Synthesis and structure-activity relationships of o-sulfonamido- arylhydrazides as inhibitors of LL-diaminopimelate aminotransferase (LL-DAP-AT)

Article abstract of DOI:10.1039/c2ob00040g

Recently, ll-diaminopimelate aminotransferase (ll-DAP-AT), a pyridoxal-5′-phosphate (PLP)-dependent enzyme, was reported to catalyze a key step in the biosynthesis of l-lysine in plants and Chlamydia. Previous screening of a 29201-compound library against

Full text of DOI:10.1039/c2ob00040g

View Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
This article is part of the
OBC 10th anniversary
themed issue
All articles in this issue will be gathered together
online at
www.rsc.org/OBC10

Dynamic Article Links
Organic &
View Online
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 5815
www.rsc.org/obc
COMMUNICATION
Synthesis and structure–activity relationships of o-sulfonamido-
arylhydrazides as inhibitors of LL-diaminopimelate aminotransferase
(LL-DAP-AT)†‡
Chenguang Fan and John C. Vederas*
Received 5th January 2012, Accepted 30th January 2012
DOI: 10.1039/c2ob00040g
semialdehyde to form L-tetrahydrodipicolinate (L-THDP) in two
enzymatic steps. Bacterial conversion of L-THDP to LL-diamino-
pimelate (LL-DAP) is generally achieved by three enzymatic
steps, acylation, transamination, and deacylation.11 At a later
stage, LL-DAP is converted by DAP epimerase to meso-DAP, a
crucial component of many Gram-negative bacterial cell
walls.12–14 Finally, L-lysine is formed upon decarboxylation of
meso-DAP. Until recently this biosynthetic pathway was believed
to be used by both plants and bacteria. However, a newly discov-
ered aminotransferase (AT) enzyme was isolated which converts
L-THDP directly to LL-DAP (Fig. 1b).15 This represents a shorter
route in the biosynthetic pathway, which normally requires three
enzymes for catalysis, but here is achieved within a single step.
This pyridoxal-5′-phosphate (PLP) dependent enzyme catalyzes
a transamination reaction, and is therefore named LL-diaminopi-
melate aminotransferase (LL-DAP-AT).16,17 It occurs in higher
plants, some bacteria, archaea and algae, including cyanobacteria
and chlamydia species.18 Specifically, recent genetic analysis has
shown that LL-DAP-AT occurs in Cyanobacteria, Desulfuromo-
nadales, Firmicutes, Bacteroidetes, Chlamydiae, Spirochaeta,
and Chloroflexi and two archaeal groups, Methanobacteriaceae
and Archaeoglobaceae.19,20
Recently, LL-diaminopimelate aminotransferase (LL-DAP-
AT), a pyridoxal-5′-phosphate (PLP)-dependent enzyme, was
reported to catalyze a key step in the biosynthesis of L-lysine
in plants and Chlamydia. Previous screening of a 29 201-com-
pound library against LL-DAP-AT identified an o-sulfonami-
doarylhydrazide as a reversible inhibitor with IC50 ∼ 5 μM.
Structure–activity relationship (SAR) studies based on this
lead compound identified key structural features essential for
enzyme inhibition and led to slightly improved inhibitors.
Preliminary studies on the mode of inhibition of LL-DAP-AT
by this class of compounds are also reported.
Introduction
Lysine biosynthesis occurs in bacteria and plants via variants of
the diaminopimelate (DAP) pathway.1,2 However, formation of
this essential amino acid is absent in mammals, and thus they
must obtain it from their diet.3 The direct precursor of lysine in
this biosynthetic route, meso-DAP, is also involved in formation
of cross-links between stem peptides of peptidoglycan in Gram-
negative bacteria. Hence it has an important role in maintaining
integrity of bacterial cell walls. Its metabolic product, L-lysine
has an analogous function in many Gram-positive bacteria.4,5
The DAP pathway is a target for a new class of antibiotics or
herbicides with low-toxicity to mammals. This has prompted
study of both the structures of the enzymes involved in DAP bio-
synthesis, as well as design of inhibitors for them.3,6–9 As lysine
intake can be limited in human diets, complete understanding of
lysine biosynthesis could also assist in engineering plants with
greater lysine content.10
In some cases, inhibitors of enzymes involved in the DAP
pathway have already been shown to have antimicrobial
activity.8,21,22 Inhibitors that directly target LL-DAP-AT could
potentially act as herbicides or as antibiotics specific to Chlamy-
diae. Substrate-based inhibitors could be very effective;
however, they often suffer from poor uptake by the bacterial
cells.8 Compound library screening provides an alternative
approach to overcome this problem.
In earlier work23 we screened nearly 30 000 compounds
against LL-DAP-AT from Arabidopsis thaliana, and found 46
compounds displaying inhibition greater than 13% at 10 μM by
both robotic and manual testing. Three pharmacophores (barbitu-
rate, thiobarbiturate, and rhodanine) were identified based on
their high occurrence in the primary hits (17%, 17%, and 28%,
respectively). From the screening, the best inhibitor (1, Table 1)
contains a hydrazide functionality, which is a potentially reactive
group. Two analogues of 1 tested earlier suggested that an
unsubstituted hydrazide moiety appears to be necessary for
inhibition.23
In most bacteria, lysine is biosynthesized by the route shown
in Fig. 1a. The pathway starts with pyruvate and L-aspartate
Department of Chemistry, University of Alberta, Edmonton, Alberta,
Canada, T6G 2G2. E-mail: john.vederas@ualberta.ca
†This article is part of the Organic & Biomolecular Chemistry 10th
Anniversary issue.
‡Electronic supplementary information (ESI) available: Detailed
enzyme assay, general synthetic procedure, and characterization data of
analogues. See DOI: 10.1039/c2ob00040g
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 5815–5819 | 5815

View Online
Fig. 1 (a) L-Lysine biosynthesis via the DAP pathway in bacteria. Enzymes involved: (i) DHDP synthase; (ii) DHDP reductase; (iii) THDP acyltrans-
ferase; (iv) N-acyl-DAP aminotransferase; (v) N-acyl-DAP deacylase; (vi) DAP epimerase; (vii) DAP decarboxylase. (b) The shortcut of the DAP
pathway in higher plants and Chlamydiae.
In the current study, additional analogues based on 1 were
synthesized and tested against LL-DAP-AT from A. thaliana. The
assay used in this work is shown in Fig. 2.15 Briefly, the assay
involves monitoring the reverse of the transamination process,
with α-ketoglutarate being used as an amino acceptor. Introduc-
tion of o-aminobenzaldehyde to the assay allows for the product
(L-THDP) to be converted into a conveniently monitored chro-
mophore. Furthermore, to test whether the hydrazide functional
group reacts with PLP or not, NMR spectroscopy and mass spec-
trometry experiments were done. Isoniazid (18) was tested with
the enzyme to examine whether this important anti-tuberculosis
drug can inhibit DAP-AT. An adduct between one of the ana-
logues and PLP was also tested against the enzyme.
Modification of the aromatic system was undertaken. Phenyl
ring analogues were synthesized and tested against the enzyme
(Table 1). The fact that compound 5 showed lower activity com-
pared to the lead compound, implies that substitution is needed
on the phenyl ring in order to retain activity. Electronic effects
were also considered, and compounds bearing chlorine and
methoxy groups para to the hydrazide moiety (compounds 6 and
7) were synthesized. The results showed these two compounds
had similar activity against the enzyme, which suggests that the
electronic effects may not play a very significant role in the inhi-
bition. Moreover, these two analogues showed slightly better
activity than the lead compound, which supports the hypothesis
mentioned above that substitution is needed on the phenyl ring.
The dimethoxy-substituted analogue 8 showed worse inhibition
than either the lead compound 1 or the mono-methoxy-substi-
tuted analogue 7. The extra methoxy group para to the sulfona-
mide moiety may be hindering the enzyme–analogue interaction
through steric effects. Since substitution is necessary on the
phenyl ring and the size of the substituent cannot be too large,
monofluoro- and difluoro-substituted analogues 9 and 10 were
synthesized and evaluated. Difluoro-substituted analogue 10
showed slightly worse inhibition than the lead compound 1,
whereas the monofluoro-substituted analogue 9 showed two-fold
improvement of activity relative to 1 and is the best inhibitor in
the current study. In addition, para-methyl-hydrazide and para-
trifluoromethyl-hydrazide analogues 11 and 12 were examined.
The trifluoromethyl group enhances the activity compared to the
methyl analogue, which indicates that an electron-withdrawing
group on the benzene ring para to the hydrazide moiety can be
beneficial for the inhibitory activity.
Results and discussion
Compound 1 has an IC50 value of 5 μM23 and was chosen as a
lead for our structure–activity relationship (SAR) analysis of
LL-DAP-AT. The simple synthetic route employed is shown in
Scheme 1. Various amino substituted carboxylic acids were first
reacted with sulfonyl chlorides under basic conditions to afford a
sulfonamide (2a–17a). Then the sulfonamide carboxylic acid
was coupled with hydrazine using carbonyl diimidazole as a
coupling reagent to form the product (2–17).
It has been shown that the terminally unsubstituted NH2 of the
hydrazide group is necessary for inhibition;23 thus, no further
modifications of this group were done. The first set of modifi-
cations focused on the naphthalene ring moiety. Compound 2,
lacking the aromatic system, was totally devoid of inhibitory
activity against the enzyme. Furthermore, L- and D-proline
derivatives (3 and 4) were synthesized and tested against the
LL-DAP-AT and both compounds showed no inhibition at
200 μM. These results suggest that a β-amino acid derived aro-
matic system is necessary for inhibition.
With the para-fluoro-hydrazide analogue as the best inhibitor
in the series of aromatic ring modifications, we then decided to
explore the sulfonamide moiety, based on the analogue 9,
Table 1 (compounds 13–17). At first, the para-chloro and
5816 | Org. Biomol. Chem., 2012, 10, 5815–5819
This journal is © The Royal Society of Chemistry 2012

View Online
Table 1 Inhibitor testing againt LL-DAP-AT
Entry
Structure
IC50/μM
Entry
Structure
IC50/μM
Entry
Structure
IC50/μM
5.3
1a
5
7
4.5
13
2
3
4
No inhibition
No inhibition
No inhibition
8
20.7
14
15
16
3.4
9
2.5
7.5
10
5.9
5.6
5
6
13.2
11
12
7.9
3.8
17
No inhibition
No inhibition
4.1
18b
a Compound 1 purchased from ChemBridge Corporation. b Compound 18 purchased from Sigma-Aldrich.
Scheme 1 General synthetic route for analogues. Reagents: (i)
R-SO2Cl, Na2CO3, H2O; (ii) carbonyl diimidazole, DMF; hydrazine
monohydrate, DMF.
Fig. 2 Assay for inhibition against LL-DAP-AT.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 5815–5819 | 5817

View Online
required. Furthermore, since the hydrazone adduct formation is a
reversible process whose rate can be catalyzed by other nucleo-
philic groups,25,26 active compounds such as 1 are completely
reversible inhibitors.
Finally, in a preliminary experiment, the preformed adduct
between compound 9 and PLP, namely 19, was also tested
against the enzyme. However, the adduct 19 only showed 40%
inhibition at 100 μM. In contrast, the free hydrazide 9 displays
∼98% inhibition at 100 μM. The adduct showed much worse
inhibition. Although it is not possible to fully stop hydrolysis of
19 to 9 in the enzyme assay, this result suggests that the cofactor
exchange is relatively slow.25,26 The mode of action of the
inhibitors to LL-DAP-AT will be studied further by X-ray crystal-
lographic analysis and detailed enzyme kinetics.
Scheme 2 Imine formation.
para-methoxy benzenesulfonamide were employed to examine
the electronic effect. Both analogues 13 and 14 showed worse
inhibition than compound 9. However, an electron-donating
group appears to be better on the para position of the benzene-
sulfonamide compared to an electron-withdrawing group.
Furthermore, para-fluorobenzenesulfonamide and para-toluene-
sulfonamide were used to construct the other two analogues 15
and 16. Ultimately, both of the analogues were reasonable inhibi-
tors, but not as effective as unsubstituted phenyl ring without
any substituents. Finally, the methanesulfonamide analogue 17
was synthesized and tested. Surprisingly, the analogue did not
show any inhibition up to 100 μM against the enzyme, indicating
that an aromatic system connected to the sulfonyl group is
necessary for the inhibition.
If the NH2 of the hydrazide moiety is capped by a ring or
acetyl group, the analogues show no activity against the enzyme,
which indicates that a free hydrazide amino group is essential for
the inhibition.23 Although 1 displays completely reversible time-
independent inhibition,23 the free hydrazide moiety is reasonably
nucleophilic and could react with the enzyme cofactor, pyri-
doxal-5′-phosphate (PLP). To determine whether the hydrazide
analogues react with PLP or not, compound 9 was mixed with
one molar equivalent of PLP, in a 1 : 1 mixture of deuterated
water and deuterated methanol solution at room temperature.
The 1H- and 13C-NMR results matched with the proposed
imine structure 19 shown in Scheme 2. Isolation of the imine
compound and high resolution mass-spectrometry (HR-MS)
confirmed its identity. These data indicate that, as expected, the
hydrazide moiety can react with PLP to form an imine (hydra-
zone) adduct.
Despite this reactivity, removal of PLP from the enzyme by
