The design of efficient and selective routes to a key 1,4-cis-substituted cyclohexylamide intermediate

Article abstract of DOI:10.1016/j.tetlet.2010.03.083

This Letter describes the synthetic challenges in synthesising key 1,4-cis-substituted cyclohexylamide intermediate 1 for our research programme. Five different routes address the major issues of selectivity to afford the cis product in isomerically pure

Full text of DOI:10.1016/j.tetlet.2010.03.083

Tetrahedron Letters 51 (2010) 2846–2848
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
The design of efficient and selective routes to a key 1,4-cis-substituted
cyclohexylamide intermediate
Christopher Barfoot a, Gerald Brooks b, David T. Davies b, John Elder a, Ilaria Giordano a, Alan Hennessy a,
Graham Jones b, Roger Markwell b, Michael McGuire c, Timothy Miles a, Neil Pearson d, Grant Spoors c,
Ravinder Sudini c, Hengxu Wei c, Jeffery Wood c
,
*
a Antibacterial Discovery Performance Unit, Infectious Diseases CEDD, GlaxoSmithKline, Gunnels Wood Road, Stevenage SG1 2NY, UK
b Antibacterial Discovery Performance Unit, Infectious Diseases CEDD, GlaxoSmithKline, Third Avenue, Harlow CM19 5AW, UK
c Synthetic Chemistry, GlaxoSmithKline, 709 Swedeland Road, King of Prussia, PA 19406, USA
d Antibacterial Discovery Performance Unit, Infectious Diseases CEDD, GlaxoSmithKline, 1250 South Collegeville Road, Collegeville, PA 19426, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
This Letter describes the synthetic challenges in synthesising key 1,4-cis-substituted cyclohexylamide
intermediate 1 for our research programme. Five different routes address the major issues of selectivity
to afford the cis product in isomerically pure form and in high yield. Major purification issues were also
encountered upon scaling some of the routes. The merits of the diverse routes are assessed and the rea-
soning given for which one was ultimately used for large-scale synthesis of 1.
Received 21 December 2009
Revised 9 March 2010
Accepted 19 March 2010
Available online 27 March 2010
Ó 2010 Elsevier Ltd. All rights reserved.
During the synthesis of a series of compounds for an anti-bacte-
rial medicinal chemistry programme, we encountered the need for
bulk supplies of the key 1,4-cis-substituted cyclohexylamide 1.1
This compound or its immediate precursors were unknown in
the literature.1 This was unlike its des-hydroxy analogue 2, which
was readily obtained by standard amide formation from the com-
mercially available carboxylic acid 3 (Fig. 1).2
quence could not be adequately scaled up. Chromatography of
the close running cis and trans isomers was extremely difficult
and made large-scale processing impractical. Recrystallisation of
this mixture also proved fruitless and after considerable efforts at
optimisation, this route was abandoned.
Synthetic Approach A
For initial supplies of compound 1, we decided to use ketone 4
as a starting material as shown in Scheme 1. Conversion into the
hydroxy-amide by reaction with KCN and then treatment with a
strong acid gave 5 with poor selectivity. Subsequent re-protection
and purification by chromatography gave 1.
This synthetic approach was attractive since it required only a
few steps from a readily available starting material. The overall
yield was modest (8%), but the major problem was that the se-
H2N
O
HO
O
(1) KCN, EtOAc, H2O
rt, 20 h
(2) conc. HCl, rt, 2 h
Crude:
~1:1 isomeric
mixture
NH2
5
NHBoc
4
(1) Boc2O, NaOH,
1,4-dioxane, H2O,
rt, 2 h
(2) careful
chromatography
H2N
O
O
H2N
HO
HO
O
H2N
HO
O
NHBoc
NHBoc
NHBoc
8%
overall
1
2
3
Figure 1. Structure of 1 and related compounds.
NHBoc
1
* Corresponding author. Tel.: +44 1438762654.
E-mail address: alan.j.hennessy@gsk.com (A. Hennessy).
Scheme 1. Synthetic approach A.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.03.083

