Synthesis of trans -2,6-disubstituted cyclohexanones through allylic substitution

Article abstract of DOI:10.1021/ol403472y

trans-2,6-Disubstituted cyclohexanones were synthesized with high regio- and stereoselectivity by allylic substitution followed by ozonolysis. Both alkyl and aryl groups were successfully installed to the cyclohexane ring. The stereochemistry of the S

Full text of DOI:10.1021/ol403472y

Letter
pubs.acs.org/OrgLett
Synthesis of trans-2,6-Disubstituted Cyclohexanones through Allylic
Substitution
Yuichi Kobayashi,* Chao Feng, Atsushi Ikoma, Narihito Ogawa, and Takayuki Hirotsu
Department of Bioengineering, Tokyo Institute of Technology, Box B52, Nagatsuta-cho 4259, Midori-ku, Yokohama 226-8501, Japan
S
* Supporting Information
ABSTRACT: trans-2,6-Disubstituted cyclohexanones were synthe-
sized with high regio- and stereoselectivity by allylic substitution
followed by ozonolysis. Both alkyl and aryl groups were successfully
installed to the cyclohexane ring. The stereochemistry of the SN2′
products was determined to be controlled by the pre-existing chirality on the ring. The present method is highlighted by the
synthesis of enantiomerically enriched cyclohexanones.
2,6-Disubstituted cyclohexanones are potentially valuable
intermediates for the synthesis of biologically active com-
pounds,1 and several methods for the synthesis of this class of
compounds have been developed.2,3 Among them, only a few
methods are stereoselective in relation to the stereochemistry of
the substituents.2 However, the methods experience racemic
synthesis or symmetrical substrates. To establish a stereo-
selective access to 2,6-disubstituted cyclohexanones in
enantiomerically enriched forms, we envisioned the strategy
illustrated in Scheme 1, which consists of allylic substitution of
and subsequent oxidation furnished the trans isomers V without
isomerization to the thermodynamically more stable cis isomers.
In addition, the transformation was applied to the enantiomeri-
cally enriched (en) picolinate en-2a successfully.
Primary picolinates 1a−d (type I substrates) in racemic
forms and secondary picolinates 2a−d (type II) with
approximately 1:1 diastereomeric ratio (dr) were synthesized
by methods delineated in Scheme 2. In brief, allylic
substitution7 of 4 with Ar2CuMgBr·MgBr2 afforded the key
intermediates 5a8−d, which were converted to 1a−d and 2a−d.
In contrast, the substitution of (R)-4 with Ph2CuMgBr·MgBr2
proceeded with almost complete racemization, whereas the use
of PhCu·MgBr2 afforded en-5a, which was converted to en-1a
with a 94:6 enantiomeric ratio (er). The CBS reduction9 of
ketone 8 afforded (1R,1S)-7a with a 91:9 dr, and subsequent
esterification with PyCO2H afforded en-2a with a high dr of
97:3 as a consequence of the separation of the minor
diastereomer by chromatography.
Scheme 1. Synthesis of 2,6-Disubstituted Cyclohexanones
For allylic substitution of the primary picolinate 1a, both
RLi- and RMgBr-based copper reagents (R = A, Me; B, Bu; C,
Ph) were examined. In the case of the RLi-based reagents,
MgBr2 was added to activate the PyCO2 leaving group. First,
MeCu species derived from MeLi and MeMgBr afforded the
desired SN2′ product 9aA (R = Me) with 93 and 94%
regioselectivity over the regioisomer 10aA (Table 1, entries 1
and 2). The trans configuration of the two substituents on the
cyclohexane ring was determined by analogy with that
determined for the substitution product of the picolinates II.
In contrast, a cuprate Me2CuMgBr·MgBr2 produced the
regioisomer 10aA as a major product (entry 3).10 Butyl copper
(BuCu·LiBr) also produced 9aB with 97% regioselectivity
(entry 4; cf. entry 5). Reaction of 1b−d (Ar = 4-MeC6H4, 4-
MeOC6H4, 4-FC6H4) with MeCu·LiBr and BuCu·LiBr resulted
in the SN2′ products with similar regioselectivities in 89−97%
yields.
I or II with copper reagents followed by the oxidation of the
SN2′ product III or IV. The use of the allylic picolinates is
based on the high SN2′ selectivity and sufficient reactivity of the
picolinates even with aryl copper reagents, which are, in
general, less reactive than alkyl copper reagents.4 In the past,
allylic substitution of simple cycloalkenylmethyl substrates has
been reported,5 although the stereochemical issue in question
was rarely studied.6
In practice, reaction of 1a−d chosen as the representative
picolinates of I with alkyl and aryl copper reagents afforded
SN2′ products III; however, they suffered from low regio- and
product selectivities. Conversely, allylic substitution of
picolinates II as exemplified by 2a−d afforded IV selectively,
Received: November 30, 2013
Published: January 22, 2014
© 2014 American Chemical Society
760
dx.doi.org/10.1021/ol403472y | Org. Lett. 2014, 16, 760−763

Organic Letters
Letter
(acac) was examined with other picolinates 1b and 1c (Ar = 4-
MeC6H4, 4-MeOC6H4) and resulted in slightly lower
regioselectivity of 86 and 88%, respectively (data not shown).
These results convinced us to switch to the investigation using
the secondary picolinates (type II), and the results are
presented in the next paragraph.10
Scheme 2. Synthesis of Allylic Picolinates
Picolinate 2a (Ar = Ph) underwent substitution with MeCu·
LiBr with 96% regioselectivity to afford 11aA as approximately
1:1 olefinic mixture (Table 2, entry 1). Alcohol 7a and the
a
Table 2. Allylic Substitution of 2
yield
bc
,
d
e
entry
2
reagent (equiv), MgBr2 (equiv)
11
11/12
(%)
1
2
2a MeCu·LiBr (2.7), MgBr2 (3.0) 11aA
96:4
81
2a Me2CuLi·LiBr (1.5), MgBr2
24:76
(3.1)
3
4
5
2a BuCu·LiBr (2.8), MgBr2 (3.0)
2a PhCu·LiBr (2.9), MgBr2 (3.0)
11aB
11aC
98:2
95:5
24:76
85
83
2a Ph2CuLi·LiBr (1.5), MgBr2
(3.0)
6
7
8
9
2a PhCu·MgBr2 (1.5)
2b PhCu·LiBr (2.7), MgBr2 (3.0)
2b 4-MeC6H4Cu·MgBr2 (1.5)
11aC
11bC
11bD
11bD
96:4
95:5
95:5
95:5
83
80
92
82
2b 4-MeC6H4Cu·MgBr(acac)
(1.5)
10
11
12
2b 2-MeC6H4Cu·MgBr2 (1.5)
2c PhCu·LiBr (2.7), MgBr2 (3.0)
2d PhCu·LiBr (2.7), MgBr2 (3.0)
11bE
11cC
83:17
94:6
94:6
67
84
81
11dC
a
b
Reactions were performed at 0 °C for 1−2 h. Each was obtained as
In contrast to the alkyl reagents, PhCu·LiBr and PhCu·
MgBr2 afforded a mixture of 9aC and regioisomer 10aC
(entries 6 and 7), whereas Ph2CuMgBr·MgBr2 predominantly
gave regioisomer 10aC (entry 8). Recently, we have introduced
ArCu·MgBr(acac) and Ar2CuMgBr·MgBr(acac) as anti-SN2′-
selective reagents.4b,c One of the reagents, PhCu·MgBr(acac),
was applied to allylic substitution of 1a to produce 9aC with
82% regioselectivity (entry 9), whereas Ph2CuMgBr·MgBr-
(acac) in the presence of ZnI2 furnished 9aC with better SN2′
regioselectivity of 92% (entry 10). Next, Ph2CuMgBr·MgBr-
approximately 1:1 olefin mixture except for 11bE (4:1) and 11cC
(3:2). The trans stereochemistry of 11 was assigned by H NMR of
the derived ketones (see the text). Determined by H NMR analysis
of the unpurified reaction mixtures. Isolated, combined yields of 11
c
1
d
1
e
and 12.
