Catalytic Asymmetric Iterative/Domino Aldehyde Cross-Aldol Reactions for the Rapid and Flexible Synthesis of 1,3-Polyols

Article abstract of DOI:10.1021/jacs.5b11192

We report here catalytic asymmetric iterative and domino cross-aldol reactions between aldehydes, endowed with a high level of robustness, flexibility, and generality. A Cu(I)-DTBM-SEGPHOS complex catalyzes an asymmetric cross-aldol reaction between acceptor aldehydes and boron enolates derived from donor aldehydes, which are generated through Ir-catalyzed isomerization of allyloxyboronates. The unit process can be repeated using the aldol products in turn as acceptor substrates for the subsequent asymmetric aldol reaction. The donor aldehydes and stereoselectivity can be flexibly switched in a stepwise manner for the double-aldol reaction. Furthermore, asymmetric triple- and quadruple-aldol reactions are possible in one-pot using the appropriate amounts of donors and amine additives, rapidly elongating the carbon skeleton with controlling up to eight stereocenters. The method should be useful for straightforward synthesis of enantiomerically and diastereomerically enriched 1,3-polyols.

Full text of DOI:10.1021/jacs.5b11192

Communication
pubs.acs.org/JACS
Catalytic Asymmetric Iterative/Domino Aldehyde Cross-Aldol
Reactions for the Rapid and Flexible Synthesis of 1,3-Polyols
Luqing Lin,†,‡,∥ Kumiko Yamamoto,†,∥ Harunobu Mitsunuma,† Yamato Kanzaki,†
Shigeki Matsunaga,*,†,§,⊥ and Motomu Kanai*,†,‡
†Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
‡ERATO, Japan Science and Technology Agency, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
§ACT-C, Japan Science and Technology Agency, Tokyo 113-0033, Japan
S
* Supporting Information
ABSTRACT: We report here catalytic asymmetric
iterative and domino cross-aldol reactions between
aldehydes, endowed with a high level of robustness,
flexibility, and generality. A Cu(I)-DTBM-SEGPHOS
complex catalyzes an asymmetric cross-aldol reaction
between acceptor aldehydes and boron enolates derived
from donor aldehydes, which are generated through Ir-
catalyzed isomerization of allyloxyboronates. The unit
process can be repeated using the aldol products in turn as
acceptor substrates for the subsequent asymmetric aldol
reaction. The donor aldehydes and stereoselectivity can be
flexibly switched in a stepwise manner for the double-aldol
reaction. Furthermore, asymmetric triple- and quadruple-
aldol reactions are possible in one-pot using the
appropriate amounts of donors and amine additives,
rapidly elongating the carbon skeleton with controlling
up to eight stereocenters. The method should be useful for
straightforward synthesis of enantiomerically and diaster-
eomerically enriched 1,3-polyols.
Figure 1. (A) Ideal approach to 1,3-polyols through iterative/domino
aldehyde cross-aldol reactions (upper row) and its potential difficulties
(bottom row). (B) Our strategy for the catalytic asymmetric iterative/
domino aldehyde cross-aldol reactions.
1,3-Polyols are ubiquitous and privileged structural motifs in
polyketide natural products and drugs.1 Polyketides are
estimated to be five times more likely to possess drug activity
than other natural product families,2a and polyketide-derived
Although the idea of iterative asymmetric aldehyde cross-aldol
pharmaceuticals comprise 20% of the top-selling small molecule
drugs.2b For the synthesis of the 1,3-polyol motifs,3 a catalytic
asymmetric aldol reaction would be a powerful unit process.
Despite marked progress, however, the development of catalytic
asymmetric aldol reactions has focused mainly on the use of
ketones and carboxylic acid derivatives as donors.4 Thus, the
installation of a second 1,3-diol unit through iterative use of aldol
reactions requires nonproductive steps; i.e., protection of the β-
hydroxy group, followed by reduction and/or oxidation of the
terminal carbonyl group to the corresponding aldehyde. An ideal
unit reaction for 1,3-polyol synthesis is the catalytic asymmetric
cross-aldol reaction between two different aldehydes,5 directly
providing an aldehyde moiety for subsequent iterative aldol
reactions (Figure 1A, upper row). Here we disclose catalytic
asymmetric iterative/domino aldehyde cross-aldol reactions for
the straightforward synthesis of enantiomerically and diaster-
eomerically enriched 1,3-polyols.
reactions is conceptually simple, it is extremely challenging for
the following reasons (Figure 1A, bottom row). First, two
aldehydes must be differentiated as either a donor or an acceptor.
