Toolbox approach to the search for effective ligands for catalytic asymmetric Cr-mediated coupling reactions

Article abstract of DOI:10.1021/ja905843e

Chromium catalysts derived from chiral sulfonamides represented by A effect the couplings of aldehydes with vinyl, allyl, or alkyl halides. With three distinct sites for structural modification, A affords access to a structurally diverse pool of chiral sulfonamides. The Cr catalysts derived from these sulfonamides exhibit a broad range of catalyst-substrate matching profiles. A strategy is presented to search for a satisfactory chiral sulfonamide for a given substrate. In order to demonstrate the generality and effectiveness of this approach, five diverse C-C bond-forming cases have been selected from the halichondrin synthesis. For each of the cases, two ligands have been deliberately searched for, to induce the formation of (R)- and (S)-alcohols, respectively, at the arbitrarily chosen efficiency level of "≥80% yield with ≥20:1 stereoselectivity in the presence of ≤20 mol % of a Cr catalyst". For 9 out of the 10 cases studied, a satisfactory catalyst has been found within this pool of sulfonamides. Even for the remaining case, a Cr catalyst inducing stereoselectivity up to 8:1 has been identified.

Full text of DOI:10.1021/ja905843e

Published on Web 10/01/2009
Toolbox Approach to the Search for Effective Ligands for
Catalytic Asymmetric Cr-Mediated Coupling Reactions
Haibing Guo, Cheng-Guo Dong, Dae-Shik Kim, Daisuke Urabe, Jiashi Wang,
Joseph T. Kim, Xiang Liu, Takeo Sasaki, and Yoshito Kishi*
Department of Chemistry and Chemical Biology, HarVard UniVersity, 12 Oxford Street,
Cambridge, Massachusetts 02138
Received July 20, 2009; E-mail: kishi@chemistry.harvard.edu
Abstract: Chromium catalysts derived from chiral sulfonamides represented by A effect the couplings of
aldehydes with vinyl, allyl, or alkyl halides. With three distinct sites for structural modification, A affords
access to a structurally diverse pool of chiral sulfonamides. The Cr catalysts derived from these sulfonamides
exhibit a broad range of catalyst-substrate matching profiles. A strategy is presented to search for a
satisfactory chiral sulfonamide for a given substrate. In order to demonstrate the generality and effectiveness
of this approach, five diverse C-C bond-forming cases have been selected from the halichondrin synthesis.
For each of the cases, two ligands have been deliberately searched for, to induce the formation of (R)-
and (S)-alcohols, respectively, at the arbitrarily chosen efficiency level of “g80% yield with g20:1
stereoselectivity in the presence of e20 mol % of a Cr catalyst”. For 9 out of the 10 cases studied, a
satisfactory catalyst has been found within this pool of sulfonamides. Even for the remaining case, a Cr
catalyst inducing stereoselectivity up to 8:1 has been identified.
Scheme 1. Cr(III)-Mediated Couplings
1. Introduction
Cr(II)-mediated addition of allyl halides/triflates to aldehydes
was reported by Hiyama, Nozaki, and co-workers in 1977.1 In
this reaction, the active nucleophiles RCrX3 were generated in
situ via oxidative addition of Cr(II) to allyl halides/triflates
(Scheme 1). Since then, new methods have been developed to
form the active nucleophiles RCrX3 from a wide range of halides
and triflates.2 Depending on the activation methods, Cr-mediated
couplings are now divided into three subgroups: (1) Ni/Cr-
mediated alkenylation, alkynylation, and arylation;3 (2) Co- or
Fe/Cr-mediated alkylation, 2-haloallylation, and propargylation;4
and (3) Cr-mediated allylation and propargylation.
Overall, the Cr-mediated C-C bond-forming reaction is
viewed as a Grignard-type carbonyl addition of halides.
However, it is noteworthy that this reaction displays a remark-
able selectivity toward aldehydes over other carbonyl com-
pounds. Activation of halides in the presence of aldehydes
provides us with not only an experimental convenience but also
an opportunity to realize chemical transformations in an
unconventional manner, cf., cyclization.5 Undoubtedly, the most
valuable feature of this reaction is its exceptional compatibility
with a wide range of functional groups. This unique potential
is appreciated most when applied to polyfunctional molecules.
There are numerous examples in which this reaction has been
successfully used at a late stage in a multistep synthesis.6
For its application to a practical synthesis, it is desirable to
develop a catalytic process for the Cr-mediated coupling
reaction. In 1996, Fu¨rstner and Shi reported a catalytic version
of this reaction, in which TMS-Cl and Mn(0) are used as a
dissociating agent of chromium alkoxides and a reducing agent
of chromium, respectively (step 4 and step 1 in Scheme 2).7
Mn(0) is the most effective agent to reduce Cr(III) and
regenerate Cr(II).8 TMS-Cl is an effective dissociating agent,
(1) (a) Okude, Y.; Hirano, S.; Hiyama, T.; Nozaki, H. J. Am. Chem. Soc.
1977, 99, 3179. (b) Takai, K.; Kimura, K.; Kuroda, T.; Hiyama, T.;
Nozaki, H. Tetrahedron Lett. 1983, 24, 5281.
