Selective activation/coupling of polyhalogenated nucleophiles in Ni/Cr-Mediated reactions: Synthesis of C1-C19 building block of halichondrin Bs

Article abstract of DOI:10.1021/jacs.5b03498

The C1-C19 building block 46 of halichondrin Bs was synthesized via a selective activation/coupling of β-bromoenone 34 with aldehyde 35 in a Ni/Cr-mediated reaction. The first phase of study was a method development to effect a coupling of a 'naked' vinylogous anion with an aldehyde. The study with the coupling of 9 + 10 ? 11 revealed: (1) β-bromoenone 9b is a better nucleophile than the corresponding β-iodo- and β-chloroenones 9a,c; (2) (Me)₂Phen(OMe)₂·NiCl₂ 13b is a better Ni-catalyst than (Me)₂Phen(H)₂·NiCl₂ 13a; and (3) a low Ni-catalyst loading, for example, 0.05-0.1 mol % Ni-catalyst against 10 mol % Cr-catalyst, is crucial for an effective coupling. The second phase of study was a method development to realize a selective activation/coupling of polyhalogenated nucleophiles such as 34. The competition experiment of 10 + 9b over 10 + 31a-c revealed: (1) (Me)₂Phen(OMe)₂·NiCl₂ 13b is more effective than (Me)₂Phen(H)₂·NiCl₂ 13a for the required selective activation/coupling; (2) a low Ni-catalyst loading, for example, 0.05-0.1 mol % Ni-catalyst against 10 mol % Cr-catalyst, is crucial for discriminating β-bromoenone 9b from the three types of vinyl iodides 31a-c. The third phase of study was an application of the developed method to execute the proposed coupling of 34 + 35 ? 36. For this application, a polyether-type Ni-catalyst 37c, readily soluble in the reaction medium, was introduced to achieve the selective activation/coupling with higher efficiency. With use of ion-exchange resin-based device, the coupling product 36 was transformed to the C1-C19 building block 46 of halichondrin Bs without purification/separation of the intermediates.

Full text of DOI:10.1021/jacs.5b03498

Article
pubs.acs.org/JACS
Selective Activation/Coupling of Polyhalogenated Nucleophiles in
Ni/Cr-Mediated Reactions: Synthesis of C1−C19 Building Block of
Halichondrin Bs
Wuming Yan, Zhanjie Li, and Yoshito Kishi*
Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, United States
S
* Supporting Information
ABSTRACT: The C1−C19 building block 46 of halichondrin
Bs was synthesized via a selective activation/coupling of β-
bromoenone 34 with aldehyde 35 in a Ni/Cr-mediated
reaction. The first phase of study was a method development
to effect a coupling of a “naked” vinylogous anion with an
aldehyde. The study with the coupling of 9 + 10 → 11
revealed: (1) β-bromoenone 9b is a better nucleophile than the
corresponding β-iodo- and β-chloroenones 9a,c; (2)
(Me)2Phen(OMe)2·NiCl2 13b is a better Ni-catalyst than
(Me)2Phen(H)2·NiCl2 13a; and (3) a low Ni-catalyst loading,
for example, 0.05−0.1 mol % Ni-catalyst against 10 mol % Cr-
catalyst, is crucial for an effective coupling. The second phase
of study was a method development to realize a selective activation/coupling of polyhalogenated nucleophiles such as 34. The
competition experiment of 10 + 9b over 10 + 31a−c revealed: (1) (Me)2Phen(OMe)2·NiCl2 13b is more effective than
(Me)2Phen(H)2·NiCl2 13a for the required selective activation/coupling; (2) a low Ni-catalyst loading, for example, 0.05−0.1
mol % Ni-catalyst against 10 mol % Cr-catalyst, is crucial for discriminating β-bromoenone 9b from the three types of vinyl
iodides 31a−c. The third phase of study was an application of the developed method to execute the proposed coupling of 34 +
35 → 36. For this application, a polyether-type Ni-catalyst 37c, readily soluble in the reaction medium, was introduced to achieve
the selective activation/coupling with higher efficiency. With use of ion-exchange resin-based device, the coupling product 36 was
transformed to the C1−C19 building block 46 of halichondrin Bs without purification/separation of the intermediates.
INTRODUCTION
RESULTS AND DISCUSSION
■
■
General Notes on Cr-Mediated Coupling Reactions.
In conjunction with the ongoing research aimed at a unified
total synthesis of the halichondrin class of marine natural
products, we have been interested in a possibility of
synthesizing the C1−C19 building block of halichondrin Bs
via a coupling of A with B, because of its high degree of
convergence (Scheme 1).1−4 To execute this plan, we have
realized that it is necessary to overcome two major obstacles:
(1) how to achieve a selective activation of the β-haloenone
over the vinyl iodide, as well as the saturated chloride, present
in the nucleophile and (2) how to realize an effective carbonyl
addition of a “naked” vinylogous acyl anion to an aldehyde, see
the structure in the bracket in Scheme 1. Based on the
considerations given in the following sections, we have felt that
a Ni/Cr-mediated coupling reaction could achieve this crucial
C−C bond formation. In this paper, we report that, through
extensive method-development efforts, it has become possible
to achieve the proposed C−C bond formation at a synthetically
useful level, thereby allowing us to achieve a convergent and
practical synthesis of the C1−C19 building block of
halichondrin Bs.
