford the needed alcohol 30 in 67% yield with recov-
ery of starting material (18%). The selective deprotection of
the PMB ether of the secondary alcohol, over that of a primary
alcohol, can be ascribed to steric crowdedness around the pri-
mary PMB ether moiety. The thioester hydrolysis into carboxyl-
ic acid was found to be troublesome. As the desired carboxylic
acid 32 was low yielding under various conditions, 30 was con-
verted into the macrolactone precursor 32 in a two-step proce-
dure consisting of Fukuyama’s reduction[16] followed by Pin-
nick–Kraus oxidation of the resulting aldehyde.[17] Yamaguchi
macrolactonization[18] of 32 successively proceeded to furnish
33 in good yield. Treatment of 33 with aqueous HF promoted
concomitant removal of the acetal protecting group and all
silyl protecting groups to afford the pentaol 34. All hydroxyl
groups thus generated were TES protected to afford 35, which
would be removed under mild conditions at the last stage of
the synthesis.
labile allyl epoxy moiety was successfully constructed by
Grieco-Nishizawa olefination at a late stage of the synthesis.
Keywords: Enders RMAP · Grieco-Nishizawa olefination · Keck
allylation · Shi-epoxidation · total synthesis
[1] K. Ochiai, S. Kuppusamy, Y. Yasui, T. Okano, Y. Matsumoto, N. R. Gupta, J.
Kobayashi, Y. Hayashi, Chem. Eur. J. 2016, 22, DOI: 10.1002/
chem.201504674.
[2] a) D. A. Evans, K. T. Chapman, E. M. Carreira, J. Am. Chem. Soc. 1988, 110,
3560; b) D. A. Evans, K. T. Chapman, Tetrahedron Lett. 1986, 27, 5939.
[3] L. Coppi, A. Ricci, M. Taddei, J. Org. Chem. 1988, 53, 911.
[4] a) G. E. Keck, K. H. Tarbet, L. S. Geraci, J. Am. Chem. Soc. 1993, 115, 8467;
b) G. E. Keck, D. Krishnamurthy, M. C. Grier, J. Org. Chem. 1993, 58, 6543;
c) G. E. Keck, L. S. Geraci, Tetrahedron Lett. 1993, 34, 7827.
[5] a) Z.-X. Wang, L. Shu, M. Frohn, Y. Tu, Y. Shi, Org. Synth. 2003, 80, 9;
b) Z.-X. Wang, Y. Tu, M. Frohn, J.-R. Zhang, Y. Shi, J. Am. Chem. Soc.
1997, 119, 11224; c) Z.-X. Wang, G.-A. Cao, Y. Shi, J. Org. Chem. 1999, 64,
7646.
[6] J. K. Stille, J. H. Simpson, J. Am. Chem. Soc. 1987, 109, 2138.
[7] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron lett. 1975, 16, 4467.
[8] a) G. Prestat, C. Baylon, M.-P. Heck, C. Mioskowski, Tetrahedron Lett.
2000, 41, 3829; b) G. Prestat, C. Baylon, M.-P. Heck, G. A. Grasa, S. P.
Nolan, C. Mioskowski, J. Org. Chem. 2004, 69, 5770.
[9] R. Takita, K. Yakura, T. Ohshima, M. Shibasaki, J. Am. Chem. Soc. 2005,
127, 13760.
[10] Regio- and stereoselective cyclization of 1,2,n-triols via the intermediacy
of a cyclic ortho ester using acid promoters was known in the literature.
T. Zheng, R. S. Narayan, J. M. Schomaker, B. Borhan, J. Am. Chem. Soc.
2005, 127, 6946.
[11] a) D. Enders, M. Voith, A. Lenzen, Angew. Chem. Int. Ed. 2005, 44, 1304;
Angew. Chem. 2005, 117, 1330; b) D. Enders, M. Voith, S. J. Ince, Synthesis
2002, 1775; c) For review, see: A. Job, C. F. Janeck, W. Bettray, R. Peters,
D. Enders, Tetrahedron 2002, 58, 2253.
[12] N. Nakajima, K. Hirota, R. Abe, O. Yonemitsu, Tetrahedron Lett. 1988, 29,
4139.
