E- and Z-Stereoselectivity in the preparation of enamides from glycidyl sulfonamides and carbamates

Article abstract of DOI:10.1039/c1ob06569f

Treatment of glycidyl sulfonamides with LDA delivers the corresponding enesulfonamide with good selectivity for the E-isomer, whereas the corresponding carbamates exhibit selectivity for the Z-enecarbamate. An E1cB elimination mechanism proceeding from a

Full text of DOI:10.1039/c1ob06569f

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2012, 10, 509
www.rsc.org/obc
COMMUNICATION
E- and Z-Stereoselectivity in the preparation of enamides from glycidyl
sulfonamides and carbamates†
a
Jack A. Brown,‡a,b Vijay Chudasama,§a Melvyn E. Giles,a Duncan M. Gill,*¶ Philip S. Keegan,a
a
William J. Kerr,b Rachel H. Munday,a Karen Griffinꢀ and Andrew Wattsa
Received 13th September 2011, Accepted 10th November 2011
DOI: 10.1039/c1ob06569f
Treatment of glycidyl sulfonamides with LDA delivers the
corresponding enesulfonamide with good selectivity for the
E-isomer, whereas the corresponding carbamates exhibit
selectivity for the Z-enecarbamate. An E1cB elimination
mechanism proceeding from a substrate–base chelate com-
Scheme 1
plex is advanced as rationalisation of the latter set of Z-
Table 1 Comparison of substratesa
selective outcomes.
Entry
Substrate
R1
R2
Conversion E : Z
Enamides are valuable synthetic intermediates,1 and their stere-
oselective preparation has attracted much recent attention from
the synthetic community. In this regard, new methods for access
to this functional unit have been introduced lately based on, inter
alia, cross coupling processes;2 hydroamination of alkynes;3 and
the Peterson reaction.4 In relation to this and based on our on-
going drug discovery programmes, we wished to prepare a range
of enamines of general structure 2, possessing different pendant
nitrogen groups. We envisaged that a base-induced epoxide–allylic
alcohol rearrangement5 of the appropriately N-functionalised
glycidyl amine 1 would represent an extremely concise entry
into this series of compounds from simple starting materials
(Scheme 1). Excepting a single example,6 there have been no
reports on this particular reaction variant with acyclic substrates.7,8
Herein, we report our studies leading to substrate dependent
stereoselective access to the desired E- and Z-product isomers.
Treatment of glycidylamine derivatives 1a–f with excess LDA
at -78 ◦C in 2-methyltetrahydrofuran (2-MeTHF), followed by
1
2
3
4
5
6
1a
1b
1c
1d
1e
1f
Ts
Ts
Boc
Boc
Boc
Boc
Me
Ph
Me
100
100
100
100
20
90 : 10
90 : 10
50 : 50
25 : 75
15 : 85
25 : 75
Ph
tBu
p-MeOC6H4
100
a Conditions: LDA (2.4 eq), 2-MeTHF, -78 ◦C to -60 ◦C, 2 h; NH4Cl (aq).
hindered substrate 1e, conversion to 2 was quantitative. However,
when we came to examine the selectivity for the resultant olefin
geometry, we made some notable observations, which led us to
consider this reaction in more detail. Glycidyl sulfonamides 1a
and 1b gave an identical E : Z ratio favouring the E-isomer. This
is consistent with the syn-elimination mechanism advanced by
Thummel and Rickborn,9 illustrated in Fig. 1, where the lithium
counterion chelates the epoxide oxygen and the amide anion.
Transition state TS-A, leading to the E-product, is favoured due to
the lack of an eclipsing interaction (cf. TS-B) between the epoxide
methylene and the sulfonamide unit.
◦
warming to -60 C over 2 h and an aqueous extractive workup,
gave the results summarised in Table 1. Other than for the sterically
aPharmaceutical Development, AstraZeneca R&D Charnwood, Bakewell
Road, Loughborough, LE11 5RH, UK
bDepartment of Pure and Applied Chemistry, University of Strathclyde, 295
Cathedral Street, Glasgow, UK, G1 1XL
† Electronic supplementary information (ESI) available: Experimental
procedures and characterisation data for compounds 1 and 2. See DOI:
10.1039/c1ob06569f
‡ Current Address: GlaxoSmithKline, Medicines Research Centre, Gun-
nels Wood Road, Stevenage, SG1 2NY, UK
§ Current Address: Department of Chemistry, University College London,
20 Gordon Street, London, WC1H OAJ, UK
¶ Current Address: Department of Chemical and Biological Sciences, Uni-
versity of Huddersfield, Queensgate, Huddersfield, HD1 3DH, UK. Email:
d.m.gill@hud.ac.uk; Fax: +44(0)1484472182; Tel: +44(0)1484422180
Fig. 1 Transition states for syn-elimination of 1a and 1b.
In contrast, glycidyl carbamates gave much higher proportions
of the Z-isomer under the same conditions. Comparison of
ꢀ Current Address: Department of Chemistry, University of Warwick,
Gibbet Hill Road, Coventry, CV4 7AL, UK
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 509–511 | 509
©

View Article Online
Table 2 Survey of conditions for reaction of 1f
Entry Solvent Base Equivalents Conversion E : Z
1
2
3
4
5
6
7
8
2-MeTHF LDA
2-MeTHF LDAa
2.4
2.4
2.4
2.4
2.4
1.0
1.5
2.0
100
100
100
100
100
50
25 : 75
80 : 20
25 : 75
60 : 40
60 : 40
25 : 75
25 : 75
25 : 75
20 : 80
25 : 75
60 : 40
75 : 25
25 : 75
THF
Toluene
Et2O
LDA
LDA
LDA
2-MeTHF LDA
2-MeTHF LDA
2-MeTHF LDA
75
100
100
100
100
100
100
9
2-MeTHF Mg(NiPr2)2 2.4
10
11
12
13
2-MeTHF LiTMP
2-MeTHF LiNEt2
2.4
2.4
Fig. 2 anti-Elimination in the presence of excess DMPU.
2-MeTHF LiNc-C4H8 2.4
explicable, a clear trend is apparent in the base comparison,
with increasing base size giving enhanced Z-selectivity. Taken
with the results in Table 1, it suggests that formation of the E-
product is increasingly disfavoured by a steric interaction between
R groups on the substrate and those connected to the base reagent
nitrogens. Furthermore, the selectivity appears to be kinetic, not
thermodynamic in origin: subjection of pure E-2c12 to the reaction
conditions did not result in the formation of observable levels of
Z-2c.
The results, showing the selective formation of the Z-
enecarbamate, are consistent with the formation of a chelated
complex between substrate and base (I or II, Scheme 2), followed
by attack of a second equivalent of base upon this intermediate, to
deliver a product that does not then dissociate the complexed base
prior to workup. Conformation I would be favoured over II due
to the less desirable flagpole-like steric interactions in the latter.
Attack of base to remove a proton from I could proceed along
trajectory a or b, removing the pseudo-equatorial proton (leading
to the E-product), or the pseudo-axial proton (leading to the Z-
alkene), respectively. Orbital overlap between the breaking C–O
bond and the abstracted proton on path a is superior to that of
path b, so this would normally be favoured on stereoelectronic
grounds. This would give rise to the E-product via an E1cB
(bordering on E2) mechanism, proceeding through intermediate
III. However, returning to intermediate I, the R group on the
2-MeTHF LiN(tamyl)2 2.4
a DMPU (10 molar equivalents) added.
1c–e (entries 3–5) illustrates that formation of the Z-isomer is
proportional to the steric bulk of the carbamate N-substituent.
That the effect is steric, and not electronic, in origin was confirmed
by comparing phenyl case 1d with that of the p-methoxy analogue
1f; exactly the same E : Z ratio was obtained in each case (entries
4 and 6).
We hypothesised that in the latter glycidyl carbamate cases some
additional chelation was responsible for the unusual stereoselec-
tivity, and chose 1f as a vehicle for further investigation. As can
be seen in Table 2, addition of 10 equivalents of DMPU (entry
2) reversed the E : Z selectivity to 80 : 20. As the presence of
excess DMPU would be expected to disrupt any metal–substrate
chelation, an alternative anti-elimination10 may be propounded,
where the carbamate subtends the methylene group in TS-D,
leading to a higher energy transition state for formation of the
Z-isomer, Fig. 2.
We next investigated reagent stoichiometry, and found that a
twofold excess of base was required for full conversion (Table
2, entries 6–8). Alternative solvents (entries 3–5) and bases
(9–13)11 were also surveyed. Whilst the solvent pattern is less
Scheme 2 Proposed course of elimination via substrate–base complex.
510 | Org. Biomol. Chem., 2012, 10, 509–511 This journal is The Royal Society of Chemistry 2012
©

View Article Online
adjacent nitrogen is closer to the proton which would be abstracted
via path a. Accordingly, as this group, or the base, is increased
in size, attack along this trajectory becomes disfavoured. On the
other hand, abstraction of the pseudo-axial proton (path b) would
give a carbon–lithium bond with poor orbital overlap towards the
breaking C–O bond (structure V). Rotation to align the bonds for
maximum orbital overlap would place the nitrogen substituent in a
syn relationship with the epoxide methylene group, and ultimately
deliver the Z-olefin by an unambiguously E1cB pathway. It is
also worth noting that the Z-product would also be obtained if
deprotonation took place on chelate II via trajectory c, driven by
both steric factors and optimal orbital overlap with the breaking
C–O bond.
An alternative pathway by which formation of Z-2 would
be favoured involves directed a-metallation of the epoxide13
leading to formation of a carbenoid species, followed by H-
migration (Scheme 3). It is not obvious why the latter process
should be selective to formation of 2 (Ha migration) over the
alternative aldehyde product 3 (Hb migration). Nevertheless we
took steps to exclude this possibility. We prepared deutero-labelled
d-1f by the sequence illustrated in Scheme 4, subjected this
species to LDA under the now standard reaction conditions, and
observed complete incorporation of deuterium in both isomers
of the product d-2f, thus ruling out any mechanism involving a-
metallation.
Scheme 4 Preparation of deuterated analogue.
of observations with a range of substrates and a selection of
bases, both possessing varying degrees of steric bulk, an E1cB
mechanism, proceeding from a chelated intermediate, has been
proposed to account for this phenomenon. We anticipate that
these accumulated outcomes will be valuable to others within the
preparative community, and especially in the planning of access to
enamides of specific double bond geometry. Further mechanistic
studies relating to this interesting reaction will be reported in due
course.
Acknowledgements
We wish to thank Drs P. Edwards (University of Manchester), S.
Eyley, G. Howell and R. Woodward (AZ) for insightful comments
during the preparation of this manuscript.
Notes and references
1 D. R. Carbery, Org. Biomol. Chem., 2008, 6, 3455.
2 J. R. Dehli, J. Legros and C. Bolm, Chem. Commun., 2005, 973.
3 L. J. Goossen, J. E. Rauhaus and G. Deng, Angew. Chem., Int. Ed.,
2005, 44, 4042.
4 A. Fu¨rstner, C. Brehm and Y. Cancho-Grande, Org. Lett., 2001, 3,
3955.
5 J. K. Crandall and M. Apparu, Organic Reactions, 1983, 29, 345.
6 T. Kim, A. Kim and Y. J. Park, Eur. J. Org. Chem., 2002, 493.
7 For an example of a cyclic substrate, see: D. M. Hodgson, T. J. Miles and
J. Witherington, Tetrahedron, 2003, 59, 9729. For a related elimination
in a cyclic system, see: M. Giles, M. S. Hadley and T. Gallagher, Chem.
Commun., 1990, 831.
8 It is also worth noting that Baldwin et al. have studied base-mediated
isomerization of equivalent epoxy isonitriles: J. E. Baldwin, D. Chen
and A. T. Russell, Chem. Commun., 1997, 2389.
9 R. P. Thummel and B. Rickborn, J. Am. Chem. Soc., 1970, 92, 2064.
10 K. M. Morgan and S. Gronert, J. Org. Chem., 2000, 65, 1461.
11 LiHMDS, KHMDS, and phosphazene P2 and P4 bases were also
surveyed and gave little or no reaction.
Scheme 3 Alternative carbenoid mechanism.
12 Obtained by DIBALH reduction of the corresponding ester; D. A.
Alonso, E. Alonso, C. Najera, D. J. Ramon and M. Yus, Tetrahedron,
1997, 53, 4835.
13 V. Capriati, S. Florio and R. Luisi, Chem. Rev., 2008, 108, 1918; A.
Ramirez and D. B. Collum, J. Am. Chem. Soc., 1999, 121, 11114.
In conclusion, upon treatment with amide bases, glycidyl
carbamates undergo elimination to give enecarbamates with se-
lectivity for the Z-isomer, in contrast to the E-selectivity observed
for the corresponding glycidyl sulfonamides. Based on a series
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 509–511 | 511
©