Unified, Efficient, and Scalable Synthesis of Halichondrins: Zirconium/Nickel-Mediated One-Pot Ketone Synthesis as the Final Coupling Reaction

Article abstract of DOI:10.1002/anie.201705523

Unified, efficient, and scalable syntheses of the halichondrin natural products are reported. A newly developed Zr/Ni-mediated one-pot ketone synthesis was used to couple the two halves of the final product at a late stage in the synthesis. With the use of a slight excess of the left halves, the desired ketones were isolated in yields of 80–90 %. The halichondrins were obtained from these ketones in two steps, namely desilylation and [5,5]-spiroketal formation. The new synthetic route was effective for the total synthesis of all members in the homohalichondrin subgroup. The scalability of this process was demonstrated with halichondrin B; 150 mg of halichondrin B (68 % overall yield) were obtained from 200 mg of the right-half precursor.

Full text of DOI:10.1002/anie.201705523

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Unified, Efficient, and Scalable Synthesis of Halichondrins,
with Use of Zr/Ni-Mediated One-Pot Ketone Synthesis as Final
Coupling Reaction
Authors: Yoshito Kishi, Kenzo Yahata, Ning Ye, Yanran Ai, and
Kentaro Iso
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201705523
Angew. Chem. 10.1002/ange.201705523
Link to VoR: http://dx.doi.org/10.1002/anie.201705523
http://dx.doi.org/10.1002/ange.201705523

10.1002/anie.201705523
Angewandte Chemie International Edition
COMMUNICATION
Unified, Efficient, and Scalable Synthesis of Halichondrins, with
Use of Zr/Ni-Mediated One-Pot Ketone Synthesis as Final
Coupling Reaction
Kenzo Yahata, Ning Ye, Yanran Ai, Kentaro Iso, and Yoshito Kishi*
For two reasons, however, we were anxious to develop an
alternative route for the final transformation. First, the enone
route was effective for a synthesis of the members in the
halichondrin/norhalichondrin sub-groups, but not for a synthesis
of the members in the homohalichondrin sub-group. Second,
isolated yields of halichondrins with preparative HPLC were not
totally satisfactory, although no distinct by-products were noticed
in the 1H NMR spectrum of crude products. Knowing good
recoveries of halichondrins in HPLC purification, we questioned
that the observed low-yield might be due to formation of a small
amount of a number of unidentified by-products, for example,
stereo-isomer at C40, by-products derived via retro-oxy-
Michael/oxy-Michael processes of B, and others.
Abstract:
A unified, efficient, and scalable synthesis of the
halichondrin class of natural products was reported. The newly
developed Zr/Ni-mediated one-pot ketone synthesis was used to
couple right halves with left halves, where Cp2ZrCl2 with Zn or Mn
played a crucial role to accelerate the coupling reaction and, at the
same time, suppress by-product formation. With use of a slight
excess of the left halves, the desired ketones were isolated in
80~90% yields. Halichondrins were obtained from these ketones
basically in two operations, i.e., desilylation and [5,5]-spiroketal
formation. For the members in the halichondrins-A and -C sub-
groups, however, an additional step was required to remove the
protecting group at the polycycle, whereas for the members in the
norhalichondrin sub-group, an additional step was used to hydrolyze
the methyl ester at C53. The new synthetic route was successfully
Me
Me
H
H
H
1
O
O
O
H
30
H
O
O
40
O
O
O
H
O
O
O
50
H
H
H
HO
10
O
applied for
a
total synthesis of all the members in the
O
H
HO
HO
Me
O
homohalichondrin sub-group as well. To demonstrate the scalability,
halichondrin B was chosen, where 150 mg of halichondrin B (68%
overall yield) was obtained from 200 mg of the right half 5.