the free amino terminus of the hydrazide moiety cannot be the
sole factor for inhibition. Compounds 2–4, 17 and isoniazid (18)
display no detectable inhibition of DAP-AT at concentrations up
to 200 μM. Some of these compounds, including the clinically
used anti-tuberculosis drug, isoniazid (18),24 are sterically much
less demanding than lead compound 1 or the best inhibitor 9.
Since the isoniazid has a hydrazide moiety with a free terminus,
the hydrazone adduct is readily formed between isoniazid and
PLP in the absence of enzyme. If the formation of the PLP-
hydrazone adduct or sequestration of the cofactor were to be the
main requirement for enzyme inhibition, then isoniazid should
show activity against LL-DAP-AT, but this is not observed.
Hydrazone adduct formation may occur after the active ana-
logues enter the enzyme active site that contains PLP bound as
an imine to lysine-270.16 However, hydrazone formation cannot
be the only factor required to inhibit to the enzyme. Additional
interactions between the inhibitor and enzyme active site are
Conclusions
In this SAR study, 16 hydrazide analogues were synthesized
based on a lead compound 1 identified from previous library
screening. The analogues were tested as inhibitors against
LL-DAP-AT from A. thaliana, and the best inhibitor was found to
be an o-sulfonamido-p-fluorophenylhydrazide 9, with an IC50
value of 2.5 μM. This compound was also found to react with
PLP readily, but the hydrazone adduct 19 was not as effective an
inhibitor of the enzyme. Testing of isoniazid (18), a hydrazide-
containing drug, showed no inhibition against LL-DAP-AT,
which indicates that the reaction between analogues and PLP
may play a role in the inhibition, but is not the key factor.
Acknowledgements
This work was supported by the Natural Sciences and Engineer-
ing Research Council of Canada (NSERC) and the Canada
Research Chair in Bioorganic and Medicinal Chemistry. The
authors thank Dr. David Dietrich for helpful suggestions.
References
1 G. Jander and V. Joshi, Mol. Plant-Microbe Interact., 2010, 3, 54.
2 J. H. Park and S. Y. Lee, Biotechnol. J., 2010, 5, 560.
3 J. C. Vederas, Can. J. Chem., 2006, 84, 1197.
4 W. Vollmer, D. Blanot and M. A. de Pedro, FEMS Microbiol. Rev., 2008,
32, 149.
5 W. Vollmer and U. Bertsche, Biochim. Biophys. Acta, 2008, 1778, 1714.
6 C. A. Hutton, M. A. Perugini and J. A. Gerrard, Mol. BioSyst., 2007, 3,
458.
7 C. A. Hutton, T. J. Southwood and J. J. Turner, Mini-Rev. Med. Chem.,
2003, 3, 115.
8 R. J. Cox, A. Sutherland and J. C. Vederas, Bioorg. Med. Chem., 2000, 8,
843.
9 N. Watanabe and M. N. James, Biochim. Biophys. Acta, 2011, 1814,
1528.
10 B. Mazur, E. Krebbers and S. Tingey, Science, 1999, 285, 372.
11 G. Scapin and J. S. Blanchard, Adv. Enzymol. Relat. Areas Mol. Biol.,
1998, 72, 279.
12 B. Pillai, V. A. Moorthie, M. J. van Belkum, S. L. Marcus, M.
M. Cherney, C. M. Diaper, J. C. Vederas and M. N. G. James, J. Mol.
Biol., 2009, 385, 580.
13 B. Pillai, M. Cherney, C. M. Diaper, A. Sutherland, J. S. Blanchard, J.
C. Vederas and M. N. G. James, Biochem. Biophys. Res. Commun., 2007,
363, 547.
14 B. Pillai, M. M. Cherney, C. M. Diaper, A. Sutherland, J. S. Blanchard, J.
C. Vederas and M. N. G. James, Proc. Natl. Acad. Sci. U. S. A., 2006,
103, 8668.
5818 | Org. Biomol. Chem., 2012, 10, 5815–5819
This journal is © The Royal Society of Chemistry 2012