(5) For examples, see: (a) Schreiber, S. S.; Meyers, H. V. J. Am. Chem.
Soc. 1988, 110, 5198. (b) Rowley, M.; Tsukamoto, M.; Kishi, Y. J. Am.
Chem. Soc. 1989, 111, 2735. (c) Namba, K.; Kishi, Y. J. Am. Chem.
Soc. 2005, 127, 15382.
(2) 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)
Fu¨rstner, A. Chem. ReV. 1999, 99, 991. (c) Wessjohann, L. A.; Scheid,
G. Synthesis 1999, 1. (d) Takai, K.; Nozaki, H. Proc. Jpn. Acad. Ser.
B 2000, 76, 123. (e) Hargaden, G. C.; Guiry, P. J. AdV. Synth. Catal.
2007, 349, 2407.
(6) For examples, see: (a) Armstrong, R. W.; Beau, J.-M.; Cheon, S. H.;
Christ, W. J.; Fujioka, H.; Ham, W.-H.; Hawkins, L. D.; Jin, H.; Kang,
S. H.; Kishi, Y.; Martinelli, M. J.; McWhorter, W. W., Jr.; Mizuno,
M.; Nakata, M.; Stutz, A. E.; Talamas, F. X.; Taniguchi, M.; Tino,
J. A.; Ueda, K.; Uenishi, J.; White, J. B.; Yonaga, M. J. Am. Chem.
Soc. 1989, 111, 7530. (b) Aicher, T. D. K.; Buszek, 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. (c)
Oddon, G.; Uguen, D. Tetrahedron Lett. 1998, 39, 1157.
(3) (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.
(4) (a) Takai, K.; Nitta, K.; Fujimura, O.; Utimoto, K. J. Org. Chem. 1989,
54, 4732. (b) Choi, H.-w.; Nakajima, K.; Demeke, D.; Kang, F.-A.;
Jun, H.-S.; Wan, Z.-K.; Kishi, Y. Org. Lett. 2002, 4, 4435. (c) Kurosu,
M.; Lin, M.-H.; Kishi, Y. J. Am. Chem. Soc. 2004, 126, 12248.
(7) (a) Fu¨rstner, A.; Shi, N. J. Am. Chem. Soc. 1996, 118, 2533. (b)
Fu¨rstner, A.; Shi, N. J. Am. Chem. Soc. 1996, 118, 12349. (c) Fu¨rstner,
A.; Wuchrer, M. Chem.sEur. J. 2006, 12, 76.
9
10.1021/ja905843e CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 15387–15393 15387

A R T I C L E S
Guo et al.
Scheme 2. Catalytic Cycle of Cr-Mediated Coupling Reactionsa
a For the activation of halides (step 2), see panels A and B in Scheme 4.
Figure 1. Structure of the first generation of (S)-sulfonamide 1 and three
X-ray structures of Cr complexes.
but we found that Zr(Cp)2Cl2 is a more effective dissociating
agent.9h The difference between TMS-Cl and Zr(Cp)2Cl2 is two-
fold. First, the catalytic Cr-mediated coupling in the presence
of TMS-Cl does not proceed to completion for enolizable
aldehydes, because of silyl enol ether formation; upon aqueous
workup, typically 10-20% of the aldehyde is recovered, which
is recyclable for many cases. However, for the enolizable
aldehydes bearing a chiral center at the carbon adjacent to the
aldehyde, the recovered starting material may not maintain its
stereochemical homogeneity. To the contrary, in the presence
of Zr(Cp)2Cl2, the coupling proceeds smoothly to completion
without presenting this problem.9h Second, the coupling rate
with Zr(Cp)2Cl2 is significantly faster than that with TMS-Cl.