Cr-mediated couplings of organic halides/triflates with
aldehydes belong to a class of 1,2-carbonyl addition reactions.
In this process, the active nucleophiles RCrX2 are generated
from halides/triflates in situ. Depending on the modes of
activation, Cr-mediated couplings are divided into three
subgroups: (1) Ni/Cr-mediated alkenylation, alkynylation and
arylation; (2) Co or Fe/Cr-mediated alkylation, 2-haloallyl-
ation and (propargylation); and (3) Cr-mediated allylation and
propargylation.5
Ni/Cr-mediated couplings of alkenyl halides/triflates with
aldehydes were originally reported by Takai, Hiyama, Nozaki,
and co-workers in 1983.6 Since then, it was shown that the
coupling was initiated by a catalytic amount of NiCl2
contaminated in CrCl2.7 It is now generally accepted that this
coupling involves: (1) oxidative addition of Ni(0), formed from
NiCl2 via reduction with CrCl2 in situ, to an alkenyl halide/
triflate to form an alkenyl Ni(II)-species; (2) transmetalation of
the resultant Ni(II)-species to Cr(II)Cl2 to form alkenyl
Received: April 3, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b03498
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
the contrary, a literature search has revealed that only a handful
of examples are known for C−C bond formation between
“naked” vinylogous acyl anions and aldehydes. Scheme 3 gives
Scheme 1. New Approach To Synthesize the C1−C19
Building Block in Halichondrin B
Scheme 3. Three Examples Relevant to the Proposed
Coupling
Cr(III)-species; and (3) carbonyl addition of the resultant
Cr(III)-species to an aldehyde to form the product Cr(III)-
alkoxide (Scheme 2). The chemistry has been developed to
achieve this coupling in a catalytic, asymmetric manner.8
three relevant examples. In 1992, we reported a coupling
reaction between methyl β-iodoacrylate 2 and aldehyde 1,
which allowed us effectively to form the C29−C30 bond of
halichondrins. Since then, we improved the original stoichio-
metric, nonasymmetric reaction to the catalytic, asymmetric
version.11 Knochel and Rao disclosed a coupling of β-
iodocyclohexenone 4 with aldehydes in the presence of CrCl2
in DMF.12 In connection with the synthetic efforts on the
taxane class of natural products, we used the Ni/Cr-mediated
coupling reaction to construct the taxane ring system, cf. 7 →
8.13,14 Notably, these successful examples relied on Cr-
organometallics. Based on this observation, we have focused
on the Ni/Cr-mediated coupling, to realize the proposed
coupling A + B → C.
Scheme 2. Catalytic Ni/Cr-Mediated Coupling Reactions:
(A) Overall Transformation and (B) Ni- and Cr-Catalytic
a
Cycles
The first phase of our study was to assess the coupling
efficiency of β-iodoenones with aldehydes, with use of model
substrates (Scheme 4). We tested the coupling efficiency
between β-iodoenone 9a and aldehyde 10 with 10 mol % Cr-
catalyst, prepared from (S)-sulfonamide 12, and 1 mol %
(Me)2Phen(H)2·NiCl2 13a in MeCN (Scheme 4).15,16 This
attempt demonstrated that expected coupling was indeed
possible. However, its efficiency (27% isolated yield) was far
below the synthetically useful level.17
We speculated that the activation of β-iodoenone 9a with Ni-
catalyst might be too fast, relative to the Cr-catalytic cycle, and
some of the resultant alkenyl Ni(II)-species might be wasted,
thereby causing a poor overall coupling efficiency. Based on this
speculation, we conducted two experiments. First, we compared
β-iodoenone 9a with β-bromo- and β-chloroenones 9b,c. We
anticipated that, because of electronic effects, 9b,c might be less
reactive than 9a and, therefore, might have a better reactivity-
balance between the Ni- and Cr-catalytic cycles. Indeed, this
experiment showed that (1) 9b,c gave a better coupling
efficiency than 9a and (2) 9b gave a slightly better coupling
efficiency than 9c.
a
Zr(Cp)2Cl2 or TMS-Cl is used to dissociate the product from Cr-
species, to regenerate the Cr-catalyst.
As mentioned, this bond-forming process is a Grignard-type
carbonyl addition reaction. However, it is noteworthy that this
reaction displays a remarkable selectivity toward aldehydes over
other carbonyl compounds. Activation of halides/triflates in the
presence of aldehydes provides not only an experimental
convenience but also an opportunity to achieve chemical
transformations in an unconventional manner, i.e., cyclization.