[13] K. Horita, T. Yoshioka, T. Tanaka, Y. Oikawa, O. Yonemitsu, Tetrahedron
1986, 42, 3021.
[14] a) K. C. Nicolaou, W. E. Brenzovich, P. G. Bulger, T. M. Francis, Org. Biomol.
Chem. 2006, 4, 2119; b) K. C. Nicolaou, P. G. Bulger, W. E. Brenzovich,
Org. Biomol. Chem. 2006, 4, 2158.
[15] D. Enders, T. Hundertmark, R. Lazny, Synlett 1998, 721.
[16] T. Fukuyama, S.-C. Lin, L. Li, J. Am. Chem. Soc. 1990, 112, 7050.
[17] a) G. A. Kraus, B. Roth, J. Org. Chem. 1980, 45, 4825; b) G. A. Kraus, M. J.
Taschner, J. Org. Chem. 1980, 45, 1175; c) B. S. Bal, W. E., Jr. Childers,
H. W. Pinnick, Tetrahedron 1981, 37, 2091.
[18] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull. Chem. Soc.
Jpn. 1979, 52, 1989.
[19] P. A. Grieco, S. Gilman, M. Nishizawa, J. Org. Chem. 1976, 41, 1485.
[20] T. W. Lee, E. J. Corey, J. Am. Chem. Soc. 2001, 123, 1872.
[21] W. P. Griffith, S. V. Ley, G. P. Whitcombe, A. D. White, J. Chem. Soc. Chem.
Commun. 1987, 1625.
[22] a) Y. Hayashi, M. Shoji, J. Yamaguchi, K. Sato, S. Yamaguchi, T. Mukaiya-
ma, K. Sakai, Y. Asami, H. Kakeya, H. Osada, J. Am. Chem. Soc. 2002, 124,
12078–12079; c) J. D. White, J. C. Amedio, Jr., S. Gut, L. Jayasinghe, J.
Org. Chem. 1989, 54, 4268; d) X.-F. Zhu, H. J. Williams, A. I. Scott, Synth.
Commun. 2003, 33, 2011; e) D. Crich, F. Hermann, Tetrahedron Lett.
1993, 34, 3385.
The final stage to complete the total synthesis of targeted 4
is shown in Scheme 9. Removal of the PMB-protecting group
was performed by treatment of 35 with DDQ in the presence
of 2,6-di-tert-pyridine to afford 36 along with recovery of the
starting material (40%). Cbz deprotection was achieved by
treatment of 36 with HCOONH4 and Pd-C in THF/EtOH. Here,
the amine generated was anticipated to react with the ester
moiety in an intramolecuar fashion to release the alcohol at C9
and 1-methyl-2-pyrrolidone. In this case, however, the C9 alco-
hol remained protected. Thus, after introduction of o-nitrophe-
nylselenyl group to give 38,[19,20] heating of the xylene solution
of 38 at 1208C under neutral conditions removed the protect-
ing group at C9 to afford alcohol 39. Next, TPAP oxidation[21]
not only led to the oxidative elimination of the selenide via its
selenoxide to the labile allyl epoxide, but also oxidized the C9
secondary alcohol to its ketone 40. Final global desilylation
with NH4F in MeOH[22] gave 7,10-epimer of the proposed struc-
ture of amphidinolide N (4).
In summary, the asymmetric total synthesis of 7,10-epimer
of the proposed structure of amphidinolide N was accom-
plished. The chiral C17–C29 subunit was stereoselectively syn-
thesized by Keck allylation, Shi epoxidation, and diastereoselec-
tive 1,3-reduction as the key steps. The chiral THF ring was
constructed by a new oxidative cyclization tactic using DDQ.
The assembly of C1–C13 and C17–C29 subunits were succes-
sively performed using Enders RAMP methodology as a C14–
C16 three carbon “linchpin” unit, thereby defining the new ste-
reogenic centers at C14 and C16. Notably, the new protecting
group 4-(N-benzyloxycarbonyl-N-methylamino)butyryl group
was found effective for the C9 alcohol position. Lastly, the
Received: November 20, 2015
Published online on February 4, 2016
Chem. Eur. J. 2016, 22, 3287 – 3291
www.chemeurj.org
3291
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim