Structure-activity relationships in a new class of non-substrate-like covalent inhibitors of the bacterial glycosyltransferase LgtC

Article abstract of DOI:10.1016/j.bmc.2018.03.006

Lipooligosaccharide (LOS) structures in the outer core of Gram-negative mucosal pathogens such as Neisseria meningitidis and Haemophilus influenzae contain characteristic glycoepitopes that contribute significantly to bacterial virulence. An important example is the digalactoside epitope generated by the retaining α-1,4-galactosyltransferase LgtC. These digalactosides camouflage the pathogen from the host immune system and increase its serum resistance. Small molecular inhibitors of LgtC are therefore sought after as chemical tools to study bacterial virulence, and as potential candidates for anti-virulence drug discovery. We have recently discovered a new class of non-substrate-like inhibitors of LgtC. The new inhibitors act via a covalent mode of action, targeting a non-catalytic cysteine residue in the LgtC active site. Here, we describe, for the first time, structure-activity relationships for this new class of glycosyltransferase inhibitors. We have carried out a detailed analysis of the inhibition kinetics to establish the relative contribution of the non-covalent binding and the covalent inactivation steps for overall inhibitory activity. Selected inhibitors were also evaluated against a serum-resistant strain of Haemophilus influenzae, but did not enhance the killing effect of human serum.

Full text of DOI:10.1016/j.bmc.2018.03.006

Accepted Manuscript
Structure-activity relationships in a new class of non-substrate-like covalent in-
hibitors of the bacterial glycosyltransferase LgtC
Yong Xu, Jon Cuccui, Carmen Denman, Tripty Maharjan, Brendan W. Wren,
Gerd K. Wagner
PII:
S0968-0896(18)30135-4
https://doi.org/10.1016/j.bmc.2018.03.006
BMC 14244
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
23 January 2018
3 March 2018
4 March 2018
Please cite this article as: Xu, Y., Cuccui, J., Denman, C., Maharjan, T., Wren, B.W., Wagner, G.K., Structure-
activity relationships in a new class of non-substrate-like covalent inhibitors of the bacterial glycosyltransferase
LgtC, Bioorganic & Medicinal Chemistry (2018), doi: https://doi.org/10.1016/j.bmc.2018.03.006
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Structure-activity relationships in a new class of non-substrate-like covalent
inhibitors of the bacterial glycosyltransferase LgtC
Yong Xu [1], Jon Cuccui [2], Carmen Denman [2], Tripty Maharjan [2], Brendan W. Wren [2]
& Gerd K. Wagner [1]*
[1]
King’s College London, Department of Chemistry, Faculty of Natural & Mathematical
Sciences, Britannia House, 7 Trinity Street, London, SE1 1DB, UK. Phone: +44 (0)20 7848
1926; e-mail: gerd.wagner@kcl.ac.uk
[2]
Faculty of Infectious and Tropical Diseases, London School of Hygiene & Tropical
Medicine
*Corresponding author
1 March 2018
Keywords: glycosyltransferase, covalent inhibitor, structure-activity relationship,
bacterial virulence, serum resistance
1

Abstract
Lipooligosaccharide (LOS) structures in the outer core of Gram-negative mucosal
pathogens such as Neisseria meningitidis and Haemophilus influenzae contain
characteristic glycoepitopes that contribute significantly to bacterial virulence. An
important example is the digalactoside epitope generated by the retaining α-1,4-
galactosyltransferase LgtC. These digalactosides camouflage the pathogen from the
host immune system and increase its serum resistance. Small molecular inhibitors of
LgtC are therefore sought after as chemical tools to study bacterial virulence, and as
potential candidates for anti-virulence drug discovery. We have recently discovered a
new class of non-substrate-like inhibitors of LgtC. The new inhibitors act via a
covalent mode of action, targeting a non-catalytic cysteine residue in the LgtC active
site. Here, we describe, for the first time, structure-activity relationships for this new
class of glycosyltransferase inhibitors. We have carried out a detailed analysis of the
inhibition kinetics to establish the relative contribution of the non-covalent binding
and the covalent inactivation steps for overall inhibitory activity. Selected inhibitors
were also evaluated against a serum-resistant strain of Haemophilus influenzae, but
did not enhance the killing effect of human serum.
2

1. Introduction
Glycoconjugates such as glycolipids and glycoproteins play a critical role for the
virulence and viability of many bacterial pathogens [1]. The biosynthetic machinery
required for bacterial glycoconjugate synthesis therefore offers a rich source of
potential targets for the development of novel anti-bacterial or anti-virulence agents.
A central role for bacterial glycoconjugate biosynthesis is played by
glycosyltransferases (GTs), a family of enzymes that catalyse the transfer of a sugar
from a glycosyl donor to a requisite acceptor [2]. It has been estimated that for most
organisms, 1-3% of their genes encode GTs and other carbohydrate-active enzymes
[3]. From the 1.6 Mb genome of Campylobacter jejuni NCTC 11168, for example,
approximately 45 GTs have been predicted [4]. Individual bacterial GTs such as the
N-acetylglucosamine transferase MurG, a key enzyme for peptidoglycan
biosynthesis, have been identified as promising targets for the discovery of novel
antibiotics [5].
Another important bacterial GT is the retaining α-1,4-galactosyltransferase LgtC
[6], which catalyses the transfer of a D-galactose moiety from a UDP-D-galactose
(UDP-Gal) donor to lactose-containing acceptors in the lipooligosaccharide (LOS)
envelope of Gram-negative pathogens such as Neisseria and Haemophilus (Fig. 1).
LgtC is a member of family GT-8 in the CAZy database of carbohydrate-active
enzymes [7], and is highly conserved across different Neisseria, Haemophilus and
Pasteurella species. The resulting digalactoside epitopes confer resistance to pre-
existing antibody and complement-mediated lysis [8] and have been associated with
increased virulence in an in-vivo model of Haemophilus influenzae infection [9]. The
expression of LgtC has also been linked directly with the high-level serum resistance
of H. influenzae R2866 [10]. Small molecular inhibitors of LgtC are therefore of
3

considerable interest as chemical tools for microbiological investigation and as
potential lead compounds for anti-virulence drug discovery [11].
Previously reported LgtC inhibitors are mainly substrate-like molecules derived
from the UDP-Gal donor [6,12]. Due to the presence of the charged diphosphate
fragment and their potentially limited stability, such donor analogues are not suitable
for applications with bacterial cultures or cells. We have recently discovered a novel
class of non-substrate-like LgtC inhibitors based on a pyrazol-3-one scaffold (Fig. 2)
[13]. These inhibitors behave as substrate mimics and react covalently, at their
Michael acceptor system, with a non-catalytic cysteine in the LgtC active site [13].
Due to their uncharged, drug-like structure and straightforward synthesis, these
pyrazol-3-ones are attractive for applications in chemical biology and drug discovery.
Fig. 1 The LgtC reaction
In the present study, we have systematically explored structural modifications
around the pyrazol-3-one scaffold. Starting from prototype inhibitor 1 [13], we first
designed and synthesised analogues with various substituents at the 5-position of
the pyrazol-3-one scaffold (Fig. 2). We investigated a range of 5-substituents with
different steric and electronic properties, in order to probe their effect on the
reactivity of the Michael acceptor system. Following the identification of a 5-CF3
substituent as advantageous, further structural optimisation was carried out on the 2-
4

and 4-position. To understand the relative contribution of the non-covalent binding
step and the covalent inactivation step to the overall inhibition of LgtC, we carried out
detailed covalent inhibition kinetics with selected derivatives. Determination of their
inhibition constants Ki and inactivation rates kinact and correlation of these parameters
with IC50 values allowed us, for the first time, to establish SAR in this series. We also
tested selected pyrazol-3-ones against H. influenzae strain R2866, both in growth
inhibition and serum survival assays. Unfortunately, these inhibitors had no effect on
serum resistance, and only a minimal effect on bacterial growth. Although these
pyrazol-3-ones therefore have probably limited potential as drug candidates, our
strategy and results provide a strong foundation for the rational design of alternative
covalent inhibitor chemotypes for this enzyme as well as related bacterial GTs.
Fig. 2 Design strategy for pyrazol-3-one derivatives investigated in this study
2. Results and discussion
2.1 Chemistry
To obtain the target molecules, one of the key steps was the synthesis of
intermediates with different substituents in the 5-position of the pyrazol-3-one
scaffold (Scheme 1A). Six intermediates (3a-3c, 4a-b, and 5) were obtained by
condensation reactions of commercially available aryl hydrazines and β-ketoesters
5

