From ATP to AZD6140: The discovery of an orally active reversible P2Y12 receptor antagonist for the prevention of thrombosis

Article abstract of DOI:10.1016/j.bmcl.2007.07.057

Starting from adenosine triphosphate (ATP), the identification of a novel series of P2Y₁₂ receptor antagonists and exploitation of their SAR is described. Modifications of the acidic side chain and the purine core and investigation of hydrophobic substituents led to a series of neutral molecules. The leading compound, 17 (AZD6140), is currently in a large phase III clinical trial for the treatment of acute coronary syndromes and prevention of thromboembolic clinical sequelae.

Full text of DOI:10.1016/j.bmcl.2007.07.057

Bioorganic & Medicinal Chemistry Letters 17 (2007) 6013–6018
From ATP to AZD6140: The discovery of an orally
active reversible P2Y12 receptor antagonist
for the prevention of thrombosis
Brian Springthorpe,* Andrew Bailey, Patrick Barton, Timothy N. Birkinshaw,
Roger V. Bonnert, Roger C. Brown, David Chapman, John Dixon, Simon D. Guile,
Robert G. Humphries, Simon F. Hunt, Francis Ince, Anthony H. Ingall, Ian P. Kirk,
Paul D. Leeson, Paul Leff, Richard J. Lewis, Barrie P. Martin, Dermot F. McGinnity,
Michael P. Mortimore, Stuart W. Paine, Garry Pairaudeau, Anil Patel, Aaron J. Rigby,
Robert J. Riley, Barry J. Teobald, Wendy Tomlinson,
Peter J. H. Webborn and Paul A. Willis
AstraZeneca R&D Charnwood, Bakewell Road, Loughborough LE11 5RH, UK
Received 8 June 2007; revised 13 July 2007; accepted 15 July 2007
Available online 19 August 2007
Abstract—Starting from adenosine triphosphate (ATP), the identification of a novel series of P2Y12 receptor antagonists and exploi-
tation of their SAR is described. Modifications of the acidic side chain and the purine core and investigation of hydrophobic sub-
stituents led to a series of neutral molecules. The leading compound, 17 (AZD6140), is currently in a large phase III clinical trial for
the treatment of acute coronary syndromes and prevention of thromboembolic clinical sequelae.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Platelet aggregation, superimposed on plaque rupture, is
a critical mechanism involved in transforming otherwise
clinically stable atherosclerotic disease into an acute,
potentially life-threatening arterial thrombotic event,
such as unstable angina, myocardial infarction or throm-
botic stroke. Platelets adhere to the ruptured arterial pla-
que and aggregate in response to a variety of local and
systemic stimuli, including mediators such as adenosine
diphosphate (ADP), adrenaline, thrombin and 5-
hydroxytryptamine. Whatever the stimulation, the final
common steps are exposure of the GPIIb/IIIa receptor
complex and subsequent cross-linking of the platelets by
fibrinogen.1 The platelet aggregate is further consolidated
by formation of fibrin, resulting in a firm, adherent clot.
age granules into the medium. The consequences of this
are that the P2Y12 receptor (formerly known as the P2T
receptor) on the platelet is stimulated, shape change oc-
curs and the GPIIb/IIIa complex is exposed. Nearby
platelets are also activated and the initial pro-aggregato-
ry signal is amplified. Therefore, the ADP/P2Y12 path-
way plays a key amplifying role in the overall platelet
response. Against this background, it is clear that inhib-
iting the effect of ADP on the P2Y12 receptor signifi-
cantly inhibits platelet aggregation and thereby
prevents the formation of a firm, cross-linked throm-
bus.2 The discovery and clinical use of the irreversible
P2Y12 antagonist clopidogrel3 (Fig. 1) has confirmed
the clinical impact of inhibiting this target. Although
this anti-platelet agent has improved the management
of acute and sub-acute coronary artery disease, several
features of this agent leave room for improvement, such
that more effective treatments are predicted to further
improve clinical outcomes.
ADP has a pivotal role in this process since the stimuli
exert much of their effect by release of ADP from stor-
Keywords: ATP; P2Y12 receptor antagonist; Thrombosis; AZD6140.
*
Corresponding author. Tel.: +44 1509 644114; fax: +44 1509
645567; e-mail: brian.springthorpe@astrazeneca.com
Using ATP, the natural antagonist (pIC50 = 3.5) of the
P2Y12 receptor, as a chemical starting point, we discov-
ered 1 (AR-C69931MX, cangrelor, Fig. 1)4 as a potent
Present address: Vertex Pharmaceuticals (Europe) Ltd 88 Milton
Park, Abingdon, Oxfordshire, OX14 4RY, UK.
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.07.057

6014
B. Springthorpe et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6013–6018
NH2
N
NH2
N
The triazolopyrimidine heterocycle has been identified
as an isostere of purine and has been successfully em-
ployed in the anticancer and antiviral areas.5 Utilising
this heterocycle led to 3 in which the P2Y12 potency
had unexpectedly increased (pIC50 = 9.5). This change
resulted in the identification of the first potent and selec-
tive non-phosphate P2Y12 antagonist 3 having activity
equivalent to that of triphosphate 1.
N
N
N
N
O
O
P
O
P
O
P
O
P
HO
O
O
O
HO
O
O
P
N
N
O
O
OH OH OH
OH OH
HO
OH
HO
OH
ATP
ADP
S
HN
In response to the potential instability of the glycosidic
bond to enzymatic cleavage we investigated replacing
the ribose sugar with a cyclopentyl unit, a structural
change that has been successful in a number of areas,
including the antiviral field.6 In compound 4 we have re-
tained the potencies of 1 and 3 but have now successfully
moved away from a classic purinergic core structure.
O
N
O
N
Cl
O
N
O
P
O
O
O
O
CF3
P
Cl
P
N
N
S
O
O
O
O
S
Cl
4Na+
HO
OH
Clopidogrel
1 (AR-C69931MX, Cangrelor)
Figure 1. Structures of ATP, ADP, clopidogrel and cangrelor.
While these changes provided improved potency over
compound 2, compounds 3 and 4 were cleared rapidly
in the rat by the biliary route. The properties of these
molecules are not favourable for oral absorption: dou-
bly negatively charged, molecular weight >500, >5
H-bond donors. Variation of the substituents on the
triazolopyrimidine core of structure 4 did not improve
potency nor, more importantly, metabolic properties,
so attention was turned once more to the acidic side
chain, with the goal of reducing its complexity.
and selective P2Y12 receptor antagonist suitable for
intravenous use only. Compound 1 is currently in phase
III clinical trials as an ultra-short acting, intravenously
administered, antithrombotic agent.
This letter describes how the physical and chemical
properties of 1 were modified in order to find potent,
selective and orally active P2Y12 antagonists.
The parallel chemistry approach employed required an
assay able to tolerate higher concentrations of DMSO
and with higher throughput than the traditional ADP-
induced aggregation in washed platelets.7 A radiola-
belled displacement binding assay was developed8 to
support this increased throughput and all subsequent
data were generated using this binding assay. The switch
from the functional to binding assay was validated with
earlier, suitably soluble compounds, which demon-
strated similar potency in both systems.
Since ATP, the endogenous antagonist of the P2Y12
receptor, differs from the agonist ADP in having a
c-phosphate group, we hypothesized that this terminal
acidic group was essential for antagonism. Efforts were
therefore directed towards finding alternative acidic
groups which could mimic the polyphosphate chain and
particularly the c-phosphate unit of ATP. This initial
strategy led to the discovery of an aspartic acid-derived
dicarboxylate 2 (Table 1), which was ꢀ300-fold less po-
tent (pIC50 = 7.0) than the triphosphate 1 (pIC50 = 9.4).
An example of the stepwise truncation from compound
4 is the monocarboxylic acid 5, which was surprisingly
potent (pKi = 8.3), a result at odds with the c-phosphate
theory. Unfortunately, compound 5 was still cleared by
the biliary route and had low bioavailability. However,
more surprising and of crucial importance was the
observation that the neutral molecules 6 and 7 (Table
2) had modest but significant P2Y12 affinity.
Further extensive variation of the substituents on the
adenosine core of 2 did not lead to improved potency
and it was decided to make more fundamental changes
to the core structure.
Table 1. P2Y12 antagonist potency of aspartic acid derivatives (2–4)
Significantly, these neutral molecules were cleared in the
rat solely by metabolic processes, which we felt able to
understand and manipulate. While the affinity of 6 and
7 was reduced compared with compound 5, these mole-
cules were more amenable to rapid parallel synthesis at
the 5- and 7-positions as a means of improving affinity
(Schemes 1 and 2: synthesis of the intermediates accord-
ing to our published procedures).9 In all, some 6000 ana-
logues were synthesised.
HN
CO2H
H
O
N
N
N
X
HO2C
N
H
N
S
Y
HO
OH
a
Compound
X
Y
hP2Y12 pIC50
2
3
4
C
N
N
O
7.0
9.5
9.3
Position 5 was varied by displacement of a sulfone-leav-
ing group by S, O and N nucleophiles. It was found that
polar substituents were not well tolerated and sulfur
linked groups were consistently more potent than their
oxygen and nitrogen analogues. From the results it
O
CH2
a ADP-induced aggregation assay using washed platelets as described
in Ref. 7.

B. Springthorpe et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6013–6018
6015
Table 2. P2Y12 affinity and rat pharmacokinetics of compounds with
simplified C40 side chains (5–10)
affinity significantly increased over that of the butyl sub-
stituent, but the enhancement found with racemic trans-
2-phenylcyclopropylamine was crucial for achieving our
objectives of potency and good preclinical metabolic
properties.
R1
HN
7
N
N
N
Separation10 of the enantiomers of trans-phenylcyclo-
propylamine showed that the 1R,2S isomer was more
potent than the 1S,2R isomer.
N
R
N
S
5
HO
OH
Not only did this discovery lead to a >10-fold increase in
affinity (5 vs 8, 6 vs 9 and 7 vs 10), but the alcohol deriv-
ative 10 showed significant oral bioavailability in the rat
(Table 2) and a sustained inhibition of ADP-induced
platelet aggregation ex vivo in the dog following oral
administration.
Compound
R
R1
hP2Y12 Rat Rat Rat Rat
pKia
Clpb Vss T1/2 F%c
b
b
5
6
7
COOH Butyl
CONH2 Butyl
CH2OH Butyl
8.3
7.7
7.1
30
21
29
0.4 0.5 <5
1.6 1.4 <5
2.3 1.2 <5
Ph
8
9
COOH
CONH2
CH2OH
9.6
8.8
8.3
16
20
16
0.5 0.5
While the alcohol 10 had acceptable P2Y12 affinity and
good preclinical pharmacokinetics in both the rat and
the dog, investigation of the metabolism of the com-
pound in hepatocytes demonstrated marked species dif-
ferences in biotransformation profiles.11 Whereas
oxidation was the predominant route of metabolism in
rat hepatocytes, glucuronidation at the C40 side chain
hydroxyl group was the major biotransformation path-
way in hepatocytes from dogs and humans.
Ph
Ph
2.2 2.0 <5
10
2.9 2.5
35
Units: Clp, mL/min/kg; Vss, L/kg; T1/2, h; F, bioavailability.
a Assay described in Ref. 8.
b Dosed at ꢀ3 mg/kg (iv).
c Dosed at 3–10 mg/kg (po).
This observation led us to set up a primary assay for
glucuronidation using UDPGA supplemented rat, dog
and human microsomes, in addition to the traditional
NADPH supplemented microsomal assays used to
measure oxidation.11,12 These assays were applied in
parallel to aid compound selection for final evaluation
in human hepatocytes. Relative figures for resistance to
glucuronidation and oxidation were determined by the
inclusion in the assay of known standards: zileuton for
the glucuronidation assay and dextromethorphan for
the oxidation assay, the aim being to identify com-
pounds with ratios of >20 · zileuton and >10 · dextro-
methorphan. In vitro–in vivo modelling had suggested
that compounds showing such figures could be
expected to have a metabolic clearance via glucuroni-
dation and oxidation equivalent to <10% hepatic
extraction (<2 mL/min/kg).
HN
N
HN
N
N
N
N
N
N
5
N
5
N
N
a, b
SO2R1
SR2
R
R
O
O
HO
OH
R = CONH2, CH2OH; R1 = Me, n-propyl
Scheme 1. Reagents and conditions: (a) NaSR2, DMF, 20 ꢁC, 1 h;
(b) TFA–H2O (9:1), 20 ꢁC, 2 h.
R2
R1
N
Cl
N
N
N
7
7
N
N
N
N
N
N
N
a, b, c
R
S
R
S
The hydroxymethyl compound 10, although having
good bioavailability, did not reach our desired criterion
for resistance to glucuronidation. Further neutral alco-
hols were examined (Table 3) and the hydroxyethoxy
11 and hydroxyethyl 13 analogues achieved the desired
metabolic stability ratios while retaining good
bioavailability.
O
O
HO
OH
R = COOH, CH2OH
R = COOH, CH2OH, CONH2
Scheme 2. Reagents and conditions: (a) R1R2NH, N,N-diisopropyl-
ethylamine, CH2Cl2, 20 ꢁC, 16 h; (b) R = (COOH ! CONH2), oxalyl
chloride, CH2Cl2, 20 ꢁC, 2 h, 35% aq ammonium hydroxide, CH2Cl2,
20 ꢁC, 2 h; (c) TFA–H2O (9:1), 20 ꢁC, 2 h.
Of particular note is compound 14 (Table 3, R = H)
which has a pKi of 8.6. This is surprising since all func-
tionality has been removed from this position and
yet affinity is only 1/10 that of the triphosphate 1. We
propose that while in the triphosphate series activity is
dominated by the electrostatic interactions of the acid
side chain, new receptor-binding interactions have been
introduced utilising the triazolopyrimidine and the
lipophilic phenylcyclopropyl group. Compound 14 is,
however, unstable to glucuronidation and overall the
was concluded that S-propyl substitution at position 5
was optimal.
Investigation of substitution at the 7-position was
accomplished by displacement of chloride by amines.
Using amines bearing polar substituents, or secondary
amines, led to reduced affinity. In only one case was

6016
B. Springthorpe et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6013–6018
Table 3. P2Y12 antagonist affinity, metabolic stability and bioavail-
ability of C40 side chain alcohols
Table 5. Rat and dog pharmacokinetics for compounds 16–18
Rat
Rat
Rat
T1/2
Rat
Dog
Dog Dog
Dog
F%b
Clpa Vss
F%b Clpa Vss
T1/2
a
a
a
a
16 11
17 21
18 12
2.7
3.8
2.1
3.0
2.6
2.5
17
24
9
16.5
9.0
5.6
3.0
4.8
2.9
NT
59
72
HN
NT
NT
NT
N
N
Units: Clp, mL/min/kg; Vss, L/kg; T1/2, h; F, bioavailability; NT, not
tested.
N
N
R
N
S
a Dosed at ꢀ3 mg/kg (iv) in rat and 1 mg/kg (iv) in dog.
b Dosed at 3–10 mg/kg (po) in rat and 1 mg/kg (po) in dog.
HO
OH
Compound 17 (AZD6140) was chosen over 16 and 18
for progression into human testing on the basis of a low-
er predicted dose (driven by a combination of potency
and pharmacokinetic properties in pre-clinical species).
Compound
R
hP2Y12 UDPGA NADPH Rat
ratioa
pKi
ratiob
F%
10
11
12
13
14
CH2OH 8.3
O(CH2)2OH 8.5
5
24
3
>30
13
35
26
50
37
38
OH
(CH2)OH
H
8.7
8.2
8.6
Stable
27
NT
AZD6140 can be synthesised using the route outlined in
Schemes 3a and 3b. Synthesis of the (1R,2S)-trans phe-
nylcyclopropylamine begins with derivatisation of
substituted cinnamic acid 19 with Oppolzer’s sultam13
to give 20. Diastereoselective cyclopropanation14 then
provides, after recrystallisation, cyclopropylamide 21
in high chiral purity which is readily saponified to acid
22. A four-step Curtius rearrangement15 gives the cyclo-
propylamine 23, which is conveniently isolated as a tar-
trate salt.
19
1
Stable, turnover below limit of quantification; NT, not tested.
a UDPGA assay: required ratio to zileuton >20.
b NADPH assay: required ratio to dextromethorphan >10.
data in Table 3 show the hydroxyl side chain has an
important effect on metabolic stability but little effect
on affinity.
Synthesis of the core utilises commercially available ace-
tate 24 which is cleanly converted to the protected amine
25 under palladium catalysis.16 Osmium catalysed cis
dihydroxylation17 occurs with good diastereoselectivity
to give, following amine deprotection, the aminotriol
26. Reprotection of the amine as a benzyl carbamate
and protection of the diol as an acetonide gives alcohol
27 suitable for alkylation. The hydroxyethyl side chain is
next incorporated by reaction with ethyl bromoacetate
followed by reduction to give key intermediate 28.
Reaction of 28 with 4,6-dichloro-2-propylthio-pyrimi-
dine-5-amine9 gives 29 which is converted to the triazol-
opyrimidine 30 under diazotization conditions. Finally,
Finally, further optimisation by fine-tuning the substitu-
tion of the phenyl ring and 5-S-propyl substituent of 10
and 11 provided compounds 15–18 (Table 4) with
acceptable affinity and metabolic stability suitable for
further progression.
Preclinical pharmacokinetic data for 3,4-difluorophenyl
compounds 16–18 are shown in Table 5.
Table 4. Optimising metabolic stability by substitution of the phenyl-
cyclopropyl group
O
a, b
F
F
O
d
HO
F
F
R2
N
SO2
HN
19
20
N
N
N
R1
15,16: R = CH2OH
17,18: R = O(CH2)2OH
N
R
N
S
c
F
F
N
SO2
HO
OH
21
O
Compound
R1
R2
H
hP2Y12 UDPGA NADPH
ratiob
F
F
F
F
e-i
pKi
ratioa
15
16
17
18
(CH2)2CF3
8.3
19
Stable
>30
24
HO2C
H2N
22
23
(CH2)2CF3 3,4-DiF 8.3
(CH2)2CH3 3,4-DiF 8.7
(CH2)2CF3 3,4-DiF 9.2
Stable
Stable
Stable
Scheme 3a. Reagents and conditions: (a) SOCl2; (b) sultam salt (80%;
two steps); (c) CH2N2, Pd(OAc)2 then recrystallise from EtOH (50%);
(d) LiOH, H2O, THF, 50 ꢁC (99%); (e) ClCO2Et, TEA, acetone, H2O;
(f) NaN3, H2O; (g) toluene, reflux; (h) 6 M HCl, reflux; (i) L-tartaric
acid, EtOH (62%, five steps).
32
Stable, turnover below limit of quantification.
a UDPGA assay: required ratio to zileuton >20.
b NADPH assay: required ratio to dextromethorphan >10.

B. Springthorpe et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6013–6018
6017
NH2.HCl
(f) identifying metabolically stable neutral compounds
by modifying the hydrophobic phenylcyclopropyl
group and the hydroxylic side chain substituent.
HO
OAc
N(BOC)2
HO
HO
a
b, c
HO
OH
24
25
26
In making these changes, the structure–activity relation-
ships moved away from dependency on the acidic side
chain, allowing identification of potent and orally bio-
available non-nucleotide reversible P2Y12 antagonists.
Compound 17 has now progressed to phase III clinical
trials in acute coronary syndromes.
HO
NHCbz
HO
NH2
O
d, e
f, g, h
i
O
O
O
O
27
28
Cl
N
Cl
N
N
N
N
H2N
HN
N
N
HO
Acknowledgments
HO
O
S
O
O
S
j
We thank Jim Murray, Dawn Adkin and Kate Harris
for formulation support. We also thank Iain Beattie,
Michael Bernstein, Kim Lawson, Hemlata Pancholi
and Andy Wright for analytical support.
O
O
O
30
29
F
F
Supplementary data
HN
N
k, l
N
N
Experimental details for the General Schemes 1 and 2
and supporting analytical data (NMR, MS and elemen-
tal analyses) for compounds 2–18 can be found in the
online version. Supplementary data associated with this
article can be found, in the online version, at
doi:10.1016/j.bmcl.2007.07.057.
N
HO
N
O
S
HO
OH
17
Scheme 3b. Reagents and conditions: (a) (BOC)2NNa, Pd(PPh3)4,
THF (92%); (b) OsO4, NMO, THF, H2O (100%); (c) HCl, MeOH,
H2O (96%); (d) dimethoxypropane, pTSA, acetone (86%); (e) CbzCl,
DIPEA, MIBK (95%); (f) ButOK, ethyl bromoacetate, THF;
(g) LiBH4, THF (86%; two steps); (h) H2, Pd/C, EtOH (99%);
(i) 4,6-dichloro-2-propylthio-pyrimidine-5-amine, DIPEA, DMF
(75%); (j) isoamylnitrite, CH3CN (88%); (k) 23, DIPEA, DCM
(99%); (l) TFA, H2O (90%).
References and notes
1. Frishman, W. H.; Burns, B.; Atac, B.; Alturak, N.;
Altajar, B.; Lerrick, K. Am. Heart J. 1995, 130, 877.
2. Leff, P.; Robertson, M. J.; Humphries, R. G. In Purinergic
Approaches in Experimental Therapeutics; Jacobson, K.
A., Jarvis, M. J., Eds.; Wiley-Liss: New York, 1997; pp
203–216.
3. (a) CAPRIE Steering Committee (CAPRIE). Lancet 1996,
348, 1329; (b) The Clopidogrel in Unstable Angina to
Prevent Recurrent Events Trial Investigators (CURE).
N. Engl. J. Med. 2001, 345, 494.
chloro displacement with amine 23 followed by depro-
tection gives 17.
4. (a) Ingall, A. H.; Dixon, J.; Bailey, A.; Coombs, M. E.; Cox,
D.; McInally, J. I.; Hunt, S. F.; Kindon, N. D.; Teobald, B.
J.; Willis, P. A.; Humphries, R. G.; Leff, P.; Clegg, J. A.;
Smith, J. A.;Tomlinson, W. J. Med. Chem. 1999, 42, 213; (b)
Humphries, R. G. Haematologica 2000, 85, 66.
5. Albert, A.. In Advances in Heterocyclic Chemistry; Katri-
tzky, A. R., Ed.; Academic Press, 1986; 39, pp 117–180.
6. Marquez, V. E.; Lim, M. L. Med. Res. Rev. 1986, 6, 1.
7. Aggregation of human washed platelets was assessed
turbidimetrically in 96-well plates as a decrease in absor-
bance (650 nm). The washed platelets were incubated with
the test compounds for 5 min before addition of 30 lM
ADP. Antagonist potency was estimated as a % inhibition
of the control ADP response to obtain a pIC50. Full assay
details are given in: Humphries, R. G.; Tomlinson, W.;
Ingall, A. H.; Cage, P. A.; Leff, P. Br. J. Pharmacol. 1994,
113, 1057.
Phase I and phase II studies have confirmed the pre-
dicted pharmacokinetic and pharmacodynamic profile
of the compound and a double blind comparison with
clopidogrel, on a background of low-dose aspirin, has
shown 17 to have superior anti-platelet activity as mea-
sured ex vivo by light transmission aggregometry.18
In summary, beginning with ATP, a poor lead for an
oral programme, we have discovered oral P2Y12 antag-
onists with clinical potential. Key elements in the medic-
inal chemical journey from ATP to 1 to 17 were:
(a) introducing affinity-enhancing 5,7-hydrophobic
substituents;
(b) replacement of the labile triphosphate group;
(c) changing the core purine to a triazolopyrimidine,
increasing affinity >100-fold;
(d) finding the first nonacidic reversible antagonists
(e.g., 6 and 7);
(e) introducing the trans-2-phenylcyclopropylamino
substituent, increasing affinity >10-fold; and
8. Binding data were obtained in washed platelets in 96-well
plates, each well containing [125I] radiolabelled P2Y12
antagonist, test compounds and washed platelets. After
30 min incubation, the reaction was terminated by filtra-
tion, washing of the platelets and the bound radioactivity
was measured and used to derive a pKi. Full assay details

6018
B. Springthorpe et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6013–6018
are given in: Kirk, I. PCT Int. Appl. WO200033080, 2000;
Chem. Abstr. 2000, 133, 14307.
13. For review, see Oppolzer, W. Pure Appl. Chem. 1990, 62,
1241.
9. (a) Bonnert, R. V.; Ingall, A. H.; Springthorpe, B.; Willis,
P. A. PCT Int. Appl. W0199828300, 1998; Chem, Abstr.
1998, 129, 95506; (b) Guile, S. D.; Ingall, A. H.;
Springthorpe, B.; Willis, P. A. PCT Int. Appl.
WO199905143, 1999; Chem, Abstr. 1999, 130, 168386.
10. Mitscher, L. A.; Sharma, P. N.; Chu, D. T. W.; Shen, L.
L.; Pernet, A. G. J. Med. Chem. 1986, 29, 2044.
11. Martin, I. J.; Lewis, R. J.; Bernstein, M. A.; Beattie, I. G.;
Martin, C. A.; Riley, R. J.; Springthorpe, B. Drug Metab.
Dispos. 2006, 34, 1502.
14. Vallgarda, J.; Hacksell, U. Tetrahedron Lett. 1991, 32,
5625.
15. Shiori, T. In Comprehensive Organic Synthesis, Selectivity,
Strategy and Efficiency in Modern Organic Chemistry;
Trost, M. T., Fleming, I., Winterfeldt, E., Eds., 1st ed.;
Pergamon Press: London, 1991; pp 795–827.
16. Jumnah, R.; Williams, J. M. J.; Williams, A. C. Tetrahe-
dron Lett. 1993, 34, 6619.
17. See Trost, B. M.; Kuo, G.-H.; Benneche, T. J. Am. Chem.
Soc. 1988, 110, 621, and references cited therein.
18. Husted, S.; Emanuelsson, H.; Heptinstall, S.; Sandset, P.
M.; Wickens, M.; Peters, G. Eur. Heart J. 2006, 27,
1038.
12. Bouska, J. J.; Bell, R. L.; Goodfellow, C. L.; Stewart, A.
O.; Brooks, C. D.; Carter, G. W. Drug Metab. Dispos.
1997, 25, 1032.