Fragmentation of 4-sulfonylbicyclo[3.1.0]hexan-2-ones as the key step in the enantioselective synthesis of (1R)-cis-chrysanthemic acid involving desymmetrization of 3,3,6,6-tetramethylbicyclo[3.1.0]hexane-2,4-dione

Article abstract of DOI:10.1055/s-2007-970747

t-BuOK-H₂O (7.6:2.3) in THF or DMSO allows the efficient Grob-type fragmentation of 4-sulfonyl-bicyclo[3.1.0]hexan-2-ones which cannot be achieved by potassium hydroxide in DMSO as we originally described. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2007-970747

LETTER
607
Fragmentation of 4-Sulfonylbicyclo[3.1.0]hexan-2-ones as the Key Step in the
Enantioselective Synthesis of (1R)-cis-Chrysanthemic Acid Involving
Desymmetrization of 3,3,6,6-Tetramethylbicyclo[3.1.0]hexane-2,4-dione
Enantioselective
S
l
ynthesi
a
s
of (1
R
)-cis-Chry
n
santhemic Acid Krief,* Adrian Kremer
Laboratoire de Chimie Organique de Synthèse, Facultés Universitaires N.-D. de la Paix, 61 Rue de Bruxelles, Namur 5000, Belgium
Fax +32(81)724536; E-mail: alain.krief@fundp.ac.be
Received 6 December 2006
The transformation of 4 to 2 is not general. For example
Abstract: t-BuOK–H2O (7.6:2.3) in THF or DMSO allows the ef-
4a and 4b which possess an endo-methyl group are very
ficient Grob-type fragmentation of 4-sulfonyl-bicyclo[3.1.0]hexan-
unstable.1a,4 Thus the effect of the stereochemistry on the
2-ones which cannot be achieved by potassium hydroxide in DMSO
Grob-type fragmentation which is known to require an
antiperiplanar arrangement of the atoms involved in the
process6 has not been systematically tested.1a,4
as we originally described.
Key words: ring opening, retro-aldol, nucleophiles, lithiation,
chirality
A few years ago Prof. N. Simpkins informed us that he
was unable to repeat7 the fragmentation of the 4-mesylate
1aexo to 2a we previously published1a and we confirmed
his finding. In fact cis-chrysanthemic acid (2a) is effec-
tively produced under conditions A, but in minute amount
(2–7%, 15 runs) beside large quantities of unidentified
polymeric material 7a (Scheme 2).
Grob-type fragmentation of bicyclo[3.1.0]hexan-2-ones
1, bearing a leaving group in 4-position, proved to be a
valuable synthetic strategy to produce cis-vinyl cyclopro-
pane carboxylic acids 2 (Scheme 1). For example, it has
been efficiently used in our laboratory as a key step in the
synthesis of cis-chrysanthemic acid (2a) in racemic form1
or as a single enantiomer1c,2,3a from 3,3,6,6-tetramethyl-
bicyclo[3.1.0]hexane-2,4-dione (3a)1,2 or from 4-tosyl-
oxy-5,5-dimethylcyclopent-2-enone (5)3 (Scheme 1).
O
OH
KO
O
OK
H
O
i
9acis
4aexo
8aexo
O
O
O
X
O
R1
R2
OH
SPh2
O
KO
OK
O
H
H
KO
R1
R1
OTs
6
R2
R2
O
R1
R2
O
3
4
1
5
9atrans
7a
R1, R2 = a: Me; b: H, Me; c: Me, H; d: H, H
n
CO2H
Scheme 2 Reagents and conditions: (i) KOH (6 equiv), DMSO–
H2O (4:1), 70 °C, 2–4 h.
i, ii
2a X = OMs 72%
1
X = OTs
X = Br
75–82%
87%
2a
2a
R2 R1
We have systematically performed the reaction between
the mesylate and tosylate 1aexo and potassium hydroxide
in DMSO under different conditions. We have changed
the temperature (50–100 °C), the concentration of the re-
agent (0.1–1 M in DMSO), the amount of water (0–50%)
and of potassium hydroxide (2.5–10 equiv) used, and
achieved little or no success. We for example observed
that the sulfonates 1aexo do not react at an appreciable rate
when the temperature is lower than 50 °C under condi-
tions A and that 2a can be obtained in 23% yield from the
tosylate 1aexo by performing the reaction with 12 equiva-
lents of potassium hydroxide at 100 °C in DMSO–H2O
(4:1).
Scheme 1 Reagents and conditions: (i) KOH (6 equiv), DMSO–
H2O (4:1), 70 °C, 2–4 h; (ii) H3O+.
It has been also successfully used for the synthesis of
related (d,l)-cis-2-desmethyl- (2b,c) and 2,2-di-desmethyl
chrysanthemic acids (2d).4 The published conditions in-
volve the reaction of aqueous potassium hydroxide solu-
tion in DMSO on the 4-mesylate,1 4-tosylate,1,3 and 4-
bromide5 1exo [conditions A: 6 equiv KOH, DMSO–H2O
(4:1), 1 M solution, 70 °C, 2–4 h,].
However, we were surprised to find that the above obser-
vations do not apply to the whole series of compounds 1.8
For example, conditions A works on some desmethyl
SYNLETT 2007, No. 4, pp 0607–0610
Advanced online publication: 21.02.2007
DOI: 10.1055/s-2007-970747; Art ID: G35906ST
© Georg Thieme Verlag Stuttgart · New York
0
1.
0
3.
2
0
0
7

608
A. Krief, A. Kremer
LETTER
derivatives (1cendo and 1dendo
)
1b,4 and especially on the 4- ‘Gassman reagent’,9 prepared from potassium tert-butox-
exo-bromo-3,3,6,6-tetramethylbicyclo[3.1.0]hexan-2-one ide and water (3.3:1), particularly efficient when used in
1aexo which produced 2a in up to 87% yield as originally excess (sulfonates 1aexo–t-BuOK–H2O, 1:7.6:2.3).
described (5/5 successful runs).5
The reaction proceeds at room temperature in less than
We suspected, after comparing the dramatic difference of 1 hour on all the exo-sulfonates 1aexo so far tested either in
reactivity between the sulfonates and the bromide 1aexo, anhydrous DMSO (conditions B: Scheme 4, Table 2
that the failure of potassium hydroxide to produce 2a from entries 1, 3, 8, 10) or in anhydrous THF (conditions C:
the sulfonates was due to a competing reaction on the Scheme 4, Table 2 entries 2, 4, 9) and produces after acid
electrophilic sulfur atom of the sulfonates. This instead hydrolysis cis-chrysanthemic acid (2a) in more than 80%
produces the potassium 4-exo-oxy-3,3,6,6-tetramethyl- yield. Conditions B, which uses DMSO proved to be more
bicyclo[3.1.0]hexan-2-one 8aexo intermediate which then efficient than conditions C, which involves instead THF
polymerizes. We effectively found, in a separate experi- (Scheme 4, Table 2 compare entries 1 to 2 and 3 to 4).
ment, that 8aexo prepared from 4-exo-hydroxy-3,3,6,6-tet- Surprisingly, however, the reverse has been observed with
ramethylbicyclo[3.1.0]hexan-2-one 4aexo and potassium the ‘hindered’ 2,4,6-trimethylbenzenesulfonate 1aexo
hydroxide in anhydrous or wet DMSO, provides a poly- (Scheme 4, Table 2, compare entry 8 to 9).
meric material postulated as 7a once heated around 70 °C,
O
as the result of a tandem retro-aldol–cis/trans-isomeriza-
tion reaction. Those steps are tentatively disclosed in
X
CO2H
i
Scheme 2.
The first strategy we tried was therefore expected to
1aexo
2a
prevent the attack of the hydroxide ion on the sulfur atom
of the sulfonate. It was achieved by increasing steric hin-
drance there using 2,4,6-trimethyl benzenesulfonate 1aexo
[R = 2,4,6-(Me)3C6H2] or 2,4,6-triisopropylbenzene-
sulfonate 1aexo [R = 2,4,6-(i-Pr)3C6H2], and proved to be
quite successful since chrysanthemic acid (2a) is pro-
duced in up to 50% yield. Nevertheless, it was of poor
synthetic value since it required too long of a reaction time
and access to the ‘sterically hindered sulfonates’ from the
ketoalcohol 4aexo was far from easy (Scheme 3, Table 1,
entries 3–5 compared to entries 1 and 2).
Scheme 4 Reagents and conditions: (i) t-BuOM–H2O, solvent,
22 °C; (ii) H3O+.
The ratio of starting materials and the solvent (anhyd
DMSO) used by Gassman to successfully cleave non-eno-
lizable cyclopropyl ketones9a–c and hydrolyze ‘hindered
esters’9d to the corresponding carboxylic acids proved to
be the most efficient combination (Table 2, entries 1–4, 8,
9). Changing the sulfonate/‘Gassman reagent’ ratio
(Table 2, compare entries 5 to 4), the potassium tert-but-
oxide/water ratio (Table 2, compare entries 6, 7, 4) or the
tert-butoxide counter ion (K to Na or Li) proved to be
highly detrimental (22 °C, metal, solvent, time, yield in
2a: Na, DMSO, 8 h, 72%; Na, THF, 36 h, 0%; Li, DMSO,
72 h, 2%).
O
O
OSO2R
OH
CO2H
i
ii, iii
1aexo
2a
4aexo
We took advantage of the efficient fragmentation reaction
described above and of the recent discovery of N.
Simpkins10,11a that the scalemic ketomesylate 4aexo can be
easily prepared (80% yield, >98% ee, Scheme 5) by
asymmetric desymmetrization of 3,3,6,6-tetramethylbicy-
clo[3.1.0]hexane-2,4-dione (3a) with scalemic lithium
(R,R)-bis(1-phenylethyl)amide in the presence of lithium
chloride (1.2 equiv) and trimethylsilyl chloride (10 equiv,
–78 °C to 20 °C, 5 h), to devise a novel efficient enantio-
selective synthesis of (1R)-cis-chrysanthemic acid (2a).
Scheme
3
Reagents and conditions: (i) RSO2Cl (1 equiv),
CH2Cl2;(ii) KOH (70 equiv), DMSO–H2O (2:1), 55 °C; (iii) H3O+.
The second strategy, which involves the use of a more
carbophilic reagent, proved to be more successful.
Although neither lithium, sodium or tetrabutyl ammoni-
um hydroxides in DMSO allow the desired transformation
of the mesylate or tosylate 1aexo to 2a, we found the
Table 1 Reaction of Compounds 4aexo to 1aexo and 2aexo under Various Conditions
Entry
R
Conditions
Yield of 1aexo Yield of 4aexo Time
Yield of 2a
(%)
86
83
76
25
35
(%)
(h)
(%)
1
2
3
4
5
Me
1.5 equiv Et3N, –10 °C, 0.75 h
–
Ph
2.5 equiv Py, 0.2 equiv DMAP, 20 °C, 24 h
2.5 equiv Py, 0.2 equiv DMAP, 20 °C, 168 h
2.5 equiv Py, 0.2 equiv DMAP, 20 °C, 108 h
2.5 equiv py, 0.2 equiv DMAP, 45 °C, 108 h
–
2,4,6-(Me)3C6H2
2,4,6-(i-Pr)3C6H2
2,4,6-(i-Pr)3C6H2
20
50
–
30
48
33
52
Synlett 2007, No. 4, 607–610 © Thieme Stuttgart · New York

LETTER
Enantioselective Synthesis of (1R)-cis-Chrysanthemic Acid
609
Table 2 Reaction of Compounds 1aexo to 2aexo Using THF and DMSO as Solvents
Entry
X
Ratio of 1a:t-BuOM:H2O
1:7.6:2.3
Solvent
DMSO
THF
Time (h)
0.4
0.3
0.3
1
Yield of 2a (%)
Yield of 1aexo (%)
1
2
MeSO2O
90
60
86
84
64
60
07
80
95
83
MeSO2O
1:7.6:2.3
3
p-TolSO2O
1:7.6:2.3
DMSO
THF
4
p-TolSO2O
1:7.6:2.3
5
p-TolSO2O
1:3.6:1.1
THF
4
08
06
64
6
p-TolSO2O
1:4.4:2.2
THF
24
7
p-TolSO2O
1:2.2:2.2
THF
52
8
2,4,6-(Me)3C6H2SO2O
2,4,6-(Me)3C6H2SO2O
2,4,6-(i-Pr)3C6H2SO2O
1:7.6:2.3
DMSO
THF
0.8
0.5
0.8
9
1:7.6:2.3
10
1:7.6:2.3
DMSO
Reduction of the resulting a-trimethylsilyl diketone 10a11a correlation unambiguously establishes the stereochemis-
by sodium borohydride in the presence of cerium trichlo- try attributed10 to scalemic 10a resulting from the enantio-
11b
ride leads to the aldol 11aexo bearing a trimethylsilyl selective desymmetrization of 3a (Scheme 5).
group at the ring junction in a-position to its carbonyl
(1R)-cis-Chrysanthemic acid has been synthesized even
more efficiently from the aldol 11aexo by changing the
group. This reaction occurs not only regioselectively on
11b
the carbonyl carbon the farest from the trimethylsilyl
order of reactions described above. Mesylation of 11aexo
group but also stereoselectively, as on related 3a,1 from
11e
leads to the ketomesylate 12aexo still bearing the tri-
the most hindered endo-face of the bicyclic dione. These
methylsilyl group [1.1 equiv MeSO2Cl (0.15 M), 1.5
features have been unambiguously established by an X-
equiv Et3N, CH2Cl2, –20 °C, 0.75 h, 86%].1a,2 Treatment
ray crystallography analysis of its p-nitrobenzoate.12
of 12aexo with the ‘Gassman reagent’ in DMSO (condi-
Desilylation of 11a {1.1 equiv [(n-Bu)4NF·3H2O], THF, tions B) followed by acid treatment, directly generates
0–20 °C, 60 h}, followed by fragmentation of the corre- scalemic (1R)-cis-chrysanthemic acid (2a),11f in a single-
sponding mesylate 1a,11c using conditions B, as disclosed pot domino reaction in which the reagent has performed
above affords (1R)-cis-chrysanthemic acid (2a)11d in up to the desilylation as well as the fragmentation reactions
76% yield almost as a single enantiomer (ee >98%). This (86% yield, ee >98%, Scheme 6).
O
O
OMs
O
O
Me3Si
Me3Si
Me3Si
OH
ii, iii
i
i
O
O
12aexo
10a
3a
11aexo
O
OH
O
OH
Me3Si
CO2H
(1R)
iv
ii, iii
4aexo
2a
11aexo
DMSO: 86%, ee >98%
THF: 44%, ee >98%
v–vii
CO2H
(1R)
Scheme 6 Reagents and conditions: (i) MsCl (1.1 equiv), 1.5 Et3N,
CH2Cl2, –20 °C, 0.75 h, 86%; (ii) t-BuOK–H2O (7.6:2.3), DMSO,
20 °C, 0.5 h; or THF, 20 °C, 0.75 h; (iii) H3O+.
2a
Scheme 5 Reagents and conditions: (i) lithium (R,R)-bis(1-phenyl-
ethyl)amide (1.2 equiv), LiCl (1.2 equiv), Me3SiCl (10 equiv), THF,
–78 °C to 20 °C, 5 h, 80%, ee >98%; (ii) CeCl3·7H2O (1 equiv),
MeOH; (iii) NaBH4 (1 equiv), MeOH, –78 °C, 5 h, 95%; (iv)
Bu4NF·3H2O (1.1. equiv), THF, 0–20 °C, 60 h, 98%; (v) MsCl (1.1
equiv), Et3N (1.5 equiv), CH2Cl2, –20 °C, 0.75 h; (vi) t-BuOK–H2O
(7.6:2.3), DMSO, 22 °C, 0.4 h; (vii) H3O+; 85–91%, >98% ee.
We suspected that the desilylation reaction occurs first but
were unable to prove it since, even when the temperature
was lowered, 2a was the only product formed whether the
reaction was carried out in anhydrous DMSO or THF. The
former conditions proved to be in this case by far the best.
Synlett 2007, No. 4, 607–610 © Thieme Stuttgart · New York

610
A. Krief, A. Kremer
LETTER
Nevertheless, we have been able to isolate the desilylated
mesylate 1aexo in more than 93% on reaction of 12aexo
with potassium hydroxide [6 equiv, DMSO–H2O (4:1),
22 °C, 0.75 h]. As expected degradation of 1aexo takes
place if the reaction is instead performed at 70 °C for four
hours.
References and Notes
(1) (a) Krief, A.; Surleraux, D.; Frauenrath, H. Tetrahedron Lett.
1988, 29, 6157. (b) Krief, A.; Surleraux, D.; Dumont, W.;
Pasau, P.; Lecomte, P. Pure Appl. Chem. 1990, 62, 1311.
(c) Krief, A. In Stereocontrolled Organic Synthesis, A
'Chemistry for the 21st Century' Monograph; Trost, B. M.,
Ed.; International Union of Pure and Applied Chemistry,
Blackwell Scientific Publications: Oxford, 1994, 337–397.