O
Me
X
Halichondrin B:
X = Y = H
54
O
20
Y
Me
Me
H
H
H
H
H
H
H
O
O
H
H
H
H
HO
O
O
40
40
O
50
50
O
O
Introduction
O
O
55
H
O
HO
O
Halichondrins are polyether macrolides, originally isolated from
the marine sponge Halichondria okadai by Uemura, Hirata, and
coworkers.[1] Several additional members, including halistatin,
were isolated from various marine sponges.[2] This class of
natural products displays interesting structure diversities at two
sites, one being the oxidation state at C10, C12, and C13 of the
C8-C14 polycycle and the other being the length of the carbon
backbone. Thus, this class of natural products is sub-grouped
into halichondrins A-C series or the norhalichondrin/halichondrin/
homohalichondrin series. Except halichondrin A, all the
members have been isolated from the natural sources (Scheme
1). Due to their intriguing structural architecture and
extraordinary antitumor activity, halichondrins have received
much attention from the scientific community.[3,4] In this
communication, we report a unified, efficient, and scalable
synthesis of halichondrins, with use of the newly developed
Zr/Ni-mediated one-pot ketone synthesis as the final coupling
reaction.
H
H
HO2C 53
Me
OH
Me
Norhalichondrin
B
Homohalichondrin B
A series
(X = Y = OH)
Note-1
B series
C series
(X = OH; Y = H)
known
(X = Y = H)
known
Halichondrin
Norhalichondrin
known
known
known
Homohalichondrin
known
known
known
Note-1: Halichondrin A was not isolated from the natural sources
thus far, but it is now available by synthesis.
Scheme 1. Halichondrin Class of Natural Products.
We recognized that these difficulties might be overcome if
ketone B were available via an alternative, well-defined route. At
the same time, we realized that the synthesis of ketone B would
present
a
challenge. For example, anion-based ketone
syntheses might be problematic for this case because of the
presence of an O-R group at C36 and C40. The “umpolung”
concept, represented by dithiane chemistry, is the historical
solution for this type of problem. Indeed, dithiane-based ketone
synthesis was successfully applied to a synthesis of complex
natural products. Nevertheless, we were interested in
developing a direct ketone synthesis, which could be effective at
a late-stage in a convergent synthesis of a complex molecule.
We believed that a radical-based, preferably one-pot, ketone
synthesis could give us the best chance for achieving this goal.
We have recently developed Zr/Ni-, Zn/Pd-, and Fe/Cu-
mediated one-pot ketone syntheses.[7] Among them, only Zr/Ni-
mediated one-pot ketone synthesis showed a potential to meet
with our needs; specifically, this method was proven effective for
In our previous synthesis, the key intermediate for
construction of the [6,6] and [5,5] spiroketals was enone 3,
which was synthesized via a Ni/Cr-mediated coupling of 1 with 2
in an excellent overall yield (Scheme 2).[5] During this
transformation, three chiral centers were introduced at C38, C40,
and C44, cf., 3
A B 4. Based on the synthetic work of
® ® ®
calcimycin, we expected the desired stereochemistry to be
preferentially formed under basic conditions.[6] Indeed, this
approach worked nicely for a synthesis of halichondrins A-Cs.
[a]
Dr. K. Yahata, Dr. N. Ye, Dr. Y. Ai, Dr. K. Iso, Prof. Dr. Y. Kishi
Department of Chemistry and Chemical Biology
Harvard University
coupling of (S)-C+(S)-D
®
(S)-E. Based on that work, we
Cambridge, Massachusetts 02138, Unites States
E-mail: kishi@chemistry.harvard.edu
proposed to synthesize the requisite ketone B from iodide 5 and
2-thiopyridine ester 6. Ketone B could also be obtained via
coupling at the C38-C39 bond, but we focused on the former
route because of the overall synthetic efficiency of 5.[8]
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
This article is protected by copyright. All rights reserved.