Overall, the Zr(Cp)2Cl2/Mn system is effective for regenerat-
ing the catalyst employed for all the subgroups of Cr-mediated
couplings. Coupled with this system, it is now possible to
achieve the Cr-mediated coupling reactions in the presence of
1 mol % of the Cr catalyst prepared from CrCl3 ·3THF and 3,3′-
dimethyl-2,2′-dipyridyl or 4,4′-di-tert-butyl-2,2′-dipyridyl.9j
Several ligands have been reported to induce asymmetric Cr-
mediated couplings under stoichiometric and/or catalytic condi-
tions.10 We have reported that the catalyst prepared from CrCl2
and (S)-sulfonamide 1 (Figure 1) is effective to achieve
asymmetric couplings under stoichiometric and catalytic
conditions.9d-g,i,l On the basis of the X-ray structure of
Cr(III)(1-)(Cl)2, Cr(III)(1-)(Me)(Cl), and Cr(II)(4-t-Bu-py)3(Cl)2
complexes (Figure 1), we proposed that the sulfonamide-based
asymmetric coupling reaction proceeds through the intermediates
depicted in Scheme 3. In this proposal, the Cr catalyst maintains
the octahedral structure throughout the catalytic cycle, and all
the events take place at only two ligation sites of the octahedral
catalyst.9d
Scheme 3. Proposed Structures of the Cr(III) Species Involved in
the Catalytic Coupling Reaction
In order to achieve asymmetric Ni/Cr-mediated coupling
under the catalytic conditions, the Ni catalyst needs to meet
two criteria. First, the Ni salt must have an efficient catalytic
cycle (panel A in Scheme 4). Second, the Ni catalyst should
not interfere with the asymmetric process induced by the chiral
Cr-sulfonamide ligand. Among many Ni(II) complexes tested,
the 2,9-dimethylphenanthroline/NiCl2 complex (NiCl2 ·DMP) is
most satisfactory. NiCl2 ·DMP is known to exist as either the
(10) For chiral ligands used in Cr-mediated couplings, see: (a) Reference
3b. (b) N-Benzoylprolinol system: Sugimoto, K.; Aoyagi, S.; Kiba-
yashi, C. J. Org. Chem. 1997, 62, 2322. (c) Salen system: Bandini,
M.; Cozzi, P. G.; Melchiorre, P.; Umani-Ronchi, A. Angew. Chem.,
Int. Ed. 1999, 38, 3357. Bandini, M.; Cozzi, P. G.; Licciulli, S.; Umani-
Ronchi, A. Synthesis 2004, 409, and references cited therein. Also
see: Berkessel, A.; Menche, D.; Sklorz, C. A.; Schro¨der, M.; Paterson,
I. Angew. Chem., Int. Ed. 2003, 42, 1032. (d) Oxazoline/sulfonamide
system: References 9c and 9f and references cited therein. (e)
Bis(oxazolinyl)carbazole system: Inoue, M.; Suzuki, T.; Nakada, M.
J. Am. Chem. Soc. 2003, 125, 1140. Inoue, M.; Suzuki, T.; Kinoshita,
A.; Nakada, M. Chem. Rec. 2008, 8, 169, and references cited
therein. Also see: McManus, H. A.; Cozzi, P. G.; Guiry, P. J. AdV.
Synth. Catal. 2006, 348, 551. Hargaden, G. C.; McManus, H. A.;
Cozzi, P. G.; Guiry, P. J. Org. Biomol. Chem. 2007, 5, 763. (f)
Oxazoline/prolineamide system: Lee, J.-Y.; Miller, J. J.; Hamilton,
S. S.; Sigman, M. S. Org. Lett. 2005, 7, 1837. Miller, J. J.; Sigman,
M. S. J. Am. Chem. Soc. 2007, 129, 2752. Also see: Hargaden, G. C.;
Mu¨ller-Bunz, H.; Guiry, P. J. Eur. J. Org. Chem. 2007, 4235. (g)
Tethered bis(8-quinolinol): Xia, G.; Yamamoto, H. J. Am. Chem. Soc.
2006, 128, 2554; 2007, 129, 496.
(8) Electrochemically driven Cr(II)-mediated couplings are also known: (a)
Grigg, R.; Putnikovic, B.; Urch, C. J Tetrahedron Lett. 1997, 38, 6307.
(b) Kuroboshi, M.; Tanaka, M.; Kishimoto, S.; Goto, K.; Tanaka, H.;
Torii, S. Tetrahedron Lett. 1999, 40, 2785. (c) Kuroboshi, M.; Tanaka,
M.; Kishimoto, S.; Tanaka, H.; Torii, S. Synlett 1999, 69. (d)
Durandetti, M.; Nedelec, J.-Y.; Perichon, J. Org. Lett. 2001, 3, 2073.
(9) For relevant studies on Cr-mediated couplings from this laboratory,
see: (a) Reference 3a. (b) Kishi, Y. Pure Appl. Chem. 1992, 64, 343.
(c) Chen, C.; Tagami, K.; Kishi, Y. J. Org. Chem. 1995, 60, 5386.
(d) Wan, Z.-K.; Choi, H.-w.; Kang, F.-A.; Nakajima, K.; Demeke,
D.; Kishi, Y. Org. Lett. 2002, 4, 4431. (e) Reference 4b. (f) Choi,
H.-w.; Demeke, D.; Kang, F.-A.; Kishi, Y.; Nakajima, K.; Nowak,
P.; Wan, Z.-K.; Xie, C. Pure Appl. Chem. 2003, 75, 1. (g) Reference
4c. (h) Namba, K.; Kishi, Y. Org. Lett. 2004, 6, 5031. (i) Namba, K.;
Cui, S.; Wang, J.; Kishi, Y. Org. Lett. 2005, 7, 5417. (j) Namba, K.;
Wang, J.; Cui, S.; Kishi, Y. Org. Lett. 2005, 7, 5421. (k) Reference
5c. (l) Zhang, Z.; Huang, J.; Ma, B.; Kishi, Y. Org. Lett. 2008, 10,
3073, and references cited therein.
9
15388 J. AM. CHEM. SOC. VOL. 131, NO. 42, 2009

Ligands for Catalytic Asymmetric Cr-Mediated Coupling
A R T I C L E S
Scheme 4. NiCl2 ·DMP and CoPc-Assisted Activation of Halidesa
Figure 2. Generic structure of sulfonamide A.
