The conformational bias of aryl, arylsulfonyl geminally substituted tertiary carbon centers: Applications in substrate-based stereocontrol

Article abstract of DOI:10.1039/b601647b

Intramolecular nitrile oxide-olefin cycloaddition to form hexahydrobenzisoxazole 14, which engenders a phenylsulfonyl, 2,5-difluorophenyl geminally substituted carbon substructure, proceeds with up to 99% ds. A rationalization of the high level of substra

Full text of DOI:10.1039/b601647b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
The conformational bias of aryl, arylsulfonyl geminally substituted tertiary
carbon centers: applications in substrate-based stereocontrol†
Jeremy P. Scott,*a Peter R. Mullens,a Sarah E. Brewer,a Karel M. J. Brands,a Jennifer R. Chilenski,b
Antony J. Davies,a Andrew D. Gibb,a David R. Lieberman,a Steven F. Olivera and Ulf-H. Dollingb
Received 2nd February 2006, Accepted 10th March 2006
First published as an Advance Article on the web 3rd April 2006
DOI: 10.1039/b601647b
Intramolecular nitrile oxide–olefin cycloaddition to form hexahydrobenzisoxazole 14, which engenders
a phenylsulfonyl, 2,5-difluorophenyl geminally substituted carbon substructure, proceeds with up to
99% ds. A rationalization of the high level of substrate-based stereo-induction observed in this and
related ketone and acrylonitrile metallohydride reductions, supported by single crystal X-ray
crystallography, is presented.
Introduction
As part of a programme at Merck Research Laboratories directed
toward the identification of c-secretase inhibitors,1 we have
prepared a number of cyclohexane derivatives 1 bearing a common
structural motif, in which one of the six membered ring carbons
is geminally substituted with an aryl and arylsulfonyl moiety
(Fig. 1). During the course of our synthesis of these derivatives, we
encountered a number of substrate-controlled diastereoselective
transformations in which unanticipated high levels of stereo-
induction were observed. Herein we document our observations
and propose a rationalization for the observed stereochemical
outcomes.
Scheme 1 Reagents and conditions: (a) NaBH4, EtOH, −10 ◦C; (b)
◦
L-SelectrideꢀR , THF, −70 ◦C; H2O2, NaOH, NaCl, H2O, 5 C. Ar1
=
4-chlorophenyl; Ar2 = 2,5-difluorophenyl.
trans-cyclohexanol product 3 was obtained preferentially with
95% ds (94% yield).2 Switching to the sterically encumbered
R
L-Selectrideꢀ reagent3 the sense of induction was reversed,
affording the cis-cyclohexanol 4 preferentially with 90% ds (98%
yield). By way of comparison, sodium borohydride reduction
of 4-(tert-butyl)cyclohexanone has been reported to afford 87%
ds (0 ◦C, MeOH) favoring the trans 4-(tert-butyl)cyclohexanol
(axial hydride addition)4 whilst L-Selectrideꢀ (−78 ◦C, THF)
R
favored the cis diastereomer to the extent of 97% ds (equatorial
hydride addition).3 Reduction of exocyclic acrylonitrile derivative
5 provides a further example of the influence of the geminal aryl,
arylsulfonyl substructure on hydride reduction stereochemistry
(Scheme 2). Under the optimized conditions which use a reaction
Fig. 1
temperature of −60 ◦C, L-Selectrideꢀ reduction proceeds in
R
a conjugate manner to afford the desired cis diastereomer 6
with >99.9 ds (94% yield). Moreover, even at T ≤ −10 ◦C the
diastereoselection remains >99 : 1.5
Results and discussion
The
4-[4-(chlorophenyl)sulfonyl]-4-(2,5-difluorophenyl)cyclo-
hexanone 2 served as a key intermediate through which
structure–activity relationships were explored (Scheme 1). Early
in our synthetic studies, we examined metallohydride reductions
of 2 and found that using NaBH4 in EtOH at −10 ◦C, the
aDepartment of Process Research, Merck Sharp & Dohme Research
Laboratories, Hertford Road, Hoddesdon, Hertfordshire, UK EN11 9BU.
E-mail: jeremy_scott@merck.com
Scheme 2 Reagents and conditions: (a) L-SelectrideꢀR , THF, −60 < T <
◦
−55 ◦C; H2O2, NaOH, NaCl, H2O, 5 C. Ar1 = 4-chlorophenyl; Ar2 =
bDepartment of Process Research, Merck Research Laboratories, P.O. BOX
2000, Rahway, New Jersey 07065, USA
2,5-difluorophenyl.
† Electronic supplementary information (ESI) available: stereochemical
assignments of 9, 10 and 14 and the single crystal X-ray structures of 2
and 4 (CIF files). Copies of 1H and 13C spectra for 3, 10 and 12. See DOI:
10.1039/b601647b
The stereochemical course of the hydride reductions of 2
and 5 is consistent with the rationale that the most reactive
conformers populated in solution have the 4-chlorophenylsulfonyl
1806 | Org. Biomol. Chem., 2006, 4, 1806–1810
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
group preferentially adopting an equatorial orientation on the
substituted cyclohexane ring, whilst the 2,5-difluorophenyl group
is axial. The sense of diastereocontrol is then dictated by the
steric bulk of the hydride source employed.6 The single crystal
X-ray structure7 of cyclohexanone 2 supports this rationale
(Fig. 2), with the cyclohexanone ring having a slightly distorted
chair conformation in which the 4-chlorophenyl ring of the
equatorial sulfone moiety effectively stacks with the axial 2,5-
difluorophenyl ring. The latter ring exposes the p-surface toward
the cyclohexanone ring, thereby minimizing steric interactions of
the ortho-hydrogen and fluorine with the syn axial hydrogens. The
X-ray structure of cis-cyclohexanol 4, which confirms the relative
stereochemical outcome of the hydride reductions of 2, demon-
strates a similar axial/equatorial conformational preference for
the 2,5-difluorophenyl and 4-chlorophenylsulfonyl substituents in
the solid state.7
Scheme 3 Reagents and conditions: (a) LHMDS, 2-(2-bromoethyl)-
1,3-dioxolane, THF, DMPU, 0 → 25 ◦C; LHMDS, allyl bromide, 0 →
25 ◦C. (b) NH2OH.HCl, CH3CN, H2O, 60 ◦C. (c) Chloramine-T, 5 ◦C.
Next we examined the impact of the introduction of a geminal
2,5-difluorophenyl group onto sulfone 7 which would lead eventu-
ally to isoxazoline 14 (Scheme 4). The required dioxolane 12 for the
cycloaddition sequence was prepared by sequential alkylations of
2,5-difluorobenzyl phenyl sulfone 1112 using LHMDS as base, in
an analogous manner to 7. In this instance, the addition of DMPU
was found to be beneficial to obtaining high conversion in the
second alkylation with allyl bromide and the sterically congested
tertiary sulfone 12 could be isolated in 74% yield. Exposure
of 12 to hydroxylamine hydrochloride in a 1 : 1 CH3CN–H2O
mixture at 60 ◦C then allowed deprotection and formation of the
oximes 13 (1 : 1 E : Z) within 4 h. These underwent oxidation
and [3 + 2] ring closure on addition of chloramine-T, with a
preparative run at 25 ◦C affording isoxazoline 14 in 84% yield.
Strikingly, the cycloaddition diastereoselection increased with the
introduction of the additional geminal 2,5-difluorophenyl moiety,
with isoxazoline 14 formed with up to 99% ds (Table 1). The
relative stereochemistry of 14 was secured by NOE studies (see
Supporting Information†) and is consistent with a cycloaddition
proceeding by way of TS-2 (Scheme 4).13
Fig. 2 X-Ray structure of cyclohexanone 2 (CCDC 236716).
We were intrigued as to whether the conformational bias
apparently operative in the hydride reductions of 2 and 3 could
be extended to control the diastereoselection of a 5-hexenyl
intramolecular nitrile oxide–olefin cycloaddition.8,9 Initially we
examined the cycloaddition sequence to prepare isoxazoline 9
(Scheme 3) in which the stereo-induction afforded by an isolated
phenylsulfonyl group was evaluated. The required cycloaddition
precursor 7 was prepared by one-pot dialkylation of phenyl-
methylsulfone using LHMDS in successive deprotonations with
2-(2-bromoethyl)-1,3-dioxolane and allyl bromide as the required
alkylating agents. Although overalkylation was a significant com-
petitive pathway,10 the required sulfone 7 could be isolated in 38%
yield following chromatography. Exposure of 7 to hydroxylamine
hydrochloride in a 1 : 1 CH3CN–H2O mixture at 60 ◦C then
allowed dioxolane deprotection and formation of the oximes 8 (1 :
1 E : Z), avoiding the necessity for isolation of the intermediate
primary aldehyde. These oximes were also not isolated but directly
oxidised with chloramine-T11 to effect nitrile oxide formation and
[3 + 2] dipolar cycloaddition. Temperature was found to impact
the observed diastereoselection in the formation of isoxazolines
The additivity of conformational energies or A values (−DG◦) in
geminally substituted cyclohexanes is generally unreliable due to
Table 1 Temperature dependence of diastereoselection in cycloadditions
to form 9 and 14
Cycloadduct
Temp/◦Ca
Diastereoselectivityb
9
5
25
50
5
25
50
93 : 7
90 : 10
90 : 10
99 : 1
98.5 : 1.5
96 : 4
9
◦
9
9 and 10 (Table 1) and a preparative run at 5 C afforded 93 : 7
14
14
14
ds (85% yield). The relative stereochemistries of these isoxazolines
were secured by NOE studies (see Supporting Information†) and
are consistent with the phenylsulfonyl group occupying a pseudo-
equatorial orientation in preferred transition state TS-1, thereby
leading to 9 as the major diastereomeric product (Scheme 3).
a Temperature at which chloramine-T was added. b Determined by reverse
phase HPLC analysis of unpurified reaction mixtures.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1806–1810 | 1807
©

View Article Online
bias toward conformer 16b becomes evident. The alternative
cyclohexane chair conformation 16a engenders a number of
energetically unfavorable interactions, notably the oxygen-inside
interaction with the syn axial hydrogens required for the sulfone to
occupy the axial orientation. This interaction has been attributed
a value of ca. 1.6 kcal mol−1.16 Eclipsing interactions of the two
aryl rings are also inevitable if the equatorial phenyl ring in 16a
is to maintain the otherwise preferred parallel alignment with the
symmetry plane of the cyclohexanone ring.18 A similar argument
can be applied to rationalize the cycloaddition diastereoselection
to form 14.
In summary, we have found that cyclohexanone 2 and acryloni-
trile 5 undergo stereocontrolled metallohydride reductions consis-
tent with the population of conformers in solution having a strong
bias for the 4-chlorophenylsulfonyl moiety to sit equatorial, with
the 2,5-difluorophenyl substituent axial. This conformational bias
is supported by the observed single crystal X-ray crystallographic
structures7 of cyclohexanone 2 and cis-cyclohexanol 4 and was
found to extend to controlling the relative stereochemical outcome
of an intramolecular nitrile–oxide olefin cycloaddition, affording
isoxazoline 14 with up to 99% ds. The geminal aryl, arylsulfonyl
moiety can thus be considered to behave analogously to the well-
documented conformational biasing ability of a tert-butyl group.
Further examples and applications of this stereochemical control
element encountered in our studies will be reported in due course.
Scheme 4 Reagents and conditions: (a) LHMDS, 2-(2-bromoethyl)-
1,3-dioxolane, THF, −10 → 25 ◦C; LHMDS, allyl bromide, THF, DMPU,
−10 → 25 ◦C. (b) NH2OH.HCl, CH3CN, H2O, 60 ◦C. (c) Chloramine-T,
25 ◦C.
Experimental
steric interactions of the two substituents not otherwise present
in the monosubstituted parents. A well-documented example
is provided by 1-methyl-1-phenylcyclohexane,14 where the axial
phenyl-equatorial methyl is preferred by 0.32 kcal mol−1, in spite
of the individual conformational values for methyl (1.74 kcal
mol−1 at 300 K)14a and phenyl (2.87 kcal mol−1 at 173 K).15 The
A value for phenylsulfonyl is reported to be 2.94 kcal mol−1 at
298 K,16 comparable to that of phenyl. Considering the effect
of the introduction of a phenylsulfonyl group onto the parent
4-phenylcyclohexanone conformers 15a and 15b (Scheme 5), in
which conformer 15a is favored,14,17 the origins of the underlying
Commercially available materials were used without further
purification. Full preparative and characterisation data for 2, 5
and 6 will be reported elsewhere.
trans-4-[4-(Chlorophenyl)sulfonyl]-4-(2,5-
difluorophenyl)cyclohexanol 3
To a stirred solution of sodium borohydride (3.9 g, 104 mmol) in
EtOH (200 mL) at −10 ◦C was added 4-[4-chlorophenyl)sulfonyl]-
4-(2,5-difluorophenyl)cyclohexanone 2 (20.0 g, 52.0 mmol) por-
tion wise such that T ≤ 0 ◦C. The reaction mixture was aged
for 2 h and then quenched by addition of 2 M HCl (50 mL).
Water (250 mL) and IPAc (200 mL) were added and the layers cut.
Filtration of the organic layer through a silica gel plug (using IPAc
as eluant) afforded the title compound (18.8 g, 94%) as a white
solid as a 95 : 5 mixture with 4. A diastereomerically pure sample
of 3 could be obtained by further crystallisation from IPAc; mp
◦
1
156–158 C (from IPAc); H NMR (400 MHz, d6-DMSO, 328
K) d 7.65–7.55 (2 H, m), 7.43–7.38 (2 H, m), 7.35–7.28 (1 H, m),
7.20–7.12 (2 H, m), 4.41 (1 H, d, J 5.0), 3.59–3.45 (1 H, m), 2.79–
2.65 (2 H, m), 2.05–1.85 (4 H, m), 1.05–0.95 (2 H, m); 13C NMR
(100 MHz, d6-DMSO, 328 K) d 159.2 (d, J 245), 158.6 (d, J 248),
140.0, 134.3, 132.1, 129.5, 121.7 (dd, J 7.0 and 11.4), 119.4 (dd,
J 4.0 and 25.0), 119.2 (dd, J 9.0 and 17.0), 118.9–118.4 (m, 2C),
70.7 (d, J 4.0), 68.2, 31.6, 28.7 (d, J 6.0); HRMS (ES) Calcd. for
C18H21ClF2NO3S (M + NH4) 404.0899. Found 404.0891.
cis-4-[4-(Chlorophenyl)sulfonyl]-4-(2,5-
difluorophenyl)cyclohexanol 4
To a stirred solution of 4-[4-chlorophenyl)sulfonyl]-4-(2,5-di-
Scheme 5
fluorophenyl)cyclohexanone 2 (2.0 g, 5.2 mmol) in THF (14 mL)
1808 | Org. Biomol. Chem., 2006, 4, 1806–1810
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
◦
R
at −70 C was added L-Selectrideꢀ (1 M in THF, 8.3 mL, 8.3 mmol)
dropwise over 30 min. The mixture was aged for 2 h and then
NaCl (1.5 g) in H2O (8.4 mL) added dropwise followed by 48%
NaOH (8 drops). 27% Aq. H2O2 (3.4 mL) was then added at
T < 5 ◦C and the mixture warmed to ambient and aged 30 min.
A solution of sodium metabisulfite (1.2 g) in H2O (10.4 mL)
was then added dropwise (T < 25 ◦C) and aged for 1 h. IPAc
(24 mL) was added and the layers cut. Brine (10% aq., 14 mL)
was added and the layers were separated. The organics were
concentrated in vacuo and then passed through a plug of silica gel
using CH2Cl2 and IPAc as eluants. Evaporation in vacuo afforded
the title compound as a 90 : 10 diastereomeric mixture of 4 and
3 respectively (1.97 g, 98%). Crystallisation from IPAc–heptane
afforded diastereomerically pure 4 for analytical purposes (Found:
C, 55.83;◦H, 4.36. C18H17ClF2O3S requires C, 55.89; H, 4.43%); mp
(50 : 50 heptane–IPAc) afforded the title compound as a 90 : 10
mixture of 9 and 10, respectively (0.76 g, 85% combined yield).
Data for 9: Found 58.60; H, 5.67; N 5.21. C13H15NO3S requires C,
58.85; H, 5.70; N 5.28%; mp 128–129 ◦C (from IPAc–heptane); 1H
NMR (400 MHz, C6D6) d 7.72–7.68 (2 H, m), 7.03–6.93 (3 H, m),
3.89 (1 H, dd, J 8.3 and 10.4), 3.22 (1 H, dd, J 8.3 and 10.3), 2.51
(1 H, m), 2.43–2.36 (1 H, m), 2.22–2.11 (1 H, m), 2.05–1.98 (1 H,
m), 1.91–1.84 (1 H, m), 1.35 (1 H, ddt, J 1.2 and 5.2 and 13.4 Hz),
1.32–1.17 (1 H, m), 1.16–1.06 (1 H, m); 13C NMR (100 MHz,
C6D6) d 156.1, 137.8, 133.5, 129.2, 129.1, 73.0, 61.2, 46.7, 30.9,
25.8, 22.9.
2-[3-Phenyl-3-(2,5-difluorophenylsulfonyl)hex-5-en-1-yl]-1,3-
dioxolane 12
1
183–184 C (from IPAc–heptane); H NMR (400 MHz, CDCl3)
A slurry of 2,5-difluorobenzyl phenyl sulfone 1112 (4.0 g,
14.9 mmol) in THF (10 mL) was cooled to −10 ◦C and LHMDS
(1 M in THF, 19.4 mL, 19.4 mmol) added over 5 min. The
resultant solution was aged 10 min. and then 2-(2-bromoethyl)-
1,3-dioxolane (2.98 g, 16.4 mmol) added dropwise over 2 min.
The solution was aged at ambient for 12 h and then quenched
with H2O (100 mL). IPAc (100 mL) was added and the layers
partitioned. The organic layer was evaporated in vacuo and the
residue dissolved in THF (12 mL) and DMPU (4 mL). This
solution was cooled to −10 ◦C and LHMDS (1 M in THF,
19.4 mmol, 19.4 mL) added over 5 min. After aging 5 min, allyl
bromide (1.81 mL, 20.9 mmol) was added and the mixture warmed
to ambient and aged 0.5 h prior to quenching with H2O (150 mL).
IPAc (150 mL) was added and the layers partitioned. The organic
layer was washed with H2O (2 × 150 mL), concentrated in vacuo
and the residue purified by silica gel chromatography (70 : 30
heptane–IPAc) to afford the title compound (4.51 g, 74%) as a
colourless oil: 1H NMR (400 MHz, CDCl3) d 7.61–7.55 (1 H, m),
7.42–7.35 (4 H, m), 7.05–6.95 (2 H, m), 6.85–6.75 (1 H, m), 5.97–
5.86 (1 H, m), 5.28–5.13 (2 H, m), 4.88 (1 H, t, J 4.4), 3.98–3.91
(2 H, m), 3.90–3.82 (2 H, m), 3.30–3.23 (1H, m), 3.15–3.05 (1 H,
m), 2.55–2.45 (2 H, m), 2.05–1.95 (1 H, m), 1.60–1.50 (1 H, m);
13C NMR (100 MHz, CDCl3) d 158.3 (2C, d, J 246 Hz), 135.2,
133.9, 132.3, 130.3, 128.4, 123.5 (dd, J 8.0 and 13), 119.4, 118.7
(dd, J 4.3 and 26), 117.9 (dd, J 8.5 and 29), 117.5 (dd, J 9.9 and
24), 103.8, 71.6, 65.0 (d, J 6.3), 36.3 (d, J 6.9), 28.3, 25.7 (d, J
5.4); HRMS (ES) Calcd. for C21H22F2O4NaS (M + Na) 431.1105.
Found 431.1100.
d 7.42–7.38 (4 H, s), 7.14–7.02 (2 H, m), 6.92–6.83 (1 H, m), 3.94
(1 H, t, J 2.7), 2.90–2.20 (4 H, m), 1.92–1.78 (3 H, m), 1.52–1.35
(2 H, m); 13C NMR (100 MHz, CDCl3) d 159.1 (d, J 245), 158.6
(d, J 242), 140.8, 133.6, 131.7, 128.8, 121.0, 119.1 (dd, J 4.2 and
25.3), 118.2 (dd, J 9.9 and 24.0), 70.9 (d, J 4.1), 63.9, 29.3, 23.7
(d, J 6.8).
2-[3-(Phenylsulfonyl)hex-5-en-1-yl]-1,3-dioxolane 7
A stirred solution of phenyl methyl sulfone (3.0 g, 19.2 mmol) in
THF (9 mL) and DMPU (3.6 mL) was cooled to 0 ◦C. LHMDS
(1 M in THF, 21.1 mL, 21.1 mmol) was added dropwise over 15 min
at T < 2 ◦C. The resulting solution was stirred for 10 min and
then 2-(2-bromoethyl)-1,3-dioxolane (2.42 mL, 20.1 mmol) added
dropwise over 10 min at T < 5 ◦C. The mixture was then warmed
to ambient and aged 1 h before recooling to −20 ◦C. LHMDS (1 M
in THF, 21.1 mL, 21.1 mmol) was added dropwise over 5 min and
the mixture aged for 10 min. Allyl bromide (2.50 mL, 28.8 mmol)
was added in one portion and the mixture warmed to ambient
and aged 1 h. H2O (45 mL) and IPAc (135 mL) were added and
the layers partitioned. The organics were washed with H2O (3 ×
20 mL) and then evaporated in vacuo. Purification by silica gel
chromatography (34 : 66 heptane–ethyl acetate) afforded the title
compound (2.14 g, 38%) as a pale yellow oil (Found C, 60.75; H,
6.80. C15H20O4S requires C, 60.79; H, 6.80%); 1H NMR (400 MHz,
CDCl3) d 7.79–7.76 (2 H, m), 7.56–7.53 (1H, m), 7.47–7.43 (2 H,
m), 5.63 (1 H, ddt, J 7.1 and 9.8 and 18.9), 4.99–4.92 (2 H, m), 4.66
(1 H, t, J 4.0), 3.80–3.62 (4 H, m), 3.09–3.02 (1H, m), 2.55–2.46
(1H, m), 2.27–2.17 (1H, m), 1.89–1.79 (1 H, m), 1.79–1.69 (1H, m),
1.69–1.58 (2H, m); 13C NMR (100 MHz, CDCl3) d 137.7, 133.7,
133.3, 129.1, 128.8, 118.3, 103.6, 64.8, 63.2, 32.2, 30.1, 21.4.
cis-5-Phenyl-5-(2,5-difluorophenylsulfonyl)-3,3a,4,5,6,7-
hexahydro-2,1-benzisoxazole 14
To a stirred solution of sulfone 12 (2.74 g, 6.7 mmol) in CH3CN
(5 mL) and H2O (5 mL) was added NH2OH·HCl (0.93 g,
13.4 mmol) and the mixture aged at 60 ◦C for 4.5 h. After cooling
to 25 ◦C, chloramine T monohydrate (2.9 g, 12.8 mmol) was added
and the mixture aged for 1 h at this temperature. The CH3CN was
removed in vacuo and then IPAc (150 mL) and aq. NaOH (2 M,
50 mL) added. The layers were partitioned and the organic washed
with aq. NaOH (2 M, 50 mL) and H2O (50 mL). The organics were
concentrated in vacuo and the residue crystallised from IPAc and
heptane to afford the title compound (2.13 g, 84%) as a white
solid (Found: C, 60.30; H, 4.52, N 3.58. C19H17F2NO3S requires
C, 60.47; H, 4.54, N 3.71); mp 188–190 ◦C (from IPAc–heptane);
cis-5-(Phenylsulfonyl)-3,3a,4,5,6,7-hexahydro-2,1-benzisoxazole 9
To a stirred solution of sulfone 7 (1.0 g, 3.4 mmol) in CH3CN
(12.5 mL) and H2O (12.5 mL) was added NH2OH·HCl (0.47 g,
6.8 mmol) and the mixture aged at 60 ◦C for 2 h. The mixture
was then cooled to 5 ◦C and chloramine-T trihydrate (1.81 g,
6.4 mmol) added in one portion. After aging at this temperature
for 2.5 h, H2O (25 mL) was added and the CH3CN evaporated in
vacuo. IPAc (60 mL) was added and the layers partitioned. The
organics were washed with NaOH (0.66 M, 12 mL), H2O (60 mL)
and evaporated in vacuo. Purification by silica gel chromatography
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1806–1810 | 1809
©

View Article Online
1H NMR (600.1 MHz, d7-DMF, 350 K) d 7.77 (1 H, m), 7.58 (2 H,
m), 7.52 (2 H, m), 7.35–7.30 (2 H, m), 7.16 (1 H, ddd, J 13.6 and
9.1 and 4.9), 4.50 (1 H, dd, J 9.8 and 7.9), 3.87 (1 H, dd, J 10.2 and
7.9), 3.32–3.22 (2 H, m), 3.17 (1 H, m), 2.84 (1 H, m), 2.23–2.15
(2 H, m), 2.09 (1 H, td, J 12.5 and 1.9); 13C NMR (150.9 MHz,
d7-DMF, 350 K) d 159.38 (dd, J 247.8 and 2.4), 158.96 (dd, J 241.1
and 2.4), 157.36, 135.35, 134.61, 130.44, 129.17, 120.74 (dd, J 11.6
and 7.3), 119.37 (dd, J 26.2 and 4.3), 118.90 (dd, J 29.3 and 8.5),
118.59 (dd, J 23.8 and 10.4),73.03, 70.62 (d, J 4.3), 45.33, 35.54
(d, J 6.1), 30.72 (d, J 7.3), 21.33.
(d) I. Churcher, K. Ashton, J. W. Butcher, E. E. Clarke, T. Harrison,
H. D. Lewis, A. P. Owens, M. R. Teall, S. Williams and J. D. J. Wrigley,
Bioorg. Med. Chem. Lett., 2003, 13, 179; (e) I. Churcher, S. Williams, S.
Kerrad, T. Harrison, J. L. Castro, M. S. Shearman, H. D. Lewis, E. E.
Clarke, J. D. J. Wrigley, D. Beher, Y. S. Tang and W. Liu, J. Med. Chem.,
2003, 46, 2275–2278.
2 In all instances, diastereoselection was determined by reverse phase
HPLC assay of unpurified samples of the reaction mixtures.
3 H. C. Brown and S. Krishnamurthy, J. Am. Chem. Soc., 1972, 94, 7159.
4 P. T. Lansbury and R. E. MacLeay, J. Org. Chem., 1963, 28, 1940.
5 Over-reduction of the nitrile functionality in the product becomes a
significant competitive pathway at −10 ◦C.
6 E. L. Eliel, S. H. Wilen and L. N. Mander, Stereochemistry of Organic
Compounds, John Wiley & Sons, New York, 1994, 735 pp.
Crystal structure determination of compounds 2 and 4
7 Crystallographic data for 2 and 4 have been deposited at the Cambridge
Crystallographic Data Centre. 2: CCDC 236716; 4: CCDC 600829.
See Experimental section and Supporting Information for additional
details of X-ray structures of 2 and 4.
Single crystals of cyclohexanone 2 suitable for X-ray diffraction
were obtained from IPAc–heptane and for cyclohexanol 4 from
acetonitrile.
8 Reviews: (a) V. Jager and P. A. Colinas, in Synthetic Applications of
1,3-Dipolar Cycloaddition Chemistry Towards Heterocycles and Natural
Products, A. Padwa and W. H. Pearson, ed., John Wiley & Sons, New
Jersey, 2003, ch. 6; (b) A. P. Kozikowski, Acc. Chem. Res., 1984, 17,
410; (c) S. Kanemasa and O. Tsuge, Heterocycles, 1990, 30, 719.
9 Related diastereocontrol in an intramolecular nitrile oxide–olefin
cycloaddition has been reported: A. P. Kozikowski and K. E. Mal-
oneyHuss, Tetrahedron Lett., 1985, 26, 5759.
Crystal data for 2. C18H15ClF2O3S, M = 384.81, orthorhom-
˚
bic, a = 15.450(10), b = 10.490(7), c = 10.828(7) A, U = 1755(2)
3
−1
˚
A , T = 298(2) K, space group Pca2(1), Z = 4, l = 0.371 mm ,
13 312 reflections measured, 3373 unique (Rint = 0.1182) which
were used in the calculations. The final wR(F2) was 0.1225 (all
data). CCDC 236717. For crystallographic data in CIF or other
electronic format see DOI: 10.1039/b601647b.
10 The main by-product identified by LCMS and 1H NMR was allylation
of 7 to give the diallylated tertiary sulfone adduct.
11 A. Hassner and K. Rai, Synthesis, 1989, 57.
12 J. P. Scott, D. C. Hammond, E. M. Beck, K. J. M. Brands, A. J. Davies,
U.-H. Dolling and D. J. Kennedy, Tetrahedron Lett., 2004, 45, 3345.
13 Cycloaddition to from the diphenyl substititued isoxazoline 17 is
also highly diastereoselective (>96% ds) but we were unable to
unambiguously establish the relative stereochemistry by NOE studies
due to overlapping aryl C–H 1H NMR resonances.
Crystal data for 4. C18H17ClF2O3S, M = 386.83, monoclinic,
˚
a = 6.9274(5), b = 8.4158(6), c = 29.301(2) A, U = 1706.3(2)
3
−1
˚
A , T = 223(2) K, space group P21/c, Z = 4, l = 0.382 mm ,
17 593 reflections measured, 3513 unique (Rint = 0.0271) which
were used in the calculations. The final wR(F2) was 0.1054 (all
data). CCDC 600829. For crystallographic data in CIF or other
electronic format see DOI: 10.1039/b601647b
Acknowledgements
14 (a) N. L. Allinger and M. T. Tribble, Tetrahedron Lett., 1971, 35, 3259;
(b) E. L. Eliel and M. Manoharan, J. Org. Chem., 1981, 46, 1959–1962;
(c) H. De Beule, D. Tavernier and M. Anteunis, Tetrahedron, 1974, 30,
3573.
We thank Robert A. Reamer for NMR structure determination
and Thomas J. Novak for mass spectrometry analyses.
15 M. E. Squillacote and J. M. Net, J. Am. Chem. Soc., 1987, 109, 198.
16 E. Juaristi, V. Labastida and S. Autunez, J. Org. Chem., 2000, 65, 969.
17 The most favorable conformation of phenylcyclohexane, when the
phenyl group is equatorial, has the phenyl parallel to the symmetry
plane of the cyclohexane ring. With the phenyl axial, a perpendicular
arrangement to the symmetry plane is preferred, as the parallel
arrangement has a steric clash of the ortho-hydrogens with the syn-
axial hydrogens of the cyclohexane ring (ref. 14a).
18 Fluorine is well documented to have isosteric requirements compared
to hydrogen and thus should not impact this rationalization. See: C. G.
Wermuth, in The Practice of Medicinal Chemistry, ed. C. G. Wermuth,
Academic Press, London, 1996, 226 pp.
References and notes
1 (a) D. E. Shaw, J. D. Best, K. Dinnell, A. Nadin, M. S. Shearman, C.
Pattison, J. Peachey, M. A. Reilly, B. J. Williams, J. D. J. Wrigley and T.
Harrison, Bioorg. Med. Chem. Lett., submitted for publication; (b) I.
Churcher, D. Beher, J. D. Best, J. L. Castro, E. E. Clarke, A. Gentry,
T. Harrison, L. Hitzel, E. Kay, S. Kerrad, H. W. Lewis, P. Morentin-
Gutierrez, R. Mortishire-Smith, P. J. Oakley, M. Reilly, D. E. Shaw,
M. S. Shearman, M. R. Teall, S. Williams and J. D. J. Wrigley, Bioorg.
Med. Chem. Lett., 2006, 16, 280; (c) M. Teall, P. Oakley, T. Harrison,
D. E. Shaw, E. Kay, J. Elliot, U. Gerhard, J. L. Castro, M. Shearman,
R. G. Ball and N. N. Tsou, Bioorg. Med. Chem. Lett., 2005, 15, 2685;
1810 | Org. Biomol. Chem., 2006, 4, 1806–1810
This journal is
The Royal Society of Chemistry 2006
©