Otherwise, undesired homo- and hetero-aldol products will be
randomly produced. A few organocatalyzed asymmetric cross-
aldol reactions of aldehydes6 were reported previously; however,
the donor/acceptor-control was based on the inherent steric
and/or electronic characteristics of the substrates. Thus,
Mukaiyama-type aldol reactions7 using preactivated donor
aldehyde enolates are a reliable way.8 Second, the intermediate
β-hydroxy aldehydes are generally unstable. Acidic, basic, and
high temperature conditions can cause undesired side reactions,
such as a retro-aldol reaction, dehydration, hemiacetal formation,
and polymerization. Third, the number of possible stereoisomers
increases exponentially as the iteration proceeds. High fidelity in
Received: October 28, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b11192
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
both enantio- and diastereoselectivity for a unit aldol reaction is
essential to avoid the formation of multiple stereoisomers.
Finally, the products of more than double-aldol reactions exist as
cyclized hemiacetal forms lacking a reactive aldehyde functional
group. To prevent the formation of the unreactive hemiacetals,
hydroxy groups of the intermediates need to be protected in
more than double-aldol reactions. Precedent examples for
iterative aldehyde cross-aldol reactions are limited to the
proline-catalyzed asymmetric aldol reaction, followed by a
diastereoselective Mukaiyama aldol reaction,9a,b the proline-
catalyzed asymmetric double-aldol reaction with switching of the
catalyst chirality to avoid microscopic reversibility,9c the
engineered enzyme-catalyzed asymmetric double-aldol reac-
tion,9d a catalytic diastereoselective (racemic) one-pot triple-
aldol reaction,5,10a−d and an enzyme-catalyzed asymmetric triple-
aldol reaction.10e There is much room left for improvement in
these pioneering examples especially with regard to the substrate
generality and practicality.
We envisioned that a copper(I) alkoxide-catalyzed asymmetric
aldol reaction between aldehydes and aldehyde-derived boron
enolates would fulfill the requirements discussed above (Figure
1B). Carreira’s group11 and our group12 independently reported
that the reactivity of silicon enolates is markedly enhanced by
copper(I) catalysis through the formation of copper enolates
from silicon enolates and that catalytic asymmetric aldol
reactions are possible under mild conditions by introducing
chiral ligands to copper catalysts. Based on the similar
characteristics of silicon and boron atoms, we hypothesized
that a catalytic asymmetric aldol reaction between aldehydes and
aldehyde-derived boron enolates is possible by copper(I)
catalysis through the generation of highly reactive aldehyde-
derived chiral copper(I) enolates. After the aldol reaction, copper
aldolate intermediates are trapped by the boron atom, generating
O-protected aldol products and thus preventing the formation of
unreactive hemiacetal in more than double-aldol reactions.
Aldehyde-derived boron enolates13 should be generated without
formation of any wastes through transition metal-catalyzed C
C double bond isomerization of allyloxyboronates, based on our
previous findings.14
We began our study by optimizing the single-aldol reaction
between 3-phenylpropanal (1a) and boron enolate 3a (R2, R3 =
H), which was generated from 2a in a different vessel prior to the
aldol reaction. Both the isomerization conditions from 2a to 3a
and copper sources as well as chiral ligands significantly affected
the result of the aldol reaction.15 We identified the optimized
conditions as isomerization of 2a to 3a using 0.5 mol % [Ir(cod)
(PPh2Me)2]PF6 complex preactivated under H2 atmosphere as a
catalyst16 in acetone at 0 °C, followed by an asymmetric aldol
reaction with 1a using 5 mol % mesitylcopper (MesCu)-(R)-
DTBM-SEGPHOS catalyst in the presence of 1 equiv of
isopropanol in THF at −60 °C, which afforded product 4a′ after
reduction in 80% yield, 97:3 dr, and 95% ee.