10.1002/anie.201705523
Angewandte Chemie International Edition
COMMUNICATION
Me
Me
H
38
I
TES
O
OTBS
OHC
1
O
O
37
35
O
H
O
48
39
O
O
H
O
O
Me
O
TES
O
O
O
O
TES
H
I
2
1
TES
1. Ni/Cr-mediated coupling
2. Dess-Martin oxidation
O
enone route
H
H
H
t-Bu
Si
O
Me
O
52
O
O
t-Bu
O
38
SPy-2
Me
O
O
O
Me
H
TES
Me
Me
OTBS
H
H
6
5
O
a
O
O
48
H
H
1. TBAF
2. PPTS
38
O
44 40
O
O
O
H
O
Me
O
TES
H
O
35
O
Me
H
H
O
O
3
Me
4
Me
O
H
TES
O
Me
Me
O
Me
OH
OTBS
O
H
H
H
TES
O
O
44 40
O
7
48
48
OH
b, c
O
35
O
Me
Me
H
H
35
H
H
H
H
H
O
O
Me
Me
1
O
O
HO
O
HO
H
H
B
A
H
O
38
HO
O
O
O
O
O
O
H
O
H
52
O
O
ketone coupling
HO
Me
ketone coupling route
TES
O
O
Me
I
11
O
Me
O
37
O
H
O
O
35
48
38 SPy-2
O
O
TES
H
D
C
Me
Me
O
Me
O
H
H
H
HO
O
37
37
O
O
O
O
O
O
+
5
+
I
TESO
TESO
SPy-2
H
H
Me
(S)-E
8
9
(S)-F
(S)-G
OTES
H
O
CO
NiII
Ni0
O
44
Scheme 2. Two Routes to Complete Total Synthesis.
6
SPy-2
OH
NiII
SPy-2
39
H
Me
10
Feasibility Study of Ketone Coupling
Being encouraged with the successful ((S)-C+(S)-D
Scheme 3. Feasibility Test of Ketone Coupling. (a) 5 (1.0 equiv), 6 (1.3 equiv),
NiBr2•(dtbbpy) (30 mol%), Cp2ZrCl2 (3 equiv), (t-Bu)2(Me)-Py (4 equiv), Zn (6
equiv) in 5:1 DMI-EtOAc (C 0.1 M), rt. (b) HF•Py (20 equiv), THF, followed by
TBAF (4 equiv), pivalic acid (2 equiv), DMF, rt. (c) PPTS (5 equiv), CH2Cl2,
®
(S)-E)-
coupling, we began the feasibility study for the proposed
synthesis. For this study, we chose to use the right half 5 of
halichondrin Bs. The C35-protecting group was selected for two
reasons, i.e., (1) the rate of ketone coupling with 5 was
significantly faster than that with the corresponding C35-TBS
substrate and (2) deprotection of the C35-TES group in the
following step was noticeably faster than that of the
corresponding C35-TBS substrate. On the other hand, we chose
the left half 6, because of its availability in a larger quantity at the
time of preliminary study. The C41-protecting group was chosen
primarily for the ease of deprotection.
We were delighted to obtain the desired product 7 in the first
attempt under the conditions used for ((S)-C+(S)-D
coupling. We then optimized the coupling conditions, thereby
revealing several critical factors. First, Cp2ZrCl2 was crucial to
accelerate the ketone coupling and, at the same time, suppress
~20 °C,
dimethyl-2-imidazolidinone;
PPTS=pyridinium p-toluenesulfonate.
2
hr. Abbreviation: TES=Et3Si-; SPy-2=2-thiopyridine; DMI=1,3-
TBAF=tetrabutylammonium fluoride;
Considering all these factors, we selected the coupling
condition specified in Legend of Scheme 3. For all the couplings,
we used the molar ratio of 5:6 = 1.0:1.3, considering the
molecular size and complexity of 5 vs. 6. Under this condition,
the ketone coupling was carried out, to furnish the desired
product 7 in 80~90% yields.
In this coupling, three by-products were isolated in very small
amounts (~3% yields). Spectroscopic analysis (1H NMR, MS)
suggested these by-products to be 8, 9, and 10, respectively.