2a-c in acetic acid at 120 °C. The condensation products were obtained in 41-75%
yield after purification by column chromatography. 3c was then converted to other
cyclised intermediates by different synthetic routes. 3c was heated at reflux in 28%
aqueous ammonia for 12 h to generate the amide derivative 3d, while a hydrolysis
reaction with 3c afforded carboxylic acid derivative 3e. With intermediates 3, 4, and 5
in hand, the benzylidene-substituted pyrazol-3-one products 6-8 were obtained by
condensation with the respective aldehyde in moderate to good yield under
microwave conditions (Scheme 1B).
3a R1 = -CF3, R2 = -H
R1
3b R1 = -CH2CO2Et, R2 = -H
4a R1 = -CF3, R2 = 4-CO2H
4b R1 = -CF3, R2 = 3-CO2H
5 R1 = -CH3, R2 = -H
N
A
N
OH
R2
(i)
(i)
H2N
H
O
O
O
N
N
NH2
+
N
R1
O
R2
(ii)
O
OH
2a R1 = -CF3
2b R1 = -CO2Et
2c R1 = -CH2CO2Et
O
3d
N
N
HO
O
OH
(iii)
N
3c
N
OH
3e
B
R1
6 R1 = -CH3
7 R1 = -CF3
8a R1 = -CO2Et
8b R1 = -CH2CO2Et
8c R1 = -CO2H
8d R1 = -C(O)NH2
R1
N
N
N
CHO
(iv)
N
+
O
R2
R3
OH
R2
R3
3-5
Scheme 1. Synthesis of pyrazol-3-one derivatives 6-8. Reagents and conditions: (i)
AcOH, 110 °C, overnight, 41-75%; (ii) aq. NH3 (28%), reflux, 38%; (iii) aq. NaOH (1
N), EtOH, rt, 91%; (iv) aldehyde, 160 °C, microwave, 15 mins, 15-80%. For
substituents R2 and R3 see Tables 1-3.
6

Compounds 10 and 12 were prepared as shown in Scheme 2. The carboxylic
acid intermediate 4b, was reacted with 2-aminoethan-1-ol in the presence of
diisopropylethylamine (DIEA) and 2-(1H-benzotriazol-1-yl)-1,1,3,3-
tetramethyluronium hexafluorophosphate (HBTU) in DMF to afford amide 9.
(Scheme 2A). For the synthesis of intermediate 11, ethyl 3-oxobutanoate was heated
to reflux with benzylhydrazine in glacial acetic acid (Scheme 2B). The target
compounds 10 and 12 were obtained from 9 and 11, respectively, via condensation
reaction with 3-bromo-4-methoxybenzaldehyde under the microwave conditions
described above.
CF3
CF3
A
CF3
N
N
N
N
N
(i)
(iii)
N
NH2
+
HO
OH
OH
Br
O
OH
O
HOOC
N
OH
10
OCH3
H
N
O
4b
H
9
CF3
B
CF3
N
N
N
O
O
(ii)
(iii)
NH
+
N
NH2
F3C
O
Br
OCH3
O
OH
11
12
Scheme 2. Synthesis of pyrazol-3-one derivatives 10 and 12. Reagents and
conditions: (i) HBTU, DIEA, DMF, rt, 75%; (ii) AcOH, 110 °C, overnight, 17%; (iii) 3-
bromo-4-methoxybenzaldehyde, 160 °C, microwave, 15 mins, 9% (10) or 28% (12).
The 1D NMR spectra of all benzylidene-substituted pyrazol-3-ones showed the
presence of only a single geometric isomer (ESI). Consistently, the peak for the H-1
and/or H-6 proton at the benzylidene ring appeared distinctly downfield (>8.5 ppm).
This downfield shift is caused by the deshielding effect of the carbonyl group on
these protons in the (Z) isomer, and is therefore diagnostic for the (Z) configuration
7

of the double bond [14]. Further evidence for the (Z) configuration was obtained from
a NOESY experiment with 5-methyl pyrazol-3-one 6 (ESI), which showed a NOE
correlation between the protons of the 5-CH3 group (δ 2.3 ppm) and the CH at the
exocyclic double bond (δ 7.75). While a similar NOESY analysis is not applicable in
the 5-CF3 series, due to the absence of protons from position 5, the (Z) configuration
of 6 is in keeping with the configuration observed in the crystal structure of a
structurally related 5-CF3 pyrazol-3-one [15].
2.2 Biochemical evaluation of inhibitors
2.2.1 Inhibition of LgtC
To assess the inhibitory activity of the pyrazolone derivatives, we used a
colorimetric biochemical assay previously reported by our group [16]. In this enzyme-
coupled assay, calf intestinal phosphatase (CIP) is used to selectively and
quantitatively release inorganic phosphate (Pi) from UDP, the secondary product of
the LgtC reaction, followed by quantification of Pi with Malachite Green. Because of
the covalent mode of action previously observed in this inhibitor series [13], we pre-
incubated the enzyme with inhibitor prior to starting the enzymatic reaction.
First, we assessed the potency of pyrazol-3-one derivatives with different
substituents at the 5-position, but identical substituents at positions 2 and 4 (Table 1).
Inhibitory activity was determined at two different concentrations of inhibitor (25 µM
and 50 µM), and with 30 mins pre-incubation time. Under these conditions, the 5-
CF3-substituted derivative 7b displayed the strongest inhibitory activity, with about 84%
inhibition at 50 µM. Practically all derivatives with an electron-withdrawing 5-
substituent were more active than the electron-donating 5-methyl congener 6, which
showed only about half the level of inhibition of 7b. These results suggested that
8

electron-withdrawing 5-substituents are advantageous for LgtC inhibition, possibly by
increasing the reactivity of the Michael acceptor system. However, caution has to be
applied when analysing the behaviour of covalent inhibitors, as their overall activity is
determined not only by the reactivity of their electrophilic warhead, but also by their
binding affinity at the target [17]. Variations in potency within this series of 5-
substituted pyrazol-3-ones may therefore result not only from differences in Michael
acceptor reactivity, but also from different non-covalent binding affinities. The
decreased activity of compounds with bulky electron-withdrawing 5-substituents (8c,
8d), for example, may be due to suboptimal non-covalent binding outweighing the
increased reactivity of the Michael acceptor.
To better understand the interplay between covalent inactivation and non-
covalent binding for overall inhibitory activity of these pyrazol-3-ones, we next
investigated how different substituents on the 2-phenyl and 4-benzylidene moieties
affect inhibitory activity in a series of analogues with the same 5-substituent. Based
on initial SAR trends, we selected 5-CF3 derivative 7b as the starting point for this
set of modifications (Table 2).
Table 1
Pyrazol-3-ones with different 5-substituents, and their inhibitory activity
against LgtC.
R1
N
N
Br
O
OCH3
inhibition (%)a
Cmpd
R1
at 50 µM
43 ± 2
at 25 µM
33 ± 0.1
80 ± 4
6
-CH3
7b
8a
-CF3
84 ± 3
-CO2Et
66 ± 3
61 ± 3
9

