Inhibition of Escherichia coli glucosamine synthetase by novel electrophilic analogues of glutamine - Comparison with 6-diazo-5-oxo-norleucine

Article abstract of DOI:10.1016/S0960-894X(00)00565-5

A series of electrophilic glutamine analogues based on 6-diazo-5-oxo-norleucine has been prepared, using novel synthetic routes, and evaluated as inhibitors of Escherichia coli. glucosamine synthetase. The γ-dimethylsulphonium salt analogue of glutamine was found to be one of the most potent inactivators of this enzyme yet reported, with an apparent second order rate constant (k₂/K(i)) of 3.5 x 10⁵ M^-1 min^-1. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00565-5

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2795±2798
Inhibition of Escherichia coli Glucosamine Synthetase by Novel
Electrophilic Analogues of GlutamineÐComparison with
6
-Diazo-5-oxo-norleucine
a,
b
a
a
Brian Walker, * Martin F. Brown, John F. Lynas, S. Lorraine Martin,
c
d
a
Andrew McDowell, Bernard Badet and Alan J. Hill
aDivision of Biomedicinal Chemistry, School of Pharmacy, The Queen's University of Belfast, Medical Biology Centre,
9
7 Lisburn Road, Belfast BT9 7BL, Northern Ireland, UK
b
School of Biology and Biochemistry, The Queen's University of Belfast, Medical Biology Centre, 97 Lisburn Road,
Belfast BT9 7BL, Northern Ireland, UK
c
Department of Bacteriology, Belfast City Hospital, Lisburn Road, Belfast BT9 7AB, Northern Ireland, UK
dLaboratoire de Bioorganique et Biotechnologies, E.N.S.C.P., 75231 Paris Cedex 05, France
Received 4 August 2000; accepted 3 October 2000
AbstractÐA series of electrophilic glutamine analogues based on 6-diazo-5-oxo-norleucine has been prepared, using novel synthetic
routes, and evaluated as inhibitors of Escherichia coli. glucosamine synthetase. The g-dimethylsulphonium salt analogue of gluta-
mine was found to be one of the most potent inactivators of this enzyme yet reported, with an apparent second order rate constant
5
� 1
� 1
(k
2
/K
i
) of 3.5Â10 M min . # 2000 Elsevier Science Ltd. All rights reserved.
The maintenance of the structural integrity of the outer
cell wall of bacteria and fungi is essential for their con-
tinued survival. Glucosamine synthetase (l-gluta-
mine:d-fructose-6-phosphate amidotransferase; EC
we wish to report the e€ectiveness of a series of alter-
native glutamine analogues as inhibitors of glucosamine
synthetase. This paper describes the synthesis and
kinetic testing of a series of analogues in which the side
2.6.1.16) (GS) plays a pivotal role in the biosynthesis of
chain amide (-CONH ) of glutamine has been replaced
2
amino sugar-containing macromolecules that constitute
the major structural outer membrane proteins of these
organisms, such as the peptidoglycans in bacteria and
chitin in fungi.1 Compounds designed to selectively
inhibit this enzyme could therefore constitute a poten-
tial source of new bactericidal and fungicidal agents.
by various electrophilic species including g-halo-
methyl ketone (-COCH Cl; -COCH Br) (2, 3), g-
2
2
nitrile (-CꢀN) (4), g-dimethylsulphonium salt (-
,2
+
COCH S (Me) ) (5) and g-diazoethyl ketone (-
2
2
COCH(CH )CHN ) (6) (Fig. 1).
3
2
One possible strategy applicable to the design of such
agents is the fact that GS belongs to the family of ami-
dotransferases, all of which possess an active site
cysteine nucleophile that can be anity labelled by the
3,4
glutamine analogue 6-diazo-5-oxo-norleucine (DON)
1). Inactivation studies with puri®ed Escherichia coli GS,
(
1
4
employing C6-labelled DON, established that the active
site cysteine residue occupied position 1 of the primary
5
sequence of the enzyme. Consequently, as part of a pro-
gramme directed towards the development of new amido-
transferase inhibitors with putative antimicrobial activity,
*Corresponding author. Fax: +44-28-9024-7794; e-mail: brian.walk-
er@qub.ac.uk
Figure 1. Electrophilic analogues of glutamine.
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00565-5

2
796
B. Walker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2795±2798
Chemistry
For time-dependent inactivation assays, glucosamine
synthetase (ꢁ17 mg) was pre-incubated in 100 mM
potassium phosphate bu€er, pH 7.5, containing 10 mM
fructose-6-phosphate and 50 mM KCl, for 15 min at
Schemes 1 and 2 outline the synthetic strategies
employed for the synthesis of the putative inhibitors
utilized in this study. Scheme 1 outlines the synthesis of
ꢂ
37 C. The reaction was initiated by addition of the
6
-diazo-5-oxo-norleucine (1), from N-a-Fmoc-glutamic
inhibitor under investigation, and at timed intervals,
20 mL aliquots were removed and diluted 40-fold into
the same bu€er, containing 6 mM (l)-glutamine. Resi-
dual activity was then determined by the colorimetric
acid-g-tert-butyl ester, and illustrates the use of this
intermediate for the preparation of halomethyl ketone
derivatives 2 and 3. The g-diazoethyl ketone analogue
7
(6) was prepared via the acylation of diazoethane
assay for glutamate described by Shiio and Ishii.
(
generated from N-nitroso-b-isobutyl methyl ketone).
Inactivation, assessed using the method of Kitz and
8
Wilson, was found to be dependent on both incubation
time and concentration of inhibitor used. A linear cor-
relation was observed from a plot of the natural loga-
rithm of residual activity against incubation time, from
which the inactivation half time (t1=2) at each inhibitor
concentration could be read directly. The apparent
pseudo ®rst-order rate constant (kobs) was calculated
according to:
kobsˆ1n2=t1=2
A hyperbolic curve was obtained when these constants
were plotted against inhibitor concentration. This beha-
viour is consistent with the formation of a reversible
enzyme±inhibitor complex prior to inactivation (1):
k‡1
k‡2
*
E ‡ I ) E:I ! E� I
ꢀ1†
Scheme 1. Synthesis of DON and halomethyl ketone analogues: (i)
ethereal diazomethane; (ii) 50% (v/v) TFA/DCM; (iii) N-methylmor-
k� 1
ꢂ
pholine, isopropenyl chloroformate at 0 C, then add ethereal diazo-
methane or diazoethane; (iv) 4 M NaOH; (v) 30% (v/v) HBr/AcOH;
where E.I is the reversible enzyme±inhibitor complex E±
I is the irreversible enzyme±inhibitor complex and k+2
is the limiting rate constant for inactivation. Assuming
that [I]>>[E] and that the reversible complex is at all
times in equilibrium with the enzyme and inhibitor, Kitz
and Wilson derived the following equation:
(
vi) ethereal HCl.
Scheme 2 outlines the synthesis of 6-dimethylsulpho-
nium-5-oxo-norleucine (5) from N-a-Boc-(l)-glutamic
acid-a-tert-butyl ester. The g-nitrile derivative (4) was
synthesized via the dehydration of (l)-glutamine, essen-
6
tially according to the method of Stu
È
ber et al. Com-
1
1
Ki
1
1
pound identity and purity was assessed by H NMR,
TLC, ESI-MS and elemental analysis.
ˆ
Â
‡
ꢀ2†
kobs
k‡2 ‰IŠ k‡2
where kobs is the apparent pseudo ®rst order rate con-
stant, k+2 is the inactivation rate constant at saturating
inhibitor concentration and K is the steady state inacti-
i
vation constant. A plot of 1/kobs against 1/[I] is a
straight line with the ordinate intercept equal to 1/k+2
and the abcissa intercept equal to � 1/K . Data obtained
i
from these studies is summarized in Table 1.
Results and Discussion
Scheme 2. Synthesis of g-dimethylsulphonium ketone analogue: (i)
N-methylmorpholine, isopropenyl chloroformate at 0 C, then add
ꢂ
Results clearly show that, in addition to DON (1), the
enzyme is inactivated by a number of other electrophilic
carbonyl analogues. Kinetic analysis of inhibition by the
dimethylsulphonium salt (5), diazoethyl- (6) and halo-
methyl ketone (2 and 3) analogues was indicative of
irreversible inhibition in each instance. One observation
of particular interest was the inactivity of the g-nitrile
derivative (4), even at concentrations as high as 200 mM.
Peptides incorporating the nitrile group have previously
ethereal diazomethane; (ii) ethereal HCl; (iii) sodium methane thiolate;
(
iv) MeI/AgBF
4
in nitromethane; (v) dimethylsulphide/TFA.
Inactivation Studies
E. coli glucosamine synthetase was puri®ed from E.
coli (strain K12), according to a previously published
6
procedure.

B. Walker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2795±2798
2797
been shown to act as good inhibitors of cysteine pepti-
9
methyl substituent a to the departing diazo moiety
enhances anity, while reducing the electrophilicity of
the methane -CH- centre, thus reducing inherent
chemical reactivity of the molecule.
dases, their lack of activity against glucosamine syn-
thetase suggests that the presence of a carbonyl group at
the g position may be one important requirement for
e€ective binding and inhibition. It is worth pointing out
however, that our nitrile-based compound (4) has a
shorter side chain (2-carbon, compared with 3-carbon,
for compounds 1±3 and 5±6) which may be an addi-
tional or alternative reason for the apparent lack of
activity of this compound.
The most promising results appeared with the dime-
thylsulphonium salt. This proved to be an exceptional
inhibitor of glucosamine synthetase, exhibiting a sec-
�
1
ond-order rate constant, (k /K ) of 320,000 M
i
+
2
�
1
min , some 2-fold greater than the bromomethyl
ketone. The improved potency of the dimethylsulpho-
nium salt can be attributed to the 7-fold higher anity
with which this compound binds to the enzyme prior to
covalent complex formation.
Table 1 lists the kinetic constants obtained for the
inactivation of E. coli GS by electrophilic glutamine
analogues prepared in this study.
It is clear from this table that, of the g-halomethyl and
diazomethyl ketone derivatives of glutamic acid, the
bromomethyl ketone analogue (3) is the most potent
inactivator of glucosamine synthetase, with an overall
�
1
second order rate constant (k /K ) of 150,000 M
2
+
i
�
1
min . This is some 2-fold more potent than DON (1)
�
1
� 1
(
70000 M min ). The improved potency of the bro-
The chief drawback with the use of glucosamine syn-
thetase inhibitors as anti-microbial compounds, is their
inherent toxicity. DON inactivates the mammalian form
of the enzyme, and is also known to block glutaminase.
One interesting approach to the utilization of glucosa-
mine synthetase inhibitors as antibiotics, is their incor-
poration into peptidase-activated prodrugs. In addition
momethyl ketone derivative over DON can be attrib-
uted to a 3-fold higher anity with which the former
binds to the enzyme, prior to covalent modi®cation,
although this is o€set somewhat by the slightly higher
rate of covalent complex formation (k+2), between the
latter and glucosamine synthetase.
1
0
to the naturally occurring tripeptides alazopeptin and
duazomycin,11 which both carry intermolecular DON
residues, the compound bacilysin (7) (tetaine), a dipep-
tide containing (l)-b-(2,3-epoxycyclohexan-4-one)-ala-
nine (anticapsin),12 which is released after the action of
an unidenti®ed intracellular aminopeptidase, is a potent
anti-fungal and anti-bacterial agent. Importantly, baci-
lysin itself has no inherent activity against glucosamine
synthetase, and as such, reduced toxicity. Thus incor-
poration of the dimethylsulphonium ketone into peptide
prodrugs, with selective susceptibility to bacterial pepti-
dases, may provide a means to target this potent
inhibitor into bacterial and/or fungal cells.
The greatly enhanced ecacy of the bromomethyl
compared with the chloromethyl ketone can similarly be
attributed to the enhanced anity of the former with
the enzyme (20-fold greater, re¯ected in the respective Ki
values for both). Additionally, the 2-fold greater rate
constant for the formation of the covalent complex also
contributes to this greater overall eciency.
Comparisons between the e€ectiveness of the diazoethyl
ketone analogue (6) and DON illustrates that the latter
�
1
� 1
is some 3.5-fold more active (20,000 M min com-
�
1
� 1
pared with 70,000 M min ). The reduced potency of
the diazoethyl ketone can be attributed to a very pro-
nounced decrease in the ®rst-order rate constant for
covalent complex formation (35-fold), compared with
DON. It is interesting to note, however, that anity for
the enzyme is actually higher (8-fold) with the diazoethyl
analogue, thus partially o€setting the loss in reactivity. It
would therefore appear that the introduction of the
Acknowledgements
This work was supported by the award of an MRC
(UK) project grant (No. 8803936) to B.W. and A.J.H.
Table 1. Steady-state inactivation rate constants (K
constants for inactivation (k+2) and second order rate constants (k+2
i
), limiting rate
References and Notes
/
K
i
) for compounds 1±6
1
. Ghosh, S.; Blumenthal, H. J.; Davidson, E.; Roseman, S. J.
Biol. Chem. 1960, 235, 1265.
. Kawamura, M.; Keim, P. S.; Goto, Y.; Zalkin, H.; Hein-
kikson, R. L. J. Biol. Chem. 1978, 253, 4659.
. Vollmer, S. J.; Switzer, R. L.; Hermodson, M. A.; Bower,
S. G.; Zalkin, H. J. Biol. Chem. 1983, 258, 582.
. Tso, J.; Zalkin, H.; Van Cleemput, M.; Yanofsky, C.;
Smith, J. M. J. Biol. Chem. 1982, 257, 3525.
. Badet, B.; Vermoote, P.; Haumont, P.-Y.; Lederer, F.; Le
Goc, F. Biochemistry 1987, 26, 1940.
6. Stuber, W.; Kosina, H.; Heimburger, N. Int. J. Pept. Prot.
Res. 1988, 31, 63.
a
i
� 1
� 1
� 1
Compounds
K
(mM)
k
+2 (min
)
k+2/K (mM min
i
2
1
2
3
4
5
6
9.5 (Æ0.40)
62.5 (Æ3.7)
2.7 (Æ0.1)
0.7 (Æ0.04)
0.19 (Æ0.007)
0.4 (Æ0.06)
0.07 (Æ0.002)
0.003 (Æ0.0001)
0.15 (Æ0.02)
3
b
b
b
NI
NI
NI
4
0.37 (Æ0.07)
0.12 (Æ0.03)
0.32 (Æ0.04)
1.1 (Æ0.1)
0.02 (Æ0.005)
0.02 (Æ0.007)
5
aValues are means of four determinations; standard deviation is given
in parentheses.
È
b
NI=no inhibition at 200 mM.

2
798
B. Walker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2795±2798
7
8
9
. Shiio, I.; Ishii, K. J. Biochem. 1969, 66, 175.
. Kitz, R.; Wilson, I. B. J. Biol. Chem. 1962, 237, 3245.
. Moon, J. B.; Coleman, R. S.; Hanzlik, R. P. J. Am. Chem.
10. Patterson, E. L.; Johnson, B. L.; DeVoe, S. E.; Bohonos,
N. Antimicrob. Agents Chemother. 1965, 115.
11. Rao, K. V. Antimicrob. Agents Chemother. 1961, 178.
12. Chmara, H.; Borowski, E. Biochem. Biophys. Res. Com-
mun. 1973, 52, 1381. Kenig, M.; Vandamme, E.; Abraham, E.
P. J. Gen. Microbiol. 1976, 94, 46.
Soc. 1986, 108, 1350. Picken, P. P.; Guthrie, D. J. S.; Walker,
B. Biochem. Soc. Trans. 1990, 18, 316. Dufour, E.; Storer, A.
C.; Menard, R. Biochemistry 1995, 34, 9136.
Â