(d) Readily available from dimethyl dimedone.
(2) Krief, A.; Surleraux, D.; Ropson, N. Tetrahedron:
Asymmetry 1993, 4, 289.
(3) (a) Krief, A.; Swinnen, D. Tetrahedron Lett. 1996, 37, 7123.
(b) Krief, A.; Swinnen, D.; Billen, D. Tetrahedron Lett.
2002, 43, 5871.
(4) Krief, A.; Surleraux, D.; Robson, M. J. Synlett 1991, 4, 276.
(5) Krief, A.; Lorvelec, G.; Jeanmart, S. Tetrahedron Lett. 2000,
41, 3871.
(6) Grob, C. A. Angew. Chem., Int. Ed. Engl. 1969, 8, 535.
(7) Once we found a good method for the fragmentation
reaction, we performed the synthesis of scalemic 2a as
originally planned by Prof. Nigel Simpkins and informed
him. We thank Prof. Nigel Simpkins, School of Chemistry,
University of Nottingham, University Park, Nottingham,
NG7 2RD, UK for his kindness.
(8) These results as well as those on the whole series of
stereoisomers disclosed on Scheme 1 will be reported later.
For example, we confirmed that the very labile endo-
mesylate 1aMs-endo does not produce chrysanthemic acid 2a
neither under conditions A nor optimized conditions B or
C.1a,4
The reasons why conditions A are not reproducible when
applied to sulfonates 1aexo still remain unanswered since
some related compounds deliver the corresponding vinyl
cyclopropane carboxylic acids. Anyhow, we have found
a particularly efficient reagent (K2O supported on t-
BuOK?) and conditions, which proved smoother than the
previous ones. We are investigating and reinvestigating
systematically the reactivity of the whole series of regio-
isomeric 1 under conditions A–C to better understand the
scope and limitations of this fragmentation reaction.
Typical Experiment
H2O (42 mg, 2.3 mmol) was added at r.t., under an atmosphere of
argon, to a solution of freshly sublimed t-BuOK (851 mg, 7.6
mmol) in anhyd DMSO (4.0 mL). The solution was stirred for 10
min before adding 4-exo-mesyloxy-3,3,6,6-tetramethylbicyc-
lo[3.1.0]hexan-2-one (1a, 246 mg, 1.0 mmol) in one portion (yel-
low coloration appeared). The reaction mixture was then stirred at
r.t. and monitored by TLC. After 45 min, the reaction mixture was
poured into an Erlenmeyer flask containing 10 mL of Et2O and
2 mL of ice and acidified to pH 2 with an aq solution of HCl (10%;
discoloration). After decantation and extraction with Et2O (4 × 10
mL), the combined organic extracts were washed with H2O (2 × 2
mL), dried (MgSO4), filtered and the solvent evaporated under re-
duced pressure to give a crude product. Purification by column
chromatography (n-pentane–Et2O, 60:40) gives the cis-chrysan-
themic acid (2a) as a white solid (153 mg, 91%).
(9) (a) Gassman, P. G.; Zalar, F. V. Tetrahedron Lett. 1964, 40,
3031. (b) Gassman, P. G.; Zalar, F. V. Tetrahedron Lett.
1964, 44, 3251. (c) Gassman, P. G.; Lumb, J. T.; Zalar, F. V.
J. Am. Chem. Soc. 1967, 946. (d) Gassman, P. G.; Schenk,
W. N. J. Org. Chem. 1977, 42, 918.
(10) Adams, D. J.; Blake, A. J.; Cooke, P. A.; Gill, C. D.;
Simpkins, N. Tetrahedron 2002, 58, 4603.
(11) (a) Compound 10a: [a]D20 –20 (c 1.20, CHCl3).
(b) Compound 11aexo: [a]D20 +12.5 (c 1.22, CHCl3).
(c) Compound 1aMs-exo: [a]D20 +55.1 (c 1.11, CHCl3).
(d) Compound 2a: [a]D20 +81.7 (c 1.0, CHCl3).
(e) Compound 12aexo: [a]D20 +4.9 (c 1.32, MeOH).
(f) Compound 2a: [a]D20 +82.2 (c 1.0, CHCl3).
(12) Norberg, B.; Wouters, J.; Krief, A.; Kremer, A. Acta
Crystallogr., Sect. C, to be published.
Synlett 2007, No. 4, 607–610 © Thieme Stuttgart · New York