A series of 2(Z)-2-benzylidene-6,7-dihydroxybenzofuran-3[2H]-ones as inhibitors of chorismate synthase

Article abstract of DOI:10.1016/S0960-894X(02)00957-5

A series of 2(Z)-2-benzylidene-6,7-dihydroxybenzofuran-3[2H]-ones was identified as potent inhibitors of bacterial chorismate synthase. The 2′-hydroxy-4′-pentoxy analogue 33 is a potent inhibitor of Streptococcus pneumoniae chorismate synthase.

Full text of DOI:10.1016/S0960-894X(02)00957-5

Bioorganic & Medicinal Chemistry Letters 13 (2003) 423–426
A Series of 2(Z)-2-Benzylidene-6,7-dihydroxybenzofuran-3[2H]-ones
as Inhibitors of Chorismate Synthase
Michael G. Thomas,* Chris Lawson, Nigel M. Allanson, Bruce W. Leslie,
Joanna R. Bottomley, Andrew McBride and Oyinkan A. Olusanya
PanTherix Ltd., Todd Campus, West of Scotland Science Park, Glasgow G20 0XA, Scotland, UK
Received 13 August 2002; revised 14 October 2002; accepted 30 October 2002
Abstract—A series of 2(Z)-2-benzylidene-6,7-dihydroxybenzofuran-3[2H]-ones was identified as potent inhibitors of bacterial
chorismate synthase. The 20-hydroxy-40-pentoxy analogue 33 is a potent inhibitor of Streptococcus pneumoniae chorismate synthase.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
The emergence of bacterial resistance to established
antibiotics in community and hospital-acquired infec-
tions is an ever-growing concern. In particular, the ‘big
three’ Gram-positive species, staphylococcus, entero-
coccus and streptococcus, have all evolved resistant
Figure 1. Structures of glyphosate 1 and 6(S)-fluoroshikimate 2.
strains over the last few years: methicillin-resistant Sta-
phylococcus aureus (MRSA) strains now account for
over 40% of S. aureus infections in hospitals;1 vanco-
mycin-resistant enterococcal (VRE) infections have also
increased substantially, with some strains resistant to all
established antibiotics;2 Streptococcus pneumoniae is the
most common cause of community-acquired pneumoniae
and is increasingly resistant to b-lactams and macro-
lides. For these reasons, there is an urgent need for new
antibacterial agents, particularly those that act on novel
targets.
Chorismate synthase (CS) is a key enzyme in the path-
way and converts EPSP to chorismate via a 1,4-trans
elimination of phosphate5 (Scheme 1). The reaction is
unusual in that there is an absolute requirement for
reduced FMN as a cofactor, although the reaction does
not involve a net redox change.
The aroC gene from S. pneumoniae was amplified by
PCR and cloned into a pET-16b expression vector sys-
tem (Novagen). The N-terminally His-tagged protein
was over-expressed in Escherichia coli by induction with
IPTG, and purified to homogeneity in three chroma-
tographic steps.
The shikimic acid pathway is essential for the synthesis
of aromatic amino acids in bacteria. Because the path-
way is absent in man, it is an attractive target for
potential antibiotics. The pathway has been validated as
an antimicrobial target by Monsanto’s herbicide gly-
phosate 1 (Roundup1)3 which acts on EPSP (5-enol-
pyruvyl shikimate-3-phosphate) synthase, and the
Zeneca compound 6(S)-fluoroshikimate 2,4 which, after
conversion to 6(S)-fluoro-EPSP, is a substrate for chor-
ismate synthase (Fig. 1).
*Corresponding author. Tel.: +44-141-581-2524; fax: +44-141-581-
6373; e-mail: michael.thomas@pantherix.co.uk
Scheme 1. Conversion of EPSP to chorismate.
0960-894X/03 - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00957-5

424
M. G. Thomas et al. / Bioorg. Med. Chem. Lett. 13 (2003) 423–426
S. pneumoniae CS (SpCS) was assayed using a phos-
phate release detection system.6 Briefly; CS, FMN,
EPSP and the compound of interest were incubated
together under anaerobic conditions. The reaction was
initiated by addition of dithionite to reduce FMN.
Phosphate was detected continuously by coupling its
production to the cleavage of 2-amino-6-mercapto-
7-methylpurine ribonucleoside by purine nucleoside
phosphorylase and measuring the increase in absor-
bance at 360 nm. For screening purposes, the enzymic
solution was overlaid with mineral oil prior to the
addition of sodium dithionite.
An extensive library of chemically diverse small mole-
cules was screened for inhibitors of SpCS. These inclu-
ded the benzofuran-3[2H]-one 3. The compound was
moderately potent, with an IC50 of 8 mM. Kinetic stud-
ies demonstrated that 3 was a competitive inhibitor with
respect to EPSP (Ki 0.65 mM), and non-competitive with
respect to the cofactor FMN (aKi 1.7 mM), suggesting
that the compound was binding in the active site of the
enzyme. The Km values for EPSP and FMN are 30 and
0.08 mM, respectively.
Scheme 3. Reagents and conditions: (a) (i) chloroacetic acid, phos-
phorus oxychloride, 70 ꢀC; (ii) sodium acetate, ethanol, reflux, 28%;
(b) HCl, ethanol, reflux, 50–95%; (c) NaOH, ethanol, reflux, 50–75%.
Biology
Modification of the substituents on the benzylidene
group of 3 led to more potent analogues. Based on the
hypothesis that the nitrogen of the 40-diethylamino group
was acting as a hydrogen bond acceptor, while the ethyl
groups were filling a hydrophobic pocket, the effect of
varying the dialkylamino group was examined (Table 1).
We initiated a chemistry programme to improve CS
potency. The 2-benzylidene group was modified, whilst the
catechol was left unchanged in order to retain solubility.
The dimethylamino analogue 4 showed a 3-fold lower
potency than 3. However, the pyrrolidine analogue 5
was slightly more potent, whilst the morpholine ana-
logue 6 was significantly less so. This indicated that
there was limited space for further substitutions around
the 40-position.
As changes to the N-alkyl substituents led to limited
improvement in potency, alternative 40-substituents were
examined (Table 2). Surprisingly, the unsubstituted ana-
logue 9 had an IC50 of 3.5 mM, suggesting that the
40-amino group did not contribute favorably to binding.
Chemistry
2-Hydroxy-4-alkoxy benzaldehydes were synthesized by
alkylation of 2,4-dihydroxybenzaldehyde in refluxing
DMF, with sodium bicarbonate as base7 (Scheme 2).
Of the 40-substituted analogues, methyl 10, nitrile 11,
carboxylate 12 and fluoro 16 proved inactive, whilst
moderate potency was seen for trifluoromethyl 13, nitro
14, hydroxy 17 and alkoxy analogues 18–20. In this latter
2-Benzylidenebenzofuran-3[2H]-ones were synthesized
according to Scheme 3. 6,7-Dihydroxybenzofuran-
3[2H]-one was synthesized from pyrogallol by treatment
with chloroacetic acid and phosphorus oxychloride to
generate the chloroacetophenone. This was cyclised
using sodium acetate8 in refluxing ethanol.
Table 1. Effect of changes to the 40-amino group on inhibition of
Streptococcus pneumoniae chorismate synthase, compounds 3–8
The condensation of 6,7-dihydroxybenzofuran-3[2H]-
one with variously substituted benzaldehydes was car-
ried out in refluxing ethanol, in the presence of either
HCl or NaOH. Condensations gave single isomers, as
shown by both LC–MS and NMR experiments. Litera-
ture precedent9 suggested that the Z-isomer would be
obtained, and this was confirmed by crystallography.
Compd
R1
R2
Activity versus SpCS
IC50, mM
3
4
5
6
7
8
N(CH2CH3)2
N(CH3)2
Pyrrolidino
H
H
H
H
OH
Me
8.5
25.9
5.5
>50
5.0
3.4
Morpholino
N(CH2CH3)2
N(CH2CH3)(CH2)2OH
Scheme 2. Reagents and conditions: (a) NaHCO3, DMF, reflux,
28–55%.

M. G. Thomas et al. / Bioorg. Med. Chem. Lett. 13 (2003) 423–426
425
Table 2. Effect of 40-substituents on inhibition of Streptococcus
pneumoniae chorismate synthase, compounds 9–20
Table 4. Inhibition of Streptococcus pneumoniae chorismate synthase
for compounds 21 and 29–45, 20-hydroxy substituted analogues
Compd
R1
Activity versus SpCS
IC50, mMa
Compd
R1
R2
Activity versus SpCS
IC50, mM
9
H
CH3
CN
3.5
n/a
n/a
24
7
4.8
21
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
H
OCH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
0.80
1.7
10
11
12
13
14
15
16
17
18
19
20
O
CH
0.70
0.51
0.45
0.22
0.32
1.0
0.58
1.0
2.5
0.86
1.6
10.3
2.6
0.65
1.1
1.1
3
COOH
CF3
NO2
N-Imidazoyl
F
OH
OCH3
OCH2CH3
O(CH2)3CH3
O(CH2)2CH3
O(CH2)3CH3
O(CH2)4CH3
O(CH2)5CH3
OCH(CH3)2
2CH(CCHH3)2
5.1
>60
15.0
14.3
5.8
O
O
2C6H5 CH
O(CH2)3OH
O(CH2)4OH
O(CH2)6OH
2COOECtH
2COOHCH
O(CH2)3COOEt
O(CH2)3COOH
O(CH2)3CN
1.5
an/a, inactive compounds.
O
O
group, we observed increasing potency with longer
chain lengths, with the 40-butoxy compound having an
IC50 of 1.5 mM.
H
Next, the effects of changes at the ortho position of the
benzylidene group were investigated (Table 3). A number
of functional groups were tolerated in the 20-position.
Methyl, chloro, nitro and oxyacetic acid analogues (22, 24,
26 and 28, respectively) all had similar IC50’s to the unsub-
stituted benzylidene compund, whereas the carboxylate
derivative 27 was less potent and the methoxy compound
23 was inactive. The best inhibitor was the 20-hydroxy
analogue 21, with an IC50 of 0.8 mM. Comparison of com-
pounds 21 (20-hydroxy) and 23 (20-methoxy) suggested that
the hydroxyl group was acting as a hydrogen bond donor.
the pentoxy-analogue 33 at 0.22mM, and decreasing
thereafter with the hexyloxy analogue 34. This represented
a 40-fold improvement in potency over the initial lead.
The effect of chain branching was investigated. The
40-isopropoxy analogue 35 showed a 2-fold lower
potency than the n-propoxy compound 31, and the iso-
butoxy 36 was less potent than the n-butoxy derivative
32. The introduction of an O-benzyl substituent was also
detrimental to potency.
Lastly, compounds were synthesized that combined the
ortho-hydroxy and para-alkoxy substituents on the
benzylidene ring (Table 4). Potency improved with
increasing alkoxy chain length, reaching a maximum with
In order to compare the binding of this series of ana-
logues with the lead compound 3, kinetic studies were
performed on compound 30. This analogue was shown
to be competitive with EPSP (Ki 0.10 mM) and non-
competitive with FMN (aKi 1.4 mM), indicating that
these compounds were binding in the same way as 3.
Table 3. Effect of 20-substituents on inhibition of Streptococcus
pneumoniae chorismate synthase, compounds 21–28
Attempts to improve activity further, by incorporation of
functionality into the alkoxy group, met with limited suc-
cess. The acids 42 and 44 were reasonable inhibitors, with
IC50’s of 2.6 and 1.1 mM, respectively. Both the alcohol 39
and the ester 43 were sub-micromolar inhibitors and were
more soluble than the unfunctionalized analogues, a factor
that may be important for later compounds in the series.
Compd
R1
Activity versus SpCS
IC50, mM
21
22
23
24
25
26
27
28
OH
CH3
OCH3
Cl
CF3
0.8
4.0
>60
5.2
17.3
2.0
10.7
1.8
NO2
COOH
OCH2COOH

426
M. G. Thomas et al. / Bioorg. Med. Chem. Lett. 13 (2003) 423–426
Conclusion
2. Low, D. E.; Keller, N.; Barth, A.; Jones, R. N. Clin. Infect.
Dis. 2001, 32 (Suppl. 2), S133.
3. Franz, J. E.; Mao, M. K.; Sikorski, J. A. ACS Monograph
Series No. 189, 1997.
A total of 70 analogues retaining the 6,7-dihydroxy
functionality were synthesized as inhibitors of CS. This
led to the identification of the 20-hydroxy-40-pentoxy
analogue 33 which was forty times more potent than the
original lead. Due to the metabolic vulnerability and
potential toxicity of the 6,7-dihydroxy functionality,
and also in an effort to increase potency, further syn-
thetic efforts have focused on replacing or removing this
group. These results will be reported subsequently.
4. Davies, G. M.; Barrett-Bee, K. J.; Jude, D. A.; Lehan, M.;
Nichols, W. W.; Pinder, P. E.; Thain, J. L.; Watkins, W. J.;
Wilson, R. G. Antimicrob. Agents Chemother. 1994, 38, 403.
5. Macheroux, P.; Schmid, J.; Amrhein, N.; Schaller, A.
Planta 1999, 207, 325.
6. Webster, S. P.; Ali, S. T.; Lewendon, A. UK Patent
2374414.
7. Binnemans, K.; Galyametdinov, Y. G.; Van Deun, R.;
Bruce, D. W.; Collinson, S. R.; Polishchuk, A. P.; Bikchan-
taev, I.; Haases, W.; Prosvirin, A. V.; Tinchurina, L.; Lityinov,
I.; Gubaidullin, A.; Rakhmatullin, A.; Uytterhoeven, K.; Van
Meervelt, L. J. Am. Chem. Soc. 2000, 122, 4335.
8. Davies, J. S. H.; Deegan, T. J. Chem. Soc. 1950, 3202.
9. Varma, R. S.; Varma, M. Tetrahedron Lett. 1992, 33,
5937.
References and Notes
1. Public Health Laboratory Service Commun. Dis. Rep. 2001,
11, 7.