C. Barfoot et al. / Tetrahedron Letters 51 (2010) 2846–2848
2847
In order to obtain larger supplies of compound 1, we decided to
employ a different approach in which we hoped to fix the stereo-
chemistry and therefore avoid the tedious chromatography step.
This approach relied on a standard Diels–Alder strategy to reach
substituted diene 8 and then a hetero Diels–Alder reaction to give
9. This approach has some literature precedent and seemed to us
an ideal way of forming compound 1.3
As can be seen in Scheme 2, the initial sequence to diene 8 pro-
ceeded in excellent yield, unfortunately, the subsequent hetero
Diels–Alder reaction occurred in a very moderate 37% yield to give
compound 9. The hetero Diels–Alder reaction of ester 11 worked
considerably better than that of amide 8, but then, two more steps
were required to access common intermediate 9. Subjecting 9 to
forcing hydrogenation conditions followed by Boc-protection gave
the desired cis stereochemically pure compound 1.
Although synthetic approaches B and C were successful in their
aims to give a scalable route, one problem was that the forcing
hydrogenation conditions resulted in partial hydrogenolysis of
the tertiary alcohol formed from 9. Therefore small amounts (1–
10%) of compound 2 were contaminating the product and leading
to downstream chemistry issues. This impurity was not removable
by recrystallisation but rather required intensive chromatography.
As with synthetic approach A, this was not suitable for a large-scale
processing.
Another approach which would control the cis stereochemistry
is described in Scheme 3 and relies on a hypervalent iodine in-
duced cyclisation.4
Starting from the commercially available 13, the key intermedi-
ate 15 was produced using standard chemistry. To our delight,
treatment of 15 with diacetoxy iodobenzene induced the desired
Synthetic Approach B
H2N
(1) Pd/C, H2
1,4-dioxane, H2O,
40 ºC, 55 psi, 68 h
(2) Boc2O, NaOH
Acrylamide,
CONH2
OAc
CONH2
CONH2
Bu4NIO4,
HO
hydroquinone
toluene
O
KOt-Bu, THF
0 ºC, 1 h
BnOCONHOH
CH2Cl2, 0 ºC, 1 h
AcO
O
N
120 ºC, 68 h
CO2Bn
>95% crude
63%
75%
37%
NHBoc
1
6
7
8
9
(1) NaOH
1,4-dioxane, H2O,
rt, 50 min
69%
(2) HOAt,
EDC-HCl, NH4HCO3
Synthetic Approach C
DMF, rt, 24 h
butyl acrylate,
CO2Bu
O
Bu4NIO4,
BnOCONHOH
CH2Cl2, 0 ºC, 1 h
OAc
CO2Bu
CO2Bu
KOt-Bu,
THF,
hydroquinone
toluene
AcO
0 ºC, 1 h
120 ºC, 68 h
>95% crude
N
CO2Bn
86%
69%
6
10
11
12
Scheme 2. Synthetic approaches B and C.
Synthetic Approach D
O
O
(1) PhI(OAc)2,
CF3CHOHCF3
0 ºC to rt, 4.5 h
O
CH2
PMBN
O
PPh3MeI, NaHMDS,
THF, -78 to 0 ºC, 3 h
(1) p-methoxybenzyl
amine, NaBH(OAc)3
OAc
76%
(2) HCl, THF
(3) (Boc)2O
60%
69%
16
O
O
NBoc
NBoc
PMB
PMB
(1) K2CO3, MeOH
13
14
15
rt, 2 h
(2) 5% Pt/C,
NaHCO3, H2O, air
75 ºC, 90 h
80%
(1) LiOH,
H2N
30% H2O2,
THF, H2O
(2) (Boc)2O
HO
O
O
O
O
(Boc)2O,
NH4HCO3,
CH3CN
CAN,
CH3CN, H2O
HN
O
PMBN
O
PMBN
O
NH2
NH2
OH
>95%
61%
O
O
O
>95%
NHBoc
1
19
18
17
Scheme 3. Synthetic approach D.

2848
C. Barfoot et al. / Tetrahedron Letters 51 (2010) 2846–2848
Synthetic Approach E
(Boc)2O,
NH4HCO3
CH3CN, rt, 4 h
O
COOH
BnO
O
COOH
O
BnO
O
CONH2
HCBr3,
KOH, BnOH
+
65% after two
recrystallisations
50%
O
O
O
O
O
22
21
13
20
HCl
THF, H2O
60 oC, 2 h
96%
(1) NH4OOCCF3, NaB(CN)H3,
HO
CONH2
BnO
CONH2
BnO
CONH2
MeOH, 0 oC
H2, Pd/C, MeOH
60 oC, 120 psi
(2) (Boc)2O, NaOH
then recrystallisation
>95%
50%
NHBoc
1
NHBoc
24
O
23
Scheme 4. Synthetic approach E.
Table 1
technical difficulties of removing the various impurities precluded
further development. Approaches D and E have similar overall
yields but with only five steps, approach E was readily scaled up
to deliver multi-gram quantities of compound 1.7
Comparison of the synthetic approaches
Approach
Number of steps
Overall yield (%)
A
B
C
D
E
3
5
7
8
5
8
17
26
15
16
In conclusion, the challenge of making cis
a-hydroxy amide 1
led to several approaches being investigated to install the key
functionalities.
Acknowledgements
intramolecular cyclisation to give the intermediate acetoxy com-
pound 16 in good yield and purity. Standard manipulation of the
acetate to a primary amide over three steps led to 18. Removal of
the PMB group, cleavage of the carbamate, followed by a final
Boc-protection step led to the desired compound 1.
We are grateful to Steve Richards for NMR support, Bill Leavens
for mass spectroscopy support and all the chemists who have been
involved in this work.
This route had none of the problems of cis/trans selectivity and
also had no issues with hydrogenolysis of the tertiary alcohol
functionality.
References and notes
1. Full
name
for
1:
1,1-dimethylethyl[cis-4-(aminocarbonyl)-4-hydrox-
ycyclohexyl]carbamate. For the syntheses of these compounds and
experimental details for the reactions in Schemes 1 and 2, see: (a) Brooks, G.;
Davies, D. T.; Jones, G. E.; Markwell, R. E.; Pearson, N. D.; WO 2003087098; Chem.
Abstr.; 2003, 139, 337959; (b) Axten, J. M.; Daines, R. A.; Davies, D. T.; Gallagher,
T. F.; Jones, G. E.; Miller, W. H.; Pearson, N. D.; Pendrak, I.; WO 2004002992;
Chem. Abstr.; 2004, 140, 94053.
Synthetic approach E adopted a different strategy and relied on
introducing the cis stereochemistry at a late stage. Starting from
readily available cyclohexanedione monoethylene ketal 13, we
decided to pursue a variant of the chemistry reported by the Kus-
umi group.5 Thus the reaction of 13 with bromoform and potas-
sium hydroxide in benzyl alcohol led to 20 in good yield along
with ꢀ5% of by-product 21. Interestingly, when water was used
as the solvent, we did not isolate any hydroxy-acid but instead
only undesired 21. Conversion of the carboxylic acid 20 into amide
22 and then deprotection to give ketone 23 proceeded smoothly.
The key reductive amination reaction was then performed, fol-
lowed by Boc-protection to give 24. This reductive amination step
took some optimisation and the best cis/trans ratio of 3:1 was
eventually achieved under the conditions shown in Scheme 4.6
We then discovered that cis 24 could be readily separated from
its trans isomer by recrystallisation from methanol. Hydrogenation
under forcing conditions cleaved the benzyl group to give the
desired compound 1.
2. For a synthesis, see: Ref. 1(b), example 3(a).
3. (a) Keck, G.; Fleming, S. Tetrahedron Lett. 1998, 39, 2059; (b) Martin, S.;
Hartmann, M.; Josey, J. Tetrahedron Lett. 1998, 39, 2059; (c) Sirisoma, N.;
Johnson, C. Tetrahedron Lett. 1998, 39, 2059.
4. (a) De Mico, A.; Margarita, R.; Mariani, A.; Piancatelli, G. Chem. Commun. 1997,
1237; (b) De Mico, A.; Margarita, R.; Mariani, A.; Piancatelli, G. Tetrahedron Lett.
1996, 37, 1889; (c) Moriarty, R.; Vaid, R.; Koser, G. Synlett 1990, 365. Treatment
of 15 with iodine (I2, NaHCO3, MeCN, rt, 20 h) led to the product 25 (below) in
80% yield. However, all attempts at elaboration of 25 to carboxylic acid 17 failed.
O
PMBN
O
I
25
.
5. Yabuuchi, T.; Kusumi, T. Chem. Pham. Bull. 1999, 47, 684–686.
6. Marui, S.; Yamamoto, T.; Sudo, K.; Akimoto, H.; Kishimoto, S. Chem. Pharm. Bull.
1995, 43, 588.
Interestingly, this final hydrogenation did not give any of the
undesired hydrogenolyis product
2 (<20 ppm), presumably
7. Data for 1: white solid; mp 210–213 °C; 1H NMR (600 MHz, DMSO-d6): d = 1.38
(s, 9H), 1.42–1.60 (m, 6H), 1.65–1.74 (m, 2H), 3.11–3.22 (m, 1H), 4.89–5.07 (m,
1H), 6.72 (br d, J = 9 Hz, 1H), 7.01 (br s, 1H), 7.16 (br s, 1H). 13C NMR (151 MHz,
DMSO-d6): d = 27.4, 28.3, 33.0, 48.6, 72.2, 77.3, 155.1, 179.5. ESI-HRMS: m/z
calcd for C12H22N2O4Na: 281.1477; found 281.1477 [M+Na]+.
because the hydrogenation in Scheme 2 goes through a potentially
more labile allylic alcohol intermediate.
In Table 1, the competing approaches are compared. Although
approaches A–C have a reasonably low number of steps, the