Stereocontrol of all-carbon quaternary centers through enantioselective desymmetrization of meso primary diols by organocatalyzed acyl transfer

Article abstract of DOI:10.1002/anie.201308268

The symmetry breaking of meso primary diols bearing a tetrahydropyran ring was employed, using catalytic asymmetric acyl transfer, to control all-carbon quaternary stereocenters. The planar chiral Fu DMAP catalyst was used in this reaction to reach a high degree of enantioselectivity (up to 97:3 e.r.) through a synergic effect combining a desymmetrization step and a kinetic resolution. Moreover, a beneficial effect was exhibited by C₆F₆ solvent, yielding the first example of an organocatalyzed asymmetric acyl transfer. The desymmetrized monoesters were then used to obtain, after a straightforward ring opening sequence, complex polyketide building blocks bearing all-carbon quaternary stereocenters. Smashing the mirror: The symmetry breaking of meso primary diols was employed to control all-carbon quaternary stereocenters using catalytic asymmetric acyl transfer. The planar chiral Fu DMAP catalyst was used to reach a high degree of enantioselectivity (up to 97:3 e.r.) through a synergic effect, combining a desymmetrization step and a kinetic resolution. Copyright

Full text of DOI:10.1002/anie.201308268

.
Angewandte
Communications
DOI: 10.1002/anie.201308268
Organocatalysis
Stereocontrol of All-Carbon Quaternary Centers through
Enantioselective Desymmetrization of Meso Primary Diols by
Organocatalyzed Acyl Transfer**
Christꢀle Roux, Mathieu Candy, Jean-Marc Pons, Olivier Chuzel,* and Cyril Bressy*
Abstract: The symmetry breaking of meso primary diols
bearing a tetrahydropyran ring was employed, using catalytic
asymmetric acyl transfer, to control all-carbon quaternary
stereocenters. The planar chiral Fu DMAP catalyst was used in
this reaction to reach a high degree of enantioselectivity (up to
97:3 e.r.) through a synergic effect combining a desymmetriza-
tion step and a kinetic resolution. Moreover, a beneficial effect
was exhibited by C6F6 solvent, yielding the first example of an
organocatalyzed asymmetric acyl transfer. The desymmetrized
monoesters were then used to obtain, after a straightforward
ring opening sequence, complex polyketide building blocks
bearing all-carbon quaternary stereocenters.
E
nantioselective access to all-carbon quaternary centers has
been clearly identified as a challenging area of research,[1]
particularly for acyclic systems. Among the catalytic processes
used to reach this goal, the classic approach consists of the
creation of the quaternary center concomitant with its
stereocontrol. An alternative to this approach is the enantio-
selective desymmetrization of molecules bearing the pre-
existing all-carbon quaternary center(s). Indeed, this power-
ful strategy separates the task of quaternary center generation
from the task of its enantiocontrol. In this field, two types of
achiral molecules could then be envisioned: prochiral pre-
cursors or meso compounds[2] (Scheme 1). Whereas prochiral
molecules hold a unique quaternary center in their plane of
symmetry, meso compounds possess several stereogenic
centers out of their symmetrical element. As a consequence,
the level of complexity reached through desymmetrization of
meso molecules is much higher than that from prochiral
molecules. Surprisingly, although several studies have been
published on the catalytic enantioselective desymmetrization
of prochiral precursors to control all-carbon quaternary
stereocenters,[3–5] none, to the best of our knowledge, has
Scheme 1. Comparison of desymmetrizing approaches using prochiral
or meso compounds in the stereocontrol of all-carbon quaternary
centers.