Undoubtedly, the most valuable feature of this coupling is its
exceptional compatibility with a wide range of functional
groups. This unique potential is appreciated most when applied
to polyfunctional molecules. Indeed, there are numerous
examples in which this bond-forming process has been used
at a late-stage in a multistep synthesis.
Coupling Efficiency of β-Haloenones with Aldehydes.
The success of the proposed synthesis of C1−C19 building
block D relies on a C−C bond formation between aldehyde A
with a vinylogous acyl anion generated from β-haloenone B.
There are several methods reported for C−C bond formation
between “masked” vinylgous acyl anions and aldehydes.9,10 To
With β-bromoenone 9b, we conducted the second experi-
ment. Namely, we tested the overall coupling efficiency with
less reactive Ni-catalyst and then with a lesser amount of the
Ni-catalyst. Based on the previous studies, we were aware that
B
DOI: 10.1021/jacs.5b03498
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 4. Model Coupling Reactions Used to Assess the
Coupling Efficiency
Table 1. Examples Tested for the Coupling of Di-Substituted
trans-β-Bromoenones with Aldehydes
a
a
a
Coupling conditions: (A) 10 mol % Cr-complex prepared from
sulfonamide 12 and 1 mol % nickel complex 13a (B) 10 mol % Cr-
complex prepared from sulfonamide 12 and 0.05 mol % nickel
complex 13b. The molar ratio of the substrates used was 9:10 = 1.5:1,
and 11 was isolated by silica gel chromatography and indicated yields
are based on aldehyde 10.
(Me)2Phen(OMe)2·NiCl2 13b is a slower activator of vinyl
iodides than (Me)2Phen(H)2·NiCl2 13a.18 Indeed, on replacing
13a with 13b, the coupling efficiency was noticeably improved.
We then optimized the ratio of Ni- over Cr-catalysts; with 10
mol % Cr-catalyst fixed, 1, 0.5. 0.1, 0.05, and 0.01 mol % Ni-
catalyst loadings were tested, thereby revealing that (1) the
coupling efficiency improved with lowering the Ni-catalyst
loading and (2) the coupling efficiency reached the plateau at
the 0.05−0.1 mol % Ni-catalyst loading. It is worthwhile noting
that the coupling reaction did not proceed without Ni-catalysts.
We used the coupling condition of “10 mol % Cr-catalyst,
prepared from sulfonamide 12, 0.05 mol % Ni-complex 13b,
Zr(Cp)2Cl2 (1.5 equiv), LiCl (2 equiv), and Mn (2 equiv) in
MeCN ([C] 0.4 M) at room temperature” for a study of
coupling efficiency. Table 1 summarizes the coupling efficiency
for disubstituted trans-β-bromoenones with aldehydes. All the
cases studied gave expected products in good yields. We should
note that the products thus obtained were stable enough to
isolate and characterize. However, on standing in benzene,
methylene chloride, and other solvents, at room temperature,
they gradually decomposed, to yield the corresponding furans.
With acid treatment (p-TSA or CSA/MeCN/RT), they gave
the furans almost instantaneously. On acylation, however, the
coupling products became stable even in the presence of acids
(aq. TFA, CH2Cl2, RT).
Table 2 summarizes our attempts to expand this coupling
reaction to other types of β-bromoenones. The first case
studied was cis-β-bromoenone 19; the attempted coupling did
not give the expected coupling product, but rather a ∼9:1
mixture of coupling product 11, identical to the coupling
product obtained from trans-β-bromoenone 9b and furan 20.
The problem encountered was two-fold. First, the geometrical
configuration of 19 was not retained in the coupling; we would
attribute it to a facile cis → trans isomerization of the alkenyl
Ni(II)-species formed from cis-β-bromoenone 19, thereby
giving a mixture of trans- and cis-coupling products.17,19
a
Coupling condition: 10 mol % Cr-catalyst, prepared from
sulfonamide 12, 0.05 mol % Ni-complex 13b, Zr(Cp)2Cl2 (1.5
equiv), LiCl (2 equiv), and Mn (2 equiv) in MeCN ([C] 0.4 M) at
room temperature for 3 h. The molar ratio of aldehyde and
nucleophile was 1:1.5; product was isolated by silica gel chromatog-
raphy; and indicated yields are based on aldehyde.
Second, the resultant cis-coupling product appeared not to
survive under the coupling condition employed, to form furan
20.
Trans-trisubstituted β-bromoenone 21 gave a 1:1 mixture of
coupling product 23 and furan 24, whereas cis-trisubstituted β-
bromoenone 22 gave only furan 24. The observed results could
be rationalized in the same term as given for the disubstituted
cis-β-bromoenone 19.