Scheme 5. Absolute Configuration of the Alcohols Preferentially
Formed in the Presence of the Cr Catalyst Derived from
(S)-Sulfonamides
a broad range of substrates.13 We first outline an effective
strategy to search for a satisfactory ligand for a given substrate.
Using the diverse C-C bond-forming cases selected from the
halichondrin synthesis,14 we then demonstrate that a satisfactory
chiral ligand can indeed be found from this pool. For each of
the cases studied, we deliberately search for two ligands, which
induce the formation of (R)- and (S)-alcohols, respectively, with
the arbitrarily chosen efficiency of “g80% yield with g20:1
stereoselectivity in the presence of e20 mol % of a Cr catalyst”.
Before discussing specific examples, we will make several
general comments. First, the absolute configuration of alcohols
formed through this process correlates to the absolute config-
uration of the ligand employed, i.e., R1 group in A; in general,
Cr catalysts derived from (S)-sulfonamides furnish secondary
alcohols with the absolute configuration depicted in Scheme 5.9e,f
Second, in terms of asymmetric induction and coupling rate,
the several hundred sulfonamides are divided into two sub-
groups, one represented by (S)-2 and the other by (S)-3 (Figure
3).15,16 It is a general trend that couplings with the Cr catalysts
derived from the 2 subgroup are faster than those with the Cr
catalysts from the 3 subgroup, but asymmetric inductions with
the former Cr catalysts are lower than those with the latter Cr
catalysts. However, it is important to emphasize that the
matching profiles, i.e., asymmetric induction and coupling rate,
depend on a given substrate; some substrate exhibits a better
matching profile with a Cr catalyst derived from a sulfonamide
in the i-Pr subgroup than a Cr catalyst from a sulfonamide in
the t-Bu subgroup or vice versa. For a given substrate, screenings
with the Cr catalysts prepared from (S)-2, (R)-2, (S)-3, and (R)-3
a Panels A and B depict the proposed activation process of halides in
the Ni/Cr- and Co/Cr-mediated couplings, respectively.
dimeric R-form or the monomeric ꢀ-form.11 For this study, we
use the R-form of NiCl2 ·DMP, prepared and purified by the
procedure given in the Supporting Information. It is also
important to add LiCl, which significantly accelerates the
coupling rate, most likely through formation of the Ni-ate
complex (panel A in Scheme 4).9e The experimental details for
the catalytic asymmetric Ni/Cr-mediated coupling are given in
the Supporting Information.
Catalytic Cr-mediated alkylation and haloallylation are initi-
ated by Co or Fe complexes.9g In our experience, Co(II)
phthalocyanine (CoPc) is the most reliable catalyst for this
purpose (panel B in Scheme 4). Mechanistically, we speculate
an involvement of a free-radical species in this process, because
an alkyl-Co(III) complex did not yield the expected coupling
product.12 Experimentally, it is important to keep the ratio of
Co to Cr catalysts low. The experimental details of the catalytic
asymmetric Co/Cr-mediated coupling are given in the Support-
ing Information.
It is worthwhile to note that the Ni- and Co-promoted
activations are specific to the types of halides, and there is no
crossover in the Ni- and Co-promoted activation. Therefore, it
is possible to activate selectively one nucleophile in the presence
of the other; the cases of C19-C20 and C23-C24 bond
formation illustrate this point. On the other hand, with no
addition of a Ni or Co salt, the active Cr nucleophiles are
generated only from allyl and propargyl halides.
2. Results and Discussions
2.1. Strategy To Search for Effective Ligands. Recognizing
their unique functional-group compatibility, we have been
interested in development of a ligand-search strategy applicable
to a broad range of substrates. With three distinct sites for
structure modification, i.e., blue-, red-, and green-coded areas,
A (Figure 2) provides us with access to structurally diverse chiral
sulfonamides. We have synthesized several hundred sulfon-
amides with the hope that they might exhibit diverse reactivity
profiles in Cr-mediated couplings, thereby allowing us to study
Figure 3. Structure of sulfonamides (S)-2 and (S)-3.
(13) The sulfonamides reported were prepared via a general synthetic route
outlined in the Supporting Information. The details of their synthesis
and reactivity profile will be disclosed in a separate account.
(14) Citations for the isolation, structure elucidation, and synthesis of the
marine natural products halichondrins will be given in our forthcoming
papers.23
(15) For the R1 group, three subgroups of ligands, represented as R1
)
(11) (a) Preston, H. S.; Kennard, C. H. L. J. Chem. Soc. A 1969, 2682. (b)
Butcher, R. J.; Sinn, E. Inorg. Chem. 1977, 16, 2334. (c) Dockum,
B. W.; Reiff, W. M. Inorg. Chem. 1982, 21, 2613.
Me, i-Pr, and t-Bu, were studied. However, 2 represents reasonably
well the reactivity profile for the R1 ) Me subgroup.
(16) The corresponding trimethoxylphenyl sulfonamide exhibits a profile
very similar to that of 3.9i
(12) This was done with a salen-Co(III)-alkyl complex at room temperature.
9
J. AM. CHEM. SOC. VOL. 131, NO. 42, 2009 15389

A R T I C L E S
Guo et al.
