4-Alkoxy-2,6-diaminopyrimidine derivatives: inhibitors of cyclin dependent kinases 1 and 2.

Article abstract of DOI:10.1016/S0960-894X(02)00884-3

The cyclin dependent kinase (cdk) inhibitor NU6027, 4-cyclohexylmethoxy-5-nitroso-pyrimidine-2,6-diamine (IC(50) vs cdk1/cyclinB1=2.9+/-0.1 microM and IC(50) vs cdk2/cyclinA3=2.2+/-0.6 microM), was used as the basis for the design of a series of 4-alkoxy-2,6-diamino-5-nitrosopyrimidine derivatives. The synthesis and evaluation of 21 compounds as potential inhibitors of cyclin-dependent kinases 1 and 2 is described and the structure-activity relationships relating to NU6027 have been probed. Simple alkoxy- or cycloalkoxy-groups at the O(4)-position were tolerated, with the 4-(2-methylbutoxy)-derivative (IC(50) vs cdk1/cyclinB1=12+/-2 microM and cdk2/cyclinA3=13+/-4 microM) retaining significant activity. Substitutions at the N(6) position were not tolerated. Replacement of the 5-nitroso substituent with ketone, oxime and semicarbazone groups essentially abolished activity. However, the derivative bearing an isosteric 5-formyl group, 2,6-diamino-4-cyclohexylmethoxy-pyrimidine-5-carbaldehyde, showed modest activity (IC(50) vs cdk1/cyclinB1=35+/-3 microM and cdk2/cyclinA3=43+/-3 microM). The X-ray crystal structure of the 5-formyl compound bound to cdk2 has been determined to 2.3A resolution. The intramolecular H-bond deduced from the structure with NU6027 bound to cdk2 is not evident in the structure with the corresponding formyl compound. Thus the parent compound, 4-cyclohexylmethoxy-5-nitrosopyrimidine-2,6-diamine (NU6027), remains the optimal basis for future structure-activity studies for cyclin-dependent kinase inhibitors in this series.

Full text of DOI:10.1016/S0960-894X(02)00884-3

Bioorganic & Medicinal Chemistry Letters 13 (2003) 217–222
4-Alkoxy-2,6-diaminopyrimidine Derivatives: Inhibitors of Cyclin
Dependent Kinases 1and 2
Veronique Mesguiche,a Rachel J. Parsons,a Christine E. Arris,b Johanne Bentley,b
F. Thomas Boyle,c Nicola J. Curtin,b Thomas G. Davies,d Jane A. Endicott,d
Ashleigh E. Gibson,a Bernard T. Golding,a Roger J. Griffin,a Philip Jewsbury,c
Louise N. Johnson,d David R. Newell,b Martin E. M. Noble,d Lan Z. Wangb
and Ian R. Hardcastlea,*
aNorthern Institute of Cancer Research and Department of Chemistry, Bedson Building, University of Newcastle,
Newcastle upon Tyne NE1 7RU, UK
bNorthern Institute of Cancer Research, Medical School, Framlington Place, Newcastle upon Tyne NE2 4HH, UK
cAstraZeneca Pharmaceuticals, Alderley Park, Cheshire SK10 4TG, UK
dLaboratory of Molecular Biophysics and Department of Biochemistry, University of Oxford, Oxford OX1 3QU, UK
Received 14 June 2002; revised 19 September 2002; accepted 10 October 2002
Abstract—The cyclin dependent kinase (cdk) inhibitor NU6027, 4-cyclohexylmethoxy-5-nitroso-pyrimidine-2,6-diamine (IC50 vs
cdk1/cyclinB1=2.9ꢀ0.1 mM and IC50 vs cdk2/cyclinA3=2.2ꢀ0.6 mM), was used as the basis for the design of a series of 4-alkoxy-
2,6-diamino-5-nitrosopyrimidine derivatives. The synthesis and evaluation of 21 compounds as potential inhibitors of cyclin-
dependent kinases 1 and 2 is described and the structure–activity relationships relating to NU6027 have been probed. Simple
alkoxy- or cycloalkoxy-groups at the O4-position were tolerated, with the 4-(2-methylbutoxy)-derivative (IC50 vs cdk1/
cyclinB1=12ꢀ2 mM and cdk2/cyclinA3=13ꢀ4 mM) retaining significant activity. Substitutions at the N6 position were not toler-
ated. Replacement of the 5-nitroso substituent with ketone, oxime and semicarbazone groups essentially abolished activity. How-
ever, the derivative bearing an isosteric 5-formyl group, 2,6-diamino-4-cyclohexylmethoxy-pyrimidine-5-carbaldehyde, showed
modest activity (IC50 vs cdk1/cyclinB1=35ꢀ3 mM and cdk2/cyclinA3=43ꢀ3 mM). The X-ray crystal structure of the 5-formyl
compound bound to cdk2 has been determined to 2.3 A resolution. The intramolecular H-bond deduced from the structure with
NU6027 bound to cdk2 is not evident in the structure with the corresponding formyl compound. Thus the parent compound, 4-
cyclohexylmethoxy-5-nitrosopyrimidine-2,6-diamine (NU6027), remains the optimal basis for future structure–activity studies for
cyclin-dependent kinase inhibitors in this series.
# 2002 Elsevier Science Ltd. All rights reserved.
The progression of cells through the cell cycle is
tightly controlled by the sequential activation and
inactivation of a family of serine-threonine kinases,
the cyclin-dependent kinases (cdks). In particular,
cdk1 controls progression from S phase through G2
and into the M phase. Similarly, progression from
G1 to S phase is controlled sequentially by cdk 4/6 and
cdk2. Cdk activity is regulated by binding to cyclin
partners, as well as by phosphorylation and depho-
sphorylation events, and the action of endogenous inhib-
itory peptides.1,2 Loss of cell cycle control leading to
uncontrolled proliferation is common in cancer and so
the identification of potent and selective cyclin depen-
dent kinase inhibitors is a priority for anti-cancer drug
discovery.
A large number of ATP-competitive cdk inhibitory
pharmacophores have been identified.3ꢁ10 First genera-
tion representatives of these compounds are in clinical
trials.11 We have identified compounds based on purine
(e.g., NU2058) and pyrimidine scaffolds which are
competitive inhibitors of both cdk1 and cdk2 with
respect to ATP.12,13 The 5-nitrosopyrimidine (NU6027,
1a) is an inhibitor of cdk1/cyclin B1 and cdk2/cyclinA3
with IC50 values of 2.9ꢀ0.1 mM and 2.2ꢀ0.6 mM,
respectively. Interestingly, the crystal structure of 1a
bound to both non-phosphorylated cdk2, in the
absence of the cyclin partner, and also to the fully
*Corresponding author. Tel.: +44-191-222-6645; e-mail: i.r.hardcastle
@ncl.ac.uk
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00884-3

218
V. Mesguiche et al. / Bioorg. Med. Chem. Lett. 13 (2003) 217–222
active phospho cdk2-cyclinA complex, shows the inhib-
itor binding in an orientation distinct from that of
the majority of the 6-aminopurine based inhibitors
such as olomoucine, roscovitine and purvalanol,14 but
in a nearly identical binding mode to that of the 6-
alkoxyguanine derivative NU2058.12 The key interac-
tions within the ATP-binding site are a triplet of H-
bonds (2-NH2 to Leu 83, N-1 to Leu 83 and 6-NH2 to
Glu 81). Preliminary structure–activity studies sug-
gested that the presence of an intramolecular
hydrogen bond between the C5-nitroso group and
the amino group at C6 of 1a was necessary to
maintain the activity in the pyrimidine series. The
intramolecular H-bond orientates 1a into a purine-
like shape, enabling an optimal interaction between the
6-amino group and Glu 81.
synthetic route (Scheme 2). Reaction of 2,4,6-tri-
chloropyrimidine with aniline or 4-methoxyaniline
in 30% aq EtOH gave the 6-arylaminopyrimidines
(4a,b; 46 and 20%, respectively), accompanied by
the 2,6-disubstituted products which were separable
by chromatography. Substitution at the 2-position
of compounds 4a,b with the masked ammonia
equivalent, bis-p-methoxybenzylamine (Et3N, n-buta-
nol, reflux), gave the disubstituted pyrimidines (5a,b) in
30–40% yield. Bis-(4-methoxybenzyl) protection of the
2-amino group allowed the alkoxide substitution at
the 4-position to proceed smoothly with sodium
cyclohexylmethoxide to give the products (6a,b) in 32
and 74% yield, respectively. Deprotection was effec-
ted with TFA at 60 ꢂC to give the 2-aminopyr-
imidines (7a,b) in good yields. The 5-nitroso
compounds (8a,b) were prepared as described above (48
and 21%, respectively).
We have designed and synthesised derivatives of
NU6027 (1a) in order to gain a better understanding of
the structural basis of inhibition by this new class of
ATP-competitive cdk inhibitor. In this paper we report
the synthesis of several O4-, N6- and C5-substituted
pyrimidines and their biological evaluation as inhibitors
of cdk1 and cdk2.
The X-ray structure of NU6027 bound to cdk2
shows evidence of an intramolecular H-bond between
the 5-nitroso group and the 6-NH. This H-bond
effectively locks the orientation of the 6-amino group
in a conformation that facilitates the donation of an
H-bond to Glu 81 of the enzyme.12 In order to
explore the scope of this interaction, the synthesis of
a variety of derivatives bearing potential H-bond
acceptors at the 5-position was carried out. The 5-for-
myl derivative (12) was prepared from 2-amino-4,6-
dichloro-5-formylpyrimidine (9)15 by sequential dis-
placement of chloride at the 6- and 4-positions with a
masked amine equivalent and cyclohexylmethoxide,
respectively, followed by deprotection. The oximes
(13a–c) were prepared by the reaction of 12 with
hydroxylamine, methoxylamine, and benzylhydroxyl-
amine, respectively, in glacial acetic acid. The
semicarbazone (13d) was prepared in 62% yield by
the reaction of 12 with semicarbazide hydrochlo-
ride (pyridine, EtOH, reflux). Reaction of formyl
compound (12) with three equivalents of phenyl
magnesium bromide gave the alcohol (14) in 74%
yield. Oxidation of 14 to the ketone (15) was
effected with Dess–Martin periodinane (82% yield)
(Scheme 3).
Variation of the substituent at the 4-position of the
pyrimidine was achieved by reaction of 4-chloro-2,6-
diaminopyrimidine with the appropriate sodium alk-
oxide (ROH, NaH, DMSO, 80 ꢂC, or ROH, Na, 150 ꢂC)
to give the corresponding O4-substituted pyrimidines
(2a–g) in 43–84% yield. Nitrosation of compound 2a–f
using sodium nitrite in 30% acetic acid/H2O gave the
required 5-nitroso derivative (1a–g) in 50–86% yield
(Scheme 1).
A series of N6-substituted derivatives (3a–c) was pre-
pared to determine the effect of substitution at this
position (Scheme 1). The cyclohexylmethyl group was
retained as the O4 substituent for this series thereby
allowing a direct comparison with NU6027 (1a). The N-
acyl and carbamate derivatives (3a–c; 30, 44 and 25%
yields, respectively) were prepared from 1a by treatment
with acetic anhydride or the appropriate chloroformate,
respectively.
Compounds were assayed for inhibitory activity against
cdk1/cyclinB1 and cdk2/cyclinA3 using the methods
previously described.12 The final ATP-concentration in
both cdk assays was 12.5 mM. Some of the compounds
assayed were insufficiently soluble for an IC50 determi-
nation. In these cases the activity at 10 mM is given.
Values for inhibition at 100 mM concentration are given
for inactive compounds. The results are summarised in
Table 1.
The introduction of hydrophobic arylamino sub-
stituents at the 6-position required
a
different
Scheme 1. Preparation of O4- and N6- substituted pyrimidines. Reagents and conditions: (a) NaH, R1OH, DMSO, 80 ꢂC or Na, R1OH, 150 ꢂC; (b)
NaNO2, 30% AcOH H2O, 80 ꢂC; (c) (CH3CO)2O, 80 ꢂC; (d) K2CO3/RCOCl, acetone, reflux. R=Me, MeO, BnO.

V. Mesguiche et al. / Bioorg. Med. Chem. Lett. 13 (2003) 217–222
219
Scheme 2. Synthesis of N6-anilino-substituted pyrimidines. Reagents and conditions: (a) ArNH2, 30% EtOH H2O, reflux; (b) HN(PMB)2, Et3N,
n-BuOH, reflux; (c) Na, cyclohexylmethanol, 150 ꢂC; (d) TFA, 60 ꢂC; (e) NaNO2, 30% AcOH, H2O, 80 ꢂC.
Scheme 3. Synthesis of C5- substituted derivatives. Reagents and conditions: (a) POCl3, DMF; (b) HN(PMB)2, Et3N, DCM; (c) NaH, cyclohexyl-
methanol, DMSO, 80 ꢂC; (d) TFA, 60 ꢂC; (e) H2NOR.HCl, AcOH, EtOH; (f) PhMgBr, THF; (g) Dess–Martin periodinane, DCM. R=H, Me, Bn.
The series of O4-substituted pyrimidines revealed that
the nature of the O4-substituent can have a major
impact on cdk inhibitory activity. Substitution at O4
with n-pentyl and n-heptyl, 1b and 1c, produced a large
decrease in activity compared with 1a and a significant
loss of solubility, suggesting that the increased length
and conformational flexibility of these groups is detri-
mental to cdk binding. Interestingly, the iso-pentyl
derivative 1d, retained solubility and displayed modest
inhibitory activity against both cdk1 and cdk2
(IC50=12ꢀ2, 13ꢀ4 mM, respectively), suggesting that a
shorter more compact alkyl chain is preferred. The
oxopyrrolidinylmethyl substituent introduces H-bond-
ing possibilities within the ribose pocket, analogous to
those predicted in the O6-alkylguanine series.13 How-
ever, as in the corresponding O6-guanine derivative any
such interactions are not observed to be favourable and
potency is lost (1e; IC50=57ꢀ14, 65ꢀ26 mM, vs cdk1
and cdk2, respectively). The O4-cyclohexenyl derivative
1f retained potency against cdk1 and cdk2
(IC50=4.0ꢀ0.3 mM and 5.0ꢀ0.1, respectively). How-
ever, in comparison with the parent compound NU6027
(1a; IC50=2.9ꢀ0.1, 2.2ꢀ0.6 mM, vs cdk1 and cdk2,
respectively) the introduction of unsaturation into the
ring is seen to have a small negative effect on potency.
The X-ray structure of 1a bound to cdk2 shows the
cyclohexylmethyl group bound within the ATP ribose
binding pocket of the enzyme.12 The cyclohexane ring
exists in a chair conformation and makes favourable
hydrophobic interactions with a non-polar patch on the
glycine loop formed principally by the sidechain of Val
18. The moderate loss of potency observed with the
unsaturated compound 1f can be explained by flattening
of the cyclohexene ring, which may have a negative
effect on packing complementarity between the inhib-
itor and kinase. This effect is even more dramatic in the
case of the O4-benzyl compound (1g; IC50=27ꢀ2,
27ꢀ3 mM, vs cdk1 and cdk2, respectively),16 suggesting
that the planar benzene ring packs poorly within the
ribose-binding site.
The necessity for the 5-nitroso group, believed to main-
tain an intramolecular H-bond with the 6-amino-group,
has been demonstrated previously with NU6034 (2a).12
This requirement was also found in the case of the O6-
cyclohexenylmethyl compound (2f) which lacks the 5-
nitroso group and showed no significant activity against
cdk1 and cdk2.

220
V. Mesguiche et al. / Bioorg. Med. Chem. Lett. 13 (2003) 217–222
Table 1. Inhibition of cdk1 and cdk2 by pyrimidine derivatives19
IC50 (mM)a or % inhibition
R1
R2
R3
cdk1
cdk2
1a
1b
1c
1d
1e
Cyclohexylmethyl
CH3(CH2)4
CH3(CH2)6
H
H
H
H
H
H
H
H
NO
NO
NO
NO
NO
NO
NO
H
H
NO
NO
NO
NO
NO
CHO
2.9ꢀ0.1
35ꢀ6%b
17ꢀ8%b
12ꢀ2
2.2ꢀ0.6
36ꢀ2%b
22ꢀ7%b
13ꢀ4
(CH3)2CH(CH2)2
(S)-3-Oxopyrrolidin-2-yl)methyl
(3-Cyclohexenyl)methyl
CH2Ph
57ꢀ14
65ꢀ26
1f
4.0ꢀ0.3
27ꢀ2
5.0ꢀ0.1
27ꢀ3
1g
2a
2f
3a
3b
3c
8a
8b
12
13a
13b
13c
13d
14
15
16
Cyclohexylmethyl
(3-Cyclohexenyl)methyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
Cyclohexylmethyl
4ꢀ5%b
16ꢀ14%b
36ꢀ4%c
8ꢀ2%c
14ꢀ6%b
23ꢀ7%c
7%c
7ꢀ3%b
8ꢀ1%b
28ꢀ2c
H
COCH3
CO2CH3
CO2Bn
6ꢀ5%c
11, 24%b
47ꢀ14%c
8%c
C6H5
p-Methoxyphenyl
H
H
H
H
H
H
H
H
35ꢀ3%b
25ꢀ5%c
13ꢀ13%b
5ꢀ6%b
61ꢀ23%c
4ꢀ5%c
21ꢀ23%b
40ꢀ3%c
43ꢀ3%b
37ꢀ23%c
13ꢀ6%b
3ꢀ1%b
44ꢀ28%c
12ꢀ4%c
6ꢀ6%b
52ꢀ6%c
CH=NOH
CH=NOCH3
CH=NOBn
CH=NNH(CO)NH2
CH(Ph)OH
C(Ph)O
NH2
a12.5 mM ATP concentration.
b% Inhibition at 10 mM.
c% Inhibition at 100 mM.
Substitution at the 6-amino group resulted in a loss of
cdk inhibitory activity (3a–c, 8a, 8c). This is in accor-
dance with the binding mode determined from the crys-
tal structure of 1a bound to cdk2. The disruption of
either the H-bond between the 6-amino NH and Glu 81
or the intramolecular H-bond to the 5-nitroso group
would be predicted to result in a significant loss of
activity.
of steric tolerance at this position. Interestingly, the 5-
amino compound (16) displayed significant activity at
10 mM against both cdks, but was sparingly soluble.
The 5-oxime derivatives (13a–c) and the 5-semi-
carbazone derivative (13d) were expected to retain the
intramolecular H-bond and allow new interactions with
the enzyme binding pocket. However, none of these
derivatives displayed good activity against the cdks and
their solubility was poor.