8b
8c
8d
-CH2CO2Et
-CO2H
67 ± 4
58 ± 1
39 ± 6
32 ± 5
35 ± 1
31 ± 6
-C(O)NH2
aLgtC was pre-incubated with inhibitor (25 µM or 50 µM) or DMSO, UDP-Gal (28
μM), MnCl2 (5 mM), CIP (10 U/mL), CEL (1 mg/mL), and Triton (0.01%) for 30 mins
at 30 °C in 13 mM HEPES buffer (pH 7.0). Lactose (2 mM) was added, and the
reactions were incubated for 20 mins at 30 °C. Each compound was tested in
triplicate; results are presented as average ± SD.
Derivative 7c with an unsubstituted 4-benzylidene moiety showed approximately
10-fold weaker inhibition than reference inhibitor 7b. Analogues of 7b in which the 3-
Br, 4-OCH3 substitution pattern was replaced with a 3-OBn (7a) or 4-Cl (7d) group
displayed a similar drop in potency. Introduction of a bulky group at the ortho-
position of the 4-benzylidene ring (7e) completely abolished inhibitory activity, which
suggested that sterically demanding substituents in this position are not tolerated by
the enzyme. In contrast, compounds with different modifications at the 3- or 4-
position of the 2-phenyl ring (7f, 7g, and 10) maintained similar inhibitory activity as
the parent 7b. These general SAR trends are consistent with previous results from
docking experiments, which indicated that the 2-phenyl ring is oriented away from
the binding pocket towards solvent [13]. This proposed binding mode can explain the
observed tolerance towards different substituents in this position. In contrast, the
insertion of an additional CH2 group between the phenyl ring and the N2 of the
pyrazol-3-one scaffold was less well tolerated, and led to a 4-fold loss in activity (12).
Table 2
5-CF3 pyrazol-3-ones with different 2- and 4-substituents, and their
inhibitory activity against LgtC.
10

R1
O
CF3
N
N
N
N
Br
O
R2
R3
OCH3
6: R1 = -CH3
7 and 10: R1 = -CF3
12
Cmpd
6
R2
R3
IC50 (µM)a
>50
H
3-Br, 4-OCH3
3-OBn
7a
7b
7c
7d
7e
7f
H
18 ± 1
H
3-Br, 4-OCH3
H
3.1 ± 0.5
28 ± 5
H
H
4-Cl
25 ± 2
H
2-CO2CH2CH3
3-Br, 4-OCH3
3-Br, 4-OCH3
3-Br, 4-OCH3
n/a
>100
3-CO2H
3.0 ± 0.3
2.9 ± 0.2
4.3 ± 0.4
12 ± 2
7g
10
12
4-CO2H
3-C(O)NHCH2CH2OH
n/a
aLgtC was pre-incubated with inhibitor (7a, 7c-7g, 10: 0.1-100 µM; 7b, 12: 0.1-50 µM)
or DMSO, UDP-Gal (28 μM), MnCl2 (5 mM), CIP (10 U/mL), CEL (1 mg/mL), and
Triton (0.01%) for 30 mins at 30 °C in 13 mM HEPES buffer (pH 7.0). Lactose (2 mM)
was added, and the reactions were incubated for 20 mins at 30 °C. Each compound
was tested in triplicate; results are presented as average ± SD.
2.2.2 Covalent inhibition kinetics for selected pyrazol-3-ones
Although IC50 values are commonly used as a measure of potency and for SAR
studies, they are not ideal for the analysis of covalent inhibitors [18]. The interaction
between a covalent inhibitor and its target can be separated into a reversible, non-
covalent binding step, and an irreversible inactivation step (Fig. S1). The first step
can be described quantitatively by the inhibition constant Ki, and the second step by
the inactivation rate kinact. Experimental protocols have been established to
determine these kinetic parameters, and thus the relative contribution of each step to
overall inhibition [17]. To fully understand the covalent inhibition kinetics in the
11

pyrazol-3-one series, we used these protocols to determine the Ki and kinact values of
eight selected derivatives (Fig. S2), and correlate them with IC50 values (Table 3).
Within this set of 5-CF3 pyrazol-3-ones, kinact values did not vary significantly
(0.9-2.5 × 10-3 s-1). This suggests that independent of the substitution pattern at the
4-benzylidene substituent, the covalent bond formation rate is of a similar order of
magnitude for all eight derivatives. In contrast, a considerable spread of Ki values
was observed for the same derivatives (5-150 µM). This indicates that different
substituents on the 4-benzylidene moiety directly affect the reversible binding affinity
of these inhibitors for LgtC. Taken together, these results suggest that it is non-
covalent binding, not covalent inactivation, that drives LgtC inhibition in this series of
pyrazol-3-ones. This interpretation is supported by the good correlation between Ki
values and IC50 values (Table 3). The rank orders of Ki and IC50 values correspond
almost perfectly, with only a single outlier (7d). This correlation is even stronger
between IC50 values and the ratio of kinact/Ki. Thus, inhibitors with a high kinact/Ki ratio
have low IC50 values, as previously observed for other covalent inhibitors [19].
The availability of Ki and kinact values also enabled us to delineate SAR for the
different substitution patterns on the 4-benzylidene moiety (Table 3). Almost all
derivatives with one or more substituents on the 4-benzylidene moiety displayed
higher kinact/Ki values, and hence better affinity for LgtC, than the derivative 7c
bearing an unsubstituted 4-benzylidene. Regarding the effect of individual
modifications, substitution of the 3-position (7a, 7j) was generally less favourable
than substitution of the 4-position. Most notably, moving a benzyloxy substituent
from position 3 to position 4 led to a 10-fold improvement, both in terms of Ki and
IC50 value (7h vs 7a). This suggests that a large substituent in position 3 is
disadvantageous for non-covalent binding affinity, which may be indicative of a
12

possible steric clash with the enzyme in this position. In contrast, substituents at
position 4 were generally very well tolerated, with the exception of a chloro
substituent (7d).
Table 3
Kinetic parameters for covalent inhibition of LgtC by selected 5-CF3
pyrazol-3-ones.
CF3
N
N
O
R3
Cmpd R3
kinact (×10-3 s-1)a Ki (µM)a
1.8 ± 0.1 50 ± 9
3-Br, 4-OCH3 1.2 ± 0.07
kinact/Ki (M-1s-1) IC50 (µM)b
7a
7b
7c
7d
7h
7i
3-OBn
36
38 ± 13
6.4 ± 1.6
63 ± 8
7.1 ± 1.1 163
153 ± 31 13
H
2.0 ± 0.2
0.9 ± 0.1
1.1 ± 0.07
1.5 ± 0.07
1.6 ± 0.3
2.5 ± 0.2
4-Cl
49 ± 18
18
80 ± 26
3.3 ± 0.6
8.9 ± 1.8
64 ± 5
4-OBn
4-OCH3
3-Cl
4.7 ± 1.0 238
19 ± 2
80
18
7j
88 ± 33
7k
4-OH
6.8 ± 1.5 370
5.8 ± 0.5
aEach compound was tested in triplicate; results are presented as average ± SD. For
details see ESI, Fig. S2. bLgtC was pre-incubated with inhibitor (7a: 0.1-100 μM; 7b,
7h, 7i, 7k: 0.1-50 μM; 7d, 7j: 0.1-200 μM; 7c: 0.1-250 μM) or DMSO, UDP-Gal (28
μM), MnCl2 (5 mM), CIP (10 U/mL), CEL (1 mg/mL), and Triton (0.01%) for 20 mins
at 30 oC in 13 mM HEPES buffer (pH 7.0). Lactose (2 mM) was added, and the
reactions were incubated for 20 mins at 30 °C. Each compound was tested in
triplicate; results are presented as average ± SD.
2.3
Microbiological evaluation of selected inhibitors
2.3.1 Growth inhibition assay
To assess the activity of this class of LgtC inhibitors against live bacteria, we
evaluated the 5-CF3 pyrazol-3-ones 7a and 7b against non-typeable H. influenzae
(NTHi) strain R2866. For direct comparison, we also included the prototype inhibitor
13

1 [13] in these experiments. NTHi strain R2866 was first isolated by Nizet from the
blood of children with a meningitis infection [20]. Despite its lack of a capsular
polysaccharide structure, this strain has been reported to be serum-resistant at a
similar level as encapsulated type b H. influenzae [10]. The unusual serum
resistance and increased virulence of the R2866 strain has been linked to the lgtC
gene and the expression of terminal digalactoside epitopes on LOS structures in the
outer membrane [9]. LgtC inhibitors may therefore be able to reduce this increased
serum resistance and potentiate serum killing. Such a mechanism would make them
useful as potential anti-virulence agents [21].
Before putting this hypothesis to the test in serum survival experiments, we first
evaluated the effect of 1, 7a and 7b on bacterial growth. To identify the optimal time
point for growth inhibition experiments, we monitored the growth of strain R2866 by
measuring the optical density (OD) over 20 hours. A single colony of the organism
was grown in the supplemented Brain Heart Infusion broth (sBHI) at 37 °C for 19
hours, and the OD590 was recorded at various time points (Fig S3). The doubling
time of H. influenzae R2866 was 2 hours. The organism reached the stationary
phase after 10-12 hours, with a maximal OD590 of 2.6, followed by a decline in
growth.
To assess the potential toxicity of pyrazol-3-ones 1, 7a and 7b against R2866,
we used a bacterial viable count method [22] to quantify viable bacteria after inhibitor
treatment. Thus, a stationary culture of H. influenza was diluted to the required OD590
and incubated for 1 h with inhibitors at various concentrations (0-200 µM), or DMSO
as the control. Samples were centrifuged, and the cell pellet was washed twice to
remove excess inhibitors. Bacterial samples were serially diluted and plated on
14

chocolate agar plates. The cultures were incubated at 37 °C, 5% CO2 for 24 hours,
and the viable count was recorded (Fig. 3).
Fig. 3 Activity of pyrazol-3-ones 1, 7a, and 7b in the H. influenzae R2866 growth
inhibition assay.a
2.0X107
7
2.0X10
2.5X107
2.0X107
1.5X107
1.0X107
0.5X107
0
1.5X107
7
7
7
1.5X10
1.0X107
1.0X10
0.5X107
0.5X10
0
0
Control
50 µM
50 µM
100 µM
100 µM
200 µM
200 µM
Control
50 µM
100 µM
200 µM
Control
7a
1
2.5X107
2.0X107
1.5X107
1.0X107
0.5X107
0
Control
50 µM
100 µM
200 µM
7b
aConditions: H. influenzae R2866 was incubated with inhibitors 1, 7a, or 7b (0-200
µM, final percentage of DMSO: 10%) at 37 °C for 1 h. Excess inhibitor was removed
by centrifugation, and the cell pellet was washed with sBHI twice. Resuspended cell
samples were plated on agar plates and incubated at 37 oC, 5% CO2 for 24 h. Viable
count was recorded and data were analysed with GraphPad Prism v6.0. Each
concentration was tested in triplicate; results are shown as the mean. Statistical
analysis was performed by an unpaired t-test; *P < 0.05.
At concentrations up to 100 µM, none of the inhibitors had a significant effect on
bacterial growth under these conditions. This is not unexpected, as LgtC is not
essential for bacterial viability. This lack of growth inhibitory activity is, in fact,
desirable, as an anti-virulence agent is, ideally, devoid of bactericidal or
bacteriostatic activity [11]. However, in cultures treated with 1 or 7a at 200 µM we did
15

observe a significant reduction in bacterial growth (P<0.05). This suggests that at
this concentration, these pyrazol-3-ones interfere with one or more targets other than
LgtC that are critical for bacterial viability.
2.3.2 Serum survival assay
To test the hypothesis that LgtC inhibitors may be able to increase the sensitivity
of H. influenzae strain R2866 to serum, we designed a serum survival experiment.
First, bacteria in the stationary phase were exposed to various concentrations of
normal human serum (NHS) for 1 h, before plating the cultures on agar plates. The
general protocol followed that of the growth inhibition assay. As shown in Fig. 4A, 10%
(v/v) NHS had no killing effect on R2866 compared to untreated bacteria. However,
the viable count dropped significantly in samples treated with 30% and 50% NHS,
with P values less than 0.05 (30%) and 0.01 (50%), respectively. These results
showed that as expected, strain R2866 was only susceptible to serum killing at high
levels of NHS.
To assess if LgtC inhibitors may enhance the killing effect of serum, we next
determined bacterial viability in the presence of both serum (30% v/v) and inhibitors.
As our goal was to attenuate bacterial serum resistance rather than kill the bacteria
outright, we tested 1, 7a, and 7b at concentrations that did not affect bacterial growth
(1 and 7a: 100 µM; 7b: 50 µM). The overnight culture of R2866 was diluted to the
required OD590, followed by incubation with both serum and the respective inhibitor
for 1 h at 37 °C. Serum and free inhibitors were removed by centrifugation. Cell pellet
was washed twice with sBHI and plated on agar plates. After 24 h incubation, the
viable count was recorded (Fig. 4B). No significant reduction in serum resistance
16

was observed in the presence of inhibitor. This suggests that none of these three
pyrazol-3-ones potentiates serum killing of H. influenzae strain R2866.
Fig. 4 Serum survival assay in the absence (a) or presence of inhibitors (b).a
(A)
(B)
77
8
8
11.5.5XX1100
6.0X10
6.0X10
7
7
8
8
1.0X10
1.0X10
4.0X10
4.0X10
6
6
8
8
5X10
5X10
2.0X10
2.0X10
0
0
0
0
0%
0%
10%
10%
30%
30%
50%
50%
Control
Control
1
7a
YX293
7b
YX303
CSG164
Serum%(v/v)
aConditions: (A) H. influenzae R2866 was incubated with different concentrations of
serum for 1 h at 37 oC, followed by centrifugation to remove excess serum. The cell
pellet was plated on agar plates and incubated for another 24 h at 37 oC, 5% CO2.
Viable count was recorded, and data were analysed by GraphPad Prism v6.0. Each
concentration was tested in triplicate; results are shown as the mean. Statistical
analysis was performed by an unpaired t-test; *P < 0.05, **P < 0.01. (B) Bacteria
were exposed to 30% NHS and inhibitor (1: 100 µM; 7a: 100 µM; 7b: 50 µM) for 1 h
at 37 oC (Control: 30% NHS). Serum and inhibitors were removed by centrifugation.
Cell pellet was plated on agar plates, and incubated for another 24 h at 37 oC, 5%
CO2. Viable count was recorded, and data were analysed by GraphPad Prism v6.0.
Each concentration was tested in triplicate; results are shown as the mean.
3.
Conclusions
We have recently identified 5-substituted pyrazol-3-ones as novel, non-
substrate-like inhibitors of the bacterial glycosyltransferase LgtC with a covalent
mode of action [13]. In this study, we have systematically explored the effect of
different substituents in position 2, 4 and 5 of the pyrazol-3-one scaffold on inhibitory
activity. We found that an electron-withdrawing CF3 substituent in position 5 is
superior not only to the electron-donating 5-CH3 substituent of the prototype inhibitor,
but also to other electron-withdrawing groups. Focusing specifically on derivatives
17