Under the optimized conditions, a variety of aliphatic, aryl, and
heteroaryl aldehydes all afforded the cross-aldol products in
moderate to excellent yields with high diastereo- and
enantioselectivity (Figure 2A). The cross-aldol products were
obtained from the combination of sterically less-hindered
propanal as an acceptor and sterically more demanding
aldehydes as donors (4e and 4f). The enamine catalysis17 did
not produce aldol products from this donor/acceptor combina-
tion. As donors, not only methyl, but also ethyl, butyl, and
dimethyl groups could be introduced at the α-position of the
product (4b−4f, 4m).
Figure 2. (A) Scope of the catalytic asymmetric cross-aldol reaction
between aldehydes. Yield and selectivity were determined after
reduction due to the instability of β-hydroxy aldehydes under analytical
conditions. See the SI for detailed reaction conditions. (B) Plausible
catalytic cycle.
A plausible catalytic cycle is depicted in Figure 2B. By mixing
MesCu, (R)-DTBM-SEGPHOS, and isopropanol, chiral CuOiPr
was generated with the extrusion of mesitylene. Transmetalation
between the copper alkoxide and 3 afforded chiral copper enolate
5, which reacted with 1 to form copper aldolate 6. Facile
protonation of 6 with isopropanol was key to promoting the
catalytic cycle because the copper aldolate could otherwise
irreversibly consume aldehyde 1 via nucleophilic attack to
produce undesired cyclic hemiacetal 7.
We then turned our attention to the double-aldol reaction.
Although the copper catalysis realized the single-aldol reaction,
the reactivity was not sufficient for the double-aldol reaction. To
B
DOI: 10.1021/jacs.5b11192
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
increase the reactivity, the diol moiety of boron enolates was
changed to pinacol, anticipating that the increased electron
density would facilitate the transmetalation step.15 The double-
aldol reaction smoothly proceeded in one-pot from aldehyde 1
and boron enolate 8 in the presence of 1 equiv of 4-
methoxyphenol, instead of isopropanol,15 to produce the
corresponding cyclized hemiacetals 9 in the reaction mixture.
After reduction with LiBH4, the desired triol 10 was obtained in
good yield and stereoselectivity (Table 1, entries 1−4). Stepwise
introduction of different donors at the first and second steps was
also possible using monoaldol products 4a and 4b, which were
synthesized by the method in Figure 2, as acceptor aldehydes
(Table 1, entries 5 and 6). Furthermore, switching the chirality of
the catalyst in the first and second aldol reactions provided
stereodivergent access to triols (Table 1, entries 7−10). Both the
enantioselectivity and diastereoselectivity were predominantly
controlled by the catalyst, not by the substrates. Two distinct
enolates were introduced in a stepwise manner with switching
the stereoselectivity as well (Table 1, entries 9 and 10).
Table 1. Copper-Catalyzed Asymmetric Double-Aldol
Reaction
a
We extended this approach to more than double-aldol
reactions. An additional difficulty with this reaction stage
comprised the facile formation of unreactive hemiacetals 9 at
the double-aldol stage, if hydroxy groups of the aldol products
were not protected. We hypothesized that hemiacetal formation
would be prevented by trapping the copper aldolate intermediate
6 as non-nucleophilic boronate 12 through a reaction with boron
enolate 8 in the catalyst turnover step (Figure 3A), which would
require aprotic conditions. Based on this hypothesis, we
examined the reactions between 1a and 8a (4.5 equiv) under
various conditions without protic additives; however, the desired
triple aldol product was produced at most in only trace amounts
(<3%). Instead, the diboronate of hemiacetal 13 was obtained as
a major product (∼50%). This unexpected result is likely due to
intramolecular boron/copper migration in copper aldolate 14
generated after double-aldol reaction, producing 15, which
instantly cyclizes to a hemiacetal.
To prevent the formation of hemiacetal 13 and/or facilitate
the desired reaction pathways, we investigated additive effects to
find that the addition of 2 equiv of triethylamine was effective,
generating triple-aldol product 17a in 71% yield, 90:10 dr (17a/
other isomers), and >99% ee after reduction (Figure 3B). The
conditions were also applicable to α,β-unsaturated aldehyde 1k
and aromatic aldehyde 1l. Furthermore, the use of N,N,N′,N′-
tetramethyl-1,4-butanediamine instead of triethylamine realized
the first catalytic asymmetric one-pot quadruple-aldol reaction,
generating one of 256 possible isomers in high diastereo- and
enantioselectivity (Figure 3C). Elucidation of the origins of
amine additive effects awaits future detailed mechanistic studies.
