Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 5005–5007
Initial synthesis of UK-427,857 (Maraviroc)
*
David A. Price, Simon Gayton, Matthew D. Selby, Jens Ahman,
Sarah Haycock-Lewandowski, Blanda L. Stammen and Andrew Warren
Pfizer Global Research and Development, Sandwich Laboratories, Ramsgate Road, Sandwich, Kent CT139NJ, UK
Received 27 April 2005; accepted 19 May 2005
Available online 9 June 2005
Abstract—The initial synthesis of UK-427,857 (Maraviroc) is described including the preparation of 4,4-difluorocyclohexanoic acid
and amide coupling utilizing a polymer supported reagent.
Ó 2005 Elsevier Ltd. All rights reserved.
UK-427,857 (Maraviroc) is a potent antagonist of the
CCR5 receptor and is currently progressing through
nucleophilic fluorinating sources. This transformation is
routinely accomplished using diethylaminosulfur trifluo-
ride (DAST) which is commercially available but does
require careful handling due to its well-known thermal
phase III clinical trials as a treatment for HIV (Fig.
1
1
). The structure includes a 4,4-difluorocyclohexyl unit,
4
a phenylpropyl motif with a benzylic stereogenic centre
and an equatorially disposed triazole in a tropane sys-
tem. The synthesis of the required 1,2,4-triazole substi-
instability. In our first attempts to use DAST to pre-
pare the target acid 5, we were disappointed to find that
treatment of ketone 1 with 1.05 equiv of DAST in
dichloromethane at 0 °C resulted in complete consump-
tion of the starting material but gave an inseparable 1:1
mixture of the required difluoro compound 2 and the
vinyl fluoride 3 in 85% yield. The formation of vinyl
fluoride co-products from treatment of ketones with
DAST is known in the literature and appears difficult
2
tuted tropane has been previously described. This
letter focuses on the challenges in preparing the
required 4,4-difluorocyclohexanoic acid and the overall
synthetic route used in the preparation of UK-427,857
(
Maraviroc) to underwrite initial studies.
5
The use of fluorine in molecules of pharmacological
interest can profoundly alter their biological properties
and there has been widespread research into methodol-
to control. Optimisation studies undertaken to influ-
ence the ratio of products including temperature,
reagent stoichiometry and solvents were unsuccessful.
3
ogies for the introduction of this atom. One of the most
commonly used techniques is the conversion of carbon–
oxygen functional groups into carbon–fluorine bonds by
The mixture of 2 and 3 was subjected to Upjohn condi-
6
tions for dihydroxylation and we were delighted to find
that after the reaction overnight the required dif-
luorocyclohexyl ester 2 could then be isolated in high
7
purity by column chromatography. The assumed prod-
O
uct 4 from oxidation of the vinyl fluoride was never iso-
lated and only identified by mass spectral evidence from
the crude reaction mixture. Saponification of 2 gave the
required 4,4-difluorocyclohexyl acid 5 which was recrys-
tallised from cyclohexane to furnish an analytically pure
material (Scheme 1).
F
H
N
F
N
N
N
N
Figure 1. UK-427,857 (Maraviroc).
With suitable quantities of 5 in hand, our attention
turned to preparing multigram quantities of the phenyl-
propyl linker. The starting material was the commer-
cially available chirally pure b-amino ester 6 which
Keywords: Heterocycles; Triazoles; Amide; Tropane; Fluorination.
*
8
6
was protected in high yield. The ester was then reduced
0
040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.05.082