with a 5-CF3 substituent, we carried out a detailed analysis of inhibition kinetics to
dissect the non-covalent binding and covalent inactivation steps that are
characteristic for covalent inhibitors [17]. This analysis showed that different
substitution patterns at the 4-benzylidene moiety have limited influence on the
reactivity of the electrophilic Michael acceptor warhead, but a strong effect on non-
covalent binding. In conjunction with IC50 values, these results indicate that non-
covalent binding rather than covalent inactivation is the driver for overall inhibitory
activity in this 5-CF3 pyrazol-3-one series of LgtC inhibitors. Non-catalytic cysteines
are a common motif in bacterial GTs [13]. The insights from this study may therefore
also be useful for the rational development of covalent inhibitors against other
bacterial GTs.
LgtC increases serum resistance in some H. influenzae strains and is therefore a
potential target for anti-virulence drug discovery [10]. This prompted us to evaluate
selected pyrazol-3-ones against the serum-resistant H. influenzae strain R2866.
While none of these LgtC inhibitors affected bacterial growth at concentrations up to
100 µM, neither did they enhance serum killing. A possible explanation for this
disappointing result is the phase variable expression of LgtC. Several genes related
to serum resistance in H. influenzae, including lgtC, are subject to phase variation
via slipped strand mispairing of DNA, a mechanism that allows bacterial pathogens
to rapidly and reversibly adapt to changing environments [8, 23]. The expression of
the digalactoside epitope in NTHi is controlled by tetranucleotide repeats [23] and, as
a consequence of slipped strand mispairing, is turned on or off at high frequency,
resulting in structural heterogeneity of the LOS structures [23]. Although H.
influenzae strain R2866 was, as expected, remarkably resistant to NHS in our serum
survival assay, under the current conditions of the assay, this serum resistance
18

cannot be linked directly to the presence of the digalactoside epitope, or the
expression of LgtC. The phase variable expression of LgtC under the conditions of
the serum survival assay therefore provides a likely explanation for the observed
lack of activity of our LgtC inhibitors in this assay.
As phase variation is an important adaptation mechanism of bacterial pathogens
in-vivo, the results from this study suggest that LgtC inhibitors have probably only
limited potential as anti-virulence agents. Due to their covalent mode of action, the 5-
CF3 pyrazol-3-ones described herein are, however, attractive templates for the
development of novel labeling reagents. Such labeling reagents would enable the
direct labeling of LgtC and related phase-variable GTs in bacterial cells and cell
lysates. Because of their minimal effect on bacterial growth, the 5-CF3 pyrazol-3-
ones described herein are ideally suited for such applications. The development of
such probes is currently underway [24].
4.
Experimental section
4.1
Chemistry
All chemical reagents were obtained commercially and used as received.
Microwave-assisted reactions were conducted on a Monowave 300 microwave
synthesis reactor from Anton Paar. Target compounds and synthetic intermediates
were purified by flash chromatography column and characterised by TLC, 1H-NMR,
13C-NMR, and ESI-MS. Flash chromatography columns were packed wet. Thin layer
chromatography (TLC) was performed on precoated aluminium plates (Silica Gel 60
F254, Merck). Compounds were visualized by exposure to UV light (254/365 nm).
NMR spectra were recorded on a Bruker BioSpin at 400 MHz (1H) or 100 MHz (13C).
Chemicals shifts (δ) are reported in ppm, and coupling constants (J) are reported in
19

Hz. The order of citation in parentheses is (i) multiplicity (s, singlet; d, doublet; t,
triplet; q, quartet and m, multiplet), (ii) coupling constant (J) quoted in Hertz to the
nearest 0.1 Hz, (iii) number of equivalent nuclei (by integration). Mass spectra were
recorded at the EPSRC National Mass Spectrometry Service Centre, Swansea.
Bromine-containing compounds 6, 7b, 7f, 7g, 8a-d, 10 and 12 contain the naturally
occurring isotope mixture of the two stable bromine isotopes Br-79 and Br-81.
Pyrazol-3-ones 1, 7a, and 7b were synthesised as previously reported [13].
Phenyl-3-(trifluoromethyl)-1H-pyrazol-5-ol (3a). [13] Phenylhydrazine (433 mg, 4
mmol) and ethyl 4,4,4-trifluoroacetoacetate (736 mg, 4 mmol) were dissolved in
glacial acetic acid. The reaction mixture was stirred at 110 oC until TLC
(hexane/EtOAc 1:1) showed complete consumption of the starting material. Upon
cooling a white solid precipitated from the solution and was filtered and washed with
ice-cold ethanol. Purification by flash column chromatography afforded the title
compound as a white solid (652 mg, 2.86 mmol, 72 %). 1H-NMR (400 MHz, DMSO-
d6, ppm) δ: 5.94 (s, 1H), 7.39 (t, J = 8.0 Hz, 1H), 7.52 (t, J = 8.0 Hz, 2H), 7.71 (d, J =
8.0 Hz, 2H), 12.49 (s, 1H) ppm; 13C-NMR (100 MHz, DMSO-d6, ppm) δ: 85.6, 121.3
(q, 1JCF = 267 Hz), 122.3, 127.2, 129.1, 137.7, 140.4 (d, 2JCF = 37 Hz), 153.7 ppm.
Ethyl 2-(5-hydroxy-1-phenyl-1H-pyrazol-3-yl)acetate (3b) [25]. Phenylhydrazine (2.5
g, 22.8 mmol) and diethyl 3-oxopentanedioate (4.6 g, 22.8 mmol) were dissolved in
EtOH, and the mixture was heated to reflux for 2 h. Upon completion, the reaction
mixture was cooled down, followed by evaporation to remove most of the solvent.
Purification by flash column chromatography afforded the title compound (2.3 g, 9.3
mmol, 41%) as a light yellow powder. 1H-NMR (400 MHz, CDCl3, ppm) δ: 1.32 (t, J =
20

7.1 Hz, 3H), 3.60 (s, 2H), 3.65 (s, 2H), 4.24 (q, J = 7.1 Hz, 2H), 7.21 (t, J = 7.4 Hz,
1H), 7.35-7.46 (m, 2H), 7.83-7.88 (m, 2H).
Ethyl 5-hydroxy-1-phenyl-1H-pyrazole-3-carboxylate (3c) [25]. The title compound
was obtained as a light yellow solid (6.7 g, 28.8 mmol, 61%) from phenylhydrazine
(5.4 g, 50 mmol) and diethyloxalacetate sodium salt (10 g, 48 mmol) under the
conditions described for 3b. 1H-NMR (400 MHz, DMSO-d6, ppm) δ: 1.29 (t, J = 7.1
Hz, 3H), 4.28 (t, J = 10.7 Hz, 2H), 5.96 (s, 1H), 7.37 (t, J = 7.4 Hz, 1H), 7.51 (t, J =
7.9 Hz, 2H), 7.73 (d, J = 7.5 Hz, 2H), 12.15 (s, 1H).
5-Hydroxy-1-phenyl-1H-pyrazole-3-carboxamide (3d). 3c (1.3 g, 5.6 mmol) was
added into 28% ammonium hydroxide aqueous solution and heated to reflux for 12 h.
Then the solution was neutralized with hydrochloric acid, followed by addition of cold
water. Precipitate was formed and filtered. The powder residue was further purified
by flash column chromatography to afford the title compound (431 mg, 2.1 mmol,
38%) as a grey powder. 1H-NMR (400 MHz, DMSO-d6, ppm) δ: 5.88 (s, 1H), 7.25 (s,
1H), 7.34 (t, J = 7.4 Hz, 1H), 7.44-7.56 (m, 3H), 7.78 (d, J = 7.5 Hz, 2H), 11.95 (s,
1H).
5-Hydroxy-1-phenyl-1H-pyrazole-3-carboxylic acid (3e). To a solution of 3c (928 mg,
4 mmol) in ethanol was added 1 N aqueous NaOH solution. The mixture was stirred
at room temperature for 3 h. Most of the organic solvent was evaporated, and the
residue was acidified with 1 N HCl solution. Precipitate was formed and filtered to
afford the title compound (745 mg, 3.6 mmol, 91%) as a white powder. 1H-NMR (400
21

MHz, DMSO-d6, ppm) δ: 5.93 (s, 1H), 7.36 (t, J = 7.4 Hz, 1H), 7.43-7.56 (m, 2H),
7.73 (d, J = 7.5 Hz, 2H), 12.05 (s, 1H), 12.75 (s, 1H).
4-(5-Hydroxy-3-(trifluoromethyl)-1H-pyrazol-1-yl)benzoic acid (4a). The title
compound was obtained as a white solid (1.4 g, 5.1 mmol, 61%) from 4-
hydrazinylbenzoic acid (1.3 g, 8.5 mmol) and ethyl 4,4,4-trifluoroacetoacetate (1.32
mL, 8.5 mmol) under the conditions described for 3a. 1H-NMR (400 MHz, DMSO-d6,
ppm) δ: 5.94 (s, 1H), 7.90 (d, J = 9.0 Hz, 2H), 7.99 (d, J = 9.0 Hz, 2H), 11.78 (s, 1H),
12.49 (s, 1H).
3-(5-Hydroxy-3-(trifluoromethyl)-1H-pyrazol-1-yl)benzoic acid (4b). The title
compound was obtained as a white solid (1.83 g, 6.7 mmol, 75%) from 3-
hydrazinylbenzoic acid (1.37 g, 9.0 mmol) and ethyl 4,4,4-trifluoroacetoacetate (1.32
mL, 9.0 mmol) under the conditions described for 3a. 1H-NMR (400 MHz, DMSO-d6,
ppm) δ: 5.98 (s, 1H), 7.65 (t, J = 7.9 Hz, 1H), 7.91-7.96 (m, 1H), 7.98-8.04 (m, 1H),
8.30 (t, J = 1.9 Hz, 1H), 12.74 (s, 1H), 13.29 (s, 1H).
5-Methyl-2-phenyl-2,4-dihydro-3H-pyrazol-3-one (5) [26]. The title compound was
obtained as a light yellow solid (343 mg, 1.97 mmol, 97%) from ethyl 3-oxobutanoate
(264 mg, 2.03 mmol) and phenylhydrazine (219 mg, 2.03 mmol) under the conditions
described for 3a. 1H-NMR (400 MHz, CDCl3, ppm) δ: 2.26 (s, 3H), 3.49 (s, 2H), 7.27-
7.32 (m, 1H), 7.48-7.54 (m, 2H), 7.96-8.01 (m, 2H); 13C-NMR (100 MHz, CDCl3, ppm)
δ: 16.9, 43.0, 118.8, 125.0, 128.8, 138.0, 156.4, 170.0.
22

(Z)-4-(3-Bromo-4-methoxybenzylidene)-5-methyl-2-phenyl-2,4-dihydro-3H-pyrazol-3-
one (6). 5 (174 mg, 1.0 mmol) and 3-bromo-4-methoxybenzaldehyde (322.5 mg, 1.5
mmol) were dissolved in glacial acetic acid. The reaction was heated to reflux until
TLC showed complete consumption of starting material. The product was purified by
flash column chromatography. The title compound was obtained as an orange solid
(199 mg, 0.54 mmol, 54%). 1H-NMR (400 MHz, DMSO-d6, ppm) δ: 2.32 (s, 3H), 3.98
(s, 3H), 7.20 (t, J = 7.4 Hz, 1H), 7.32 (d, J = 8.8 Hz, 1H), 7.39-7.51 (m, 2H), 7.76 (s,
1H), 7.91 (d, J = 7.6 Hz, 2H), 8.56 (dd, J = 8.8, 2.1 Hz, 1H), 9.25 (d, J = 2.1 Hz, 1H).
13C-NMR (100 MHz, DMSO-d6, ppm) δ: 13.1, 56.8, 110.7, 112.6, 118.4, 124.6, 124.9,
127.3, 128.8, 136.6, 138.0, 138.1, 146.6, 151.7, 159.2, 161.7. ESI-MS: m/z 371.03
(100%) [M+H]+; HR-MS: m/z 371.0386 (100%), 372.0419 (20%), 373.0364 (100%),
374.0396 (20%) [M+H]+, [C18H16BrN2O2]+ calcd for 371.0390, 372.0423, 373.0369,
374.0403.
(Z)-4-Benzylidene-2-phenyl-5-(trifluoromethyl)-2,4-dihydro-3H-pyrazol-3-one (7c)
[15]. 3a (141 mg, 0.5 mmol) and benzaldehyde (106 mg, 0.75 mmol) were placed in
a microwave-proof glass tube and heated for 15 mins at 160 oC in a commercial
microwave apparatus. The reaction was cooled to room temperature. The reaction
product was precipitated by addition of ethyl acetate and hexane, collected by
filtration, and recrystallized from hexane and ethyl acetate. The title compound was
obtained as an orange solid (93 mg, 0.30 mmol, 59%). 1H-NMR (400 MHz, CDCl3,
ppm) δ: 7.28-7.34 (m, 1H), 7.44-7.51 (m, 2H), 7.57 (t, J = 7.5 Hz, 2H), 7.65 (t, J = 7.4
Hz, 1H), 7.80 (s, 1H), 7.94 (dd, J = 8.7, 1.1 Hz, 2H), 8.55 (d, J = 7.3 Hz, 2H). 13C-
NMR (100 MHz, CDCl3, ppm) δ: 119.8 (q, 1JCF = 270 Hz), 120.0, 121.5, 126.3, 128.9,
129.0, 132.5, 134.5, 134.6, 137.5, 140.7 (q, 2JCF = 37 Hz), 150.6, 161.1. ESI-MS:
23

m/z 317.1 (100%) [M+H]+; HR-MS: m/z 317.0900 [M+H]+, [C17H12F3N2O]+ calcd for
317.0896.
(Z)-4-(4-Chlorobenzylidene)-2-phenyl-5-(trifluoromethyl)-2,4-dihydro-3H-pyrazol-3-
one (7d) [15]. The title compound was obtained as an orange solid (35 mg, 0.1 mmol,
20%) from 3a (114 mg, 0.5 mmol) and 4-chlorobenzaldehyde (106 mg, 0.75 mmol)
under the conditions described for 7c. 1H-NMR (400 MHz, CDCl3, ppm) δ: 7.31 (t, J
= 7.4 Hz, 1H), 7.45-7.51 (m, 2H), 7.54 (d, J = 8.7 Hz, 2H), 7.91 (d, J = 7.6 Hz, 2H),
7.72 (s, 1H), 8.52 (d, J = 8.6 Hz, 2H). 13C-NMR (100 MHz, CDCl3, ppm) δ: 119.7 (q,
1JCF = 270 Hz), 120.0, 121.8, 126.4, 129.0, 129.4, 130.9, 135.8, 137.3, 140.6 (d, 2JCF
= 39 Hz), 141.3, 148.8, 161.1. ESI-MS: m/z 351.1 (100 %) [M+H]+; 383.1 (95%)
[M+MeOH+H]+; 405.1 (55%) [M+MeOH+Na]+; HR-MS: m/z 351.0506 [M+H]+,
[C17H11ClF3N2O]+ calcd for 351.0507.
Ethyl (Z)-2-((5-oxo-1-phenyl-3-(trifluoromethyl)-1,5-dihydro-4H-pyrazol-4-
ylidene)methyl)-benzoate (7e). The title compound was obtained as an orange solid
(15 mg, 0.04 mmol, 15%) from 3a (54 mg, 0.25 mmol) and ethyl 2-formylbenzoate
(66.8 mg, 0.38 mmol) under the conditions described for 7c. 1H-NMR (400 MHz,
CDCl3, ppm) δ: 1.42 (t, J = 7.1 Hz, 3H), 4.42 (q, J = 7.1 Hz, 2H), 7.25 (d, J = 7.4 Hz,
1H), 7.43 (t, J = 8.0 Hz, 2H), 7.60-7.69 (m, 2H), 7.86 (d, J = 7.7 Hz, 2H), 8.04 (d, J =
7.1 Hz, 1H), 8.19 (d, J = 7.5 Hz, 1H), 8.62 (s, 1H). 13C-NMR (100 MHz, CDCl3, ppm)
δ: 14.2, 61.9, 119.6, 119.7 (q, 1JCF = 270 Hz), 121.2, 126.2, 129.0, 129.9, 131.0,
131.2, 131.7, 132.0, 132.9, 137.3, 140.5 (d, 2JCF = 37 Hz), 151.5, 160.9, 166.1. ESI-
MS: m/z 389.1 (85%) [M+H]+, 421.1 (100%) [M+MeOH+H]+, 443.1 (56%)
[M+MeOH+Na]+; HR-MS: m/z 389.1110 [M+H]+, [C20H16F3N2O3]+ calcd for 389.1108.
24

(Z)-3-(4-(3-Bromo-4-methoxybenzylidene)-5-oxo-3-(trifluoromethyl)-4,5-dihydro-1H-
pyrazol-1-yl)benzoic acid (7f). The title compound was obtained as an orange solid
(118 mg, 0.25 mmol, 50%) from 4b (114 mg, 0.5 mmol) and 3-bromo-4-
methoxybenzaldehyde (161 mg, 0.75 mmol) under the conditions described for 7c.
1H-NMR (400 MHz, DMSO-d6, ppm) δ: 4.03 (s, 3H), 7.39 (d, J = 8.9 Hz, 1H), 7.65 (t,
J = 8.0 Hz, 1H), 7.88 (d, J = 7.9 Hz, 1H), 7.94 (s, 1H), 8.10 (d, J = 8.1 Hz, 1H), 8.44
(t, J = 1.8 Hz, 1H), 8.71 (dd, J = 8.9, 2.1 Hz, 1H), 9.31 (d, J = 2.1 Hz, 1H), 13.19 (s,
1H). ESI-MS: m/z 468.98 (5%) [M+H]+, 499.01 (100%) [M+MeOH+H]+; HR-MS: m/z
467.0015 (100%), 467.9850 (30%), 468.9824 (50%), 469.9879 (5%) [M-H]-,
[C19H11BrF3N2O4]- calcd for 466.9860, 467.9891, 468.9840, 469.9872.
(Z)-4-(4-(3-Bromo-4-methoxybenzylidene)-5-oxo-3-(trifluoromethyl)-4,5-dihydro-1H-
pyrazol-1-yl)benzoic acid (7g) [27]. The title compound was obtained as an orange
solid (66 mg, 0.14 mmol, 28%) from 4a (114 mg, 0.5 mmol) and 3-bromo-4-
methoxybenzaldehyde (161 mg, 0.75 mmol) under the conditions described for 7c.
1H-NMR (400 MHz, DMSO-d6, ppm) δ: 4.02 (s, 3H), 7.35 (d, J = 8.9 Hz, 1H), 7.90 (s,
1H), 8.00 (d, J = 8.9 Hz, 2H), 8.06 (d, J = 8.9 Hz, 2H), 8.64 (d, J = 8.9 Hz, 1H), 9.28
(s, 1H). ESI-MS: m/z 467.0 (100 %) [M-H]-; HR-MS: m/z 466.9854 (100%), 467.9889
(15%), 468.9833 (100%), 469.9867 (15%) [M-H]-, [C19H11BrF3N2O4]- calcd for
466.9860, 467.9893, 468.9839, 469.9873.
(Z)-4-(4-(Benzyloxy)benzylidene)-2-phenyl-5-(trifluoromethyl)-2,4-dihydro-3H-
pyrazol-3-one (7h). The title compound was obtained as an orange solid (31 mg,
0.07 mmol, 30%) from 3a (57 mg, 0.25 mmol) and 4-(benzyloxy)benzaldehyde (64
25

mg, 0.3 mmol) under the conditions described for 7c. 1H-NMR (400 MHz, DMSO-d6,
ppm) δ: 5.31 (s, 2H), 7.27 (d, J = 9.0 Hz, 2H), 7.33 (t, J = 7.4 Hz, 1H), 7.38 (d, J =
7.1 Hz, 1H), 7.43 (t, J = 7.2 Hz, 2H), 7.51 (t, J = 7.7 Hz, 4H), 7.84 (d, J = 7.8 Hz, 2H),
7.91 (s, 1H), 8.81 (d, J = 9.0 Hz, 2H). 13C-NMR (100 MHz, DMSO-d6, ppm) δ: 70.0,
115.4, 117.56, 119.8 (q, 1JCF = 270 Hz), 120.0, 125.9, 126.2, 128.0, 128.2, 128.5,
129.0, 136.0, 137.3, 138.2, 139.9 (d, 2JCF = 36 Hz), 150.0, 161.1, 164.3. ESI-MS:
m/z 423.1 (100 %) [M+H]+; HR-MS: m/z 423.1312 [M+H]+, [C24H18F3N2O2]+ calcd for
423.1315.
(Z)-4-(4-Methoxybenzylidene)-2-phenyl-5-(trifluoromethyl)-2,4-dihydro-3H-pyrazol-3-
one (7i) [28]. The title compound was obtained as an orange solid (68 mg, 0.20
mmol, 39%) from 3a (114 mg, 0.5 mmol) and 4-methoxybenzaldehyde (102 mg, 0.75
mmol) under the conditions described for 7c. 1H-NMR (400 MHz, CDCl3, ppm) δ:
3.96 (s, 3H), 7.06 (d, J = 9.0 Hz, 2H), 7.29 (t, J = 7.4 Hz, 1H), 7.44-7.51 (m, 2H),
7.72 (s, 1H), 7.95 (dd, J = 8.7, 1.1 Hz, 2H), 8.67 (d, J = 9.0 Hz, 2H). 13C-NMR (100
MHz, CDCl3, ppm) δ: 55.82, 114.65, 118.50, 119.9 (q, 1JCF = 270 Hz), 120.04,
126.11, 126.14, 128.96, 137.73, 138.05, 140.9 (d, 2JCF = 36 Hz), 150.05, 161.65,
165.25. ESI-MS: m/z 347.1 (100 %) [M+H]+; 379.1 (10%) [M+MeOH+H]+; 401.1 (20%)
[M+MeOH+Na]+; HR-MS: m/z 347.1002 [M+H]+, [C18H14F3N2O2]+ calcd for 347.1002.
(Z)-4-(3-Chlorobenzylidene)-2-phenyl-5-(trifluoromethyl)-2,4-dihydro-3H-pyrazol-3-
one (7j) [28]. The title compound was obtained as an orange solid (74 mg, 0.21
mmol, 42%) from 3a (114 mg, 0.5 mmol) and 3-chlorobenzaldehyde (105 mg, 0.75
mmol) under the conditions described for 7c. 1H-NMR (400 MHz, CDCl3, ppm) δ:
7.31 (t, J = 7.4 Hz, 1H), 7.49 (m, 3H), 7.58-7.64 (m, 1H), 7.70 (s, 1H), 7.91 (d, J =
26