Overall, the disclosed coupling reaction between an aldehyde
and a “naked” vinylogous acyl anion is synthetically useful at
least for disubstituted trans-β-bromoenones. Interestingly, the
developed method meets our need to achieve the proposed
coupling reaction A + B → C (Scheme 1).
Selective Activation of β-Bromoenone over Vinyl
Iodide and Saturated Chloride. To the best of our
knowledge, there is no literature precedent to support directly
the proposed selective activation/coupling of a polyhalogenated
nucleophile in the Ni/Cr-mediated coupling reactions. Perhaps,
two examples shown in Scheme 5 are relevant, although
remotely, to the proposed plan. The first example shows that a
selective activation/coupling is possible with the use of selective
activator in the Cr-mediated couplings, namely, cobalt- and
iron-salts are known to activate saturated halides, but not vinyl
C
DOI: 10.1021/jacs.5b03498
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
that a selective activation/coupling of a β-bromoenone is
possible without disturbing the vinyl iodide present in 34.
In order to test the feasibility of the proposed selective
activation/coupling, we conducted competition experiments.
Namely, aldehyde 10 was coupled with a 1:1 mixture of β-
bromoenone 9b and vinyl iodide 31a, b, or c in the presence of
a different amount of Ni-catalysts 13a,b, followed by ratio-
analysis of the two expected products 11 and 32 (1H NMR)
(Table 3). The competition experiments demonstrated that (1)
Table 2. Attempted Coupling with Other Types of β-Bromo-
enones
a
Table 3. Competition Experiment of trans-β-Bromoenone 9b
a
over Vinyl Iodides 31a−c
a
Coupling condition: see Table 1.
Scheme 5. Two Examples Remotely Relevant to the
Proposed Synthetic Plan
a
Coupling conditions: nucleophile: a 1:1 mixture of 9b and 31a, b, or
c. Reagents: 10 mol % Cr-catalyst, prepared from sulfonamide 12, 0.05
mol % Ni-complex 13a or 13b, Zr(Cp)2Cl2 (1.5 equiv), LiCl (2
equiv), and Mn (2 equiv) in MeCN ([C] 0.4 M) at room temperature
1
for 3 h. The product ratio was estimated from H NMR analysis of
crude coupling products. b* and c* are the coupling products derived
from 31b and 31c, respectively.
0.05 and 0.1 mol % Ni-catalyst loadings, against 10 mol % Cr-
catalyst loading, allow selectively to activate/couple β-
bromoenone 9b over all the three types of vinyl iodides
31a−c and (2) Ni-catalyst 13b gives a better discrimination of
β-bromoenone 9b over vinyl iodides 31a−c than Ni-catalyst
13a. Interestingly, 0.05 and 0.1 mol % Ni-catalyst loadings
coincided with the Ni-catalyst amount ideal for β-bromoenone
couplings (see the previous section).
Encouraged by the results of competition experiments, we
began the coupling study of 34 and 35 (Scheme 7). Requisite
nucleophile 34 was readily prepared from the previously
reported, optically pure aldehyde 33 (Scheme 6).22 With
respect to the electrophile, we tested several possible protecting
groups at C8 and C9, thereby showing that the cyclohexylidene
is the best option.23
halides.20,21 The second example shows that a selective
activation of iodoacetylene in the Ni/Cr-mediated reaction is
possible without disturbing the vinyl iodide present in the
electrophile.3c
In this work, we propose to activate selectively a β-
bromoenone over a vinyl iodide as well as a saturated chloride.
As long as the Ni/Cr-mediated coupling is employed, the
selective activation of a β-bromoenone over a saturated chloride
should not be an issue. Thus, the central question is the
feasibility of activating selectively a β-bromoenone over a vinyl
iodide. We assume that, because of an electronic effect, a β-
bromoenone is more reactive toward Ni(0) than a vinyl iodide,
thus giving a hope for realizing the proposed selective
activation/coupling. The experimental results discussed in the
previous section are very encouraging, because the activation/
coupling of β-bromoenones can be achieved with a trace
amount of Ni-catalyst. However, it does not necessarily ensure
Aldehyde 35 was subjected to the Ni/Cr-coupling reaction
(10 mol % Cr-catalyst, prepared from sulfonamide 12, and 0.05
Scheme 6. Synthesis of Polyhalogenated Nucleophile 34
D
DOI: 10.1021/jacs.5b03498
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
mol % Ni-catalyst 13b), to furnish a single coupling product in
46% yield. The spectroscopic analysis (HR-MS, H NMR, and
Table 4. Examples Tested for the Coupling of
Polyhalogenated Nucleophile 34 with Various Aldehydes
1
a
13C NMR) demonstrated that the isolated product was the
desired coupling product 36. In particular, the C10−C11
vicinal proton spin-coupling constant (1.0 Hz) allowed us to
assign the desired β-configuration to the newly introduced
alcohol. Based on the previous examples similar to the present
case, we anticipated the desired diastereomer to be formed in a
high stereoselectivity with the Cr-catalyst prepared from (S)-
sulfonamide 12.24,25 Nevertheless, we were delighted to observe
the perfect stereocontrol for the coupling.
Naturally, we were anxious to improve the coupling
efficiency. In this connection, we should comment on one
technical difficulty encountered, i.e., controlling a very small
amount of Ni-catalyst 13b, which is virtually insoluble in the
reaction medium. We hoped that an improvement of Ni-
catalyst solubility may not only overcome the technical
difficulty but also improve the coupling efficiency, because
the Ni-catalyst can be homogeneously maintained at a very low
concentration in the reaction medium. Along with this line of
consideration, we prepared three polyether-type phenanthrene·
NiCl2 complexes 37a−c (Scheme 7). As hoped, the solubility of
a
Coupling conditions: 10 mol % Cr-catalyst, prepared from
sulfonamide 12, 0.05 mol % Ni-complex 37c, Zr(Cp)2Cl2 (1.5
equiv), LiCl (2 equiv), and Mn (2 equiv) in MeCN ([C] 0.4 M) at
room temperature for 3 h; the molar ratio of aldehyde and nucleophile
was 1:1.5; product was isolated by silica gel chromatography, and
indicated yields are based on aldehyde.
Scheme 7. Ni/Cr-Mediated Coupling of 34 with 35, with Use
a
of Polyether-Type Phenanthrene·NiCl2 Complex 37c
acylation of the resultant allylic alcohol. Among several acyl
groups tested, we chose p-nitrobenzoate, because it was found
to be stable under the aq. TFA condition required for
hydrolysis of the C8,C9-cyclohexylidene group, cf., step 2 in
Scheme 8.
On treatment with aqueous Na2CO3, the p-nitrobenzoate
group of aq. TFA-hydrolysis product was smoothly hydrolyzed,
followed by an oxy-Michael reaction of the C9 hydroxyl group
to the α,β-unsaturated ketone, to furnish a ∼1:2 mixture of 44
and 45 (Scheme 8). In the previous studies, we learned the
Scheme 8. Completion of Synthesis of the C1−C19 Building
a
Block 46 of Halichondrin B
a
Coupling conditions: 10 mol % Cr-catalyst, prepared from
sulfonamide 12, 0.05 mol % Ni-complex 37c, Zr(Cp)2Cl2 (1.5
equiv), LiCl (2 equiv), and Mn (2 equiv) in MeCN ([C] 0.4 M) at
room temperature for 3 h. Yield: chromatographically isolated yield in
7.1 g aldehyde scale.
these complexes, particularly 37b and 37c, was vastly improved.
To our delight, with the use of 37c, the coupling yield was also
dramatically improved from 46% to 87%.26
Table 4 summarizes the coupling of β-bromoenone 34 with
various aldehydes under the optimized condition. The coupling
efficiency of all the cases studied was very good. Among them,
the result with aldehyde 33 was interesting to demonstrate that
a selective activation of a β-bromoenone over a vinyl iodide is
indeed high; note that the activation of vinyl iodide in 33
induces cyclization with the aldehyde.
Synthesis of C1−C19 Building Block of Halichondrin
Bs. As mentioned earlier, the coupling product 36 was prone to
furan formation, but this instability could be overcome by
a
Yield: chromatographically isolated yield in 11.4 g scale of coupling
product 36.
E
DOI: 10.1021/jacs.5b03498
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
chemical behaviors of these oxy-Michael products, including:
(1) PPTS treatment allows the C12-β oxy-Michael product 45
to convert to the desired polycycle, cf., 46; (2) undesired C12-
α oxy-Michael product 44 can be recycled via retro oxy-
Michael/oxy-Michael equilibration under basic condition; (3)
an ion-exchange resin-based device allows the mixture of oxy-
Michael products to convert to the desired polycycle without
isolation and recycling of the undesired oxy-Michael product.27
With this background, we attempted to convert oxy-Michael
products 44 and 45 into polycycle 46, thereby revealing that
(1) transformation of 45 into 46 under the PPTS condition was
clean and facile, but (2) isomerization of 44 to 45 under the
previously established basic conditions or ion-exchange-resin
protocol was problematic; one problem identified was the
elimination of HCl to form iodo-diene (see the lower half of
Scheme 8). With this information, we searched for a reaction
condition to establish the equilibrium between two oxy-Michael
products without elimination of HCl and eventually found that
the equilibration can be established with DBU or tetrame-
thylguanidine in isopropanol or ethanol at room temperature,
without the undesired elimination.28
iodides 31a−c. The third phase of study was an application of
the developed method to execute the proposed coupling of 34
+ 35 → 36. For this application, a polyether-type Ni-catalyst
37c, readily soluble in the reaction medium, was introduced to
achieve the selective activation/coupling with higher efficiency.
With use of ion-exchange resin-based device, the coupling
product 36 was transformed to the C1−C19 building block 46
of halichondrin Bs without purification/separation of the
intermediates.
Lastly, it is worthwhile noting that all of the halogens present
in the nucleophile 34 are designed selectively to achieve specific
bond formation in a controlled manner, as illustrated in the
synthesis of right-half of halichondrin A (Scheme 9).3c Namely,
Scheme 9. Summary of the Synthesis of the Right-Half of
Halichondrin A
Naturally, we hoped to translate some of these conditions to
ion-exchange resin-based device and found that polymer-bound
guanidine base, coupled with polymer-bound PPTS, was
effective directly to convert a mixture of oxy-Michael products
44 and 45 to polycycle 46 in a high yield without isolation/
separation/equilibration (Figure 1).29−31 The structure of C1−
Figure 1. Polymer-bound guanidine and PPTS used for ion-exchange
resin-based device.
C19 building block 46 thus synthesized was fully supported by
1
spectroscopic data (HR-MS, H and 13C NMR), which was
further confirmed by X-ray analysis of its derivative.32
The synthesis reported is easy to scale; the overall yield of 46
from 36 was 69% in a 11.4 g scale.
C19 vinyl bromide or iodide was used for the Ni/Cr-mediated
coupling stereoselectively to form the C19−C20 bond, whereas
C17 chloride allowed stereospecifically to form the tetrahy-
drofuran ring in an SN2 fashion.
CONCLUSION
■
The C1−C19 building block 46 of halichondrin Bs was
synthesized via a selective activation/coupling of β-bromoe-
none 34 with aldehyde 35 in a Ni/Cr-mediated reaction. The
first phase of study was a method development to effect a
coupling of a “naked” vinylogous anion with an aldehyde. The
study with the coupling of 9 + 10 → 11 revealed: (1) β-
bromoenone 9b is a better nucleophile than the corresponding
β-iodo- and β-chloroenones 9a,c; (2) (Me)2Phen(OMe)2·
NiCl2 13b is a better Ni-catalyst than (Me)2Phen(H)2·NiCl2
13a; and (3) a low Ni-catalyst loading, for example, 0.05−0.1
mol % Ni-catalyst against 10 mol % Cr-catalyst, is crucial for an
effective coupling. The second phase of study was a method
development to realize a selective activation/coupling of
polyhalogenated nucleophiles such as 34. The competition
experiment of 10 + 9b over 10 + 31a−c revealed: (1)
(Me)2Phen(OMe)2·NiCl2 13b is more effective than
(Me)2Phen(H)2·NiCl2 13a for the required selective activa-
tion/coupling; (2) a low Ni-catalyst loading, for example, 0.05−
0.1 mol % Ni-catalyst against 10 mol % Cr-catalyst, is crucial for
discriminating β-bromoenone 9b from the three types of vinyl
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, characterization data, copies of
spectra, and crystallographic data (CIF). The Supporting
Information is available free of charge on the ACS Publications
website at DOI: 10.1021/jacs.5b03498.
AUTHOR INFORMATION
Corresponding Author
*kishi@chemistry.harvard.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support from the Eisai USA Foundation is gratefully
acknowledged. We thank Dr. Shao-Liang Zheng for his help
with the X-ray data collection and structure determination.
F
DOI: 10.1021/jacs.5b03498
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
C.-G.; Kim, D.-S.; Urabe, D.; Wang, J.; Kim, J. T.; Liu, X.; Sasaki, T.;
Kishi, Y. J. Am. Chem. Soc. 2009, 131, 15387.
(12) Knochel, P.; Rao, C. J. Tetrahedron 1993, 49, 29.
(13) Kress, M. H.; Ruel, R.; Miller, W. H.; Kishi, Y. Tetrahedron Lett.
1993, 34, 6003.
REFERENCES
■
(1) For isolation/structure elucidation of the halichondrins from the
marine sponge Halichondria okadai by Uemura and Hirata, see:
(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 isolation of the halichondrins from different species of
sponges, see: (a) Pettit, G. R.; Herald, C. L.; Boyd, M. R.; Leet, J. E.;
Dufresne, C.; Doubek, D. L.; Schmidt, J. M.; Cerny, R. L.; Hooper, J.
(14) Our research efforts in this area were recorded in three
dissertations: (a) Kress, M. H. Applications of the Ni(II)/Cr(II)-
Mediated Coupling Reaction to the Synthesis of the Taxane
Diterpenes, Ph.D. Thesis, Harvard University, January, 1995.
(b) Sheng, X. C. Total Synthesis of (+)-Cinnamoyltaxicin-I Triacetate,
Ph.D. Thesis, Harvard University, September, 1998. (c) Jongwon, L.
Total Synthesis of Taxol, Ph.D. Thesis, Harvard University, July, 2000.
(15) The coupling condition routinely used for ligand optimization
with the toolbox approach, cf., ref 11d.
(16) We anticipated that (R)-11 was predominantly formed because
the Cr-catalyst, prepared from an optically active sulfonamide (S)-12,
was used. 1H NMR analysis of its (R)- and (S)-Mosher esters indicated
that its optical purity was ∼70% ee. Because our goal of this study was
to demonstrate the efficiency of coupling, we did not attempt to
improve the asymmetric induction, for example, by adopting the
toolbox approach (ref 11d). For the same reason, we did not examine
the optical purity for the coupling products listed in Tables 1, 2, and 4.
(17) In spite of only 27% isolated yield, no distinct byproduct(s) was
detected, i.e., TLC showed a streaking line, besides 11.
(18) Liu, X.; Li, X.; Chen, Y.; Hu, Y.; Kishi, Y. J. Am. Chem. Soc. 2012,
134, 6136.
(19) For cis → trans isomerization of alkenyl Ni(II)-species, for
example see: (a) Huggins, J. M.; Bergman, R. G. J. Am. Chem. Soc.
1981, 103, 3002. (b) Hirata, Y.; Tanaka, M.; Yada, A.; Nakao, Y.;
Hiyama, T. Tetrahedron 2009, 65, 5037.
(20) Takai, K.; Nitta, K.; Fujimura, O.; Utimoto, K. J. Org. Chem.
1989, 54, 4732.
(21) (a) Wan, Z.-K.; Choi, H.-w.; Kang, F.-A.; Nakajima, K.; Demeke,
D.; Kishi, Y. Org. Lett. 2002, 4, 4431. (b) ref 11d.
(22) Liu, S.; Kim, J. T.; Dong, C.-G.; Kishi, Y. Org. Lett. 2009, 11,
4520.
(23) C8,C9-Protecting groups tested included acetonide, anisylidene,
bis-TBS, carbonate, bis-Ac, bis-Bz, and bis-MOM.
(24) (a) Aicher, T. D.; Kishi, Y. Tetrahedron Lett. 1987, 28, 3463. (b)
ref 11d.
(25) Six chiral sulfonamides were tested. All the Cr-catalysts prepared
from them gave the desired 11β-alcohol stereospecifically in the
indicated yields. (S)-Sulfonamide 12 was chosen, because it gave the
best coupling yield.
(26) The coupling yields with 37a and 37b were 59% and 75%,
respectively. We also prepared a Ni-catalyst with n-dodecyloxy
substituents, i.e., X = n-C12H25O in 13, but found that its solubility
was roughly same as that of 13b and the coupling yield with this Ni-
catalyst was 60%.
(27) (a) Namba, K.; Jun, H.-S.; Kishi, Y. J. Am. Chem. Soc. 2004, 126,
7770. (b) Kaburagi, Y.; Kishi, Y. Org. Lett. 2007, 9, 723.
(28) DBU, Triton B(OMe), and tetramethylguanidine were tested.
DBU and tetramethylguanidine established an equilibrium in
isopropanol or ethanol at RT without an elimination of HCl, whereas
caused an elimination of HCl in DMF and MeCN. Triton B(OMe)
caused unknown decomposition of the oxy-Michael products.
(29) Basic ion-exchange resins tested included Amberlite IRA-400,
Amberlite IRA-402, Amberlite IRA-900, Amberlite A-21, Amberlite A-
26, and Amberlite A-27. Acidic ion-exchange resins tested included
Rexyn 101, Amberlite IR-120, Amberlite 15, and Amberlite IRC-86.
(30) Both purchased from Aldrich: polymer-bound guanidine:
#358754; polymer-bound PPTS: #82817.
N. A.; Rutzler, K. C. J. Med. Chem. 1991, 34, 3339. (b) Pettit, G. R.;
̈
Tan, R.; Gao, F.; Williams, M. D.; Doubek, D. L.; Boyd, M. R.;
Schmidt, J. M.; Chapuis, J.-C.; Hamel, E.; Bai, R.; Hooper, J. N. A.;
Tackett, L. P. J. Org. Chem. 1993, 58, 2538. (c) Litaudon, M.; Hart, J.
B.; Blunt, J. W.; Lake, R. J.; Munro, M. H. G. Tetrahedron Lett. 1994,
35, 9435. (d) Litaudon, M.; Hickford, S. J. H.; Lill, R. E.; Lake, R. J.;
Blunt, J. W.; Munro, M. H. G. J. Org. Chem. 1997, 62, 1868.
(e) Hickford, S. J. H.; Blunt, J. W.; Munro, M. H. G. Bioorg. Med.
Chem. 2009, 17, 2199.
(3) For the synthetic work from this group, see: (a) Halichondrin
Bs: 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) Halichondrin Cs: Yamamoto,
́
A.; Ueda, A.; Bremond, P.; Tiseni, P. S.; Kishi, Y. J. Am. Chem. Soc.
2012, 134, 893. (c) Halichondrin As: Ueda, A.; Yamamoto, A.; Kato,
D.; Kishi, Y. J. Am. Chem. Soc. 2014, 136, 5171 and references cited
therein.
(4) For synthetic efforts by Salomon, Burke, Yonemitsu, Phillips, and
Yadav see: (a) Henderson, J. A.; Jackson, K. L.; Phillips, A. J. Org. Lett.
2007, 9, 5299. Jackson, K. L.; Henderson, J. A.; Motoyoshi, H.;
Phillips, A. J. Angew. Chem., Int. Ed. 2009, 48, 2346 and the references
cited therein. (b) Burke, S. D.; Buchanan, J. L.; Rovin, J. D.
Tetrahedron Lett. 1991, 32, 3961. Lambert, W. T.; Hanson, G. H.;
Benayoud, F.; Burke, S. D. J. Org. Chem. 2005, 70, 9382 and the
references cited therein. (c) Kim, S.; Salomon, R. G. Tetrahedron Lett.
1989, 30, 6279. Cooper, A. J.; Pan, W.; Salomon, R. G. Tetrahedron
Lett. 1993, 34, 8193 and the references cited therein. (d) Horita, K.;
Hachiya, S.; Nagasawa, M.; Hikota, M.; Yonemitsu, O. Synlett 1994,
38. Horita, K.; Nishibe, S.; Yonemitsu, O. Phytochem. Phytopharm.
2000, 386 and the references cited therein. (e) Yadav, J. S.; Reddy, C.
N.; Sabitha, G. Tetrahedron Lett. 2012, 53, 2504.
(5) For reviews on Cr-mediated carbon−carbon bond-forming
reactions, see: (a) Saccomano, N. A. In Comprehensive Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol.
1, p 173. (b) Furstner, A. Chem. Rev. 1999, 99, 991. (c) Wessjohann, L.
̈
A.; Scheid, G. Synthesis 1999, 1. (d) Takai, K.; Nozaki, H. Proc. Japan
Acad. Ser. B 2000, 76, 123. (e) Takai, K. Org. React. 2004, 64, 253.
(e) Hargaden, G. C.; Guiry, P. J. Adv. Synth. Catal. 2007, 349, 2407.
(6) Takai, K.; Kimura, K.; Kuroda, T.; Hiyama, T.; Nozaki, H.
Tetrahedron Lett. 1983, 24, 5281.
(7) (a) Jin, H.; Uenishi, J.-I.; Christ, W. J.; Kishi, Y. J. Am. Chem. Soc.
1986, 108, 5644. (b) Takai, K.; Tagashira, M.; Kuroda, T.; Oshima, K.;
Utimoto, K.; Nozaki, H. J. Am. Chem. Soc. 1986, 108, 6048.
(8) For the work from this laboratory, see: (a) Ref 7a. (b) Ref 21a.
(c) Choi, H.-w.; Nakajima, K.; Demeke, D.; Kang, F.-A.; Jun, H.-S.;
Wan, Z.-K.; Kishi, Y. Org. Lett. 2002, 4, 4435. (d) Ref 11c. (e) Ref
11d. (f) Liu, X.; Henderson, J. A.; Sasaki, T.; Kishi, Y. J. Am. Chem. Soc.
2009, 131, 16678. (g) Ref 18 and references cited therein.
(9) For “masked” vinylogous acyl anion equivalents, see: (a) Cohen,
T.; Bennett, A. D.; Mura, A. J., Jr. J. Org. Chem. 1976, 41, 2506.
(b) Debal, A.; Cuvigny, T.; Larcheveque, M. Tetrahedron Lett. 1977,
18, 3187. (c) Piers, E.; Morton, H. E. J. Org. Chem. 1979, 44, 3437.
(d) Ren, H.; Krasovskiy, A.; Knochel, P. Org. Lett. 2004, 6, 4215.
(10) For a review on related subjects, see: Chinchilla, R.; Najera, C.
Chem. Rev. 2000, 100, 1891.
(31) As ethanol was used as the solvent, an ester exchange was
noticed if the reaction was run over 1 day. However, it did not present
an issue for preparative purpose, as the conversion was usually
complete within 12 h.
(11) (a) Aicher, T. D.; Buszek, K. R.; Fang, F. G.; Forsyth, C. J.; Jung,
S. H.; Kishi, Y.; Scola, P. M. Tetrahedron Lett. 1992, 33, 1549.
(b) Chen, C.; Tagami, K.; Kishi, Y. J. Org. Chem. 1995, 60, 5386.
(c) Namba, K.; Kishi, Y. Org. Lett. 2004, 6, 5031. (d) Guo, H.; Dong,
(32) See Supporting Information.
G
DOI: 10.1021/jacs.5b03498
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX