Exploring 8-benzyl pteridine-6,7-diones as inhibitors of glutamate racemase (MurI) in Gram-positive bacteria

Article abstract of DOI:10.1016/j.bmcl.2008.10.022

A successful scaffold-hopping approach gave a novel series of inhibitors of bacterial glutamate racemase (MurI). Early SAR studies of the 8-benzyl pteridine-6,7-diones led to compounds with micromolar enzyme potency and antibacterial activity.

Full text of DOI:10.1016/j.bmcl.2008.10.022

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6100–6103
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Exploring 8-benzyl pteridine-6,7-diones as inhibitors of glutamate racemase
(MurI) in Gram-positive bacteria
*
Gloria A. Breault , Janelle Comita-Prevoir, Charles J. Eyermann, Bolin Geng, Randy Petrichko, Peter Doig,
Elise Gorseth, Brian Noonan
Infection Discovery, AstraZeneca R&D Boston, 35 Gatehouse Drive, Waltham, MA 02451, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A successful scaffold-hopping approach gave a novel series of inhibitors of bacterial glutamate racemase
(MurI). Early SAR studies of the 8-benzyl pteridine-6,7-diones led to compounds with micromolar
enzyme potency and antibacterial activity.
Received 29 August 2008
Accepted 6 October 2008
Available online 8 October 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Glutamate racemase
MurI inhibitor
8-Benzyl pteridine-6,7-dione
UDP-MurNAc-Ala
COOH
MurI
COOH
H
H2N
COOH
MurD
H2N
COOH
H
L-Glutamate
D-Glutamate
UDP-MurNAc-Ala-(D)Glu
peptidoglycan
Scheme 1. The role of glutamate racemase (MurI) in the peptidoglycan precursor biosynthetic pathway.
The increasing prevalence of infections caused by resistant
bacteria has added urgency to the search for antibacterial agents
with novel mechanisms of action.1,2 The bacterial enzyme glutamate
racemase (MurI) has potential as a target for the development of
such novel antibacterial agents. MurI, an enzymeinvolvedin thebio-
synthesis of bacterial peptidoglycan (see Scheme 1), has been shown
to be essential in several bacterial species.3–5 Moreover, the MurI
gene is conserved in all species that produce peptidoglycan.6
High-throughput screening of the AstraZeneca compound col-
lection identified a number of hits including the 9-benzyl purines
shown in Figure 1. Our work on this series has been described in
a recent publication.7 Herein we describe a successful ‘scaffold-
hopping’ endeavor from the 9-benzyl purines to the 8-benzyl pte-
ridinediones and our hit-to-lead work on this series.
Inhibition data on the 9-benzyl purines showed good activity
against Enterococcus faecalis (E. fa) and Enterococcus faecium
(E. fm) MurI. However no activity against the Staphylococcus aureus
(S. au) isozyme could be detected. As S. aureus was seen as a key
pathogen, these results prompted us to consider alternate strate-
gies for identification of a novel lead series.
Many scaffold-hopping studies have been reported in the recent
literature.8 Several useful reviews have been published that clarify
the diverse strategies.9 These strategies have encompassed a range
* Corresponding author. Tel.: +1 781 839 4299; fax: +1 781 839 4230.
E-mail address: gloria.breault@astrazeneca.com (G.A. Breault).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.022

G. A. Breault et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6100–6103
6101
NH2
N
NH2
H
N
scaffold hop
O
O
N
N
N
N
S
N
N
S
1
2
O2N
Figure 1. 9-Benzyl purine hit (1) and 8-benzyl pteridinedione (2).
functionalized pyrimidine (compound 3) as shown in Figure 2.
Searching of the AstraZeneca corporate collection using this sub-
5
NH2
N
NH2
4
structure yielded
a number of different potential scaffolds.
H
N
6
Among those tested was 2-benzylthio 8-benzyl pteridinedione,
2. This compound showed activity against the S. aureus enzyme
as well as E. faecalis and E. faecium (see Table 1). Our initial goals
were to improve the enzyme potency, achieve antibacterial
activity and improve the physico-chemical properties. Our strat-
egy to achieve these goals was to investigate variation of the
groups at the 2- and 8-positions initially.
O
NH2
3
N
N
2
7
H
N
S
S
N
1
N
O
8
3
2
The synthesis of the 2-position variants began with the prepara-
tion of the 2-alkylthio pteridinedione core (Scheme 2).11 Treatment
of 2-thio-4,6-diamino pyrimidine with an alkyl bromide under ba-
sic conditions gave the 2-alkylthio-4,6-diamino pyrimidine. Nitro-
sation was accomplished by treatment with sodium nitrite in
water. Catalytic hydrogenation of the nitroso gave 2-butylthio-
4,5,6-triamino pyrimidine. Cyclization to the pteridinedione was
achieved by treatment with ethyl chloro oxalate. Alkylation at
the 8-position was accomplished by subjecting the core to benzyl
bromide and triethyl amine in DMF.
The results for the 2-position variants are shown in Table 2.
Introduction of small cyclic groups at the 2-position retained E. fae-
calis activity but led to a 10-fold loss of potency against S. aureus
enzyme (compound 11 where R2 is S-cyclopentyl). Extending the
group at the 2-position, compound 12 R = S-phenethyl, also led to
unfavorable interaction for the E. faecalis enzyme. From our inves-
tigation, we selected S-n-butyl as the best group to use at the 2-po-
sition for further optimization studies. This choice was based on
the favorable enzyme activity as well as the favorable solubility ob-
served for compound 10.
Figure 2. Ring opened core used to identify novel scaffold.
Table 1
Comparison of purine and pteridinedione series
Compound
E. fa MurI IC50
M)
E. fm MurI IC50
M)
S. au MurI IC50
M)
Solubilityb
(lM)
a
a
a
(
l
(
l
(
l
1
2
7
21
18
nt
>400
7.6
10
3
a
Activity against MurI isozymes from several bacterial species was assessed by
IC50 determination. For assay details see Refs. 6 and 10. nt, not tested.
b
Solubility (nephelometry) measured in the assay medium.
of approaches from ligand-based virtual screening to receptor-
based screening and combined methods, which make use of
increasingly complex computational techniques. The ligand-based
approaches have employed similarity searches based on substruc-
ture, shape, fingerprint and 3-D potential pharmacophore point
pair descriptors. The receptor-based methods often use automated
docking of potential scaffolds to the enzyme or receptor pocket
from X-ray co-crystal structures or homology models. Combination
methods may use multiple elements from either ligand-based or
structure-based approaches.
We then began to modify the 8-position, holding the 2-position
constant. The synthesis of the 8-position derivatives was analogous
to the sequence used earlier substituting n-butyl bromide in the
first step and using the appropriate alkylating agent in the last step.
The results obtained for the 8-position variants are shown in Table
3. Replacing the benzyl with n-butyl (compound 13) led to weaker
enzyme activity. Substitution of the benzyl group with electron-
Our approach was ligand-based and relied largely on
substructure searching. We considered our starting scaffold to
be a 5,6-bicyclic system bearing two side chains. Fragmenting
the 5-membered ring gave a monocyclic template, the highly
NH2
NH2
NH2
N
R2 Br
a
O
N
N
N
R2
R2
b
c
H2N
N
S
H2N
N
SH
S
N
H2N
4
5
O
6
NH2
NH2
NH2
O
H
N
H
Cl
Et
H2N
O
O
O
N
O
N
N
N
R2
R2
R2
e
d
O
N
H
N
S
H2N
N
S
N
N
S
R8
7
8
9
Scheme 2. General route to 8-substituted pteridinediones. Reagents and conditions: (a) NaOH, methanol, R2Br, DMF; (b) NaNO2, water; (c) H2, PtO2, ethanol; (d) NMP, 0 °C;
(e) BzBr or other R8 electrophiles, Et3N, DMF.

6102
G. A. Breault et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6100–6103
Table 2
Table 3
R2 variations
R8 variations
NH2
NH2
H
N
H
N
O
O
O
O
N
N
N
N
S
N
N
R2
R8
Compound
R8
E. fa
S. au
Solubilityb
M)
S. au
MurI
MurI
(
l
MICc
a
a
IC50
IC50
(lg/mL)
Compound
R2
E. fa
S. au
Solubilityb
M)
S. au
(
l
M)
(lM)
MurI
MurI
(l
MICc
a
a
13
10
11
nt
25
25
>64
>64
IC50
IC50
(lg/mL)
(lM)
(lM)
2.4
6.4
S
S
2
21
7.6
3
nt
O
10
11
2.4
6.4
25
12
3
>64
>64
nt
14
15
1.7
14
50
50
>64
O
F
1.9
52
11
S
S
1.5
3.1
>64
>64
12
50
F
a
Activity against MurI isozymes from several bacterial species was assessed by
IC50 determination. For assay details see Ref. 10, nt, not tested.
Cl
b
16
0.65
<1.5
3.1
Solubility (nephelometry) measured in the assay medium.
MIC values were measured according to NCCLS guidelines.
c
Cl
a
Activity against MurI isozymes from several bacterial species was assessed by
IC50 determination. For assay details see Ref. 10. nt, not tested.
donating groups, for example 3,5-dimethoxy as in compound 14,
gave a similar level of potency to the unsubstituted benzyl. Intro-
duction of electron-withdrawing groups, such as 2,6-difluoro,
compound 15, was well tolerated. Increased potency was obtained
with the 3,4-dichloro substitution, compound 16.
b
Solubility (nephelometry) measured in the assay medium.
MIC values were measured according to NCCLS guidelines.
c
S. aureus enzymes and solubility of 32 lM. Substituting with a sin-
gle methyl on the R4 nitrogen (compound 26) led to a compound
with similar enzymatic potency and solubility and an MIC of
The exploration of the 2- and 8-positions suggested that lipo-
philic groups were preferred for optimal biochemical potency. At
this point we needed to improve the physical properties and to
make further gains in enzyme and antibacterial potency. We
turned to structural information for insights into possible other
directions. Enzyme-inhibitor co-crystal structures had been ob-
tained with a number of the purine derivatives.7 Attempts to ob-
tain co-crystal structures for the pteridinedione analogs were
unsuccessful. Docking of a pteridinedione compound12 suggested
that the 4-position pointed toward the solvent exposed region of
the binding site (Fig. 3). We hypothesized that the 4-position could
thus be used to modulate physical properties.
8
l
g/mL against S. au. Changing to the dimethyl amine at the R4
position, (Compound 27), gave reasonable enzyme activity but
very much reduced solubility. Compound 27 had an MIC of 4 g/
l
mL. Introduction of a small cyclic amine, 3-hydroxy azetidine, gave
compound 28, which showed good enzyme activity but poor solu-
bility and loss of antibacterial activity. Solubility could be restored
by inserting di-basic groups such as 4-methyl piperazine (com-
pound 29). However, we observed a loss of enzyme potency for
both isozymes for this analog. In a similar way, introduction of
2-morpholino ethylamine at the R4 position gave improved solu-
The general route used to prepare these 4-position analogs is
shown in Scheme 3. Compound 17, 6-amino 4-hydroxy 2-thiopy-
rimidine, was treated with n-butyl bromide under basic conditions.
Preparation of the triflate was accomplished by treatment with tri-
flic anhydride. Displacement with amines was carried out in meth-
anol at 60 °C. Treatment of compound 20 with sodium nitrite and
acetic acid in water gave the nitroso intermediates 21. Reduction
with zinc and acetic acid gave the triamino compounds 22. Cycliza-
tion was then accomplished by treating with ethyl chloro oxalate
in N-methyl pyrrolidone at 0 °C with warming to 120 °C. This core
could then be alkylated using the appropriate benzylic halide and
triethylamine in dimethyl formamide (DMF). The results for the
4-position analogs are shown in Table 4.
For the study of the 4-position we selected the 4-fluorobenzyl
substituent at the 8-position. This group was held constant for
the analogs shown in Table 4. The parent molecule, compound
25, (R4 = NH2), showed good potency against E. faecalis and
Figure 3. 8-(4-Fluorobenzyl) pteridine-6,7-dione docked into the MurI allosteric
binding site.

G. A. Breault et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6100–6103
6103
OH
N
OTf
OH
Tf2O
b
Br
N
N
N
a
H2N
SH
R1
N
H2N
S
H2N
S
N
17
18
19
R1
R2
R1
R2
N
N
R2
N
N
H
d
N
e
N
O
c
H2N
N
S
S
N
H2N
20
21
O
O
R2
R1
R2
N
R1
R2
R1
Cl
N
N
N
N
N
N
O
H
N
H
N
g
H2N
O
O
O
N
N
f
S
S
H2N
N
H
O
S
N
R8
22
23
24
Scheme 3. General route to 4-substituted pteridinediones. Reagents and conditions: (a) NaOH, MeOH; (b) Tf2O; (c) R1R2NH, MeOH, 60 °C; (d) HOAc, H2O, NaNO2, 0 °C; (e) Zn,
HOAc, 70 °C; (f) NMP, 0–120 °C; (g) 4F-BzBr, Et3N, DMF.
identified analogs with good enzyme potency and improved phys-
ical properties. Some of the most potent analogs showed modest
antibacterial activity.
Table 4
R4 variations (compound 28)
R4
N
H
N
O
O
N
References and notes
N
S
F
1. Bush, K. Clin. Microbiol. Infect. 2004, 10, 10.
2. Payne, D. J.; Gwynn, M. N.; Holmes, D. J.; Rosenberg, M. Methods Mol. Biol. 2004,
266, 231.
3. (a) Doublet, P.; van Heijenoort, J.; Mengin-Lecreulx, D. J. Bacteriol. 1992, 174,
5772; (b) Glavas, S.; Tanner, M. E. Biochemistry 2001, 40, 6199.
4. Doublet, P.; van Heijenoort, J.; Mengin-Lecreulx, D. Biochemistry 1994, 33, 5285.
5. Van Heijenoort, J. Nat. Prod. Rep. 2001, 18, 503.
6. For investigation of structural and regulatory diversity of glutamate racemases,
see Lundqvist, T.; Fisher, S. L.; Kern, G.; Folmer, R. H. A.; Xue, Y.; Newton, D. T.;
Keating, T. A.; Alm, R.; de Jonge, B. L. M. Nature 2007, 447, 817.
Compound R4
E. fa
MurI
IC50
S. au
Solubilityb S. au
M)
MICc
g/
MurI
(l
a
a
IC50
(l
7. Geng, B.; Breault, G.; Comita-Prevoir, J.; Petrichko, R.; Eyermann, J. C.; Doig, P.;
Gorseth, E.; Noonan, B. Bioorg. Med. Chem. Lett. 2008, 18, 4368.
(
l
M)
(
l
M)
mL)
25
26
27
NH2
NHCH3
N(CH3)2
3.5
0.9
1.1
1.1
3.8
nt
32
25
0.8
>64
8
4
8. For case studies of scaffold-hopping, see (a) Hall, A.; Billinton, A.; Brown,
S. H.; Chowdhury, A.; Giblin, G. M. P.; Goldsmith, P.; Hurst, D. N.; Naylor,
A.; Patel, S.; Scoccitti, T.; Theobald, P. J. Bioorg. Med. Chem. Lett. 2008, 18,
2684; (b) Feher, M.; Gao, Y.; Baber, J. C.; Shirley, W. A.; Saunders, J.
Bioorg. Med. Chem. 2008, 16, 422; (c) Larbig, G.; Pickhardt, M.; Lloyd, D.
G.; Schmidt, B.; Mandelkow, E. Curr. Alzheimer Res. 2007, 4, 315; (d)
O’Meara, J. A.; Jakalian, A.; LaPlante, S.; Bonneau, P. R.; Coulombe, R.;
Paucher, A.-M.; Suse, I.; Landry, S.; Racine, J.; Simoneau, B.; Thavonek, B.;
Yoakim, C. Bioorg. Med. Chem. Lett. 2007, 17, 3362; (e) Bostroem, J.;
Berggren, K.; Elebring, T.; Greaslye, P. J.; Wilstermann, M. Bioorg. Med.
Chem. 2007, 15, 4077.
9. For reviews of scaffold-hopping, see (a) Zhao, H. Drug Discovery Today 2007, 12,
149; (b) Schneider, G.; Schneider, P.; Renner, S. QSAR Comb. Sci. 2006, 25, 1162;
(c) Brown, N.; Jacoby, E. Mini-Rev. Med. Chem. 2006, 6, 1217.
10. Compound potency was based on IC50 measurements determined from
reactions performed in the presence of ten different compound
OH
28
29
30
2.1
6.1
2.5
4.0
21
25
1.6
25
50
>64
16
N
N
N
O
64
N
N
H
a
Activity against MurI isozymes from several bacterial species was assessed by
IC50 determination. For assay details see Ref. 10. nt, not tested.
concentrations. Compound IC50
aureus glutamate racemase at pH 8 in the presence of Tris buffer containing
polyethylene glycol, dithiothreitol and -glutamate. The conversation of
glutamate to -glutamate was monitored by HPLC using chiral separation
s were determined for E. faecalis and S.
b
Solubility (nephelometry) measured in the assay medium.
MIC values were measured according to NCCLS guidelines.
D
D-
c
L
of the glutamate enantiomers on
column.
a Phenomenex Chirex (D)-Penicillamine
bility (comp4ound 30) but a reduction in activity against the S. aur-
eus enzyme.
In summary, using scaffold-hopping techniques we were able to
transfer from a lead series with restricted spectrum to a novel ser-
ies with activity against the isozymes of the important pathogenic
bacterial species. By exploration of the 2-, 4- and 8-positions we
11. (a) Forrest, H. S.; Hull, R.; Rodda, H. J.; Todd, A. R. J. Chem. Soc. 1951, 3; (b)
Bonnert, R. V.; Cage, P. A.; Hunt, S. F.; Walters, I. A. S.; Austin, R. P. PCT Int. Appl.
WO 2003024966, 2003.; (c) Walters, I.; Austin, Caroline; Austin, Rupert;
Bonnert, Roger; Cage, P.; Christie, M.; Ebden, M.; Gardiner, S.; Grahames, C.;
Hill, S.; Hunt, F.; Jewell, R.; Lewis, S.; Martin, I.; Nicholls, D.; Robinson, D. Bioorg.
Med. Chem. Lett. 2008, 18, 798.
12. McMartin, C.; Bohacek, R. S. J. Comput. Aided Mol. Des. 1997, 11, 333.