View Online
15 A. O. Hudson, B. K. Singh, T. Leustek and C. Gilvarg, Plant Physiol.,
2006, 140, 292.
16 N. Watanabe, M. M. Cherney, M. J. van Belkum, S. L. Marcus, M.
D. Flegel, M. D. Clay, M. K. Deyholos, J. C. Vederas and M. N. James,
J. Mol. Biol., 2007, 371, 685.
17 N. Watanabe, M. D. Clay, M. J. van Belkum, M. M. Cherney, J.
C. Vederas and M. N. James, J. Mol. Biol., 2008, 384, 1314.
18 A. J. McCoy, N. E. Adams, A. O. Hudson, C. Gilvarg, T. Leustek and A.
T. Maurelli, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 17909.
19 A. O. Hudson, C. Gilvarg and T. Leustek, J. Bacteriol., 2008, 190, 3256.
20 D. E. Graham and H. K. Huse, FEBS Lett., 2008, 582, 1369.
21 D. Gillner, N. Armoush, R. C. Holz and D. P. Becker, Bioorg. Med.
Chem. Lett., 2009, 19, 6350.
22 R. J. Cox and P. S. H. Wang, J. Chem. Soc., Perkin Trans. 1, 2001, 2022.
23 C. Fan, M. D. Clay, M. K. Deyholos and J. C. Vederas, Bioorg. Med.
Chem., 2010, 18, 2141.
24 G. S. Timmins and V. Deretic, Mol. Microbiol., 2006, 62, 1220.
25 A. Dirksen, S. Dirksen, T. M. Hackeng and P. E. Dawson, J. Am. Chem.
Soc., 2006, 128, 15602.
26 A. Dirksen and P. E. Dawson, Bioconjugate Chem., 2008, 19, 2549.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 5815–5819 | 5819