starting picolinate 2a were not detected by 1H NMR and TLC
analyses. The trans stereochemistry between Me and Ph was
determined by NOE analysis (Figure 1) and by conversion to
the known ketone 13aA (vide infra). In contrast, Me2CuLi·LiBr
a
Table 1. Allylic Substitution of 1a
b
c
entry
reagent (equiv)
MeCu·LiBr (2.5)
MeCu·MgBr2 (1.5)
Me2CuMgBr·MgBr2 (1.5)
BuCu·LiBr (2.9)
Bu2CuMgBr·MgBr2 (1.7)
PhCu·LiBr (3.0)
additive (equiv)
MgBr2 (3.1)
9/10/6a/1a
9/10
yield (%)
1
2
90:7:3:0
88:6:6:0
6:78:8:8
94:3:2:1
13:81:6:0
53:35:12:0
46:54:0:0
3:97:0:0
80:18:2:0
84:7:5:4
93:7
90
89
94:6
3
7:93
4
MgBr2 (3.0)
MgBr2 (3.0)
97:3
82
5
14:86
60:40
46:54
3:97
6
7
PhCu·MgBr2 (1.4)
8
Ph2CuMgBr·MgBr2 (1.5)
PhCu·MgBr(acac) (1.5)
Ph2CuMgBr·MgBr(acac) (1.5)
9
82:18
92:8
d
e
10
ZnI2 (1.5)
81
a
b
Reactions were performed at 0 °C for 1−2 h to produce 9aA/10aA (R = Me), 9aB/10aB (R = Bu), and 9aC/10aC (R = Ph). Determined by 1H
c
d
e
NMR analysis of the unpurified reaction mixtures. Isolated, combined yields of 9 and 10. At −40 °C for 12 h. 9aC/10aC = 22:78 without ZnI2.
761
dx.doi.org/10.1021/ol403472y | Org. Lett. 2014, 16, 760−763

Organic Letters
Letter
In the case of 13aC, the 1H NMR spectrum was consistent with
that reported for the trans isomer.2f
The present method culminated in the synthesis of ketones
in enantiomerically enriched forms. As delineated in Scheme 4,
Scheme 4. Synthesis of Enantiomerically Enriched
a
Cyclohexanones
Figure 1. NOE analysis of (Z)- and (E)-11aA and en-13aA.
produced a mixture of the regioisomers (entry 2). Similarly,
substitution of 2a with BuCu·LiBr produced 11aB with 98%
regioselectivity in 85% yield (entry 3).
Next, phenyl addition of 2a was examined. To our delight,
PhCu·LiBr and PhCu·MgBr2 afforded 11aC with 95 and 96%
regioselectivities, respectively, in good yields (entries 4 and 6;
cf. entry 5), demonstrating higher SN2′ preference for the
secondary picolinate 2a than that of the primary picolinate 1a
(Table 1, entries 6 and 7). In a similar way, substitution of 2b
(Ar = 4-MeC6H4) with PhCu·LiBr and 4-MeC6H4Cu·X (X =
MgBr2, MgBr(acac)) produced 11bC and 11bD, respectively,
with 95% regioselectivity in good yields (entries 7−9). A
sterically congested reagent, 2-MeC6H4Cu·MgBr2, lowered the
regioselectivity and yield of 11bE (entry 10). In addition, upon
reaction with PhCu·LiBr, 2c (Ar = 4-MeOC6H4) and 2d (Ar =
4-FC6H4) produced 11cC and 11dC, respectively (entries 11
and 12).
a
THF, 0 °C, 1 h for the first step.
substitution of enantiomerically enriched picolinate en-2a with
MeCu·LiBr produced (Z)-en-11aA with 98% Z selectivity.11
This product was subsequently oxidized to ketone en-13aA
with a >98:2 trans/cis ratio and a 98:2 er. The trans
Next, the second step of the strategy illustrated in Scheme 1
was investigated. Ozonolysis of 11aB (Ar = Ph, R = Bu) at −78
°C followed by reductive workup afforded trans-2,6-disub-
stituted cyclohexanone 13aB in 86% yield with a 98:2 trans/cis
ratio (Scheme 3), and exposure of this ketone to t-BuOK in
stereochemistry was confirmed by comparing the H and 13C
1
NMR spectra with those reported for the trans isomer3f and by
NOE analysis, as well (Figure 1). This result indicates that the
R chirality at the α carbon of the allylic moiety is transferred to
the Z olefin, whereas the stereochemical course of the reaction
is determined independently of this chirality. This indication
was also observed in the production of (Z)-en-11aB with 93%
Z selectivity and then en-13aB with a >98:2 trans/cis ratio.
Similarly, transformation of en-2a with PhCu·MgBr2 furnished
en-13aC with a >98:2 trans/cis ratio and 96:4 er. In addition,
ozonolysis of en-9aB (R = Bu) derived from en-1a produced en-
13aB selectively (equation not shown).