8.6 Hz, 2H), 8.38 (d, J = 7.8 Hz, 1H), 8.61 (t, J = 1.8 Hz, 1H). 13C-NMR (100 MHz,
CDCl3, ppm) δ: 119.7 (q, 1JCF = 270 Hz), 120.0, 122.7, 126.5, 129.1, 130.1, 132.4,
133.6, 133.8, 134.1, 135.0, 137.3, 140.4 (d, 2JCF = 36 Hz), 148.5, 160.8. ESI-MS:
m/z 351.1 (80 %) [M+H]+; 383.1 (100%) [M+MeOH+H]+; 405.1 (45%)
[M+MeOH+Na]+; HR-MS: m/z 351.0506 [M+H]+, [C17H11ClF3N2O]+ calcd for
351.0507.
(Z)-4-(4-Hydroxybenzylidene)-2-phenyl-5-(trifluoromethyl)-2,4-dihydro-3H-pyrazol-3-
one (7k). The title compound was obtained as an orange solid (60 mg, 0.18 mmol,
35%) from 3a (114 mg, 0.5 mmol) and 4-hydroxybenzaldehyde (92 mg, 0.75 mmol)
under the conditions described for 7c. 1H-NMR (400 MHz, DMSO-d6, ppm) δ: 6.99 (d,
J = 8.9 Hz, 2H), 7.31 (t, J = 7.4 Hz, 1H), 7.42-7.55 (m, 2H), 7.79-7.89 (m, 3H), 8.74
(d, J = 8.9 Hz, 2H), 11.34 (s, 1H). 13C-NMR (100 MHz, DMSO-d6, ppm) δ: 116.10,
116.27, 119.87, 119.9 (q, 1JCF = 270 Hz), 124.56, 126.07, 128.98, 137.40, 138.95,
140.0 (d, 2JCF = 36 Hz), 150.34, 161.18, 165.08. ESI-MS: m/z 333.1 (100 %) [M+H]+;
355.1 (20%) [M+Na]+; 387.1 (25%) [M+MeOH+Na]+; HR-MS: m/z 333.0847 [M+H]+,
[C17H12F3N2O2]+ calcd for 333.0845.
Ethyl (Z)-4-(3-bromo-4-methoxybenzylidene)-5-oxo-1-phenyl-4,5-dihydro-1H-
pyrazole-3-carboxylate (8a). The title compound was obtained as an orange solid
(109 mg, 0.26 mmol, 51%) from 3c (116 mg, 0.5 mmol) and 3-bromo-4-
methoxybenzaldehyde (161 mg, 0.75 mmol) under the conditions described for 7c.
1H-NMR (400 MHz, CDCl3, ppm) δ: 1.49 (t, J = 7.1 Hz, 3H), 4.03 (s, 3H), 4.50 (q, J =
7.1 Hz, 2H), 7.03 (d, J = 8.8 Hz, 1H), 7.30 (t, J = 7.4 Hz, 1H), 7.44-7.50 (m, 2H), 7.94
(d, J = 7.5 Hz, 2H), 8.65 (s, 1H), 8.69 (dd, J = 8.8, 2.2 Hz, 1H), 8.91 (d, J = 2.2 Hz,
27

1H). 13C-NMR (100 MHz, CDCl3, ppm) δ: 14.3, 56.7, 61.9, 111.5, 112.0, 120.8, 122.1,
126.4, 127.7, 128.9, 136.6, 137.6, 139.9, 140.1, 151.2, 160.4, 161.2, 162.1. ESI-MS:
m/z 429.0 (100%) [M+H]+, 451.0 (25%) [M+Na]+, 483.0 (35%) [M+MeOH+Na]+; HR-
MS: m/z 429.0444 (100%), 430.0478 (20%), 431.0423 (100%), 432.0456 (20%)
[M+H]+, [C20H18BrN2O4]+ calcd for 429.0444, 430.0478, 431.0424, 432.0458.
Ethyl (Z)-2-(4-(3-bromo-4-methoxybenzylidene)-5-oxo-1-phenyl-4,5-dihydro-1H-
pyrazol-3-yl)acetate (8b). The title compound was obtained as an orange solid (93
mg, 0.21 mmol, 53%) from 3b (123 mg, 0.5 mmol) and 3-bromo-4-
methoxybenzaldehyde (161 mg, 0.75 mmol) under the conditions described for 7c.
1H-NMR (400 MHz, CDCl3, ppm) δ: 1.31 (t, J = 7.1 Hz, 3H), 3.78 (s, 2H), 4.03 (s, 3H),
4.25 (q, J = 7.1 Hz, 2H), 7.03 (d, J = 9.3 Hz, 1H), 7.23 (t, J = 7.4 Hz, 1H), 7.38 (s,
1H), 7.44 (t, J = 8.0 Hz, 2H), 7.97 (d, J = 8.5 Hz, 2H), 8.72-8.77 (m, 2H). 13C-NMR
(100 MHz, CDCl3, ppm) δ: 14.2, 34.6, 56.6, 61.8, 111.5, 111.9, 119.4, 124.9, 125.2,
127.3, 128.8, 135.6, 138.2, 139.3, 146.1, 147.2, 159.9, 162.0, 168.8. ESI-MS: m/z
443.1 (100%) [M+H]+, 475.1 (45%) [M+Na]+, 479.1 (15%) [M+MeOH+Na]+; HR-MS:
m/z 443.0598 (100%), 444.0633 (20%), 445.0576 (100%), 446.0610 (20%) [M+H]+,
[C21H20BrN2O4]+ calcd for 443.0601, 444.0633, 445.0581, 446.0614.
(Z)-4-(3-Bromo-4-methoxybenzylidene)-5-oxo-1-phenyl-4,5-dihydro-1H-pyrazole-3-
carboxylic acid (8c). The title compound was obtained as an orange solid (160 mg,
0.40 mmol, 80%) from 3f (103 mg, 0.5 mmol) and 3-bromo-4-methoxybenzaldehyde
(161 mg, 0.75 mmol) under the conditions described for 7c. 1H-NMR (400 MHz,
DMSO-d6, ppm) δ: 3.99 (s, 3H), 7.31 (t, J = 8.0 Hz, 2H), 7.50 (t, J = 8.0 Hz, 2H), 7.87
(d, J = 7.6 Hz, 2H), 8.39 (dd, J = 8.9, 2.1 Hz, 1H), 8.57 (s, 1H), 9.25 (d, J = 2.1 Hz,
28

1H), 13.79 (s, 1H). 13C-NMR (100 MHz, DMSO-d6, ppm) δ: 57.0, 110.9, 112.6, 119.9,
121.8, 126.0, 127.1, 129.0, 137.5, 137.6, 138.1, 141.1, 150.5, 159.9, 161.7, 162.0.
ESI-MS: m/z 401.0 (100%) [M+H]+; HR-MS: m/z 401.0132 (100%), 402.0166 (20%),
403.0110 (100%), 404.0144 (20%) [M+H]+, [C18H14BrN2O4]+ calcd for 401.0131,
402.0165, 403.0111, 404.0145.
(Z)-4-(3-Bromo-4-methoxybenzylidene)-5-oxo-1-phenyl-4,5-dihydro-1H-pyrazole-3-
carboxamide (8d). The title compound was obtained as an orange solid (60 mg, 0.15
mmol, 30%) from 3d (102 mg, 0.5 mmol) and 3-bromo-4-methoxybenzaldehyde (161
mg, 0.75 mmol) under the conditions described for 7c. 1H-NMR (400 MHz, DMSO-d6,
ppm) δ: 4.01 (s, 3H), 7.27-7.35 (m, 2H), 7.46-7.53 (m, 2H), 7.77 (s, 1H), 8.01 (d, J =
7.6 Hz, 2H), 8.17 (s, 1H), 8.38 (dd, J = 8.9, 2.1 Hz, 1H), 8.74 (s, 1H), 9.31 (d, J = 2.1
Hz, 1H). 13C-NMR (100 MHz, DMSO-d6, ppm) δ: 57.0, 110.9, 112.7, 119.5, 122.0,
125.6, 127.2, 128.9, 137.6, 137.7, 138.0, 142.8, 151.0, 159.8, 161.8, 162.6. ESI-MS:
m/z 400.0 (100%) [M+H]+; HR-MS: m/z 400.0293 (100%), 401.0326 (20%), 402.0273
(100%), 403.0306 (20%) [M+H]+, [C18H15BrN3O3]+ calcd for 400.0291, 401.0325,
402.0271, 403.0305.
3-(5-Hydroxy-3-(trifluoromethyl)-1H-pyrazol-1-yl)-N-(2-hydroxyethyl)benzamide (9).
To a solution of 4b (272 mg, 1 mmol) in DMF, HBTU (758 mg, 2.0 mmol), DIEA (387
mg, 3.0 mmol), and 2-aminoethan-1-ol (92 mg, 1.5 mmol) were added. The reaction
mixture was stirred at rt for 16 h. Most of the DMF was then evaporated. The residue
was diluted with water and extracted with ethyl acetate for 3 times. The combined
organic layer was dried over Na2SO4. The crude product was purified by flash
chromatograph to afford the title compound (235 mg, 0.76 mmol, 75%). 1H-NMR
29