For the moment, we confirmed that addition of triethylamine
increased the concentration of reactive aldehyde form 16a (R1 =
Ph(CH2)2, R2 = Me) of the double-aldol intermediate, based on
NMR studies of the reaction mixture.15
a
In conclusion, we developed copper(I)-catalyzed asymmetric
iterative/domino cross-aldol reactions between aldehydes and
aldehyde-derived boron enolates, which are generated from
allyloxyboronates via iridium-catalyzed double bond isomer-
ization. This method leads to the rapid production of a range of
enantiomerically and diastereomerically enriched 1,3-polyols.
Two novel findings in this study are (1) flexible and stepwise
switching of the donors and the stereoselectivity in the first and
second steps of the iterative double-aldol reaction; and (2)
robust and one-pot double, triple, and quadruple catalytic
asymmetric domino aldol reactions using the appropriate
Yield refers to the combined yield of all diastereomers. Diastereo-
meric ratio (dr) refers to the ratio of 10/11/other isomers. Two
equivalents of triethylamine was added. See the SI for experimental
details.
b
amounts of donors and amine additives. These findings
demonstrate that the Cu(I)-catalyzed asymmetric iterative
cross-aldol reactions of aldehydes could serve as an ideal method
for rapid 1,3-polyol synthesis.
C
DOI: 10.1021/jacs.5b11192
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Corporation for generous donation of DTBM-SEGPHOS. L.L.
thanks the Uehara Memorial Foundation and ERATO for the
fellowship. K.Y. and H.M. thank JSPS for the fellowship.
REFERENCES
(1) Rychnovsky, S. D. Chem. Rev. 1995, 95, 2021.
(2) (a) Rohr, J. Angew. Chem., Int. Ed. 2000, 39, 2847. (b) Weissman, K.
■
J.; Leadlay, P. F. Nat. Rev. Microbiol. 2005, 3, 925.
(3) (a) Bode, S. E.; Wolberg, M.; Muller, M. Synthesis 2006, 2006, 557.
̈
(b) Li, J.; Menche, D. Synthesis 2009, 2009, 2293. (c) Dechert-Schmitt,
A.-M. R.; Schmitt, D. C.; Gao, X.; Itoh, T.; Krische, M. J. Nat. Prod. Rep.
2014, 31, 504. (d) Harrison, T. J.; Ho, S.; Leighton, J. L. J. Am. Chem. Soc.
2011, 133, 7308.
(4) (a) Schetter, B.; Mahrwald, R. Angew. Chem., Int. Ed. 2006, 45,
7506. (b) Modern Aldol Reactions; Mahrwald, R., Ed.; Wiley-VCH:
Weinheim, Germany, 2004. (c) Groger, H.; Vogl, E. M.; Shibasaki, M.
̈
Chem. - Eur. J. 1998, 4, 1137. (d) Trost, B. M.; Brindle, C. S. Chem. Soc.
Rev. 2010, 39, 1600.
(5) Brady, P. B.; Yamamoto, H. In Modern Methods in Stereoselective
Aldol Reactions; Mahrwald, R., Ed.; Wiley-VCH: Weinheim, Germany,
2013; p 269.
(6) (a) Northrup, A. B.; MacMillan, D. W. C. J. Am. Chem. Soc. 2002,
124, 6798. (b) Mangion, I. K.; Northrup, A. B.; MacMillan, D. W. C.
Angew. Chem., Int. Ed. 2004, 43, 6722. (c) Kano, T.; Yamaguchi, Y.;
Tanaka, Y.; Maruoka, K. Angew. Chem., Int. Ed. 2007, 46, 1738.
(d) Hayashi, Y.; Itoh, T.; Aratake, S.; Ishikawa, H. Angew. Chem., Int. Ed.
2008, 47, 2082. (e) Markert, M.; Scheffler, U.; Mahrwald, R. J. Am.
Chem. Soc. 2009, 131, 16642. (f) Kano, T.; Sugimoto, H.; Maruoka, K. J.