been reported starting from meso compounds bearing qua-
ternary centers out of the plan of symmetry.[6]
To fill this gap, we targeted the desymmetrization of meso
diols bearing revealable stereocenters. Among the enantio-
selective catalytic methods of desymmetrization of meso diols,
asymmetric acyl transfer takes a central place with a long
hegemony of biocatalysis.[7] This transformation furnished
several valuable building blocks for total synthesis.[8] We
recently reported the desymmetrization of meso primary diols
with a high level of enantioselection using an enzyme,
Rhizomucor meihei lipase.[9] However, this catalyst was not
effective with hindered meso compounds bearing all-carbon
quaternary stereocenters, such as 2a (Scheme 2). To circum-
vent this absence of reactivity, we envisioned a chemical
alternative requiring a nucleophilic organocatalysis approach
as a valuable answer to the major drawbacks encountered
with enzymes, namely the lack of tolerance towards substrates
and the access to only one out of two enantiomers. Never-
theless, few studies have been conducted on prochiral primary
diols bearing a quaternary center[10] (Scheme 2, Eq. (1)). As
for the organocatalyzed desymmetrization of meso primary
diols by acyl transfer, a unique case was reported by Oriyama
et al.[11] (Scheme 2, Eq. (2)). We assume that the difficulty in
reaching a good level of enantioselection[12] mainly explains
this lack of studies on meso primary diols. As a consequence,
these substrates can be considered as challenging ones for
non-enzymatic desymmetrization by acyl transfer.[13] In this
paper, we aim to report the first efforts towards the
stereocontrol of all-carbon quaternary centers by desymmet-
[*] C. Roux, Dr. M. Candy, Prof. Dr. J.-M. Pons, Dr. O. Chuzel,
Dr. C. Bressy
Aix Marseille Universitꢀ, Centrale Marseille, CNRS, iSm2 UMR 7313
13397 Marseille (France)
E-mail: cyril.bressy@univ-amu.fr
Homepage: http://ism2.univ-amu.fr/
[**] We warmly thank Dr. N. Vanthuyne, M. Jean (HPLC), and Dr. M.
Giorgi (X-ray diffraction). Aix-Marseille University, CNRS, the
Conseil Gꢀnꢀral des Bouches du Rhꢁne, COST ORCA Action 0905
(Organocatalysis), and ANR (Agence Nationale de la Recherche)
(Orcademe project ANR-10-JCJC-0710) are also gratefully acknowl-
edged for funding.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201308268.
766
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 766 –770

Angewandte
Chemie
Table 1: Optimization of the non-enzymatic desymmetrization of 2a.
Entry Cat.
Solvent
Conv. [%][a] 3a [%][b] 4a [%] e.r.[c]
1[d]
2[d]
3
4
5
1a
1b
1a
1a
1a
1a
1a
1a
t-AmOH
t-AmOH
t-AmOH
CHCl3
90
53
96
98
7
60 (57) 30
41 (23) 12
72 (68) 24
92:8
50:50
94:6
77:23
95:5
91:9
90:10
90:10
–
93 (88)
7 (5)
57 (53)
65 (58)
83 (75)
32
5
0
3
0
7
[e]
C6F6
6
7
C6F6/t-AmOH[f] 60
[f]
C6F6/CHCl3
C6F6/CHCl3
65
90
50
0
8[g]
9
[f]
none t-AmOH
none C6F6/CHCl3
none C6F6/CHCl3
18
0
Scheme 2. Desymmetrization with prochiral or meso primary diols.
[f]
[f]
10
0
–
–
11[g]
7
7
0
[a] Determined by 1H NMR analysis of the crude reaction mixture.
rization of meso primary diols using nucleophilic organo-
catalysis (Scheme 2, Eq. (3)).
[b] Yields of isolated products are given in parentheses. [c] Determined
by HPLC analysis on a chiral stationary phase (Chiralpak AS-3),
n-hexane/isopropanol 80:20, 1 mLminÀ1, DAD and polarimeter. [d] Et3N
was employed as base. [e] Performed at 48C. [f] 1:1 ratio. [g] Ac2O
(3 equiv) and iPrNEt2 (3 equiv) were employed.
Among the variety of organocatalysts able to desymme-
trize meso diols,[14] chiral dialkylaminopyridines derivatives[15]
are very popular for enantioselective acyl transfer, owing to
their efficiency.[16] The commercially available and versatile
ferrocenyl Fu catalysts[17] exhibited a single excellent result
with a meso secondary diol,[18] but no study was reported on
meso primary diols with these catalysts. Moreover, this
catalyst is known to be highly selective in the case of
enantioselective acyl transfer on secondary alcohols bearing
b substituents with p substituents, such as aromatic rings or
multiple bonds in the vicinity of the hydroxy group.[17]
Based on these considerations, we started the optimiza-
tion[19] with catalysts 1a and 1b, using 2a as model substrate.
The first attempts, using catalyst 1a under the conditions
described by Fu and using 2-methyl-2-butanol (tert-amyl
alcohol, t-AmOH) as solvent gave encouraging results.
Indeed, monoacetate 3a was obtained in 57% yield with
a very good enantioselection (e.r. = 92:8; Table 1, entry 1)
accompanied by a significant amount of diacetate 4a. Catalyst
1b led to a racemic mixture (entry 2), thus showing the strong
influence of the cyclopentadienyl (Cp) ring substituents.
Replacing Et3N as the base by iPrNEt2 slightly improved the
yield and the enantioselectivity up to 94:6 e.r. (entry 3). A
solvent screen highlighted the high selectivity obtained for
monoacetate 3a in CHCl3 (88%), but also it revealed a drop
in enantioselectivity for this solvent (entry 4). Hexafluoro-
benzene gave monoacetate 3a with excellent enantioselec-
tivity (e.r. = 95:5), but poor conversion (entry 5).
amounts of monoacetate 3a and diacetate 4a after 16 h in t-
AmOH, but no acylation reaction occured in the C6F6/CHCl3
solvent system, except when increasing the amount of
acylating reagent and base (entry 11).
To explain the high level of enantioselection during this
non-enzymatic symmetry breaking, we proposed that the
present desymmetrization belongs to a particular class of
reactions characterized by the synergic combination of a true
desymmetrization reaction and
a
kinetic resolution
(Scheme 3),[20] with the nucleophilic catalyst being involved
in both processes. The formation of the initially undesired
diacetate 4a could be beneficial, if it arises from the
consumption of the minor enantiomer of 3a. Typically, the
case where k1 > k2 and k4 > k3 represents a favorable scenario
to obtain monoacetate 3a with high enantioselectivity. To
experimentally underscore this synergic connection, a follow-
ing of the enantioselectivity of the product 3a versus time was
conducted using 1a as catalyst (Scheme 3). Under condi-
tions A (using t-AmOH as solvent) the enantiomeric ratio
increased over the course of the reactions from 81:19 after
30 min to 93:7 after 46 h. The same tendency was observed
under conditions B (with a solvent system combining C6F6
and CHCl3) starting from an e.r. of 75:25 after 30 min to 90:10
ratio after 48 h. These observations validated our hypothesis
in both solvent systems (Scheme 3). It is important to notice
that the enantioselectivity obtained under conditions A
(Table 1, entry 3) results from a desymmetrization reaction/
kinetic resolution process, whereas the enantioselectivity in
C6F6 (e.r. = 95:5) comes from a true desymmetrization
In an attempt to combine the excellent conversions in t-
AmOH or CHCl3 with the high enantioselectivity in C6F6, we
found that a mixture of these solvents in a 1:1 ratio was
effective (entries 6 and 7), and avoided the production of
diester 4a. Increasing the equivalents of base and acylating
reagent allowed to improve the yield and retain the level of
enantioselectivity. Various acylating reagents were also tested
in this reaction, but they did not improved both the reactivity
and the enantioselectivity.[19]
Interestingly, control experiments without catalyst
(entries 9 and 10) exhibited the formation of significant
Angew. Chem. Int. Ed. 2014, 53, 766 –770
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
767

.
Angewandte
Communications
Scheme 3. Experimental evidence for the combination of a desymmetri-
zation reaction coupled to a kinetic resolution by monitoring of the
^
enantioselectivity over time. Conditions A ( ), conditions B (&).
Scheme 4. Scope and limitation of the non-enzymatic desymmetriza-
tion. Unless otherwise specified, reactions performed with 2 on
a 0.2 mmol scale. [a] Performed on a 600 mg scale. [c] Performed on
a 300 mg scale. Ac=acetyl, Bn=benzyl, Bz=benzoyl.
reaction (Table 1, entry 5). Owing to the aromatic structure of
the catalyst 1a, p–p complexation(s) could be postulated
between C6F6 and the chiral DMAP derivative 1a (DMAP =
dimethylaminopyridine).[21] The role of this solvent remains
unclear, but it seems that the observed enhancement in
enantioselectivity would result from a hindered supramolec-
ular entity more discriminating than the catalyst without C6F6.
This hypothesis could also explain the lower conversion.
Additionally, no evolution of the enantiomeric ratio of
monoacetate 3a was observed under conditions A without
acetic anhydride, which demonstrates the irreversibility of the
acylation reaction (thus avoiding an eventual racemization
process during the transformation).
inversion of the prostereogenic center in the 4 position of the
THP ring of diol 2g led to a significant loss of enantioselec-
tivity for monoester 3g, as compared to its diastereomer 3a,
which was formed from 2a. The lower reactivity and
selectivity for 2h could be explained by a dramatic increase
in steric hindrance around the quaternary centers bearing
here phenyl and benzyl groups. Interestingly, the desymmet-
rization conditions also gave good results with substrate 2i,
which was successfully employed in reactions with Rhizomu-
cor meihei lipase.[9] Lower yields were obtained under both
conditions owing to the higher reactivity of diol 2i with less
steric hindrance in the 3 and 5 positions of the THP ring,
which gave larger amounts of diester. Additionally, the
desymmetrizations of ent-3a and 3d were performed on
larger scales (respectively on 600 mg and 300 mg) under
conditions A with reproducible yields and enantioselectivi-
ties.
With these building blocks in hand, we decided to convert
the two diastereomeric enantioenriched monoacetates 3a and
3d, which contain the THP systems in acyclic stereotetrads
(Scheme 5). The three-step sequence (iodination of the
hydroxy group, Zn-mediated reductive ring opening, and
mild hydrolysis) provided diastereomeric diols 5a and 5d
from 3a and 3d, respectively, without erosion in enantiose-
lectivity. This global transformation offers privileged access to
stereotetrads using an enantioselective organocatalytic
method. A high level of complexity was obtained with
building blocks bearing three differentiable hydroxy func-
tions, and four contiguous stereogenic centers, including two
The absolute configuration of the desymmetrized mono-
acetate ent-3a, obtained using (R)-ferrocenyl-DMAP ent-1a
as catalyst, was determined by X-ray diffraction of the
corresponding crystalline tosylate[23] (see the Supporting
Information).
With these optimized conditions, several meso primary
diols were submitted to both desymmetrization conditions A
and B (Scheme 4). Generally, slightly better enantioselectiv-
ity was observed under conditions A than under conditions B,
but higher yields were obtained with the latter. The sub-
stitution pattern on the a-face of the two quaternary centers
has a significant effect on the stereoselectivity. Indeed, when
this face was blocked by methyl groups in the 3 and 5 positions
of the tetrahydropyran (THP) ring (3a, 3b, and 3c) very good
selectivities (up to 94:6 e.r.) were observed. The level of
enantioselectivity was improved by exchanging the methyl
groups for ethyl groups, as in 3d, 3e, and 3 f, reaching up to
97:3 e.r. This transformation also tolerates several hydroxy
protecting groups on the prostereogenic 4 position, such as
methylether (3a), benzylether (3b), or benzoate (3i). The
768
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 766 –770

Angewandte
Chemie
Iosub, A. R. Haerl, J. Leung, M. Tang, K. Vembaiyan, M. Parvez,
T. G. Back, J. Org. Chem. 2010, 75, 1612.
[4] For selected references, see: a) E. S. Sattely, S. J. Meek, S. J.
Malcolmson, R. R. Schrock, A. H. Hoveyda, J. Am. Chem. Soc.
2009, 131, 943; b) D. H. T. Phan, K. G. M. Kou, V. M. Dong, J.
Am. Chem. Soc. 2010, 132, 16354; c) A. K. Mourad, J. Leutzow,
C. Czekelius, Angew. Chem. 2012, 124, 11311; Angew. Chem. Int.
Ed. 2012, 51, 11149.
[5] For selected references, see: a) Z. G. Hajos, D. R. Parrish, J. Org.
Chem. 1974, 39, 1615; b) C. A. Leverett, V. C. Purohit, D. Romo,
Angew. Chem. 2010, 122, 9669; Angew. Chem. Int. Ed. 2010, 49,
9479; c) F. Liu, X. Bugaut, M. Schedler, R. Frçhlich, F. Glorius,
Angew. Chem. 2011, 123, 12834; Angew. Chem. Int. Ed. 2011, 50,
12626; d) M. Wilking, C. Mꢀck-Lichtenfeld, C. G. Daniliuc, U.
Hennecke, J. Am. Chem. Soc. 2013, 135, 8133.
Scheme 5. Access to complex polyketides through ring-opening reac-
tions of desymmetrized THPs 3a and 3d.
all-carbon quaternary ones. In both cases, the quaternary
stereocenters bear methyl and ethyl groups, which are quite
difficult to discriminate through other approaches.[23]
[6] We have found a couple of examples of catalytic enantioselective
desymmetrization of meso compounds bearing an all-carbon
quaternary center in the plan of symmetry; see: a) M. H. Wu,
K. B. Hansen, E. N. Jacobsen, Angew. Chem. 1999, 111, 2167;
Angew. Chem. Int. Ed. 1999, 38, 2012; b) A. C. Spivey, S. J.
Woodhead, M. Weston, B. I. Andrews, Angew. Chem. 2001, 113,
791; Angew. Chem. Int. Ed. 2001, 40, 769.
In conclusion, in this preliminary communication we have
described the first cases of catalytic enantioselective non-
enzymatic desymmetrizations of meso compounds bearing
all-carbon quaternary centers. This was achieved through
a challenging transformation, an asymmetric acyl transfer to
primary diols, with an enantioselective ratio of up to 97:3. This
is the first example of the use of a chiral Fu DMAP to
desymmetrize meso primary diols. From a mechanistic point
of view, we showed that this symmetry breaking was in fact
the result of the combination of a desymmetrization reaction
and a kinetic resolution. Moreover, it was shown that building
blocks containing up to five stereogenic centers can be easily
converted into complex polyketide fragments, which could be
useful in the synthesis of analogues of natural products.
During this study, important solvent effects were noted with
C6F6 which, we assume, interacts with the catalyst to produce
a supramolecular entity that is more efficient, in terms of
enantioselectivity, than the catalyst alone. This hypothesis will
be examined in further detail in due course.
[7] E. Garcꢁa-Urdiales, I. Alfonso, V. Gotor, Chem. Rev. 2011, 111,
PR110.
[8] Recent applications in total synthesis by our group: for
Crocacin C, see: M. Candy, G. Audran, H. Bienaymꢂ, C.
Bressy, J.-M. Pons, J. Org. Chem. 2010, 75, 1354; for a synthetic
approach to Callystatin A, see: M. Candy, L. Tomas, S. Parat, V.
Heran, H. Bienaymꢂ, J.-M. Pons, C. Bressy, Chem. Eur. J. 2012,
18, 14267.
[9] M. Candy, G. Audran, H. Bienaymꢂ, C. Bressy, J.-M. Pons, Org.
Lett. 2009, 11, 4950.
[10] J. Y. Lee, Y. S. You, S. H. Kang, J. Am. Chem. Soc. 2011, 133,
1772.
[11] D. Terakado, T. Oriyama, Org. Synth. 2006, 83, 70.
[12] Two factors seem to be the origins of the lower enantioselectiv-
ity: first, the higher nucleophilicity of primary alcohols, com-
pared to secondary ones, which is partially due to the higher
accessibility of the reactive center; second, lower asymmetric
induction with meso primary diols could be expected than with
secondary meso diols because of the distance between the
preexisting stereogenic centers and the reactive ones.
Received: September 20, 2013
Revised: October 25, 2013
Published online: December 6, 2013
Keywords: desymmetrization · meso compounds ·
.
organocatalysis · quaternary centers · tetrahydropyranes
[13] For recent reviews see: a) C. E. Mꢀller, P. R. Schreiner, Angew.
Chem. 2011, 123, 6136; Angew. Chem. Int. Ed. 2011, 50, 6012;
b) A. Enrꢁquez-Garcꢁa, E. P. Kꢀndig, Chem. Soc. Rev. 2012, 41,
7803; c) M. D. Dꢁaz-de-Villegas, J. A. Galvez, R. Badorrey, M. P.
Lopez-Ram-de-Viu, Chem. Eur. J. 2012, 18, 13920.
[14] a) T. Oriyama, K. Imai, T. Hosoya, T. Sano, Tetrahedron Lett.
1998, 39, 397; b) E. Vedejs, O. Daugulis, N. Tuttle, J. Org. Chem.
2004, 69, 1389; c) V. B. Birman, H. Jiang, X. Li, Org. Lett. 2007, 9,
3237; d) P. E. Kꢀndig, A. Enriquez Garcia, T. Lomberget, P.
Perez Garcia, P. Romanens, Chem. Commun. 2008, 3519.
[15] For a review, see: R. P. Wurz, Chem. Rev. 2007, 107, 5570.
[16] For mechanistic aspects, see: a) A. C. Spivey, S. Arseniyadis,
Angew. Chem. 2004, 116, 5552; Angew. Chem. Int. Ed. 2004, 43,
5436; b) A. C. Spivey, S. Arseniyadis, Top. Curr. Chem. 2010, 291,
233; c) N. De Tycke, F. Couty, O. R. P. David, Chem. Eur. J. 2011,
17, 12852.
[1] For reviews, see: a) A. Y. Hong, B. M. Stoltz, Eur. J. Org. Chem.
2013, 2745; b) J. Prakash Das, I. Marek, Chem. Commun. 2011,
47, 4593; c) C. Hawner, A. Alexakis, Chem. Commun. 2010, 46,
7295; d) M. Bella, T. Gasperi, Synthesis 2009, 1583.
[2] Confusion between meso compounds and prochiral molecules
exists and is persistent in current literature. In the “Basic
Terminology of Stereochemistry”, IUPAC defines a meso com-
pound as “the achiral member(s) of a set of diastereomers which
also includes one or more chiral members” (G. P. Moss, Pure
Appl. Chem. 1996, 68, 2193). This definition is not applicable to
prochiral compounds such as the 2-methylpropane-1,3-diol, for
example.
[3] For selected references, see: a) T. Honda, T. Koizumi, Y.
Komatsuzaki, R. Yamashita, K. Kanai, H. Nagase, Tetrahedron:
Asymmetry 1999, 10, 2703; b) S. Akai, T. Tsujino, E. Akiyama,
K. Tanimoto, T. Naka, Y. Kita, J. Org. Chem. 2004, 69, 2478; c) V.
[17] a) G. C. Fu, Acc. Chem. Res. 2000, 33, 412; b) G. C. Fu, Acc.
Chem. Res. 2004, 37, 542; c) S. Yunmi Lee, J. M. Murphy, A.
Angew. Chem. Int. Ed. 2014, 53, 766 –770
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
769

.
Angewandte
Communications
Ukai, G. C. Fu, J. Am. Chem. Soc. 2012, 134, 15149, and
references therein.
Lattanzi, C. De Fusco, A. Russo, A. Poater, L. Cavallo, Chem.
Commun. 2012, 48, 1650.
[18] J. C. Ruble, J. Tweddell, G. C. Fu, J. Org. Chem. 1998, 63, 2794.
[19] The optimization process is fully described in the Supporting
Information.
[22] CCDC 942839 contains the supplementary crystallographic data
available at the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[20] S. L. Schreiber, M. T. Goulet, G. Schulte, J. Am. Chem. Soc.
1987, 109, 4718.
[21] A similar amplification of enantioselectivity was recently
observed in the organocatalytic Michael addition; see: A.
[23] For elegant diastereoselective approaches, see: a) Y. Minko, M.
Pasco, L. Lercher, M. Botoshansky, I. Marek, Nature 2012, 490,
522; b) P.-O. Delaye, D. Didier, I. Marek, Angew. Chem. 2013,
125, 5441; Angew. Chem. Int. Ed. 2013, 52, 5333.
770
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 766 –770