The first two by-products were derived from 5, formation of
which was not surprising in light of the results discussed in the
method-development work.[7a] The third by-product 10 was
obviously derived from 6, which was, we speculated, formed via
a Ni-mediated decarbonylation, cf., the transformation depicted
in Scheme 3.[10]
®
(S)-E)-
by-product formation via
a (I®Spy)-displacement at C37.
Second, a 5:1 mixture of DMI and EtOAc was found to be the
best solvent. Third, the coupling proceeded well at 0.1 M
concentration, although a higher concentration, for example 0.4
M, was better. Fourth, both Zn and Mn metals were effective.
Fifth, the method of catalyst preparation was important to
achieve
a
good reproducibility.[9] Sixth, 2,6-di-tert-butyl-4-
Ketone 7 also served for a model study on the second stage
of synthesis, i.e., deprotection of the silyl groups, followed by
acid-catalyzed [5,5]-spiroketal formation. As expected, the
C50/C52-dioxasilinane group in 7 was readily removed on a
methylpyridine was added to avoid partial deprotection of the
TES groups during the reaction and/or workup. Lastly, as
expected, the coupling efficiency depended on the molar ratio of
5 and 6, for example 84% yield with 5:6 = 1.0:1.4; 84% yield with
1.0:1.3; 62% with 1.0:1.0; 71% with 1.0:1.2.
treatment with HF•Py, to give the corresponding diol.
A
treatment of the resultant C50/C52-diol with TBAF buffered with
pivalic acid gave the completely deprotected product within 6
.
This article is protected by copyright. All rights reserved.

10.1002/anie.201705523
Angewandte Chemie International Edition
COMMUNICATION
hours, thereby confirming the ease of deprotection of the two
TES group at C35 and C41.[11]
Right Halves of Halichondrins
H
O
O
The completely desilylated product was treated with an acid,
to give 11; namely, PPTS in CH2Cl2 at room temperature gave a
~5:1 mixture of 11 and C38-epi-11, which were separated by
reverse phase, medium-pressure column chromatography, to
furnish 11 (67% overall yield from 5) and C38-epi-11 (13%
overall yield from 5). With the method previously reported,[3c]
C38-epi-11 was isomerized to give additional 11 (9% isolated
yield), thereby making the overall yield of 11 from 5 76%. The
H
10
O
O
O
O
Me
I
H
H
1
O
O
37
30
O
H
O
35
O
O
H
O
C6H4OMe-p
O
O
10
TES
12: Right-Half of Halichondrin As
O
O
O
Me
1
O
20
H
structure of 11 was concluded from spectroscopic analysis; H
O
and 13C NMR spectra were found nicely to correspond to those
of norhalichondrin B.
O
H
10
O
5: Right-Half of Halichondrin Bs
O
O
O
Unified Synthesis of Halichondrins
The results given in the previous section made a convincing
case that the Zr/Ni-mediated one-pot ketone synthesis should
lead us to a development of a unified synthesis of the
halichondrin class of natural products. To demonstrate
experimentally, we synthesized three types of right- and left-
halves, respectively (Scheme 4).[12] Combinations of these right-
and left-halves should give all the nine halichondrins (Scheme 6).
The first stage in this approach was to apply the Zr/Ni-
mediated one-pot ketone synthesis for each combination. The
ketone coupling was conducted under the previously specified
condition, to furnish the expected products in 80~90% isolated
yields. The results were virtually identical with those found for
13: Right-Half of Halichondrin Cs
Left Halves of Halichondrins
H
TESO
O
50
Me
TES
O
O
O
H
H
H
53
O
40
O
MeO2C
Me
SPy
38
50
TESO
TBSO
15: Left-Half of Norhalichondrins A-C
O
Me
H
H
5+6
®
7, including coupling rates, isolated yields, and detected
54
O
TBSO
O
14: Left-Half of Halichondrins A-C
50
by-products. For example, the (5+14)-coupling was carried out in
a 200-mg scale of 5, to give the expected, desired ketone in
88% isolated yield, along with three by-products 8, 9, and one
corresponding to 10 in small amounts (~3%). Noteworthily, the
C12 allyl group of halichondrins-C was found intact in the time-
scale of ketone synthesis.
55
TBSO
O
H
H
TESO
Me
16: Left-Half of Homohalichondrins A-C
Scheme 4. Right- and left-halves of halichondrins
The second stage was deprotection of the silyl protecting
groups, followed by [5,5]-spiroketal formation under acidic
conditions. Halichondrin-B synthesis was first studied, where
deprotection of the silyl groups and formation of the [5,5]-
spiroketal were effected with TBAF buffered with pivalic acid in
DMF and then PPTS in CH2Cl2, to give a ~5:1 mixture of
halichondrin B and its C38-epimer. Reverse-phase medium-
pressure column chromatography was adopted for separation/
isolation, to furnish halichondrin B and C38-epimer in an
excellent overall yield; for example, 200 mg of 5 gave 133 mg
(68%) and 25 mg (13%) of halichondrin B and C38-epi-
halichondrin B, respectively. With the TMSOTf-induced
Similarly, the synthesis of halichondrin A, cf., 12+14
®
20, was
carried out. In this series, an additional step was required to
remove the C12/C13 anisylidene, i.e., PPTS treatment in a
mixture of isopropanol and 2,2-dimethyl-1,3-propandiol. During
the acid-treatment, the ratio of halichondrin A and its C38-
epimer changed from ~5:1 down to ~3:1.[13] As before, C38-
epimer was isomerized, furnishing halichondrin A in 61% total
yield from 12. Spectroscopic comparison confirmed that
halichondrin A was identical to the authentic sample.
The synthesis of halichondrin C, cf., 13+14
®
23 was also
method,[3c] C38-epi-halichondrin
B was isomerized to give
carried out. In this series, an additional step was required to
remove the allyl group at C12, which was uneventfully achieved
with the method used in the previous synthesis.[3b] Synthetic
halichondrin C and C38-epimer were isolated in 55% and 11%
yields, respectively. Spectroscopic comparison confirmed that
halichondrin C was identical to the authentic sample. Notably,
attempted TMSOTf-induced isomerization in CH2Cl2 did not yield
halichondrin C. This phenomenon was observed for all the
members in the halichondrin-C sub-group, but not for any
member in other sub-groups, thereby indicating that the reason
for the unsuccessful isomerization was due to the chemical
property of halichondrin-C polycycle. Spectroscopic analysis of a
product formed during the attempted isomerization suggested a
rearrangement of the halichondrin-C polycycle to the C12 ketal
reported in the halichondrin C synthesis, cf., see ketal 17 in
reference 3b.
additional 17 mg (9%) of halichondrin B. Thus, the overall yield
of halichondrin B was 77% from 5. Spectroscopic comparison
(HR-MS, 1H and 13C NMR) confirmed that halichondrin B was
identical to the authentic sample. The reproducibility of overall
transformation was excellent and no potential issue was noticed
for scaling.
Synthesis in the norhalichondrin series proceeded equally well,
although an extra step was required to hydrolyze the methyl
ester at C53, which was achieved under the condition used in
the previous work.[3] It should be noted that, for synthesis of
norhalichondrin C, base-induced hydrolysis of the methyl ester
This article is protected by copyright. All rights reserved.

10.1002/anie.201705523
Angewandte Chemie International Edition
COMMUNICATION
was done before deprotection of the ally group, because of the
base-instability of halichondrin-C polycycle. Spectroscopic
comparison established that norhalichondrins A-C thus obtained
were identical to the authentic samples.
150 mg of halichondrin B (77% yield) was obtained from 200 mg
of the right half 5. In turn, 5 was synthesized from commercial D-
galactal in 21.1% overall yield in 24 steps.[12] Thus, the overall
yield from commercial D-galactal to halichondrin B was 14.3%,
i.e., 1.07 g halichondrin B (MW = 1111) from 1.0 g D-galactal
(MW = 148). The reproducibility of overall transformation was
excellent and no potential issue was noticed for scaling.
Halichondrin-B Series
a-1; yields: i = 88%; ii = 77%
5
5
5
+
+
+
14
15
16
Halichondrin B (17)
a-2; yields: i = 84%; ii = 72%
a-3; yields: i = 82%; ii = 75%
Norhalichondrin B (18)
Homohalichondrin B (19)
Halichondrin-A Series
b-1; yields: i = 86%; ii = 61%
b-2; yields: i = 87%; ii = 64%
b-3; yields: i = 85%; ii = 71%
12
12
12
+
+
+
14
15
16
Halichondrin A (20)
Norhalichondrin A (21)
Homohalichondrin A (22)
Halichondrin-C Series
c-1; yields: i = 85%; ii = 55%
c-2; yields: i = 85%; ii = 58%
c-3; yields: i = 85%; ii = 65%
Figure 1. X-ray structure of halichondrin C
13
13
13
+
+
+
14
15
16
Halichondrin C (23)
Norhalichondrin C (24)
Homohalichondrin C (25)
Scheme 5. Synthesis of halichondrins. For all the cases, step #1 was ketone
coupling under the condition specified in Scheme 3; step #2 was TBAF (10
equiv), pivalic acid (5 equiv), DMF, rt, 3~8 h; step #3 was PPTS, CH2Cl2,
~20 °C, 2~4 h. Isomerization of C38-epi-halichondrins was done with TMSOTf,
CH2Cl2, -78 °C. For the halichondrin-A or -C series, these steps were followed
by PPTS, 2,2-dimethylpropan-1,3-diol, i-PrOH, rt, overnight or by Pd(PPh3)4,
dimedone, CH2Cl2, rt, 4~8 h, respectively. In the norhalichondrin series, the
methyl ester at C53 was hydrolyzed by treatment with aq. LiOH (excess), THF,
rt, 2 h. Numbers after i and ii indicate the yield for ketone couplings and overall
yield for after ketone coupling, respectively.
Acknowledgements
Financial support from the Eisai USA Foundation is gratefully
acknowledged. K.Y. thanks the Naito Foundation for a fellowship.
We thank Dr. Shao-Liang Zheng, Department of Chemistry and
Chemical Biology, Harvard University, for the X-ray data
collection and structure determination.
Keywords: natural products • total synthesis • ketone coupling •
Lastly, the ketone route was applied to the homohalichondrin
series. It is noteworthy that the previous enone route was not
effective for a synthesis of homohalichondrins; it was successful
only for homohalichondrin A, but with a very low efficiency (~5%
isolated yield). To our delight, the new synthetic route was found
effective for a total synthesis of all the homohalichondrins; the
overall efficiency in the homohalichondrin series was
comparable to that in the halichondrin and norhalichondrin
halichondrins • Zr/Ni-catalytic system
[1] a) D. Uemura, K. Takahashi, T. Yamamoto, C. Katayama, J. Tanaka, Y.
Okumura, Y. Hirata, J. Am. Chem. Soc. 1985, 107, 4796; b) Y. Hirata,
D. Uemura, Pure Appl. Chem. 1986, 58, 701.
[2] For isolation of the halichondrins from different species of sponges, see
reference 2 of reference 3c.
[3] a) T. D. Aicher, K. R. Buszek, F. G. Fang, C. J. Forsyth; S. H. Jung, Y.
Kishi, M. C. Matelich, P. M. Scola, D. M. Spero, S. K. Yoon, J. Am.