Scheme 3. Synthesis of 2,6-Disubstituted Cyclohexanones
Taking the anti-SN2′ pathway into consideration for the
substitution of allylic picolinates with copper reagents, TS-A
and TS-B were conceived as transition state models for the
reaction of en-2a with PhCu (Scheme 5), in which the Ph
Scheme 5. Pathway Analysis for the Substitution of en-2a
THF afforded the cis isomer in 73% yield with an 81:19 cis/
1
trans ratio determined by H NMR spectroscopy. Similarly,
ozonolysis of other olefins 11 provided trans-ketones 13 with
high stereoselectivity (Scheme 3), which was determined by
comparison with the cis isomers synthesized by isomerization.
762
dx.doi.org/10.1021/ol403472y | Org. Lett. 2014, 16, 760−763

Organic Letters
Letter
(l) Wender, P. A.; Erhardt, L. M.; Letendre, L. J. J. Am. Chem. Soc.
1981, 103, 2114−2116. (m) Ireland, R. E.; Marshall, J. A. J. Org. Chem.
1962, 27, 1615−1619.
group is projected to the pseudoequatorial space with a
perpendicular angle to the cyclohexenyl ring by the steric
reason. Among these models, the severe steric repulsion is
conceived between the Ph group and Cu−Ph that approaches
the olefin from the α side to form the π complex, thus
disfavoring TS-B. Consequently, TS-A is the predominant
pathway to produce (Z)-en-11aC. This consideration is
consistent with the fact that the present substitution selectively
afforded 11 with the trans stereochemistry between the R and
Ar (Table 2). Thus, stereodefined construction of the chirality
on the α carbon possessing PyCO2 is unnecessary for the
purpose of this two-step synthesis of the 2,6-disubstituted
ketones in enantiomerically enriched form from en-2a.12
In summary, we developed a stereoselective method to
obtain trans-2,6-disubstituted cyclohexanones. Furthermore, we
clarified that (1) high anti-SN2′ selectivity is secured with the
Me group, and (2) the stereochemical course is definitely
dictated by the pre-existing chirality on the ring.13
(4) (a) Feng, C.; Kobayashi, Y. Eur. J. Org. Chem. 2013, 6666−6676.
(b) Feng, C.; Kobayashi, Y. J. Org. Chem. 2013, 78, 3755−3766.
(c) Wang, Q.; Kobayashi, Y. Org. Lett. 2011, 13, 6252−6255.
(d) Kaneko, Y.; Kiyotsuka, Y.; Acharya, H. P.; Kobayashi, Y. Chem.
Commun. 2010, 46, 5482−5484. (e) Hyodo, T.; Kiyotsuka, Y.;
Kobayashi, Y. Org. Lett. 2009, 11, 1103−1106. (f) Kiyotsuka, Y.;
Acharya, H. P.; Katayama, Y.; Hyodo, T.; Kobayashi, Y. Org. Lett.
2008, 10, 1719−1722. (g) Kiyotsuka, Y.; Katayama, Y.; Acharya, H. P.;
Hyodo, T.; Kobayashi, Y. J. Org. Chem. 2009, 74, 1939−1951.
(h) Kiyotsuka, Y.; Kobayashi, Y. J. Org. Chem. 2009, 74, 7489−7495.
(5) (a) Falciola, C. A.; Tissot-Croset, K.; Alexakis, A. Angew. Chem.,
Int. Ed. 2006, 45, 5995−5998. (b) Whittaker, A. M.; Rucker, R. P.;
Lalic, G. Org. Lett. 2010, 12, 3216−3218.
(6) (a) Gais, H.-J.; Muller, H.; Bund, J.; Scommoda, M.; Brandt, J.;
̈
Raabe, G. J. Am. Soc. Chem. 1995, 117, 2453−2466. (b) Flemming, S.;
Kabbara, J.; Nickisch, K.; Westermann, J.; Mohr, J. Synlett 1995, 183−
185.
(7) Soorukram, D.; Knochel, P. Angew. Chem., Int. Ed. 2006, 45,
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures and spectral data of compounds
described herein. This material is available free of charge via the
Internet at http://pubs.acs.org.
3686−3689.