The reduction in inhibitory activity resulting from
replacement of the 5-nitroso group with an isosteric
formyl group in 12 (% inhibition vs cdk and cdk2 at
10 mM=35ꢀ3 and 43ꢀ3%, respectively) was surpris-
ing. The formyl group was expected to be able to form
an intramolecular H-bond with the 6-amino group and
so restrict the 6-amino group in the active conforma-
tion. In order to investigate this further, the crystal
structure of 12 bound to the phosphorylated cdk2-
cyclinA complex was determined at 2.3 A (Fig. 1).17
Examination of this complex showed that the formyl
group was not coplanar with the 6-amino group. The
orientation of the formyl group could not be attributed
to an intermolecular H-bond with the enzyme or any
other structural feature and so is unexpected. This
apparent absence of the anticipated intramolecular H-
bond may explain the significant loss in activity
observed for 12 compared with the nitroso-compound
1a. The 5-phenylmethanol derivative (14) and the 5-
phenylketone derivative (15) showed a significant
decrease in activity compared with 1, suggesting a lack
Recently, 4-amino-6-(40-sulfanilyl)pyrimidine has been
described as an ATP-competitive cdk2/cyclinE inhibitor
with a Ki of 2 mM.18 When soaked into cdk2 crystals,
like 1a, this 4-aminopyrimidine formed H-bonds with
Glu 81 and Leu 83, confirming the utility of the amino-
pyrimidine pharmacophore in structure based cdk inhib-
itor design.
In summary, a series of 4-alkoxy-2,6-diaminopyrimidine
derivatives has been prepared and evaluated as inhibi-
tors of cdk1 and cdk2. The ATP-ribose binding pocket
of the enzyme is occupied by the 4-alkoxy substituent
and the structural requirements are rigorous. The
cyclohexylmethoxy group is optimal, with minor modi-
fications to the ring resulting in loss of potency. The 5-
nitroso group is required for activity, with the isosteric
5-formyl derivative surprisingly lacking potency. Modi-
fications at the 6-amino group were not tolerated. In
conclusion, the parent compound NU6027 (1a) is the

V. Mesguiche et al. / Bioorg. Med. Chem. Lett. 13 (2003) 217–222
221
Figure 1. (A) Crystal structure of compound 12 bound to the phospho-cdk2-cyclinA complex. Compound 12 shown in cyan, phospho-cdk2 in
purple, and cyclinA in lime. (B) Stereoview of the active site showing two H-bonds between compound 12 (green) and cdk2 with the structure of
NU6027 (grey) superimposed.
Rusnak, D.; Shewchuk, L.; Veal, J. M.; Walker, D. H.; Kuy-
per, L. F. J. Med. Chem. 2001, 44, 4339.
6. Dreyer, M. K.; Borcherding, D. R.; Dumont, J. A.; Peet,
N. P.; Tsay, J. T.; Wright, P. S.; Bitonti, A. J.; Shen, J.; Kim,
S.-H. J. Med. Chem. 2001, 44, 524.
most potent inhibitor of cdk1 and cdk2 discovered so
far in this series and represents the optimal starting
point for cdk inhibitory structure activity studies in the
future.
7. Killday, K. B.; Yarwood, D.; Sills, M. A.; Murphy, P. T.;
Hooper, J. N. A.; Wright, A. E. J. Nat. Prod. 2001, 64, 525.
8. Lane, M. E.; Yu, B.; Rice, A.; Lipson, K. E.; Liang, C.;
Sun, L.; Tang, C.; McMahon, G.; Pestel, R. G.; Wadler, S.
Cancer Res. 2001, 61, 6170.
9. Sielecki, T. M.; Johnson, T. L.; Liu, J.; Muckelbauer, J. K.;
Grafstrom, R. H.; Cox, S.; Boylan, J.; Burton, C. R.; Chen,
H.; Smallwood, A.; Chang, C.-H.; Boisclair, M.; Benfield,
P. A.; Trainor, G. L.; Seitz, S. P. Bioorg. Med. Chem. Lett.
2001, 11, 1157.
Acknowledgements
The authors thank Cancer Research UK, the
BBSRC (Studentship to R.J.P.), the MRC, and
AstraZeneca Pharmaceuticals for financial support.
Richard Davison and Paula Mackley are gratefully
acknowledged for technical support.
10. Hardcastle, I. R.; Golding, B. T.; Griffin, R. J. Annu. Rev.
Pharmacol. Toxicol. 2002, 42, 325.
11. Sausville, E. A.; Zaharevitz, D.; Gussio, R.; Meijer, L.;
Louarn-Leost, M.; Kunick, C.; Schultz, R.; Lahusen, T.;
Headlee, D.; Stinson, S.; Arbuck, S. G.; Senderowicz, A.
Pharmac. Ther. 1999, 82, 285.
12. Arris, C. E.; Boyle, F. T.; Calvert, A. H.; Curtin, N. J.;
Endicott, J. A.; Garman, E. F.; Gibson, A. E.; Golding, B. T.;
Grant, S.; Griffin, R. J.; Jewsbury, P.; Johnson, L. N.; Lawrie,
A. M.; Newell, D. R.; Noble, M. E. N.; Sausville, E. A.;
Schultz, R.; Yu, W. J. Med. Chem. 2000, 43, 2797.
13. Gibson, A. E.; Arris, C. E.; Bentley, J.; Boyle, F. T.;
Curtin, N. J.; Davies, T. G.; Endicott, J. A.; Golding, B. T.;
Grant, S.; Griffin, R. J.; Jewsbury, P.; Johnson, L. N.; Mes-
guiche, V.; Newell, D. R.; Noble, M. E. N.; Tucker, J. A.;
Whitfield, H. J. J. Med. Chem. 2002, 45, 3381.
References and Notes
1. Norbury, C.; Nurse, P. A. Ann. Rev. Biochem. 1992, 61,
441.
2. Morgan, D. O. Nature 1995, 374, 131.
3. Sielecki, T. M.; Boylan, J. F.; Benfield, P. A.; Trainor, G. L.
J. Med. Chem. 2000, 43, 1.
4. Barvian, M.; Boschelli, D. H.; Cossrow, J.; Dobrusin, E.;
Fattaey, A.; Fritsch, A.; Fry, A.; Harvey, P.; Keller, P.; Gar-
rett, M.; La, F.; Leopold, W.; McNamara, D.; Quin, M.;
Trumpp-Kallmeyer, S.; Toogood, P.; Wu, Z.; Zhang, E. J.
Med. Chem. 2000, 43, 4606.
5. Bramson, H. N.; Corona, J.; Davis, S. T.; Dickerson, S. H.;
Edelstein, M.; Frye, S. V.; Gampe, R. T., Jr.; Harris, P. A.;
Hassell, A.; Holmes, W. D.; Hunter, R. N.; Lackey, K. E.;
Lovejoy, B.; Luzzio, M. J.; Montana, V.; Rocque, W. J.;
14. Gray, N. S.; Wodicka, L.; Thunnissen, A.-M. W. H.;
Norman, T. C.; Kwon, S.; Espinoza, F. H.; Morgan, D. O.;

222
V. Mesguiche et al. / Bioorg. Med. Chem. Lett. 13 (2003) 217–222
Barnes, G.; LeClerc, S.; Meijer, L.; Kim, S.-H.; Lockhart,
D. J.; Schultz, P. G. Science 1998, 281, 533.
17. PDB code: 1h3s. Crystallographic R-factor=23%; free R-
factor=27%.
15. Bell, L.; McGuire, H. M.; Freeman, G. A. J. Heterocyclic.
Chem. 1983, 20, 41.
16. Griffin, R. J.; Arris, C. E.; Bleasdale, C.; Boyle, F. T.;
Calvert, A. H.; Curtin, N. J.; Dalby, C.; Kanugula, S.; Lem-
bicz, N. K.; Newell, D. R.; Pegg, A. E.; Golding, B. T. J. Med.
Chem. 2000, 43, 4071.
18. Clare, P. M.; Poorman, R. A.; Kelly, L. C.; Watenpaugh,
K. D.; Bannow, C. A.; Leach, K. L. J. Biol. Chem. 2001, 276,
48292.
19. All compounds gave satisfactory spectroscopic (NMR,
IR, UV), EI or ESI MS, and combustion analyses.