Tài liệu Báo cáo khoa học: Inhibition of cobalamin-dependent methionine synthase by substituted benzo-fused heterocycles pptx

Inhibition of cobalamin-dependent methionine synthase
by substituted benzo-fused heterocycles
Elizabeth C. Banks
1
, Stephen W. Doughty
2,
*, Steven M. Toms
1
, Richard T. Wheelhouse
1
and Anna Nicolaou
1
1 School of Pharmacy, University of Bradford, UK
2 School of Pharmacy, Centre for Biomolecular Sciences, University of Nottingham, UK
Methionine synthase (MetS) (5-methyltetrahydrofolate-
homocysteine transmethylase)
1
(EC.2.1.1.13) is one of
two established mammalian enzymes that utilize a bio-
logically active cobalamin derivative [methylcobalamin
(CH
3
-Cbl)] as a cofactor [1]. MetS catalyses the transfer
of the methyl group from 5-methyltetrahydrofolate to
homocysteine via the CH
3
-Cbl cofactor, with cycling
of cobalamin between the +1 [Cbl(I)] and +3 [Cbl(III)]
valency states (Fig. 1). Studies on the Escherichia coli
and Homo sapiens cobalamin-dependent MetS have
revealed that it is a large, conformationally flexible pro-

tein, consisting of four functional domains arranged
in a linear manner. Each one of these domains binds a
different substrate or cofactor. In detail, the N-terminal
module is the homocysteine (Hcy)-binding domain; the
second domain binds 5-methyltetrahydrofolate, the third
domain binds CH
3
-Cbl; and the fourth domain (C-term-
inal module) binds S-adenosyl-methionine (S-AdoMet),
an allosteric cofactor required for reductive reactivation
[2]. X-ray crystal structures of the cobalamin-, S-AdoMet-
and 5-methyltetrahydrofolate-binding sites have only
been reported for the bacterial enzyme [3–5].
The reaction products methionine and tetrahydro-
folate are further metabolized through the one-carbon
methionine transmethylation and folate cycles. MetS
is therefore intimately linked to important biochemical
Keywords
benzimidazole; benzothiadiazole; inhibition;
methionine synthase; molecular modelling
Correspondence
A. Nicolaou, School of Pharmacy, University
of Bradford, Richmond Road, Bradford BD7
1DP, UK
Fax: +44 1274 235600
Tel: +44 1274 234717
E-mail:
*Present address
Faculty of Health and Biological Sciences,
School of Pharmacy, University of Notting-

ham Malaysia Campus, Jalan Broga, 43500
Semenyih, Selangor Darul Ehsan, Malaysia
(Received 14 July 2006, revised 7 November
2006, accepted 9 November 2006)
doi:10.1111/j.1742-4658.2006.05583.x
The cobalamin–dependent cytosolic enzyme, methionine synthase
(EC.2.1.1.13), catalyzes the remethylation of homocysteine to methionine
using 5-methyltetrahydrofolate as the methyl donor. The products of this
remethylation – methionine and tetrahydrofolate – participate in the active
methionine and folate pathways. Impaired methionine synthase activity has
been implicated in the pathogenesis of anaemias, cancer and neurological
disorders. Although the need for potent and specific inhibitors of methion-
ine synthase has been recognized, there is a lack of such agents. In this
study, we designed, synthesized and evaluated the inhibitory activity of a
series of substituted benzimidazoles and small benzothiadiazoles. Kinetic
analysis revealed that the benzimidazoles act as competitive inhibitors of
the rat liver methionine synthase, whilst the most active benzothiadiazole
(IC
50
¼ 80 lm) exhibited characteristics of uncompetitive inhibition. A
model of the methyltetrahydrofolate-binding site of the rat liver methionine
synthase was constructed; docking experiments were designed to elucidate,
in greater detail, the binding mode and reveal structural requirements for
the design of inhibitors of methionine synthase. Our results indicate that
the potency of the tested compounds is related to a planar region of the
inhibitor that can be positioned in the centre of the active site, the presence
of a nitro functional group and two or three probable hydrogen-bonding
interactions.
Abbreviations
CH

3
-Cbl, methylcobalamin; DHPS, dihydropteroate synthase; Hcy, homocysteine; IC
50
, half-inhibitory concentration; MeTr, methyltransferase
protein; MetS, methionine synthase; S-AdoMet, S-adenosylmethionine.
FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS 287
pathways. These include the reactions of trans-sulfura-
tion through the production of homocysteine, biologi-
cal methylations of DNA, lipids and proteins, and
polyamine biosynthesis through the production of
methionine and S-AdoMet [6,7]. Furthermore, MetS is
the only human enzyme that metabolizes methyltetra-
hydrofolate to tetrahydrofolate, thereby facilitating the
recycling of this major form of folates to other bio-
active folates that provide one-carbon units for purine
and pyrimidine synthesis. Impaired function of MetS
has been linked to megaloblastic anaemias and neuro-
logical disorders [8], atherosclerosis [7] and carcinogen-
esis [9,10].
Although the need to develop inhibitors of MetS as
drug candidates has long been recognized [11], there
are a limited number of reports on agents inhibiting
this enzyme. The anaesthetic gas N
2
O is possibly the
only selective inhibitor of MetS reported to date, its
action mediated through the oxidation of the cobala-
min cofactor [12]. Other compounds that have been
shown to inhibit this enzyme are the cell-signalling
molecule nitric oxide [13,14], chloroform and carbon

tetrachloride [15], methylmercury [16], ethanol and
acetaldehyde [17], hydrazine [18], S-AdoMet deriva-
tives [19] and a series of cobalamin analogues [20].
Polyamines have been shown to stimulate MetS
activity [21], whilst methotrexate has been shown to
indirectly inhibit the enzyme in vivo through depletion
of its substrate, 5-methyltetrahydrofolate [22].
In a new strategy for discovering specific inhibitors
of MetS, drug-like, benzo-fused heterocycles that
mimic substructures of 5-methyltetrahydrofolate have
been evaluated in a cell-free system. The inhibitory
activity and mechanism of action have been probed by
kinetic studies using purified rat liver enzyme, whilst a
structure–activity relationship study has been discerned
using a model of the 5-methyltetrahydrofolate-binding
site constructed by homology modelling.
Results and Discussion
The test compounds 1a–k and 2a–c (Table 1) were
designed to mimic the pteridine substructure of
5-methyltetrahydrofolate, one of the two substrates of
MetS
2
(Fig. 1), and carry functionalities that may facili-
tate molecular recognition.
The synthesis of compounds 1c–k followed adapta-
tions of known methodologies. Substituted phenylene-
diamines were prepared by selective reduction of
nitroanilines [23] and cyclized with formic acid [24] to
give the desired substituted benzimidazoles. The benz-
imidazoles 1a–b and the benzothiadiazoles 2a–c were

commercially available.
All the compounds were tested against highly puri-
fied rat liver MetS [14] and the half-inhibitory con-
centrations (IC
50
) are presented in Table 1. The
benzimidazoles 1c, 1h and 1k, and the nitrobenzothiad-
iazole 2b, gave IC
50
values close to or below 100 lm,
with 2b being the most potent inhibitor (IC
50
¼
80 lm). From these results it was apparent that the
presence of a nitro group at the 5-position was associ-
ated with stronger inhibition (1c compared with 1d; 1h
with 1i) and this was positively associated with the
presence of the 3-methoxy group (1c compared with
1h). However, the aminobenzimidazole 1k showed a
marginally stronger inhibition than the corresponding
nitrobenzimidazole 1j. This result may indicate that
there is more than one mode of interaction with the
active site that affects the activity of more-highly sub-
stituted molecules such as 1j and 1k. Furthermore, the
presence of an N-methyl group on the benzimidazole
ring (position similar to the one in the substrate
5-methyltetrahydrofolate; Fig. 1) was detrimental to
the IC
50
(comparing 1c with 1f and 1h with 1j).

Finally, the inhibitory activity of the benzothiadiazoles
was improved by the nitro substitution ( 2b compared
with 2a or 2c).
To explore further the molecular mechanism of
action of those two classes of substituted benzohetero-
cycles, the kinetic parameters of inhibition were meas-
ured. Compounds 1c and 2b were chosen as being
representative of each class as a result of their good
inhibitory activity and availability. Figure 2 shows the
Lineweaver–Burk, Dixon and Cornish–Bowden plots
for the uninhibited and inhibited reactions. The K
m
values for the uninhibited reaction were calculated to
be 25 lm for 5-methyltetrahydrofolate and 0.6 lm
for homocysteine, both results being in fair agreement
Fig. 1. The cobalamin-dependent methionine synthase catalysed
reaction. Cbl(I), cob(I)alamin; CH
3
-Cbl, methylcobalamin; R, ptero-
glutamate.
Methionine synthase inhibitors E. C. Banks et al.
288 FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS
with previously published data for the pig liver enzyme
(16.8 and 2.16 lm, respectively) [25]. The Lineweaver–
Burk plots for the inhibited reactions showed that the
nitrobenzimidazole 1c exhibits the characteristics of
mixed inhibition (Fig. 2A) (K
i
¼ 26 lm), whilst the
nitrobenzothiadiazole 2b is an uncompetitive inhibitor

of MetS with respect to 5-methyltetrahydrofolate
(K
i
¼ 17 lm) (Fig. 2B). These findings were confirmed
by the Dixon (Fig. 2C,D) [26] and Cornish–Bowden
(Fig. 2E,F) [27] plots for the two compounds. When
1c and 2b were assessed with homocysteine as the vari-
able substrate, the Lineweaver–Burk, Dixon and Corn-
ish–Bowden plots indicated that both compounds
exhibited characteristics of mixed inhibition (Fig. 3),
with the nitrobenzothiadiazole 2b presenting a strong
component of uncompetitive inhibition (Fig. 3D). It
must also be noted that both compounds were very
weak inhibitors when assessed with respect to homo-
cysteine, with detectable inhibition noted mainly at
high concentrations of the inhibitors (0.5 and 1 mm;
Fig. 3A).
The results of these studies suggest that the two clas-
ses of substituted benzo-fused heterocycles may act by
two distinct mechanisms. The mixed inhibition exhib-
ited by the nitrobenzimidazole 1c is a pattern usually
observed in multisubstrate enzyme-catalysed reactions
such as MetS. However, the uncompetitive inhibition,
shown by the nitrobenzothiadiazole 2b, indicates that
there is no reversible link between the inhibitor and
the variable substrates. Plausible rationalization
includes the possibility that the relatively small nitro-
benzothiadiazole may displace the dimethylbenzimidaz-
ole side chain of the cobalamin-cofactor or that it may
act on the binding site of the MetS allosteric cofactor,

S-AdoMet, both effects which could explain the
observed uncompetitive inhibition. Furthermore, clo-
sely related 1,2,3-benzothiadiazoles and 1,2,4-thiazoles
act as potent electron acceptors in biological systems.
Thus, 1,2,4-thiadiazoles can be used to trap cysteine
residues by mixed disulfide formulation [28]. Alternat-
ively, 1,2,3-benzothiazoles have been shown to inhibit
cytochrome P450 metabolites by interference with elec-
tron transport within the catalytic cycle of cytochrome
P450 [29]. Details of the potential binding and electron
transfer events that may account for the uncompetitive
inhibition of MetS by nitrobenzothiadiazole 2b are the
subject of continuing investigation in this laboratory.
To elucidate further the mechanism of action of the
substituted benzo-fused heterocycles, to explore the
interactions occurring at the binding site, and to
develop a tool that could assist further optimization
of inhibitors, a molecular model of the methyltetra-
hydrofolate-binding domain of the rat liver MetS was
constructed. In the absence of a high-resolution struc-
ture of the methyltetrahydrofolate-binding site for the
mammalian enzyme, a model based on the X-ray crys-
tal structure of the methyltetrahydrofolate corrinoid
iron-sulfur methyltransferase protein (MeTr) from
Clostridium thermoaceticum, as determined by Doukov
et al. [30], was constructed. It has been suggested that
Table 1. Structures and half inhibitory concentrations (IC
50
) of the series 1 and 2 substituted benzo-fused heterocycles. IC
50

values were
determined using highly purified rat liver methionine synthase.
Series Compound IC
50
(lM) Series Compound IC
50
(lM)
1a R¼HX¼HY¼H > 150 2a Z¼H > 150
1b R¼CH
3
X¼HY¼H > 150 2b Z¼NO
2
80 ± 6
1c R¼HX¼HY¼NO
2
120 ± 8 2c Z¼NH
2
> 150
1d R¼HX¼HY¼NH
2
> 150
1e R¼HX¼HY¼OCH
3
> 150
1f R¼CH
3
X¼HY¼NO
2
> 150
1g R¼CH

3
X¼HY¼NH
2
> 150
1h R¼HX¼OCH
3
Y¼NO
2
100 ± 13
1i R¼HX¼OCH
3
Y¼NH
2
> 150
1j R¼CH
3
X¼OCH
3
Y¼NO
2
150 ± 9
1k R¼CH
3
X¼OCH
3
Y¼NH
2
95 ± 17
N
N

R
X
Y
1
5
7
N
S
N
Z
E. C. Banks et al. Methionine synthase inhibitors
FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS 289
the methyltetrahydrofolate-binding domain of MetS,
and indeed of other methyltransferases, share architec-
tural similarities [30]. We therefore constructed the rat
liver MetS methyltetrahydrofolate-binding site (resi-
dues 359–639) model based on the homology of this
protein with that of MeTr (residues 1–262). Using
information from the Brookhaven protein data bank,
sequence homologues of the two proteins were
obtained (Fig. 4). This led to the deduction of a back-
bone structure that was modelled to the conserved
TIM barrel fold (triose phosphate isomerase type
structure), a common feature for globular proteins that
B
A
D
C
F
E

Fig. 2. Lineweaver–Burk plots for nitrobenzimidazole 1c (A) and nitrothiadiazole 2b (B), with respect to methyltetrahydrofolate (MTHF), at
inhibitor concentrations of 1000 l
M (·), 500 lM (m), 100 lM (d) and 0 lM (j ). Dixon plots for nitrobenzimidazole 1c (C) and nitrothiadiazole
2b (D), and Cornish–Bowden plots for nitrobenzimidazole 1c (E) and nitrothiadiazole 2b (F), with respect to the inhibitor, at methyltetrahydro-
folate (MTHF) concentrations of 11 l
M (·), 22 lM (m), 67 lM (d) and 224 lM (j).
Methionine synthase inhibitors E. C. Banks et al.
290 FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS
has been predicted to occur in the pterin-binding site
of related methyltransferases [30]. The nonconserved
sequences were altered and any insertions or deletions
were applied using the molecular modelling program,
sybyl
3
, to construct and refine the model. A gradual
refinement of the resulting structure was performed
using minimization through application of the
charmm
4
program and force field [31].
A
C
E
B
D
F
Fig. 3. Lineweaver–Burk plots for nitrobenzimidazole 1c (A) and nitrothiadiazole 2b (B), with respect to homocysteine (Hcy), at inhibitor con-
centrations of 1000 l
M (·), 500 lM (m), 100 lM (d) and 0 lM (j). Dixon plots for nitrobenzimidazole 1c (C) and nitrothiadiazole 2b (D), and
Cornish–Bowden plots for nitrobenzimidazole 1c (E) and nitrothiadiazole 2b (F), with respect to the inhibitor, at homocysteine (Hcy) concen-

trations of 1.1 l
M (·), 2.2 lM (m), 6.5 lM (d) and 11 lM (j).
E. C. Banks et al. Methionine synthase inhibitors
FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS 291
Docking of the substrate 5-methyltetrahydrofolate in
the model of the active site was based on the approach
followed by Doukov et al. [30] to identify the inter-
actions between the pterin cofactor and MeTr. This
approach was based on the assumption that the strong
structural homology of MeTr and the dihydropteroate
synthases (DHPS) allows the prediction of interactions
between the pterin ring and MeTr based on the
binding of hydroxymethylpterin pyrophosphate to
DHPS. Following the same approach, and in order to
identify the orientation of the substrate when bound
to the active site, we superimposed the rat liver MetS
model on the experimentally determined structure of
hydroxymethlypterin pyrophosphate bound to DHPS
[32]. The result of this approach indicated the orienta-
tion of 5-methyltetrahydrofolate in the MetS active
site. Figure 5 shows the superimposed structures of
DHPS and 5-methyltetrahydrofolate. The model of the
methyltetrahydrofolate-binding site of MetS with one
molecule of the substrate included was further opti-
mized using charmm, and the ligand was parameter-
ized using partial atomic charges and other parameters
obtained from quantum mechanic modelling (Hartree-
Fock 6–31G* within the Spartan PCPro package)
5
of

the ligand structure. Electrostatic surfaces of the meth-
yltetrahydrofolate-binding site domain were generated
to show the size of the active site (Fig. 6). The negat-
ively charged areas may indicate the need of the inhib-
itor to have positively charged regions for favourable
interactions to take place. Figure 7 highlights the
amino acyl residues that are proposed to interact with
5-methyltetrahydrofolate, according to this model.
Calculations using interaction potentials produced
predicted values for the percentage inhibition of each of
the tested compounds. These data were then compared
with the experimentally determined data (percentage
inhibition at 100 lm), and the results are presented in
Fig. 8. The predicted activities of seven heterocycles
(1j, 1k, 1f, 1g, 1a, 2a , 2c) were found to have good cor-
relation with the experimentally determined inhibition
Fig. 4. Sequence homology of the template sequence of MeTr and the methyltetrahydrofolate-binding domain of rat liver MetS. The
residues shown in bold indicate conserved homology between the MeTr and MetS proteins. Marked with a cross (+) are the residues
that have a high degree of similarity so that although the sequence is not identical, the function of the residues is expected to remain the
same.
Fig. 5. The orientation of hydroxymethylpterin pyrophosphate
(HMPP) when bound to dihydropteroate synthase, and the super-
imposed structure of methyltetrahydrofolate showing the proposed
orientation in the methionine synthase active site. HMPP is repre-
sented as a stick structure; methyltetrahydrofolate is represented
as a wire structure.
Methionine synthase inhibitors E. C. Banks et al.
292 FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS
(percentage inhibition ± 10, Fig. 8), including two of
the five most active compounds (i.e. the benzimidazoles

1j and 1k). Interestingly, 2b, the most active compound
of this series, was not predicted to have strong inhibitory
activity. This finding is consistent with the kinetic evalu-
ation for this inhibitor that suggests a different mode of
action (i.e. through noncompetitive inhibition, as shown
in Fig. 2) unrelated to direct binding to the methyltetra-
hydrofolate site.
Overall, the biological evaluation and molecular
modelling studies indicate two routes for the develop-
ment of the next generation of inhibitors. Specifically,
the molecules need to be relatively small or not carry
bulky substituents in order to enter the active site.
They require a planar region that can be positioned in
the centre of the active site, a nitro functional group
and two or three possible hydrogen-bonding groups.
Further refinement of this model could assist the dis-
covery of the next generation of inhibitors for MetS.
Moreover, the observed noncompetitive inhibition pat-
tern, with respect to the methyltetrahydrofolate-bind-
ing site of MetS, implies the existence of other binding
sites that may also be investigated for the development
of inhibitors, whilst the potential reactivity of the
benzothiadiazole ring opens the possibility for design-
ing mechanism-based inhibitors. Overall, this approach
may lead to the identification of compounds with
potential therapeutic value, in particular as chemo-
therapeutic agents for methionine-dependent cancers in
combination with methionine-depleted treatments [33].
As the enzyme and its related metabolites have been
involved in many disorders, including cardiovascular

disease, neurodegenerative diseases and cancer, potent
and specific inhibitors will also be valuable tools for
defining the exact role of MetS in the pathophysiology
of these diseases.
Experimental procedures
dl-homocysteine, S-AdoMet (iodide salt), 5-methyl-
tetrahydrofolic acid (barium salt), diothiothreitol,
hydroxycobalamin, dimethylsulfoxide, ascorbic acid, phenyl-
methanesulfonyl fluoride, Na-p-tosyl-l-lysylchloromethyl
ketone, trypsin inhibitor, aprotinin, DEAE-cellulose, and
phosphate buffers were purchased from Sigma (Poole, UK).
5-[
14
C]-methyl]methyltetrahydrofolic acid (barium salt)
(56 mCiÆmmol
)1
) was purchased from Amersham (Little
Chalfont, UK). AG1-X8 resin (200–400 mesh chloride form)
6
and the Protein Assay kit were from Bio-Rad (Hemel Hemp-
stead, UK). Q-Sepharose Fast Flow and Hydroxyapatite
were from Pharmacia
7
(Chalfont St Giles, UK). Optiphase
HiSafe 3 scintillation cocktail was from Fisher Scientific
(Leicester, UK). Amicon ultrafiltration membranes, of
30 kDa, were purchased from
8
Millipore (Watford, UK).
Benzimidazole (1a), 1-methylbenzimidazole (1b), 2,1,3-ben-

zothiadiazole (2a), 4-nitro-2,1,3-benzothiadiazole (2b),
4-amino-2,1,3-benzothiadiazole (2c) and 4-methoxyaniline
were obtained from Aldrich (Poole, UK); and 1,3-dinitroben-
zene was from Avocado (Hewsham, UK). Solvents were
of the highest purity commercially available and were
A
B
C
Fig. 6. Snapshot pictures showing the electrostatic surfaces of the
methyltetrahydrofolate-binding domain of rat liver MetS (A) with
5-methyltetrahydrofolate bound showing the open cleft binding side
from a side angle, (B) from a reverse angle, and (C) from above.
Red indicates negatively charged surfaces, and blue indicates posi-
tively charged surfaces.
E. C. Banks et al. Methionine synthase inhibitors
FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS 293
purchased from Sigma or BDH (Poole, UK). TLC plates (sil-
ica gel 60 F
254
) and silica gel (particle size 40–63 lm) for
chromatography were from Merck (Beeston, UK). Deuterat-
ed solvents were from Goss (Glossop, UK). Melting points
were determined using an Electrothermal IA9200 digital melt-
ing point apparatus. IR spectra were recorded on a Perkin
Elmer
9
(Paragon 1000) FT-IR Spectrophotometer (Perkin
Elmer, Seer Green, UK).
1
H and

13
C NMR spectra were
acquired at 270.05 and 67.80 MHz, respectively, on a JEOL
GX270 spectrometer (JEOL UK, Welwyn, UK)
10
;
13
C assign-
ments were made using the DEPT135 experiment. Mass spectra
were obtained from the EPSRC National Mass Spectrometry
Service Centre, University of Wales (Swansea, UK).
Synthesis of the substituted benzimidazoles 1c–k
1,3,5-Trinitrobenzene [34]
1,3-Dinitrobenzene 50 g (0.297 mol) was dissolved in fum-
ing nitric acid (130.5 mL) and fuming sulphuric acid
(243.5 mL), then heated under reflux at 150 °C for 7 days.
The reaction was cooled slowly to room temperature. On
addition to ice-cold distilled water, a solid precipitated
which was collected by filtration and recrystallized from
glacial acetic acid to give 1,3,5-trinitrobenzene (61.01 g,
97%), melting point (m.p.) 118–119 °C, literature
11
122 °C
[34].
1
H NMR (CDCl
3
) d: 9.41 (s
12
, 2-H, 4-H, 6-H).

13
C
NMR (CDCl
3
) d: 149.6 (C-NO
2
), 124.4 (CH-Ar). MS (EI):
213 (M
+
). IR v
max
Æcm
)1
3104s
13;14
(C-H aromatic), 1624s (C¼C
aromatic), 1475m
13;14
(C¼C aromatic), 1544s (N¼O, asym-
metric), 1345s (N¼O, symmetric), 900s (C-H bend).
Fig. 7. Detailed view into the 5-methyltetrahydrofolate-binding pocket of rat liver MetS. Atoms within 7 A
˚
of the docked ligand are shown:
hydrophobic amino acid residues are coloured bronze, and hydrophilic acid residues are coloured blue. Blue lines indicate the putative hydro-
gen bonding interactions with ASN100. H-bond lengths are shown on the drawing, deviations from linearity are < 15° for both.
Fig. 8. Correlation of experimentally determined inhibition (percent-
age inhibition at 100 l
M) with computer-predicted inhibition, based
on the calculation of interaction potentials.
Methionine synthase inhibitors E. C. Banks et al.

294 FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS
3,5-Dinitroanisole [35]
1,3,5-Trinitrobenzene (5 g, 0.023 mol) was dissolved in
methanol (75 mL) with gentle heating. To this hot solution,
a hot solution of potassium bicarbonate (0.5 mol, 7.5 g) in
water (30 mL) and methanol (20 mL) was added. The mix-
ture was heated at reflux for 2.5 h, cooled to room tem-
perature and the methanol evaporated under reduced
pressure. The aqueous residue was extracted with chloro-
form (3 · 40 mL), the chloroform extracts combined, dried
over MgSO
4
and the solvent evaporated. The product was
recrystallized from ethanol to give 3,5-dinitroanisole
(3.36 g, 74%), m.p.
15
98–100 °C, lit. 104–106 °C [35].
1
H
NMR (CDCl
3
) d: 8.65 (d
16
, J ¼ 2 Hz, 1H, 4-H), 8.06 (d,
J ¼ 2 Hz, 2H, 2-H, 6-H), 4.01 (s, 3H, OCH
3
).
13
C NMR
(dimethylsulfoxide) d: 164.3 (C-1), 152.7 (C-3,5), 118.9

(C-2,6), 114.3 (C-4), 61.07 (CH
3
). MS (EI): 198 (M
+
). IR,
v
max
Æcm
)1
: 3098s (C-H aromatic), 2862w
17
(C-H sp
3
), 1600m
(C¼C aromatic), 1544s (N¼O, asymmetric), 1345s (N¼O,
symmetric), 1080s (C-O).
2,3,5-Trinitroanisole [36]
3,5-Dinitroanisole (1.5 g, 0.007 mol) was dissolved in con-
centrated sulfuric acid (20 mL) with gentle heating. The
solution was placed in an ice bath and fuming nitric acid
(4.2 mL) was added dropwise over a period of 10 min.
The mixture was kept on ice for 20 min, with constant
stirring, and monitored by TLC. The reaction was
stopped by the addition of distilled water (50–100 mL)
and the product was extracted into ether (3 · 40 mL).
The ether extracts were combined, dried over MgSO
4
and
the solvent evaporated. The product was recrystallized
from ethanol to give 2,3,5-trinitroanisole (1.6 g, 96%),

m.p. 100–104 °C, lit. 104 °C [36].
1
H NMR (CDCl
3
) d:
8.59 (d, J ¼ 2 Hz, 1H, 4-H), 7.99 (d, J ¼ 2 Hz, 1H,
6-H), 4.06 (s, 3H, OCH
3
).
13
C NMR (dimethylsulfoxide-
d
6
) d: 155.1 (C-1), 152.5 (C-5), 143.9 (C-3), 140.0 (C-2),
116.5 (C-6), 107.4 (C-4), 60.2 (CH
3
). MS (EI): 243 (M
+
).
IR v
max
Æcm
)1
: 3117m (C-H aromatic), 2992w (C-H sp
3
),
1600m (C¼C aromatic), 1469m (C¼C aromatic), 1046s
(C-O symmetric).
2-Amino-3,5-dinitroanisole [37]
2,3,5-Trinitroanisole (1 g, 0.004 mol) was dissolved in abso-

lute ethanol (54 mL), cooled to 2 °C, and concentrated
NH
3
(6 mL) was added. The mixture was heated under
reflux for 3 h, cooled to room temperature and the solvent
evaporated under reduced pressure. After isolation by flash
chromatography [chloroform ⁄ petroleum ether (70 : 30;
v ⁄ v)], the product was recrystallised from ethanol to give
2-amino-3,5-dinitroanisole (0.86 g, 80%), m.p. 182–184 °C,
lit. 180 °C [37].
1
H NMR (CDCl
3
) d: 8.82 (d, J ¼ 2 Hz,
1H, 4-H), 9.0–5.0 (br
18
, 2H, NH
2
), 7.72 (d, J ¼ 2 Hz, 1H,
6-H), 4.05 (s, 3H, OCH
3
). MS (EI): 214(M
+
). IR
v
max
⁄ cm
)1
: 3465s (NH asymmetric), 3322s (NH symmetric),
3098w (C-H aromatic), 2992w (C-H sp

3
), 1600m (C¼C aro-
matic), 1456m (C¼C aromatic), 1550s (N¼O asymmetric),
145s (N¼O symmetric), 1059s (C-O).
2-N-Methylamino-3,5-dinitroanisole [37]
2,3,5-Trinitroanisole (1 g, 0.004 mol) was dissolved in
tetrahydrofuran (THF) (5 mL) and methylamine in THF
19
(10 mL, 2 m), then the solution was heated in a Young’s
tube for 4 h, cooled and the solvent evaporated under
reduced pressure. The product was isolated by flash chro-
matography [diethyl ether ⁄ hexane (60 : 40, v ⁄ v)] to give
2-aminomethyl-3,5-dinitroanisole (0.8 g, 89%), 220–222 °C,
lit. 230 °C [37].
1
H NMR (CDCl
3
) d: 8.76 (d, J ¼ 2 Hz,
1H, 4-H), 8.49 (br, 1H, NH), 7.65 (d, J ¼ 2 Hz, 1H, 6-H),
3.94 (s, 3H, OCH
3
), 3.37 (d, J ¼ 6 Hz, 3H, NCH
3
).
13
C
NMR (CDCl
3
) d: 149.6 (C-1), 148.4 (C-5), 138.8 (C-3),
137.2 (C-2), 124.3 (C-6), 118.9 (C-4), 57.2 (OCH

3
), 33.6
(NCH
3
). MS (EI): 198 (M
+
).
2,3-Diamino-5-nitroanisole, 3-amino-2-methylamino-
5-nitroanisole [37]
The same method was
20
applied to both 2-amino-3,5-dinitro-
anisole and 2-N-methylamino-3,5-dinitroanisole. The appro-
priate compound (1.0 mmol) was dissolved in methanol
(30 mL) and water (2 mL). Ammonium chloride (573 mg,
10 mmol) and ammonium carbonate (350 mg, 3.64 mmol)
were added to the solution. A hot solution of sodium
sulfide (110 mg, 1.41 mmol in 1.5 mL water) was added
dropwise over a time-period of 5 min and the solution
heated at reflux for 2 h. The reaction mixture was allowed
to cool slowly to room temperature, the volatile solvent
evaporated under reduced pressure and the product was
isolated by flash chromatography, eluted with methanol:
c. NH
3
: chloroform (1 : 1 : 98). 2,3-Diamino-5-nitroanisole:
(0.150 mg, 77%), m.p. 173–175 °C, lit.165–167 °C [37].
1
H
NMR (CDCl

3
) d: 7.41 (d, J ¼ 2 Hz, 1H, 4-H), 7.39 (d,
J ¼ 2 Hz, 1H, 6-H), 4.05 (s, 2H, NH
2
), 3.93 (s, 3H,
OCH
3
), 3.79 (s, 2H, NH
2
). 3-Amino-2-methylamino-
5-nitroanisole: (0.181 mg, 80%), m.p. 158–160 °C.
1
H
NMR (CDCl
3
) d: 7.32 (d, J ¼ 2 Hz, 1H, 4-H), 7.24 (d,
J ¼ 2 Hz, 1H, 6-H), 4.1 (br, s, 3H, NH), 3.87 (s, 3H,
OCH
3
), 2.82 (s, 3H, NCH
3
).
5-Nitro-7-methoxybenzimidazole (1h) and 1-methyl-
5-nitro-7-methoxybenzimidazole (1j) [37]
The same method was
21
applied to both 2,3-diamino-5-nit-
roanisole and 3-amino-2-methylamino-5-nitroanisole. The
compound (1.0 mmol) was dissolved in formic acid (5 mL)
and heated at reflux for 2 h. The reaction was removed

E. C. Banks et al. Methionine synthase inhibitors
FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS 295
from the heat, cooled to room temperature, and toluene
(20 mL) and water (1 mL) were added. The volatile sol-
vents were evaporated under reduced pressure and the resi-
due poured into water (30 mL) and extracted with ethyl
acetate (3 · 30 mL). The combined organic extracts were
washed with water (20 mL), dried over MgSO
4
and evapor-
ated under reduced pressure. The product was recrystallized
from ethyl acetate. 1h: (216 mg, 96%), m.p. 254–256 °C, lit.
258–260 °C [37].
1
H NMR (dimethylsulfoxide-d
6
) d: 13.25
(br, 1H, NH), 8.44 (s, 1H, 2-H), 8.13 (d, J ¼ 2 Hz, 1H,
4-H), 7.57 (d, J ¼ 2 Hz, 1H, 6-H), 4.04 (s, 3H, OCH
3
). MS
(EI): 193 (M
+
). 1j: (202 mg, 98%), m.p. 146–148 °C.
1
H
NMR (dimethylsulfoxide-d
6
) d: 8.35 (s, 1H, 2-H), 8.18
(d, J ¼ 2 Hz, 1H, 4-H), 7.06 (d, J ¼ 2 Hz, 1H, 6-H), 4.03

(s, 3H, OCH
3
), 3.29 (s, 3H, NCH
3
). HRMS (ES) (M + H)
208.0717, C
9
H
10
N
3
O
3
requires 208.0717.
5-Amino-7-methoxybenzimidazole (1i) [37] and
5-amino-7-methoxy-N1-methylbenzimidazole (1k) [38]
The same method was
22
applied to both 5-nitro-7-methoxy-
benzimidazole (1h) and 1-methyl-5-nitro-7-methoxybenzimi-
dazole (1j). The compound (0.6 mmol) was dissolved in
ethanol (30 mL) with 2 drops of concentrated HCl, and
10% weight of palladium on a carbon catalyst was added.
The system was evacuated and the mixture stirred vigo-
rously under a hydrogen atmosphere until the reaction was
complete (approximately 2–3 h by TLC). The catalyst was
removed by filtration through celite and washed with copi-
ous amounts of ethanol. The solvent was evaporated under
reduced pressure and the product was recrystallized from
ethanol and ethyl acetate. An alternative method involved

using eight equivalents of ammonium formate as the hydro-
gen source and reacting the mixture, as above, for 2 h in an
evacuated system. 1i: (0.08 mg, 82%), m.p. 220–223 °C, lit.
216–218 °C [37].
1
H NMR (dimethylsulfoxide-d
6
) d: 8.72 (s,
1H, 2-H), 8.20 (d, J ¼ 2 Hz, 1H, 4-H), 7.65 (d, J ¼ 2 Hz,
1H, 6-H), 5.81–6.12 (br, 3H, NH, NH
2
) 4.07 (s, 3H,
OCH
3
). 1k: (0.081 mg, 76%), m.p. 180–182, lit. 178 °C
[38].
1
H NMR (dimethylsulfoxide-d
6
) d: 8.64 (s, 1H, 2-H),
8.26 (d, J ¼ 2 Hz, 1H, 4-H), 7.12 (d, J ¼ 2 Hz, 1H, 6-H),
5.23–5.65 (br, 2H, NH
2
), 4.01 (s, 3H, OCH
3
), 3.26 (s, 3H,
NCH
3
).
5-Nitrobenzimidazole (1c) and 5-aminobenzimidazole

(1d) [39]
These compounds were synthesized, according to the meth-
ods described above, from 2,4 dinitroaniline. 1c: (2.5 g,
88%), m.p. 203–204 °C, lit. 204–205 °C [39].
1
H NMR
(dimethylsulfoxide-d
6
) d: 8.54 (s, 1H, 2-H), 8.51 (d, J ¼ 2 Hz,
1H, 4-H), 8.44 (s, 1H, NH), 8.13 (dd
23
, J ¼ 2 Hz, J ¼ 8 Hz,
1H, 6-H) 7.09 (d, J ¼ 8 Hz, 1H, 7-H). MS (EI): 164(M
+
).
1dÆ2HClÆ0.2H
2
O: (3.1 g, 90%), m.p. decomposition
24
>
230 °C, lit. 165–166 °C (free base) [39].
1
H NMR (dimethyl-
sulfoxide-d
6
) d: 9.46 (s, 1H, 2-H), 7.80 (d, J ¼ 8 Hz, 1H, 7-
H), 7.58 (br, s, 1H, 4-H), 7.33 (dd, J ¼ 2 Hz, J ¼ 8 Hz, 1H,
6-H), 5.6–3.4 (br, s, 3H, NH + NH
2
). MS (EI): 134(M

+
).
Found: C, 40.53; H, 4.44; N, 19.55. C
7
H
7
N
3
Æ2HClÆO.2H
2
O
requires: C, 40.10; H, 4.52; N, 20.04%.
5-Methoxybenzimidazole (1e)
2-Amino-4-methoxy-2-aniline hydrochoride (0.8 g, 4.60
mmol) was dissolved in formic acid (50 mL) and heated at
120 °C for 12 h. The mixture was evaporated to dryness
and partitioned between ethyl acetate (100 mL) and concen-
trated NH
3
(20 mL). The layers were separated and the
aqueous layer further extracted with ethyl acetate
(2 · 30 mL). The combined organic extracts were dried
over MgSO
4
and evaporated under reduced pressure. The
residue was dissolved in isopropanol, treated with 5 mL of
concentrated HCl, evaporated twice from isopropanol then
recrystallized from isopropanol-ether to yield a grey solid,
1eÆHCl: (0.50 g, 59%), m.p. 202–206 °C, lit. 199–202 °C
[40].

1
H NMR (dimethylsulfoxide-d
6
) d: 15.01 (br, 2H,
2 · NH), 9.43 (d, J ¼ 5 Hz, 1H, 2-H), 7.71 (d, J ¼ 9 Hz,
7-H), 7.23 (d, J ¼ 2 Hz, 1H, 4-H), 7.14 (dd, J ¼ 9Hz, J ¼
2 Hz, 1H, 6-H), 3.83 (s, 3H, OCH
3
). MS (EI) (free base):
148 (M
+
).
1-Methyl-5-nitro-benzimidazole (1f)
Chlorodinitrobenzene (0.81 g, 4.0 mmol), dissolved in THF
(5 mL) and methylamine (10 mL · 2 m THF)
25
, was heated
for 12 h at 90 °C in a Young’s tube. The reaction was
monitored using TLC with diethyl ether as the eluant. The
mixture was cooled to room temperature and the solvent
evaporated under reduced pressure. The resulting diamino
compound was then cyclized in formic acid (5 mL) heated
at reflux for 2 h, after which the reaction was cooled to
room temperature and toluene (20 mL) and water (1 mL)
were added. The volatile solvent was evaporated under
reduced pressure and the residue was poured into water
(30 mL) and extracted with ethyl acetate (3 · 30 mL). The
combined organic extracts were washed with water
(20 mL), dried over MgSO
4

and evaporated under reduced
pressure. The product was recrystallized from ethyl acetate
(0.23 g, 32%), m.p. 213–215 °C, lit. 209–211 °C [41].
1
H
NMR (dimethylsulfoxide-d
6
) d: 8.73 (d, J ¼ 2 Hz, 1H,
4-H), 8.25 (d, J ¼ 9 Hz, 2H, 6-H), 8.04 (s, 1H, 2-H), 7.44
(d, J ¼ 9 Hz, 1H, 7-H), 3.92 (s, 3H, NCH
3
). MS (EI):
178 (M
+
).
1-Methyl-5-amino-benzimidazoleÆ2HClÆ0.2H
2
O (1g)
1-Methyl-5-nitro-benzimidazole (1f) (0.097 g, 0.548 mmol)
was dissolved in ethanol (20 mL), 10% palladium on carbon
Methionine synthase inhibitors E. C. Banks et al.
296 FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS
catalyst (13 mg) and two drops of concentrated HCl were
added, the system was evacuated and the mixture stirred
vigorously under a hydrogen atmosphere until the reaction
was complete ( 3 h). The catalyst was removed by filtra-
tion through celite and washed with copious amounts of
ethanol. The solvent was evaporated under reduced pres-
sure and the product recrystallized from ethanol and ethyl
acetate (0.12 g, 85%), m.p. decomp

26
244–246 °C, lit. 158–
195 °C (free base) [42].
1
H NMR (dimethylsulfoxide-d
6
) d:
9.33 (s, 1H, 2-H), 7.75 (d, J ¼ 9 Hz, 1H, 7-H), 7.32 (s, 1H,
6-H), 7.17 (d, J ¼ 9 Hz, 1H, 7-H), 4.8–4.2 (br, 3H, NH,
NH
2
), 3.98 (s, 3H, NCH
3
). MS (EI): 185 (M
+
). Found: C,
43.19; H, 5.00; N, 18.31. C
8
H
9
N
3
Æ2HClÆO.2H
2
O requires:
C, 42.95; H, 5.14; N, 18.78%.
Enzyme purification
MetS was purified from rat liver, as previously described
[14]. Briefly, rat liver homogenate was prepared in ice-cold
50 mm potassium phosphate buffer, pH 7.0, containing

0.1 m NaCl and protease inhibitors (14.4 mg of Na-p-tosyl-
l-lysylchloromethyl ketone, 100.32 mg phenylmethanesulfo-
nyl fluoride, 49.2 mg of trypsin inhibitor and 10.56 mg of
aprotinin) and centrifuged at 1000 g
27
for 15 min. The super-
natant was centrifuged at 27 000 g
28
for 30 min followed by
a second centrifugation step at 100 000 g
29
for 60 min to give
the cytosolic fraction. This was then run through DEAE-
cellulose (batch chromatography) equilibrated with 20 mm
sodium phosphate buffer, pH 7.2. Unbound protein was fil-
tered under vacuum, whereas bound proteins (including
MetS) were removed by 20 mm sodium phosphate buffer,
pH 7.2, containing 500 mm NaCl. The active enzyme eluate
from this step was diluted with water to reduce the concen-
tration of NaCl. The preparation was further purified using
a Q-sepharose column equilibrated with 20 mm sodium
phosphate buffer, pH 7.2, and run with a linear gradient of
0–1 m NaCl. MetS was eluted at 500 mm NaCl. The active
fractions were pooled, concentrated and desalted using an
Amicon ultrafiltrator fitted with a 30 kDa membrane. The
enzyme preparation was further purified on a hydroxyapa-
tite column equilibrated with 20 mm potassium phosphate
buffer, pH 7.2. The enzyme was eluted using a stepwise gra-
dient from 20 to 500 mm potassium phosphate buffer; the
active fractions eluted at 150 mm phosphate. The enzyme-

active fractions were pooled, desalted and concentrated by
ultrafiltration, as described above. The final enzyme pre-
paration was stored at )20 °C, in 20% (v ⁄ v) glycerol.
Handling of enzyme solutions was performed at low
temperature, out of direct light.
Methionine synthase assay
MetS activity was determined using the assay described by
Kenyon et al. [43]. Briefly, reactions contained 50 mm phos-
phate buffer (pH 7.4), 227 lm
14
C-5-methyltetrahydrofolate
[2077 disintegrations per min (dpm)
30
Ænmol
)1
], 23 mm dio-
thiothreitol, 40 lm S-AdoMet, 60 lm hydroxycobalamin,
the enzyme source and (when applicable) dimethylsulfoxide
solutions of the inhibitors (maximum volume 5 lL) in a
total volume of 300 lL. Incubations were performed in
light-excluding sealed serum vials under nitrogen. The reac-
tion mixture was pre-incubated for 5 min, the reaction was
initiated by the addition of 500 lm (dl)-homocysteine and
incubated at 37 °C for a further 30 min, unless otherwise
stated. The reaction was terminated by the addition of ice-
cold water (400 lL). The reaction mixture was passed
through a 0.5 · 5 cm AG1-X8 resin column, [
14
C] methion-
ine was eluted with 2 mL of water and quantified using

a liquid scintillation counter (Packard Tricarb 1900CA;
Perkin Elmer)
31
.
Determination of K
m
and V
max
for 5-methyltetra-
hydrofolate
Assays were incubated for 10 min at a fixed concentration
of Hcy (500 lm) and varying concentrations of 5-methyl-
tetrahydrofolate (i.e. 0, 50, 100, 500 and 1000 lm). During
preliminary experiments, the enzyme concentration and the
incubation time were varied in order to establish conditions
for linear kinetics. The concentration of protein used was
then adjusted to ensure that the initial velocities were cor-
rectly estimated.
Determination of K
m
and V
max
for homocysteine
Assays were incubated for 10 min at a fixed concentration
of 5-methyltetrahydrofolate (67.2 nmol per assay) and vary-
ing concentrations of Hcy (i.e. 0, 0.5, 1.1, 2.2, 4.3, 6.5 and
11 lm). Preliminary experiments, similar to those described
for 5-methyltetrahydrofolate, were performed with respect
to time and enzyme concentration to achieve linear kinetics.
Protein assay

Protein content was determined using the Protein Assay kit
based on the method of Bradford [44]. Standards and sam-
ples were assayed in triplicate, according to the manufac-
turer’s instructions. Sample absorbances were read against
a BSA standard curve to determine protein content.
Molecular model construction
Protein sequences were identified using the Human Gen-
ome Mapping Project Centre and the SWISSPROT data-
bases (). Sequence alignment
was performed using clustalw dynamic programming.
The model was constructed using a Silicon Graphics
workstation with sybyl software (Tripos Inc.) for model
building, charmm minimization and molecular dynamics
E. C. Banks et al. Methionine synthase inhibitors
FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS 297
[31] for structure optimization. Ligands were parameter-
ized using partial atomic charges and other values
obtained from quantum mechanic modelling (Hartree-
Fock 6–31G*) of the ligand structure using pc spartan
pro (Wavefunction Inc.).
References
1 Banerjee RV & Matthews RG (1990) Cobalamin-depen-
dent methionine synthase. FASEB J 4, 1450–1459.
2 Ludwig ML & Matthews RG (1997) Structure-based
perspectives on B12-dependent enzymes. Annu Rev Bio-
chem 66, 269–313.
3 Evans JC, Huddler DP, Hilgers MT, Romanchuk G,
Matthews RG & Ludwig ML (2004) Structures of the
N-terminal modules imply large domain motions during
catalysis by methionine synthase. Proc Natl Acad Sci

USA 101, 3729–3736.
4 Drennan CL, Huang S, Drummond JT, Matthews RG
& Lidwig ML (1994) How a protein binds B12: a 3.0 A
X-ray structure of B12-binding domains of methionine
synthase. Science 266, 1669–1674.
5 Dixon MM, Huang S, Matthews RG & Ludwig M (1996)
The structure of the C-terminal domain of methionine
synthase: presenting S-adenosylmethionine for reductive
methylation of B12. Structure 4, 1263–1275.
6 Finkelstein JD (1990) Methionine metabolism in mam-
mals. J Nutr Biochem 1, 228–237.
7 Fowler B (2005) Homocysteine: overview of biochemis-
try, molecular biology, and role in disease processes.
Semin Vasc Med 5, 77–86.
8 Allen RH, Stabler SP, Savage DG & Lindenbaum J
(1993) Metabolic abnormalities in cobalamin (vitamin
B12) and folate deficiency. FASEB J 7, 1344–1353.
9 Shi Q, Zhang Z, Li G, Pillow PC, Hernandez LM, Spitz
MR & Wei Q (2005) Polymorphisms of methionine
synthase and methionine synthase reductase and risk of
lung cancer: a case-control analysis. Pharmacogenet
Genomics 15, 547–555.
10 Mason JB & Choi SW (2005) Effects of alcohol on
folate metabolism: implications for carcinogenesis.
Alcohol 35, 235–241.
11 Matthews RG, Drummond JT & Webb HK (1998)
Cobalamin-dependent methionine synthase and serine
hydroxymethyltransferase: targets for chemotherapeutic
intervention? Adv Enzyme Regul 38, 377–392.
12 Frasca V, Riazzi BS & Matthews RG (1986) In vitro

inactivation of methionine synthase by nitrous oxide.
J Biol Chem 261, 15823–15826.
13 Nicolaou A, Waterfield CJ, Kenyon SH & Gibbons
WA (1997) The inactivation of methionine synthase in
isolated rat hepatocytes by sodium nitroprusside. Eur
J Biochem 244 , 876–882.
14 Nicolaou A, Kenyon SH, Gibbons JM, Ast T & Gib-
bons WA (1996) In vitro inactivation of mammalian
methionine synthase by nitric oxide. Eur J Clin Invest
26, 167–170.
15 Alston TA (1991) Inhibition of vitamin B12-dependent
methionine biosynthesis by chloroform and carbon
tetrachloride. Biochem Pharmacol 42, R25–R28.
16 Smith JR & Smith JG (1990) Effects of methylmercury
in vitro on methionine synthase activity in various rat
tissues. Bull Environ Contam Toxicol 45, 649–654.
17 Kenyon SH, Nicolaou A & Gibbons WA (1998) The
effect of ethanol and its metabolites upon methionine
synthase activity in vitro. Alcohol 15, 305–309.
18 Kenyon SH, Waterfield CJ, Asker DS, Kudo M, Moss
DW, Bates TE, Nicolaou A, Gibbons WA & Timbrell
JA (1999) Effect of hydrazine upon vitamin B12-depen-
dent methionine synthase activity and the sulphur
amino acid pathway in isolated rat hepatocytes. Bio-
chem Pharmacol 57, 1311–1319.
19 Kenyon SH, Alves M, Neubert H, Nicolaou A, Del
Olmo E & Gibbons WA (1996) Sulphur amino acid
derivatives as inhibitors of vitamin B12 dependent
methionine synthase. Biochem Soc Trans 24, 265S.
20 Stabler SP, Brass EP, Marcell PD & Allen RH (1991)

Inhibition of cobalamin-dependent enzymes by
cobalamin analogues in rats. J Clin Invest 87,
1422–1430.
21 Kenyon SH, Nicolaou A, Ast T & Gibbons WA (1996)
Stimulation in vitro of vitamin B-12-dependent methio-
nine synthase by polyamines. Biochem J 316, 661–665.
22 Fiskerstrand T, Ueland PM & Refsum H (1997) Folate
depletion induced by methotrexate affects methionine
synthase activity and its susceptibility to inactivation by
nitrous oxide. J Pharmacol Exp Ther 282, 1305–1311.
23 Devi TK (1998) Synthesis of 6-Alkyl ⁄ Aralkyl-5a, 11-
dihydro-9-nitro[1]benzopyranol[2,3-b][benzodiazepin-13-
ones as possible antipsychotic agents. J Indian Chem
Soc LXV
32
, 567–570.
24 Katrisky A & Rachwal S (1994) New routes to selective
methylated benzimidazoles. J Hetero Chem 31, 775–779.
25 Chen Z, Crippen K, Gulati S & Banerjee R (1994) Puri-
fication and kinetic mechanism of a mammalian methio-
nine synthase from pig liver. J Biol Chem 269, 27193–
27197.
26 Dixon M (1953) The determination of inhibitor con-
stants. J Biochem 55, 170–172.
27 Cornish–Bowden A (1974) A simple graphical method
for determining the inhibition constants of mixed,
uncompetitive and non-competitive inhibitors. Biochem
J 137, 143–144.
28 Leung-Toung R, Wodzinska J, Li W, Lowrie J, Kukreja
R, Desilets D, Karimian K & Tam TF (2003) 1,2,4-thia-

diazole: a novel Cathepsin B inhibitor. Bioorg Med
Chem 11, 5529–5537.
29 Pedemonte J, Olate J, Cervantes P, Oberti C & Gil L
(1981) Alterations in mouse liver monooxygenases by
benzothiadiazoles. Biochem Pharmacol 30, 1483–1495.
Methionine synthase inhibitors E. C. Banks et al.
298 FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS
30 Doukov T, Seravalli J, Stezowski JJ & Ragsdale SW
(2000) Crystal structure of a methyltetrahydrofolate-
and corrinoid-dependent methyltransferase. Structure
Fold Des 8, 817–830.
31 Brooks BR, Brucoleri RE, Olafson BD, States DJ,
Swaminathan S & Karplus M (1983) CHARMM: a pro-
gram for Macromolecular Energy, Minimization and
Dynamics Calculations. J Comp Chem 4, 187–217.
32 Achari A, Somers DO, Champness JN, Bryant PK,
Rosemond J & Stammers DK (1997) Crystal structure
of the anti-bacterial sulfonamide drug target dihydrop-
teroate synthase. Nat Struct Biol 4, 490–497.
33 Pavillard V, Nicolaou A, Double JA & Phillips RM
(2006) Methionine dependence of tumours: a biochem-
ical strategy for optimizing paclitaxel chemosensitivity
in vitro. Biochem Pharmacol 71, 772–778.
34 Jenkins OH (1936) The dipole moments of certain poly-
nitro-compounds. J Chem Soc 00, 862–867.
33
35 Izzo P (1959) Bicarbonate-catalyzed displacement of a
nitro group of 1,3,5-trinitrobenzene. J Org Chem 24,
2026–2028.
36 Blanksma JJ (1904) Nitration du dinitroanisol symetri-

que. Rec Trav Chim 23, 111–118.
37 Gillespie HB, Engelman M & Graff S (1954) Benzimi-
dazoles and benzotriazoles as growth antagonists. JAm
Chem Soc 76, 3531–3533.
38 Horner S (1953) Zum isoteren austausch des pyrimidn-
rings gegn den benzolken im coffein. Justus Liebigs Ann
Chem 579, 204–210.
39 Rabinowitz J & Wangner E (1951) Restriction of tau-
toerism in the amide system by hydrogen bonding. The
case of 4(7)-nitrobenzimidazole. J Am Chem Soc 73,
3030–3037.
40 Kanaoka Y, Sekine T, Machida M, Soma Y, Tanizawa
K & Ban Y (1964) Studies on protein-sulfhydryl
reagents 1. Synthesis of benzimidazole derivatives of
maleimide – fluorescent labeling of maleimide. Chem
Pharm Bull 12, 127–134.
41 Bergman J & Sand P (1984) A new simple procedure
for alkylation of nitrogen heterocycles using
dialkyl oxalates and alkoxides. Tet Lett 25, 1957–
1960.
42 Montanari F (1955) Ricerche sui Benzimidazoli-Nota
VI. Coloranti Azoici. Gazz Chim Ital 85, 981–988.
43 Kenyon SH, Waterfield CJ, Timbrell JA & Nicolaou A
(2002) Methionine synthase activity and sulphur amino
acid levels in the rat liver tumour cells HTC and Phi-1.
Biochem Pharmacol 63, 381–391.
44 Bradford M (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein util-
izing the principle of protein-dye binding. Anal Biochem
72, 248–254.

E. C. Banks et al. Methionine synthase inhibitors
FEBS Journal 274 (2007) 287–299 ª 2006 The Authors Journal compilation ª 2006 FEBS 299