■
(8) Christl, M.; Schreck, M.; Fischer, T.; Rudolph, M.; Moigno, D.;
Fischer, H.; Deuerlein, S.; Stalke, D. Chem.Eur. J. 2009, 15, 11256−
11265.
S
(9) Corey, E. J.; Helal, C. J. Angew. Chem., Int. Ed. 1998, 37, 1986−
2012.
(10) The α regioselectivity observed with the cuprates (Table 1,
entries 3, 5, and 8) might stem in part from the increased
nucleophilicity of the reagents, which prefer reaction at less sterically
congested carbon.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: ykobayas@bio.titech.ac.jp.
Notes
(11) Substitution of the diastereomer of en-2a is predictable on the
basis of the results of the substitutions using en-2a (Scheme 4) and 2a
(Table 1, entry 1).
The authors declare no competing financial interest.
(12) Allylic substitution of picolinate II (Ar = Bu, R2 = Me) with
PhCu afforded a mixture of the products, among which the desired
ACKNOWLEDGMENTS
1
■
SN2′ product was confirmed by H NMR spectroscopy.
This work was supported by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Science, Sports, and
Culture, Japan.
(13) Substitution of the cyclopentenyl picolinate corresponding to 2a
with PhCu·MgBr2 proceeded with 95% regioelectivity and 80%
stereoselectivity.
REFERENCES
■
(1) (a) Bellina, F.; Rossi, R. Chem. Rev. 2010, 110, 1082−1146.
(b) Turner, T. C.; Shibayama, K.; Boger, D. L. Org. Lett. 2013, 15,
1100−1103. (c) Iwama, T.; Rawal, V. H. Org. Lett. 2006, 8, 5725−
5728. (d) MacKay, J. A.; Bishop, R. L.; Rawal, V. H. Org. Lett. 2005, 7,
3421−3424.
(2) (a) Miyoshi, T.; Miyakawa, T.; Ueda, M.; Miyata, O. Angew.
Chem., Int. Ed. 2011, 50, 928−931. (b) Miyoshi, T.; Sato, S.; Tanaka,
H.; Hasegawa, C.; Ueda, M.; Miyata, O. Tetrahedron Lett. 2012, 53,
4188−4191. (c) Gao, S.; Tu, Y. Q.; Song, Z.; Wang, A.; Fan, X.; Jiang,
Y. J. Org. Chem. 2005, 70, 6523−6525. (d) Henderson, K. W.; Kerr, W.
J.; Moir, J. H. Tetrahedron 2002, 58, 4573−4587. (e) Bozzini, S.;
Gratton, S.; Lisini, A.; Pellizer, G.; Risaliti, A. Tetrahedron 1982, 38,
1459−1464. (f) Bozzini, S.; Cova, B.; Gratton, S.; Lisini, A.; Risaliti, A.
J. Chem. Soc., Perkin. Trans. 1 1980, 240−243.
(3) (a) Hatakeyama, T.; Ito, S.; Nakamura, M.; Nakamura, E. J. Am.
Chem. Soc. 2005, 127, 14192−14193. (b) Hatakeyama, T.; Ito, S.;
Yamane, H.; Nakamura, M.; Nakamura, E. Tetrahedron 2007, 63,
8440−8448. (c) Ooi, T.; Goto, R.; Maruoka, K. J. Am. Chem. Soc.
2003, 125, 10494−10495. (d) Iwama, T.; Birman, V. B.; Kozmin, S.
A.; Rawal, V. H. Org. Lett. 1999, 1, 673−676. (e) Witt, O.; Mauser, H.;
Friedl, T.; Wilhelm, D.; Clark, T. J. Org. Chem. 1998, 63, 959−967.
(f) Molander, G. A.; McKie, J. A. J. Org. Chem. 1993, 58, 7216−7227.
(g) Chen, K.; Koser, G. F. J. Org. Chem. 1991, 56, 5764−5767.
(h) Peyman, A.; Beckhaus, H.-D.; Ruchardt, C. Chem. Ber. 1988, 121,
̈
1027−1031. (i) Rathke, M. W.; Vogiazoglou, D. J. Org. Chem. 1987,
52, 3697−3698. (j) Birch, A. J.; Kelly, L. F.; Narula, A. S. Tetrahedron
1982, 38, 1813−1823. (k) Dana, D. E. Synthesis 1982, 164−165.
763
dx.doi.org/10.1021/ol403472y | Org. Lett. 2014, 16, 760−763