Am. Chem. Soc. 2011, 133, 18130.
(7) Mukaiyama, T.; Narasaka, K.; Banno, K. Chem. Lett. 1973, 1011.
(8) (a) Mukaiyama, T.; Banno, K.; Narasaka, K. J. Am. Chem. Soc. 1974,
96, 7503. (b) Denmark, S. E.; Ghosh, S. K. Angew. Chem., Int. Ed. 2001,
40, 4759. (c) Boxer, M. B.; Yamamoto, H. J. Am. Chem. Soc. 2006, 128,
48.
(9) (a) Northrup, A. B.; Mangion, I. K.; Hettche, F.; MacMillan, D. W.
C. Angew. Chem., Int. Ed. 2004, 43, 2152. (b) Northrup, A. B.;
MacMillan, D. W. C. Science 2004, 305, 1752. (c) Casas, J.; Engqvist, M.;
Figure 3. Copper-catalyzed asymmetric triple- and quadruple-aldol
reactions. (A) Working hypothesis for the reaction progress to the
triple-aldol reaction. (B,C) Catalytic asymmetric triple- and quadruple-
domino aldol reactions. *Diastereoselectivity refers to the ratio of the
major diastereomer to minor diastereomers.
Ibrahem, I.; Kaynak, B.; Cor
1343. (d) Szekrenyi, A.; Garrabou, X.; Parella, T.; Joglar, J.; Bujons, J.;
Clapes, P. Nat. Chem. 2015, 7, 724.
́
dova, A. Angew. Chem., Int. Ed. 2005, 44,
́
(10) (a) Albert, B. J.; Yamamoto, H. Angew. Chem., Int. Ed. 2010, 49,
2747. (b) Albert, B. J.; Yamaoka, Y.; Yamamoto, H. Angew. Chem., Int.
Ed. 2011, 50, 2610. (c) Brady, P. B.; Yamamoto, H. Angew. Chem., Int.
Ed. 2012, 51, 1942. (d) Brady, P. B.; Albert, B. J.; Akakura, M.;
Yamamoto, H. Chem. Sci. 2013, 4, 3223. (e) Gijsen, H. J. M.; Wong, C.-
H. J. Am. Chem. Soc. 1995, 117, 7585.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.5b11192.
■
S
(11) (a) Kruger, J.; Carreira, E. M. J. Am. Chem. Soc. 1998, 120, 837.
̈
Experimental details and characterization data (PDF)
X-ray crystallographic data for 10a (CIF)
X-ray crystallographic data for ent-17a (CIF)
(b) Pagenkopf, B. L.; Kruger, J.; Stojanovic, A.; Carreira, E. M. Angew.
̈
Chem., Int. Ed. 1998, 37, 3124.
(12) Oisaki, K.; Zhao, D.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc.
2006, 128, 7164.
(13) (a) Wulff, G.; Hansen, A. Angew. Chem., Int. Ed. Engl. 1986, 25,
AUTHOR INFORMATION
Corresponding Authors
*smatsuna@pharm.hokudai.ac.jp
*kanai@mol.f.u-tokyo.ac.jp
■
560. (b) Hoffmann, R. W.; Ditrich, K.; Froch, S. Liebigs Ann. Chem.
̈
1987, 1987, 977.
(14) (a) Lin, L.; Yamamoto, K.; Matsunaga, S.; Kanai, M. Angew.
Chem., Int. Ed. 2012, 51, 10275. (b) Lin, L.; Yamamoto, K.; Matsunaga,
S.; Kanai, M. Chem. - Asian J. 2013, 8, 2974.
(15) For full details, see Supporting Information.
(16) Ohmura, T.; Yamamoto, Y.; Miyaura, N. Organometallics 1999,
18, 413.
Present Address
⊥Faculty of Pharmaceutical Sciences, Hokkaido University, Kita-
ku, Sapporo 060-0812, Japan.
Author Contributions
(17) Mukherjee, S.; Yang, J. W.; Hoffmann, S.; List, B. Chem. Rev. 2007,
107, 5471.
∥These authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported in part by ERATO (to M.K.) and ACT-
C (to S.M.) from JST. We thank Takasago International
■
D
DOI: 10.1021/jacs.5b11192
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX