[2,3]-Wittig Rearrangement as a Formal Asymmetric Alkylation of α-Branched Ketones

Article abstract of DOI:10.1021/acs.orglett.9b01495

The enantioselective [2,3]-Wittig rearrangement of cinnamyloxycyclopentanone derivatives was performed in the presence of a Cinchona-based primary amine. The described method provides synthetically valuable α-hydroxy ketones with quaternary stereogenic centers in excellent enantiomeric purities. Relying on the X-ray crystal structure of the product and the DFT calculations, we propose that the rearrangement is promoted by an intramolecular hydrogen bond between the substrate and the catalyst.

Full text of DOI:10.1021/acs.orglett.9b01495

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
[2,3]-Wittig Rearrangement as a Formal Asymmetric Alkylation of
α‑Branched Ketones
†
̌
̈
̃
Mariliis Kimm, Maksim Oseka, Sandra Kaabel, Andrus Metsala, Ivar Jarving, and Tonis Kanger*
Department of Chemistry and Biotechnology, Tallinn University of Technology, Akadeemia tee 15, 12618 Tallinn, Estonia
S
* Supporting Information
ABSTRACT: The enantioselective [2,3]-Wittig rearrangement of
cinnamyloxycyclopentanone derivatives was performed in the presence
of a Cinchona-based primary amine. The described method provides
synthetically valuable α-hydroxy ketones with quaternary stereogenic
centers in excellent enantiomeric purities. Relying on the X-ray crystal
structure of the product and the DFT calculations, we propose that the
rearrangement is promoted by an intramolecular hydrogen bond between
the substrate and the catalyst.
he α-alkylation of ketones represents a fundamental
method for the creation of a new C−C bond, giving
homoallyl alcohols that makes it possible to generate a new
C−C bond and to insert stereocomplexity into a structure.9
The Wittig rearrangement has been applied as a key step for
the synthesis of many natural products.10 On the other hand,
formed rearranged products are substituted α-hydroxy ketones,
also known as acyloins. Acyloins are compounds of great
synthetic importance, and can be found in natural products
and biologically active compounds. The development of
efficient catalytic methods to construct this functionality
remains challenging.11
T
access to the formation of a new stereogenic center.1 Although
the asymmetric α-alkylation of ketones has been the subject of
intense discussion for decades, a general method for direct
catalytic asymmetric α-alkylation has not yet been discovered.2
Asymmetric alkylation of α-branched ketones that leads to the
formation of a quaternary chiral center is an even more
challenging task due to the steric hindrance and inductive
effect of the alkyl substituent. The number of such trans-
formations is limited. Transition metal catalysis is the most
common tool for this purpose. However, the scope of these
transformations is usually limited to α-allylations lacking
examples of others alkyl electrophiles and substrates needing
prefunctionalization in many cases.3 Organocatalysis is another
strategy for the asymmetric alkylation of α-branched ketones.
The reported methods were mediated by phase-transfer
catalysts,4 bifunctional organocatalysts,5 chiral phosphoric
acids,6 and primary amines.7 Over the past decade, the area
of photocatalysis has rapidly developed opening access to
reaction pathways, which were previously impossible to reach
under classic thermochemical conditions, but still photo-
catalytic methods were only applied to α-branched β-
ketocarbonyl compounds.8
To date, only a limited number of asymmetric catalytic
examples of [2,3]-Wittig rearrangement have been re-
ported.12,13 In 2006, Gaunt demonstrated the first example
of asymmetric organocatalytic rearrangement on allyloxy
ketones.14 Nine years later, Denmark and co-workers obtained
moderate enantioselectivities (up to 54% ee), when they
applied phase-transfer catalysis to allyloxy oxindole and
tetralone derivatives.15 Simultaneously, our group reported a
squaramide catalyzed rearrangement of oxindole derivatives
with excellent enantiomeric control (dr up to 2.7:1; ee up to
94%/97%).16 As a continuation of this research, in 2017 our
group described [2,3]-Wittig rearrangement of allyloxy
malonates catalyzed by a chiral calcium complex (ee up to
85%) and highly basic cyclopropenimine (ee up to 59%).17 In
the same year, Jacobsen and co-workers reported an effective
synergistic ion-binding catalysis concept for the rearrangement
of allyloxy-1,3-dicarbonyl compounds (ee up to 92%).18 In
2018, Feng’s research group described the first asymmetric
nickel-catalytic rearrangement of propargyloxyoxindoles, af-
fording products in high enantiomeric purities (up to 99%
Since the direct asymmetric alkylation of α-branched
ketones is an unmet target, every new method, even the
indirect alkylation method, has great importance and expands
the existing alkylation methodology. The [2,3]-Wittig rear-
rangement of α-branched ketones can be seen as a formal
alkylation method (Scheme 1). The [2,3]-Wittig rearrange-
ment is a transformation of allylic or propargylic ethers to
ee).19 Very recently, Sebesta et al. reported an aminocatalytic
̌
rearrangement of allyoxyketones with moderate enantioselec-
tivity (ee 60%).20 Additionally, Aleman recently demonstrated
Scheme 1. Wittig Rearrangement of α-Branched Ketones
́
a highly diastereoselective (dr up to 98:2) rearrangement of
Received: April 29, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b01495
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
allylic and propargylic sulfoxides induced by a stoichiometric
amount of strong bases.21
conversion and selectivity were obtained when para-nitro-
benzoic acid (p-NBA) was used as an acidic additive.
Unfortunately, the diastereoisomeric ratio of the products
could not be determined by the 1H NMR analysis of the crude
mixture due to the incomplete starting material conversion. A
significant increase in the conversion was observed when the
rearrangement reaction was catalyzed by Cinchona derived
amine B and amino alcohol C with high enantioselectivity in
the case of catalyst B (Table 1, entries 5 and 6). A brief solvent
screening showed that the rearrangement is the most efficient
in deuterochloroform (selected examples Table 1, entries 5, 7,
and 8; see the Supporting Information for other examples). We
managed to push the rearrangement to the full conversion by
running the reaction at higher temperature with only a slight
decrease in the selectivity (Table 1, entry 9). In the reaction
catalyzed by amine D, the lower conversion and reversed
enantioselectivity were observed (Table 1, entry 11). Although
simple chiral α-methylbenzylamine is a very attractive catalyst
to use, it provided the product with poor selectivity (Table 1,
entry 12). We tried to further improve the reaction by varying
acidic additives and their amount, but the best results both in
terms of reactivity and selectivity were obtained with 40 mol %
of p-NBA (Table 1, entries 5 and 13−18). Finally, in order to
simplify the reaction setup, we prepared a salt of p-NBA and
Cinchona derived amine B (ratio 2:1). The same conversion
and selectivity was obtained when the preformed catalyst salt
was used for the [2,3]-rearrangement. These conditions were
chosen to investigate the scope of the reaction (Table 1, entry
19).
Inspired by Gaunt’s pioneering work in the field, we
examined the [2,3]-Wittig rearrangement of cyclic allyloxy
ketones as a formal alkylation of α-branched ketones catalyzed
by chiral primary amines. The superior ability of primary
amines over secondary ones to effectively catalyze the
functionalization of ketones and α-branched aldehydes was
caused by steric factors that inhibited the condensation of the
secondary amine to bulky carbonyl compounds and the
enamine formation, which is the key intermediate of the amino
catalyzed Wittig rearrangement.22 Usually, an acidic cocatalyst
is used to accelerate the condensation of the primary amine
with the carbonyl moiety. With this knowledge in hand, we
chose several primary amines for our model reaction of a [2,3]-
Wittig rearrangement of cinnamyloxycyclopentanone 1a under
acidic conditions (Table 1).
The preliminary screening for the optimal reaction
conditions revealed that a Cinchona alkaloid derived amine A
catalyzes the [2,3]-Wittig rearrangement under acidic con-
ditions providing products in high enantioselectivities (Table
1, entries 1−4). The most promising results in terms of
a
Table 1. Screening of the Catalysts and Optimization
With optimal conditions in hand, the influence of
substituents at the aromatic core of cinnamyloxycyclo-
pentanone 1 was investigated (Scheme 2). A [2,3]-Wittig
rearrangement of model substrate 1a proceeded smoothly at
1.0 mmol scale and provided α-hydroxyketone 2a in high yield
and excellent enantioselectivity of the major diastereoisomer.
The chlorine atom at the para- and meta-positions of
cinnamyloxycyclopentanone 1 did not affect the reaction
noticeably, and the rearranged products 2b and 2c were
obtained in good yields and excellent enantiomeric purities.
However, the rearrangement of ortho-chlorosubstituted cyclo-
pentanone 1d was very slow and full conversion was not
achieved even under a longer reaction time. When substrate 1e
with an electron-donating methoxy group was used, the
enantioselectivity of the major diastereoisomer remained
excellent and the diastereoisomeric ratio slightly decreased
compared to the unsubstituted substrate 1a. The strong
electron-withdrawing group lowered the reactivity of the
substrate 1f and the selectivity. Next, we studied the influence
of different aromatic groups on the [2,3]-rearrangement of
cyclopentanones. The substrate bearing a naphthyl group
underwent a smooth rearrangement and provided α-hydrox-
yketone 2g in good yield with the best diastereoselectivity in
the series. A heteroaromatic 2-thienyl substituent in cyclo-
pentanone 1h caused a decrease in reactivity and selectivity.
Transformations of substituted cyclohexanone and cyclo-
butanone derivates showed that, in the case of cinnamyl-
oxycyclohexanone 1i, the yield and enantioselectivity for the
major diastereoisomer decreased drastically, while only the
decomposition of cyclobutanone 1m was observed by NMR
spectroscopy. Finally, similarly to our previous results, no
rearrangement of cis-cinnamyloxy- 1k and allyloxy cyclo-
pentanone 1l was observed.16 An additional methyl substituent
temp
conv
b
c
entry cat.
acid
solvent
(°C)
(%)
ee (%)
1
2
3
4
5
6
7
8
9
A
A
A
A
B
C
B
B
B
B
D
E
B
B
B
B
B
B
TFA
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
toluene
DCE
CDCl3
CHCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
50
50
50
50
50
50
50
50
60
60
60
60
60
60
60
60
60
60
60
33
45
45
35
78
81
59
62
99
99
76
96
85
60
92
14
71
86
99
97/70
93/26
98/52
86/13
96/35
10/17
93/26
94/19
94/19
94/14
PhCOOH
p-NBA
EtCOOH
p-NBA
d
p-NBA
p-NBA
p-NBA
p-NBA
p-NBA
p-NBA
p-NBA
PhCOOH
TFA
f
f
10
11
12
13
14
15
16
17
18
−90/−7
63/16
85/11
95/56
56/5
p-NO2-PhOH
−
nd
d
p-NBA
p-NBA
90/25
94/16
94/17
e
f g
19 ,
B and 2·p-NBA salt
a
Reaction conditions: 0.1 mmol scale, 20 mol % of cat., 40 mol % of
b
acid, solvent (0.5 mL). Conversion determined by 1H NMR analysis
c
d
of the crude mixture. Determined by chiral HPLC analysis. 20 mol
% of acid. 60 mol % of acid. dr 3.5:1, determined by H NMR
analysis of the crude mixture. Isolated yield 65%.
e
f
1
g
B
DOI: 10.1021/acs.orglett.9b01495
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Scope of the Reaction
Figure 1. X-ray structure of [2,3]-rearranged product 2a (major
diastereoisomer).
Scheme 3. (a) Proposed Catalytic Cycle of the [2,3]-Wittig
Rearrangement; (b) DFT Calculated Transition State of the
[2,3]-Wittig Rearrangement
a
Reaction conditions: 0.2 mmol scale, 20 mol % of cat. B, CDCl3 (1
mL), 60 °C, 24 h; ee determined by chiral HPLC and dr by 1H NMR
analysis of the isolated product. NMR yield (in parentheses)
determined by H NMR analysis of the crude mixture. 1.0 mmol
b
1
c
reaction scale, reaction time 30 h. Reaction was stopped after 72 h.
d
Reaction was stopped after 48 h.
at the double bond of substrate 1j also suppressed the reaction
completely.
The relative and absolute stereochemistry of the major
diastereoisomer of the [2,3]-Wittig rearrangement product 2a
was unambiguously assigned by single crystal X-ray diffraction
(Figure 1). The major enantiomer is a syn-isomer with an
(R,R)-configuration of stereogenic centers. The configurations
of other compounds in the series were assigned by analogy.
A proposed catalytic cycle of a [2,3]-Wittig rearrangement is
depicted in Scheme 3a. The iminium ion is formed upon the
condensation of aminocatalyst B with cinnamyloxycyclo-
pentanone 1a under acidic conditions followed by the
formation of enamine.22 The unfavorable equilibrium between
imine and enamine is directed to the more stable imine, and
thus the concentration of enamine is reduced. However, the
enamine undergoes a concerted sigmatropic rearrangement
through a five-membered transition state that leads to the
cleavage of the allylic C−O bond and to the formation of a
new C−C bond. Finally, α-hydroxyketone 2a and catalyst B
are released upon hydrolysis.
Based on the absolute configuration of α-hydroxyketone 2a,
we have designed and proven by DFT calculation the
transition state of the rearrangement of cinnamyloxycyclo-
pentanone 1a leading to the formation of a major isomer
(Scheme 3b). According to the calculations the five-membered
transition state of enamine was stabilized by a strong hydrogen
bond (1.564 Å, 5.567 kcal/mol, depicted in blue color)
between the ether oxygen atom of the substrate and the
protonated nitrogen atom of the quinuclidine moiety of the
catalyst. This observation is in excellent agreement with our
previous experimental results which demonstrated the
C
DOI: 10.1021/acs.orglett.9b01495
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
importance of hydrogen-bond activation for promoting [2,3]-
Wittig rearrangement.17
In order to further investigate the rate and the mechanism of
the [2,3]-Wittig rearrangement of cinnamyloxycyclopentanone
1a, we performed a kinetic study by taking crude samples for
NMR analysis over time (Figure 2). A rapid consumption of
and moderate yields and diastereoselectivities. We have proved
by DFT calculations the importance of hydrogen bonding in
the transition state stabilization. α-Hydroxyketones were
further successfully derivatized to functionalized bicyclic
products without a loss of stereoselectivity to demonstrate
the utility of the [2,3]-rearranged products as building blocks.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.9b01495.
Synthesis of starting compounds, optimization of the
procedures, copies of 1H and 13C spectra, HPLC
chromatograms (PDF)
Accession Codes
CCDC 1908273 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
Figure 2. Kinetic study of a [2,3]-Wittig rearrangement of
cinnamyloxycyclopentanone 1a.
the starting material was observed during the first hour, which
corresponded to imine formation by the condensation
reaction; nevertheless, only traces of the products were
detected. This observation was probably caused by the
imine-enamine equilibrium and indicates that the unfavorable
formation of the enamine intermediate was the rate-
determining step. Although there was no starting material
left after stirring the reaction for 23 h, the NMR yield of the
products stayed at 75% even at the extended reaction time.
The lowered reaction yield can be explained by side reactions,
and the partial decomposition of the starting material, at higher
temperature.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: tonis.kanger@taltech.ee.
ORCID
̌
Maksim Oseka: 0000-0002-7095-4497
nis Kanger: 0000-0001-5339-9682
To
̃
Present Address
†Department of Chemistry, McGill University, 801 Sherbrooke
Street West, Montreal, Quebec H3A 0B8, Canada
Notes
Finally, to demonstrate the utility of the α-hydroxyketones
we decided to derivatize [2,3]-rearrangement product 2a
(Scheme 4).23 The reaction sequence started with the
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the Estonian Ministry of Education and
Research (Grant Nos. IUT 19-32, IUT19-9, PUT 692, and
PUT 1468) and the Centre of Excellence in Molecular Cell
Engineering (2014-2020.4.01.15-0013) for financial support.
Authors thank Dr. Jasper Adamson (NICP, Tallinn, Estonia)
for his assistance in NMR studies.
Scheme 4. Derivatization of [2,3]-Rearrangement Product
2a
REFERENCES
■
(1) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry Part B:
Reactions and Synthesis; Springer: New York, 2007; pp 1−62.
(2) Cano, R.; Zakarian, A.; McGlacken, G. P. Angew. Chem., Int. Ed.
2017, 56, 9278.
esterification of the tertiary hydroxyl group followed by an
intramolecular Horner−Wadsworth−Emmons reaction.
Formed bicyclic product 3a was isolated in high yield, and
no decrease in enantiomeric excess was observed. Next,
catalytic hydrogenation proceeded with high stereoselectively
and provided saturated bicyclic lactone 4a with three adjacent
stereogenic centers as a single diastereoisomer. The relative
configuration of 4a was determined by 2D NOESY NMR
experiment.
In conclusion, we have developed a formal asymmetric
organocatalytic alkylation method of cyclic α-branched ketones
based on [2,3]-Wittig rearrangement. The rearrangement was
catalyzed by a Cinchona-derived amine, and cyclic α-
hydroxyketones were obtained in excellent enantioselectivities
(3) For recent examples, see: (a) Jiang, X.; Chen, W.; Hartwig, J. F.
Angew. Chem., Int. Ed. 2016, 55, 5819. (b) Pupo, G.; Properzi, R.; List,
B. Angew. Chem., Int. Ed. 2016, 55, 6099. (c) Alexy, E. J.; Zhang, H.;
Stoltz, B. M. J. Am. Chem. Soc. 2018, 140, 10109. (d) Liu, W.; Ali, S.
Z.; Ammann, S. E.; White, M. C. J. Am. Chem. Soc. 2018, 140, 10658.
(4) (a) Nibbs, A. E.; Baize, A.-L.; Herter, R. M.; Scheidt, K. A. Org.
Lett. 2009, 11, 4010. (b) Kano, T.; Hayashi, Y.; Maruoka, K. J. Am.
Chem. Soc. 2013, 135, 7134. (c) Teng, B.; Chen, W.; Dong, S.; Kee,
C. W.; Gandamana, D. A.; Zong, L.; Tan, C.-H. J. Am. Chem. Soc.
2016, 138, 9935.
(5) (a) Dong, X.-Q.; Teng, H.-L.; Tong, M.-C.; Huang, H.; Tao, H.-
Y.; Wang, C.-J. Chem. Commun. 2010, 46, 6840. (b) Kang, J. Y.;
Carter, R. G. Org. Lett. 2012, 14, 3178. (c) Kang, J. Y.; Johnston, R.
D
DOI: 10.1021/acs.orglett.9b01495
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
C.; Snyder, K. M.; Cheong, P. H.-Y.; Carter, R. G. J. Org. Chem. 2016,
81, 3629. (d) Urruzuno, I.; Mugica, O.; Oiarbide, M.; Palomo, C.
Angew. Chem., Int. Ed. 2017, 56, 2059.
(6) (a) Felker, I.; Pupo, G.; Kraft, P.; List, B. Angew. Chem., Int. Ed.
2015, 54, 1960. (b) Yang, X.; Toste, F. D. Chem. Sci. 2016, 7, 2653.
(7) (a) Horinouchi, R.; Kamei, K.; Watanabe, R.; Hieda, N.;
Tatsumi, N.; Nakano, K.; Ichikawa, Y.; Kotsuki, H. Eur. J. Org. Chem.
2015, 2015, 4457. (b) Liu, S.; Tong, M.; Yu, Y.; Xie, H.; Li, H.; Wang,
W. Chem. Commun. 2015, 51, 11221. (c) Zhu, Y.; Zhang, W.-Z.;
Zhang, L.; Luo, S. Chem. - Eur. J. 2017, 23, 1253.
́
(8) (a) Wozniak, Ł.; Murphy, J. J.; Melchiorre, P. J. Am. Chem. Soc.
2015, 137, 5678. (b) Zhu, Y.; Zhang, L.; Luo, S. J. Am. Chem. Soc.
2014, 136, 14642. (c) Wang, D.; Zhang, L.; Luo, S. Org. Lett. 2017,
19, 4924. (d) Zhang, W.; Zhu, Y.; Zhang, L.; Luo, S. Chin. J. Chem.
2018, 36, 716.
(9) For recent reviews, see: (a) Wolfe, J. P. In Comprehensive Organic
Synthesis II, Carbon−Carbon σ-Bond Formation; Marek, I., Ed.;
Elsevier: Amsterdam, 2014; Vol. 3, pp 1038−1072. (b) West, T.
H.; Spoehrle, S. S. M.; Kasten, K.; Taylor, J. E.; Smith, A. D. ACS
Catal. 2015, 5, 7446.
(10) Rycek, L.; Hudlicky, T. Angew. Chem., Int. Ed. 2017, 56, 6022.
(11) Scheidt, K. A.; O’Bryan, E. A. in Comprehensive Organic
Synthesis II, Acyloin Coupling Reactions; Marek, I., Ed.; Elsevier:
Amsterdam, 2014; Vol. 3, pp 621−653.
(12) For a recent review, see: Wu, H.; Wang, Q.; Zhu, J. Eur. J. Org.
Chem. 2019, 2019, 1964.
(13) For related ammonium ylide rearrangements, see: (a) West, T.
H.; Daniels, D. S. B.; Slawin, A. M. Z.; Smith, A. D. J. Am. Chem. Soc.
2014, 136 (12), 4476. (b) West, T. H.; Walden, D. M.; Taylor, J. E.;
Brueckner, A. C.; Johnston, R. C.; Cheong, P. H.-Y.; Lloyd-Jones, G.
C.; Smith, A. D. J. Am. Chem. Soc. 2017, 139, 4366.
(14) McNally, A.; Evans, B.; Gaunt, M. J. Angew. Chem., Int. Ed.
2006, 45, 2116.
(15) Denmark, S. E.; Cullen, L. R. J. Org. Chem. 2015, 80, 11818.
̌
̈
(16) Oseka, M.; Kimm, M.; Kaabel, S.; Jarving, I.; Rissanen, K.;
Kanger, T. Org. Lett. 2016, 18, 1358.
̌
̈
(17) Oseka, M.; Kimm, M.; Jarving, I.; Lippur, K.; Kanger, T. J. Org.
Chem. 2017, 82, 2889.
(18) Kennedy, C. R.; Guidera, J. A.; Jacobsen, E. N. ACS Cent. Sci.
2016, 2, 416.
(19) Xu, X.; Zhang, J.; Dong, S.; Lin, L.; Lin, X.; Liu, X.; Feng, X.
Angew. Chem., Int. Ed. 2018, 57, 8734.
̌
̌
́
́
(20) Penaska, T.; Meyer Mojzes, M.; Filo, J.; Jurdakova, H.;
̌
̌
́
Meciarova, M.; Sebesta, R. Eur. J. Org. Chem. 2019, 2019, 605.
́
́
(21) Rodríguez, R. I.; Ramírez, E.; Fernandez-Salas, J. A.; Sanchez-
́
́
Obregon, R.; Yuste, F.; Aleman, J. Org. Lett. 2018, 20, 8047.
(22) Melchiorre, P. Angew. Chem., Int. Ed. 2012, 51, 9748.
(23) Ma, N.; Yao, Y.; Zhao, B.-X.; Wang, Y.; Ye, W.-C.; Jiang, S.
Chem. Commun. 2014, 50, 9284.
E
DOI: 10.1021/acs.orglett.9b01495
Org. Lett. XXXX, XXX, XXX−XXX