Scheme 6. Catalytic Asymmetric Ni/Cr-Mediated Couplings To
Form the C19-C20 Bond19
Figure 4. Representative aromatic and aliphatic R2 groups in A.
of halichondrins (Scheme 6). Screening with the Cr catalysts
prepared from (S)-2, (R)-2, (S)-3, and (R)-3 revealed that (1)
all the Cr catalysts override the chirality present in the
nucleophile and (2) all the Cr catalysts show a very small match/
mismatch effect toward this chiral substrate, cf., dr with (S)-2
over (R)-2 and dr with (S)-3 over (R)-3. The observed results
of asymmetric induction itself might suggest using (S)-3 and
(R)-3 as the leads for sulfonamide optimization. However, we
noticed that the coupling rate with the Cr catalysts derived from
(S)-3 or (R)-3 was slowssignificantly slower than that with the
Cr catalysts derived from (S)-2 or (R)-2sthereby raising a
concern about the overall efficiency of the catalytic process.
For this reason, we decided to use (S)-2 and (R)-2 as the leads
for sulfonamide optimization. Replacing R2 ) Me with R2 )
3,5-Cl2Ph or 3,5-(CF3)2Ph resulted in a significant improvement
in stereoselectivity for both C20 ꢀ- and R-series. With the R2
group being fixed with 3,5-Cl2Ph and 3,5-(CF3)2Ph, a final
refinement was then conducted on the R3 group, yielding several
satisfactory ligands.18 In terms of asymmetric induction, they
were very similar, but the Cr catalyst prepared from (R)-ligand
4 was best in terms of chemical yield. Coupling in the presence
of the Cr catalyst derived from (R)- or (S)-ligand was very clean,
selectively furnishing either the C20 R-alcohol (dr ) 24:1) or
the C20 ꢀ-alcohol (dr ) 1:21) in >90% yields.
2.2.2. C29-C30 Bond Formation. The coupling depicted in
Scheme 7 was originally achieved under stoichiometric non-
asymmetric conditions.20 Since then, this coupling was extended
to a stoichiometric, asymmetric process9c and then to a catalytic
asymmetric process.9h
In this work, we first screened the Cr catalysts derived from
(S)-2, (R)-2, (S)-3, and (R)-3, revealing that (1) all the Cr
catalysts override the chirality present in the substrate, (2) the
Cr catalysts in the (S)-series of sulfonamides exhibit a better
matching profile than the Cr catalysts in the (R)-series (cf., the
dr’s observed for (S)-2 and (S)-3 over those for (R)-2 and (R)-
3, respectively), and (3) the Cr catalysts from (S)-3 and (R)-3
Figure 5. Representative R3 groups in A.
shed light on two aspects of catalyst-substrate matching profile,
i.e., (1) whether the Cr catalysts derived from (S)-2, (R)-2, (S)-
3, and (R)-3 override the effects from the chirality present in a
given substrate and (2) which subgroup of sulfonamides yields
Cr catalysts with better catalyst-substrate matching profiles.
Third, for the R2 group of A, we have studied a large number
of aromatic and aliphatic sulfonamides, revealing the following.
Aromatic sulfonamides are more effective in changing the
overall matching profile of Cr catalysts for a given substrate
than aliphatic sulfonamides. Representative aromatic sulfon-
amides are listed in Figure 4, with the (apparent) order of steric
bulkiness. A given substrate often gives a bell-shaped optimum
against the order of steric bulkiness of these aromatic sulfon-
amides.17 Therefore, for a given substrate, we test first phenyl
and 3,5-dichlorophenyl to obtain a rough catalyst-substrate
matching profile for further ligand optimization. In addition,
R2 ) Me and CH2Ph(Me)2-2,6 are chosen to test the effective-
ness of ligand optimization within the aliphatic sulfonamides.
Fourth, for the R3 group of A, we have made five types of
structure modifications (Figure 5). Among them, Cr catalysts
derived from the sulfonamides with R3 ) OR exhibited an
interesting trend: replacement of the O-Me group with O-i-Pr,
O-c-C6H11, or O-cis,cis-3,5-dimethylcyclohexyl results in a
noticeable change of asymmetric induction without significantly
affecting the coupling rate. As reported previously, the Cr
catalysts derived from sulfonamides with R3 ) H perform well
for 2-haloallylation and allylation.9g,l
With this background, we use the following examples to
address two key questions. First, is the sulfonamide pool
obtained through structure modification at the three sites of A
large enough to match with a broad range of substrates? Second,
is the outlined strategy effective to find a satisfactory ligand
for a given substrate?
2.2. Examples. 2.2.1. C19-C20 Bond Formation. The first
example is Ni/Cr-mediated coupling to form the C19-C20 bond
(18) These include (R)-i-Pr/PhCl2/OMe (dr ) 21:1), (R)-i-Pr/PhCl2/O-i-Pr
(dr ) 24:1), (R)-i-Pr/PhCl2/O-unnat-menthol (dr ) 22:1), (R)-i-Pr/
PhCl2/O-c-Hex (dr ) 24:1), and (R)-i-Pr/Ph(CF3)2/O-c-Hex(Me)2 (dr
) 24:1).