Chem. Soc. 1992, 114, 3162; b) A. Yamamoto, A. Ueda, P. Brémond,
P. S. Tiseni, Y. Kishi, J. Am. Chem. Soc. 2012, 134, 893; c) A. Ueda, A.
Yamamoto, D. Kato, Y. Kishi, J. Am. Chem. Soc. 2014, 136, 5171 and
references cited therein.
series. For instance, 100 mg
5
furnished 72 mg
homohalichondrin
B
(75% overall yield). Spectroscopic
comparison (HR-MS, 1H and 13C NMR) confirmed that
homohalichondrins A-C were identical to the authentic samples.
[4] For synthetic work by Salomon, Burke, Yonemitsu, Phillips, and Yadav
see reference 3 of reference 3c.
[5] The best combination of protecting groups at C35, C41, and C48 was
recently identified to be TES, TBS, and TES, respectively.
[6] D. P. Negri, Y. Kishi, Tetrahedron Lett. 1987, 28, 1063.
In summary, a unified, efficient, and scalable synthesis of the
halichondrin class of natural products was completed. Newly
developed Zr/Ni-mediated one-pot ketone synthesis was used
for coupling of right halves with left haves, where Cp2ZrCl2 with
Zn or Mn was found crucial to accelerate the coupling rate and,
at the same time, suppress by-product formation. Halichondrins
were obtained from these ketones basically in two operations,
i.e., desilylation and [5,5]-spiroketal formation. Notably, the new
synthetic route was successfully applied for a total synthesis of
all the homohalichondrins. All the halichondrins thus synthesized
were isolated as crystalline solids. We succeeded in growing a
single crystal for X-ray analysis for some of them; thus far, the
analysis completed for halichondrin C (Figure 1), which was the
first successful X-ray analysis of intact halichondrins.[9] To
demonstrate the scalability, halichondrin B was chosen, where
[7]
a) Y. Ai, N. Ye, Q. Wang, K. Yahata, Y. Kishi, submitted for publication;
b) J. H. Lee, Y. Kishi, J. Am. Chem. Soc. 2016, 138, 7178; c) V. P.
Kumar, V. S. Babu, K. Yahata, Y. Kishi, Org. Lett. 2017, 19, 2766.
[8] The feasibility of this disconnection was demonstrated with use of the
combination of CH2I at C40 with C(=O)SPy at C38.
[9] For the detailed procedure, see Supporting Information.
[10]
It was not clear whether 10 formed before or after 5 was consumed.
[11] This transformation was also done in one step, i.e., treatment of 7
directly with TBAF, buffered with pivalic acid.
[12]
a) K. Yahata, N. Ye, K. Iso, S. R. Naini, S. Yamashita, Y. Ai, Y. Kishi,
submitted for publication; b) K. Yahata, N. Ye, K. Iso, Y. Ai, J. Lee, Y.
Kishi, submitted for publication.
[13] Without addition of 2,2-dimethylpropan-1,3-diol, some of liberated
anisaldehyde was trapped with the triols at C51-C54.
This article is protected by copyright. All rights reserved.

10.1002/anie.201705523
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 1:
COMMUNICATION
1. Zr/Ni-mediated one-
pot ketone synthesis
A unified, efficient, and scalable
Kenzo Yahata, Ning Ye, Yanran
Ai, Kentaro Iso, and Yoshito Kishi*
I
2. TBAF
3. PPTS
synthesis of the halichondrin
class of natural products was
reported. The newly developed
Zr/Ni-mediated one-pot ketone
synthesis was used to couple
right halves with left halves,
where Cp2ZrCl2 with Zn or Mn
played a crucial dual role.
O
37
Me
35
H
H
H
O
Page No. – Page No.
O
O
H
H
38
O
44 40
TES
O
O
Unified, Efficient, and Scalable
Synthesis of Halichondrins,
with Use of Zr/Ni-Mediated One-
Pot Ketone Synthesis as Final
Coupling Reaction
H
TES
Me
O
O
Me
O
H
O
48
38 SPy
H
Me
This article is protected by copyright. All rights reserved.