(19) For the coupling conditions, see the Supporting Information.
(20) 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.
(17) Aromatic sulfonamides with electron-donating groups, as well as other
substitution patterns, were tested, but they were not as effective as
those shown in Figure 4.
9
15390 J. AM. CHEM. SOC. VOL. 131, NO. 42, 2009

Ligands for Catalytic Asymmetric Cr-Mediated Coupling
A R T I C L E S
Scheme 7. Catalytic Asymmetric Ni/Cr-Mediated Couplings To
Form the C29-C30 Bond19
Scheme 9. Catalytic Asymmetric Ni/Cr-Mediated Couplings To
Form the C26-C27 Bond19
Scheme 8. Ni/Cr-Mediated Coupling Used in the First-Generation
Synthesis of Halichondrins6b
and validate, that the unique functional-group compatibility is
maintained under the catalytic asymmetric conditions. Ideally,
we wished to study the reactivity profile for all the stereoisomers
possible for both coupling partners. However, considering the
total number of possible stereoisomers, we opted to address this
question by using the C27-C35 aldehyde of E7389 as the
electrophile toward the nucleophiles representing all the possible
stereoisomers at C20-C26, i.e., I-IV.
The standard screening with the Cr catalysts derived from
(S)-2, (R)-2, (S)-3, and (R)-3 revealed that all the Cr catalysts
override the substrate chirality in all of the combinations
(Scheme 9). The observed degrees of asymmetric induction
indicated that the Cr catalysts derived from (S)-2 and (R)-2
exhibit a better catalyst-substrate matching profile than that
from (S)-3 and (R)-3. In addition, the coupling rate was found
to be significantly slower in the (S)-3/(R)-3 series than in the
(S)-2/(R)-2 series. These observations suggested that (S)-2 and
(R)-2 are better leads for sulfonamide optimization.
In the presence of the Cr catalysts derived from (S)-2 and
(R)-2, all the couplings proceeded smoothly to furnish the
expected products in excellent yields and stereoselectivities;
indeed, the arbitrarily chosen g20:1 stereoselectivity was met
for 7 out of 8 possible substrate combinations. Ironically, the
remaining combination was the one representing the stereoiso-
mer required for the synthesis of halichondrins and E7389, i.e.,
substrate II with (S)-2. Therefore, ligand optimization was
conducted for this combination with the usual procedure, thereby
resulting in several satisfactory catalysts.22 Among them, the
Cr catalyst derived from ligand (S)-6 was most effective (dr )
ca. 47:1). In our forthcoming papers,23 we use this catalyst for
the key step in the synthesis of the C14-C38 building block of
halichondrins as well as the C14-C35 building block of E7389.
give a greater asymmetric induction than the Cr catalysts from
(S)-2 and (R)-2, respectively (Scheme 7).
For a selective synthesis of the C30 ꢀ-alcohol, the Cr catalyst
derived from (S)-3 gave an asymmetric induction already well
above the arbitrarily chosen level. In contrast, a Cr catalyst
derived from either (R)-2 or (R)-3 failed to yield C30 R-alcohol
at satisfactory selectivity, thereby requiring ligand optimization.
As noted earlier, the Cr catalysts derived from the R1 ) t-Bu
sulfonamide subgroup generally give a higher asymmetric
induction than those from the R1 ) i-Pr sulfonamide subgroup.
Unlike the case of C19-C20 bond formation, the coupling rate
with the catalyst from (R)-3 was not significantly slower than
that from (R)-2. Therefore, we chose to conduct structure
optimization of the R2 and R3 groups in (R)-3, in the usual
stepwise manner, yielding ligand (R)-5; the catalyst derived from
(R)-5 gave C30 R-alcohol in >90% yield with satisfactory
selectivity (dr ) ca. 1:27). Notably, the catalyst from (S)-5 was
also found to improve the stereoselectivity for C30 ꢀ-alcohol
formation (dr ) ca. 60:1; >90% yield).21
2.2.3. C26-C27 Bond Formation. In the first-generation
synthesis of halichondrins,6b we relied on the stoichiometric non-
asymmetric Ni/Cr-mediated reaction to couple B and C (Scheme
8). A stoichiometric asymmetric version was then developed
with use of the first-generation sulfonamide (S)-1 (dr ) 20:1).9d,e
This Ni/Cr-mediated coupling utilizes the electrophile and
nucleophile bearing seven and four stereogenic centers, respec-
tively, thereby presenting an interesting opportunity to study,
(21) For the C30 ꢀ-alcohol series, several additional sulfonamides, including
(S)-t-Bu/PhCl2/O-c-Hex(Me)2 (dr ) ca. 30:1), (S)-i-Pr/PhCl2/O-i-Pr
(dr ) 21:1), and (S)-i-Pr/3,5-(CO-phenothiazine)2Ph/OMe (dr ) ca.
37:1), gave satisfactory Cr catalysts.
(22) (S)-i-Pr/PhCl2/O-i-Pr also gave the satisfactory result (dr ) ca. 26:1).
9
J. AM. CHEM. SOC. VOL. 131, NO. 42, 2009 15391

A R T I C L E S
Guo et al.
Scheme 10. Catalytic Asymmetric Co/Cr-Mediated Couplings To
Form the C23-C24 Bond19,24
Scheme 11. Catalytic Asymmetric Ni/Cr-Mediated Couplings To
Form the C11-C12 Bond19
2.2.4. C23-C24 Bond Formation. In this coupling (Scheme
10), selective activation of the alkyl iodide bond over the vinyl
iodide bond is achieved with a catalytic amount of Co
phthalocyanine (CoPc).9e Using the Cr catalyst prepared from
the first-generation sulfonamide (S)-1, we demonstrate its
feasibility under catalytic asymmetric conditions (73% yield;
dr ) 5.3:1).
With this background, we screened the Cr catalysts derived
from (S)-2, (R)-2, (S)-3, and (R)-3, thereby revealing that (1)
all the Cr catalysts override the effects of the chirality present
in both nucleophile and electrophile and (2) the Cr catalysts
derived from (S)-2 and (R)-2 exhibit a better substrate matching
profile than those from (S)-3 and (R)-3.24 Using (S)-2 as the
lead, we conducted ligand optimization first for the C24
ꢀ-alcohol series, yielding several satisfactory sulfonamides; the
Cr catalyst derived from ligand (S)-4 (Scheme 6) was most
effective for the asymmetric induction (ca. 33:1), but the Cr
catalyst from ligand (S)-7 was best in asymmetric induction (dr
) 22:1) and yield (82%) combined.25
It was significantly more challenging to find a satisfactory
Cr catalyst for the C24 R-alcohol series. Using (R)-2 as the
lead,26 we performed ligand optimization, resulting in modest
improvement; for example, an (R)-sulfonamide with R1 ) i-Pr,
R2 ) 3,5-Cl2Ph, and R3 ) OMe gave dr ) 1:6.2. Under this
circumstance, we decided to conduct a broader structural
modification, through which ligand (R)-8 emerged as the leading
ligand (dr ) 1:8.1).27 However, its optimization has not yet
yielded a sulfonamide to meet the arbitrarily chosen standard.
2.2.5. C11-C12 Bond Formation. The transformation sum-
marized in Scheme 11 was originally used under stoichiometric
non-asymmetric conditions, to furnish the desired C11 ꢀ-alcohol
in 75% yield with >10:1 stereoselectivity.6b
With use of the Cr catalysts derived from (S)-2, (R)-2, (S)-3,
and (R)-3, we conducted a standard catalyst screening, revealing
that the Cr catalysts derived from (R)-2 or (R)-3 cannot override
the effects from the chirality present in the aldehyde. Given
the high stereoselectivity observed under the stoichiometric non-
asymmetric conditions, we were not surprised by the observed
catalyst-substrate matching profile. Overall, the observed
catalyst-substrate matching profile predicted that ligand opti-
mization might be relatively straightforward for the C11
ꢀ-alcohol series but challenging for the C11 R-alcohol series.
For the C11 ꢀ-alcohol series, considering the stereoselectivity
achieved in the presence of the Cr catalyst from (S)-2 over that
from (S)-3, we first used (S)-2 as the lead for ligand optimization,
yielding several acceptable ligands,28 but none of them per-
formed better than (S)-2 (dr ) ca. 42:1). Intriguingly, optimiza-
tion of (S)-3 resulted in the best ligand, (S)-9: in the presence
of its Cr complex, the coupling gives the desired C11 ꢀ-alcohol
virtually as a single diastereomer quantitatively.
(23) (a) Kim, D.-S.; Dong, C.-G.; Kim, J. T.; Guo, H.; Huang, J.; Tiseni,
P. S.; Kishi, Y. J. Am. Chem. Soc. 2009, 131, in press (ja9058475).
(b) 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, in press (ja9058487) .
(24) For the Co/Cr-mediated coupling, it is important to keep the ratio of
Co and Cr catalysts low. To measure an amount of CoPc with an
acceptable accuracy, we used 30 mol % of Cr catalyst for this
screening. For preparative purposes, 20 mol % catalyst loading was
sufficient to effect this coupling.
(25) These include (S)-i-Pr/Ph(CF3)2/O-c-Hex(Me)2 (dr ) ca. 32:1), (S)-
i-Pr/PhCl2/benzo-1,4-dioxane (dr ) ca. 26:1), (S)-i-Pr/Ph-(CO2Me)2/
Me (dr ) ca. 25:1), (S)-i-Pr/3,5-(CO-phenothiazine)2Ph/Me (dr ) 25:
1), and (S)-i-Pr/2-naphthyl/O-c-Hex(Me)2 (dr ) 22:1).
(26) For this ligand optimization, we used also the first-generation (R)-
sulfonamide (R1 ) i-Pr, R2 ) Me, and R3 ) Me; dr ) 1:1.9) as the
lead, resulting in a modest improvement; for example, dr ) 1:2.5 was
observed for the catalyst from (R)-i-Pr/2-naphthyl/i-Pr.
(27) Several sulfonamides, including (R)-i-Pr/PhOMe-p/naphthyl (dr )
1:8.0) and (R)-i-Pr/2-naphthyl/naphthyl (dr ) 1:7.4), gave ap-
proximately the same level of asymmetric induction as ligand (R)-8.
(28) These include (S)-i-Pr/Ph(CF3)2/OMe (dr ) ca. 29:1), (S)-i-Pr/PhCl2/
OMe (dr ) ca. 28:1), and (R)-i-Pr/PhCl2/O-i-Pr (dr ) 24:1).
9
15392 J. AM. CHEM. SOC. VOL. 131, NO. 42, 2009

Ligands for Catalytic Asymmetric Cr-Mediated Coupling
A R T I C L E S
As predicted, it proved a major challenge to override the
chirality inherent in the aldehyde; despite extensive efforts, we
were unable to find a Cr catalyst for the coupling to yield the
C11 R-alcohol as the major product.29 With these results, we
turned our attention to the catalyst-substrate matching profile
for the same aldehyde but bearing different protecting groups.
Namely, we became interested in testing whether the Cr catalysts
derived from (S)-2, (R)-2, (S)-3, and (R)-3 could override the
chirality inherent in the aldehyde with a minor structural
modification. Thus, aldehydes bearing the acetonide, bis-
methoxymethyls (MOMs), and bis-tert-butyldimethylsilyls (TBSs)
were studied, without any sign of success until the bisbenzoate
aldehyde was tested. To our delight, this substrate exhibited
the long-awaited catalyst-substrate matching profile: the C11
R-alcohol was the major product formed in the presence of the
Cr catalyst from (R)-2 and (R)-3, whereas the C11 ꢀ-alcohol
was the major product in the presence of the Cr catalyst from
(S)-2 and (S)-3 (Scheme 11).
Encouraged by this result, we conducted standard ligand
optimization. For the C11 ꢀ-alcohol series, we first used (S)-2,
rather than (S)-3, as the lead. In parallel with the cyclohexylidene
series, however, the best ligand, (S)-10 (dr ) ca. 56:1), emerged
again from the ligand optimization of (S)-3. For the C11
R-alcohol series, considering both the observed asymmetric
induction and coupling rate, we used (R)-2 as the lead for ligand
optimization. Through this effort, (R)-sulfonamide 11 was found
as a satisfactory ligand (dr ) 1:23).30
sulfonamides. The Cr catalysts derived from these sulfonamides
exhibit a broad range of catalyst-substrate matching profiles,
suggesting the possibility that this class of sulfonamide-based
catalysts could match with diverse substrates. A strategy has
been presented to search for a satisfactory sulfonamide for a
given substrate. In order to demonstrate the generality and
effectiVeness of this approach, we have used five diverse C-C
bond-forming cases from the halichondrin synthesis. For each
of the cases studied, we have deliberately searched for two
ligands, which induce the formation of (R)- and (S)-alcohols,
respectively, at the arbitrarily chosen efficiency level of “g80%
yield with g20:1 stereoselectivity in the presence of e20 mol
% of a Cr catalyst”. For 9 out of the 10 cases studied, we have
been able to find a satisfactory catalyst. Even for the remaining
case, we have found a Cr catalyst inducing stereoselectivity up
to 8:1. Encouraged by the successful case studies on structurally
diverse substrates, we have developed an optimistic prospect
for this approach. In our forthcoming papers,23 we will report
that these catalytic asymmetric Cr-mediated couplings have
major impacts on the syntheses of halichondrins and E7389;
the newly developed methods allow us to form the key
carbon-carbon bonds in a highly stereoselective manner,
improve the overall synthetic efficiency, reduce the total number
of synthetic steps, and also incorporate a high degree of
flexibility in synthesis.
Acknowledgment. We are grateful to the National Institutes
of Health (CA 22215) and to the Eisai Research Institute for
generous financial support. D.U. gratefully acknowledges a post-
doctoral fellowship from Japan Society for the Promotion of Science
(JSPS).
3. Conclusion and Outlook
In summary, catalytic asymmetric Cr-mediated couplings are
effected by catalysts prepared from CrCl2 and chiral sulfon-
amides represented by A. With three distinct sites for structure
modification, A provides access to structurally diverse chiral
Supporting Information Available: Experimental details,
including syntheses of the substrates for the coupling reactions
and chiral sulfonamides, estimation of the diastereomer ratios
by 1H NMR, and synthesis/characterization of the chiral
sulfonamides. This material is available free of charge via the
Internet at http://pubs.acs.org.
(29) Interestingly, trans-1-iodohex-1-ene exhibits a similar catalyst-
substrate profile.
(30) For this optimization, we studied sulfonamides in the R1 ) Me
subgroup. Some of them, e.g., (R)-Me/PhCl2/O-c-Hex(Me)2 (dr )
1:15), exhibited a promising matching profile, but their overall behavior
was found to be similar to that of R1 ) i-Pr subgroup sulfonamides.
JA905843E
9
J. AM. CHEM. SOC. VOL. 131, NO. 42, 2009 15393