Tetralone derivatives are MIF tautomerase inhibitors and attenuate macrophage activation and amplify the hypothermic response in endotoxemic mice

Article abstract of DOI:10.1080/14756366.2021.1916010

Macrophage migration inhibitory factor (MIF) is a pro-inflammatory cytokine playing crucial role in immunity. MIF exerts a unique tautomerase enzymatic activity that has relevance concerning its multiple functions and its small molecule inhibitors have been proven to block its pro-inflammatory effects. Here we demonstrate that some of the E-2-arylmethylene-1-tetralones and their heteroanalogues efficiently bind to MIF’s active site and inhibit MIF tautomeric (enolase, ketolase activity) functions. A small set of the synthesised derivatives, namely compounds (4), (23), (24), (26) and (32), reduced inflammatory macrophage activation. Two of the selected compounds (24) and (26), however, markedly inhibited ROS and nitrite production, NF-κB activation, TNF-α, IL-6 and CCL-2 cytokine expression. Pre-treatment of mice with compound (24) exaggerated the hypothermic response to high dose of bacterial endotoxin. Our experiments suggest that tetralones and their derivatives inhibit MIF’s tautomeric functions and regulate macrophage activation and thermal changes in severe forms of systemic inflammation.

Full text of DOI:10.1080/14756366.2021.1916010

Journal of Enzyme Inhibition and Medicinal Chemistry
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/ienz20
Tetralone derivatives are MIF tautomerase
inhibitors and attenuate macrophage activation
and amplify the hypothermic response in
endotoxemic mice
János Garai, Marcell Krekó, László Őrfi, Péter Balázs Jakus, Zoltán Rumbus,
Patrik Kéringer, András Garami, Eszter Vámos, Dominika Kovács, Viola
Bagóné Vántus, Balázs Radnai & Tamás Lóránd
To cite this article: János Garai, Marcell Krekó, László Őrfi, Péter Balázs Jakus, Zoltán
Rumbus, Patrik Kéringer, András Garami, Eszter Vámos, Dominika Kovács, Viola Bagóné
Vántus, Balázs Radnai & Tamás Lóránd (2021) Tetralone derivatives are MIF tautomerase
inhibitors and attenuate macrophage activation and amplify the hypothermic response in
endotoxemic mice, Journal of Enzyme Inhibition and Medicinal Chemistry, 36:1, 1357-1369, DOI:
10.1080/14756366.2021.1916010
To link to this article: https://doi.org/10.1080/14756366.2021.1916010
© 2021 The Author(s). Published by Informa
UK Limited, trading as Taylor & Francis
Group.
View supplementary material
Submit your article to this journal
View related articles
Published online: 06 Jul 2021.
Article views: 266
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ienz20

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
2021, VOL. 36, NO. 1, 1357–1369
https://doi.org/10.1080/14756366.2021.1916010
RESEARCH PAPER
Tetralone derivatives are MIF tautomerase inhibitors and attenuate macrophage
activation and amplify the hypothermic response in endotoxemic mice
a
b
c
d
d
}
ꢀ
ꢀb
, Marcell Kreko
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Janos Garai
, Laszlo Orfi
, Peter Balazs Jakus
, Zoltan Rumbus
, Patrik Keringer
,
d
c
c
c
c
ꢀ
ꢀ
ꢀ
ꢀ ꢀ ꢀ
, Viola Bagone Vantus
ꢀ
Andras Garami
, Eszter Vamos
, Dominika Kovacs
, Balazs Radnai
and
c
ꢀ ꢀ
Tamas Lorand
a
b
ꢀ
ꢀ
Department of Pathophysiology, Institute for Translational Medicine, University of Pecs, Medical School, Pecs, Hungary; Department of
c
ꢀ
Pharmaceutical Chemistry, Semmelweis University, Budapest, Hungary; Department of Biochemistry and Medical Chemistry, University of Pecs,
Medical School, Pecs, Hungary; Department of Thermophysiology, Institute for Translational Medicine, University of Pecs, Medical School,
Pecs, Hungary
d
ꢀ
ꢀ
ꢀ
ABSTRACT
ARTICLE HISTORY
Received 1 February 2021
Revised 30 March 2021
Accepted 6 April 2021
Macrophage migration inhibitory factor (MIF) is a pro-inflammatory cytokine playing crucial role in immun-
ity. MIF exerts a unique tautomerase enzymatic activity that has relevance concerning its multiple func-
tions and its small molecule inhibitors have been proven to block its pro-inflammatory effects. Here we
demonstrate that some of the E-2-arylmethylene-1-tetralones and their heteroanalogues efficiently bind to
MIF’s active site and inhibit MIF tautomeric (enolase, ketolase activity) functions. A small set of the syn-
thesised derivatives, namely compounds (4), (23), (24), (26) and (32), reduced inflammatory macrophage
activation. Two of the selected compounds (24) and (26), however, markedly inhibited ROS and nitrite
production, NF-jB activation, TNF-a, IL-6 and CCL-2 cytokine expression. Pre-treatment of mice with com-
pound (24) exaggerated the hypothermic response to high dose of bacterial endotoxin. Our experiments
suggest that tetralones and their derivatives inhibit MIF’s tautomeric functions and regulate macrophage
activation and thermal changes in severe forms of systemic inflammation.
KEYWORDS
MIF; MIF inhibitors; 2-
arylmethylene-1-tetralones;
tautomerase; inflammation
Introduction
catalytic site is located between each of two adjacent monomers.
Acidic pKa of the N-terminal proline of each MIF monomer is
thought to play a crucial role in the keto-enol tautomerisation reac-
tion7. A direct link between cytokine activity and tautomerase cata-
lytic site of MIF has been reported14, although a “true” endogenous
small molecule ligand has yet to be found. Blocking of endogenous
MIF by a small molecule such as “ISO-1” and neutralisation of MIF by
anti-MIF antibodies or by plant-derived MIF inhibitors reduces the
manifestations of inflammatory conditions such as type II collagen-
induced arthritis, immunologically induced kidney disease, experi-
mental autoimmune encephalomyelitis, experimental allergic neuritis,
immunoinflammatory diabetes, experimental autoimmune myocardi-
tis, irradiation-induced acute pneumonitis, sepsis, and ischemia–re-
perfusion injury15–18. Therefore, inhibitors of MIF tautomerase hold
promise for prospective clinical use in many pathologic conditions19.
Development of individualised therapy targeting MIF in these condi-
tions is expected based on the human genetic data supporting the
role of high-expression MIF alleles in the clinical severity and end-
organ complications of a number of inflammatory disorders20.
Macrophage migration inhibitory factor (MIF) was the first repre-
sentative of those polypeptide immune mediators that have been
grouped later as “cytokines”1,2. Since its description 50 years ago
MIF has accumulated a bewildering variety of immune and non-
immune functions3. MIF has been also considered to be a missing
link between inflammation and tumorigenesis4.
MIF structure shares no homology with other known cytokines.
Its structural relatives are the mammalian enzyme, D-dopachrome-
tautomerase (DDT)5, and the prokaryotic enzymes: chorismate
mutase, 5-carboxymethyl-2-hydroxymuconate-isomerase (CHMI),
trans- and cis-3-chloroacrylic acid dehalogenase (CaaD and cis-
CaaD, respectively), and 4-oxalocrotonate-tautomerase (4-OT)6,7
.
MIF exists in a homotrimeric form of 12.4 kD monomers. In each
monomer, two a-helices are packed against a four-stranded
b-sheet8. There are reports, however, of the occurrence of a
dimeric form of MIF9 as well as of other oligomerisation states10
and also of heteromers11.
More than twenty years ago Rorsman et al. reported that recom-
binant MIF catalyses the tautomerisation of the non-naturally occur-
ring D-dopachrome, transforming the coloured compound to the
colourless dihydroindole carboxylic acid (DHICA)12, however, L-dopa-
chrome methyl ester (a melanin precursor) have also been found a
suitable substrate. This was soon followed by the identification of
phenylpyruvate and p-hydroxyphenylpyruvate as alternate substrates
Molecular modelling techniques have been used lately to find
“designer” inhibitors of promising potential21–23. The reported MIF
enzyme inhibitors include long-chain fatty acids, E-2-fluoro-p-
hydroxycinnamate, dopachrome analogues, tryptophan and tyro-
sine derivatives24, or coumarin- and chromenone derivatives19,25
.
A large variety of natural compounds show inhibitory effect of
the MIF enzyme including e.g. ellagic acid26. According to previ-
for the tautomeric activity of MIF13. In the homotrimeric MIF, the ous results, caffeic acid proved to be an efficient inhibitor of MIF
ꢀ
ꢀ ꢀ
Department of Biochemistry and Medical Chemistry, University of Pecs, Medical School, Pecs, Hungary.
CONTACT Balazs Radnai
balazs.radnai@aok.pte.hu
Supplemental data for this article can be accessed here.
ß 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.

1358
J. GARAI ET AL.
and catalyst) of the synthetic method were a little bit different. See
the detailed methods below. (Their structure is depicted in
Supplementary material’s Figure S1 and in Table 1; the synthesis
method is in Ref. [34].) Scheme 1 describes the general synthetic
method. The preparation of the E-2-phenylmethylene-1-indanone
(1), E-2-phenylmethylene-1-benzosuberone (3) and E-2-arylmethy-
lene-1-tetralones (2, 4–26) was carried out at room temperature in
ethanol. The synthesis of the E-3-arylmethylenechroman-4-ones and
E-3-arylmethylene-1-thiochroman-4-ones (27–36) was performed at
140 ꢀC with piperidine as a catalyst without solvent. The 4-thiochro-
manone-1-oxide (36) and ꢁ1,1-dioxide (37) was obtained according
to literature methods35. All of the compounds were purified by
recrystallisation from methanol and with column chromatography.
Their structural characterisation is based on FT-IR methods and pre-
viously published NMR data36–53. Thus, all of the test compounds
studied here have been published elsewhere and characterised, see
the references above. All of the compounds were proved as E-iso-
mer based on the NMR measurements54.
Figure 1. General formula of the test compounds (1–38).
tautomerase when using either dopachrome27 or phenylpyru-
vate28 as the substrate. Phenylpyruvate tautomerase was
also inhibited by curcumin28, a substance known long for its anti--
inflammatory and cancer chemopreventive properties – a constitu-
ent of the spice turmeric29. Therefore, the selection of our
potential MIF inhibitors was based on the unsaturated ketone –
structural motif – found in several natural or synthetic MIF inhibi-
tors such as coumarines and chromenones and cinnamates such
as caffeic acid, curcumin, and N-acetyl-p-benzoquinone19. In add-
ition, according to our previous studies, some tetralone derivatives
proved to be efficient MIF inhibitors30.
Based on these findings, we have focused on the group of cyc-
lic a,b-unsaturated ketones to study their inhibitory effects
towards the MIF tautomerase activity. To find compounds with
good inhibitory effect, we have previously selected some different
model compounds as 2-arylidenecycloalkanones, 2,5-diarylidene-
cyclopentanone and few 2-arylidene-1-tetralones30. The enol-keto
tautomeric conversion of phenylpyruvate (ketonase reaction) was
investigated by spectrophotometric method. From the substances
investigated, several proved to be good inhibitors of the tauto-
merase activity of the MIF enzyme.
This paper deals with a large molecular library of E-2-aryl-
methylene-1-tetralones and their hetero-analogues related to nat-
urally occurring flavonoids. Some representatives of this latter
group have previously been proven as potent MIFtautomerase
inhibitors28,31. At this end, potential analogies between the bio-
logical roles of flavonoids and melanins (or precursors thereof)
deserve particular attention32 concerning possible implication of
MIF in (neuro-) melanogenesis33.
The molecule library we have used here shows large structural
diversity. We have chosen cyclic ketones of different size and with
different heteroatoms as skeletal compounds, and applied various
aryl side chains to find an optimal lead compound. We have var-
ied the size of the ketone ring, the aromatic substituent C-ring
(with electron withdrawing and electron donating groups), and
also changed the type of the B-ring (see Figure 1 and Table 1).
Through in silico methods, we aimed to identify the key interac-
tions required for MIF inhibitory activity. In addition, we tested
the biological activity of the best tautomerase inhibitors from the
molecular library in lipopolysaccharide-(LPS)-induced macrophages
and also in LPS-treated mice.
Tautomerase assay
The enol-keto tautomeric conversion of phenylpyruvate (ketonase
reaction) was assessed at room temperature by monitoring the
decrease in absorbance at 288 nm on a dual path Shimadzu 2100
UV spectrophotometer (Shimadzu, Kyoto, Japan) according to
Taylor et al.55 with minor modifications. Briefly, the 0.72ml reaction
mixture consisted of the enzyme at a final concentration of 0.4 mg/
ml and the substrate in a 50 mM sodium phosphate buffer (pH ¼
6.5). The phenylpyruvate substrate was dissolved fresh in absolute
ethanol to yield a final concentration of 100mM. The reaction was
monitored for 75s at room temperature. A recombinant human
MIF (from ATGen Co. Ltd., Seoul, Republic of Korea) was used as
the enzyme source. Inhibitors were dissolved fresh in ethanol or
DMSO. Ethanol or DMSO did not affect the enzyme reaction at the
amounts applied. The compounds were diluted to achieve final
concentrations of 200, 100, 50, 20, 5, 1, and 0.5mM in the reaction
mixture. All data presented have been derived from consonant
measurements repeated at least three times. As reference inhibitor
caffeic acid (ketonase and enolase IC50: 0.5 and 2.0 mM respectively)
was used28. The areas under the curves obtained with and without
the inhibitors were used to compare the conversion rates of the
substrate. The IC50 value – the concentration of the given com-
pound required to achieve 50% inhibition56 –was obtained by a
nonlinear curve fitting (Sigma Plot 2000, SPSS Inc., Chicago, IL, USA)
on the data obtained with the different concentrations of the
inhibitor using the equation: f ¼ minþ(max ꢁ min)/(1þ(x/EC50)nHill).
Molecular modelling
Molecular modelling was applied as a means of gathering informa-
tion on the structure’s binding mode. Glide docking was conducted
on various MIF-ligand crystallographic structures in Extra Precision
mode without constraints. Dockings with core restrictions using
two types of docking output structures were examined to assess
the favoured binding mode. Covalent docking was used to visualise
the hypothesised adduct formation with Pro1. Molecular Dynamics
simulations were used to assess the stability of the reversible com-
plexes and explore the nature of the binding site interactions.
The crystallographic structure of MIF in complex with an inhibitor
was obtained from Protein Data Bank (PDB 6B1K). All calculations
Material and methods
Reagents
All reagents were purchased from Sigma-Aldrich (St. Louis, MO)
unless otherwise indicated. A molecular library of 38 structurally
related known compounds has been studied. Based on their struc-
tural features, this library could be further divided into two sub-
groups (Table 1). The group I consists of a 1-indanone derivative
(1), substituted arylmethylene–tetralones (2, 4–18), heteroaryl–me-
thylene–tetralones (19–26) and a benzosuberone derivative (3).
Group II contains chromanones and their thioanalogues (27–38). All
€
were carried out with the modules of Schrodinger Suites 2019–2
€
of the compounds investigated here have been prepared by base (Schrodinger, LLC, New York, NY) in Maestro. The protein was pre-
catalysed aldol condensation, the conditions (temperature, solvent pared by adding hydrogens and missing side chains. Water

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
1359
Table 1. Inhibitor activity of the test compounds (1–38) in the enol-keto and keto-enol tautomeric conversion of phenylpyruvate.
Inhibition of
Inhibition of
Compounds
X
Ar
ketonase (IC50, lM)
enolase (IC50, lM)
1
2
3
4
5
6
7
8
–
CH2
(CH2)2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
O
Phenyl
Phenyl
Phenyl
82.8 16.5
127 20
20.2 3.9
113 16
129 19
2290 343
33.4 8.8
42.8 3.1
458 73
232 54
3350 424
42.1 4.3
35.5 5.9
91.5 9.0
301 42
37.7 16.3
495 101
43.2 12.4
302 63
213 31
25.4 4.7
52.3 7.6
125 39
27.8 5.6
2.89 0.75
28.6 7.4
28.6 6.0
209 77
1740 483
3620 829
443 112
425 70
205 32
1260 159
84.0 15.3
16.5 3.4
57.0 14.1
59.1 9.7
2640 406
52.2 11.5
7340 1024
72.4 3.4
58.9 10.3
2620 696
16.9 4.2
187 32
40- CH3-phenyl
40-OCH3-phenyl
40-Cl-phenyl
20-Cl-phenyl
40-Br-phenyl
9
30-Br-phenyl
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
40-F-phenyl
20,60-(Cl)2-phenyl
30,40-(Cl)2-phenyl
20,40-(Cl)2-phenyl
40-COOH-phenyl
40-(NCH3)2--phenyl
40-CN-phenyl
30,40-(OCH3)2 -phenyl
30,40-(OCH2O)-phenyl
20-Furyl
90.2 15.1
146 17
585 94
50.7 6.0
57.7 12.6
54.0 5.5
142 31
23.8 3.8
58.8 10.6
5.63 0.94
20.3 5.5
21.0 6.9
267 59
20-Thienyl
20-Pyrrolyl
N-Methyl-20-pyrrolyl
30-Indolyl
20-Pyridyl
30-Pyridyl
40-Pyridyl
Phenyl
O
O
O
O
O
O
O
40- CH3-phenyl
30- CH3-phenyl
40-OCH3-phenyl
30-OCH3-phenyl
20-OCH3-phenyl
40-Br-phenyl
102 19
56.1 8.8
96.0 29.9
97.0 20.7
4.31 1.34
311 77
1890 438
3200 887
20-Br-phenyl
98.3 16.0
419 113
–
(Because of low solubility data are not available)
O
O
36
37
38
S
SO
SO2
Phenyl
Phenyl
Phenyl
208 42
94.6 25.4
160 38
633 108
193 59
151 27
4-one (RCSB ligand ID 47X, co-crystallized in MIF PDB entry 3L5R)
with extra precision and without using constraints. Four viable poses
were found, all of them similar to the binding mode observed in
3L5R, with docking scores ranging between ꢁ9.975 and
ꢁ10.533 kcal/mol. Ligand RMSD values ranging from 0.7428 to
1.6388 Å for all heavy atoms were calculated, while values between
0.3643 and 0.7292 Å were calculated for the 7-hydroxy-4H-chromen-
4-one core residing in the solvent inaccessible binding pocket of
MIF. RMSD values for the ligand were calculated after binding site
alignment considering residues within 5 Å range of the ligand, and
Scheme 1. General synthetic method for the preparation of the title compounds
1–38. Reagents and conditions (i) base as NaOH or piperidine, temperature:
ambient temperature or 140 ꢀC.
molecules, sulphate ions and glycerol molecules were removed, as 3L5R was used as the reference structure in its native form.
Covalent dockings were run in CovDock within Maestro. The
following custom covalent mechanism protocol was created for
this purpose:
well as ligands in the active sites formed by chains A–B and B–C.
Hydrogen bonds were optimised at pH ¼ 7.4, followed by Impref
minimisation using OPLS3e force field. The 3D structures of the
ligands were determined by LigPrep at pH ¼ 7.4 using OPLS3e force
field. The grid box for docking was centred on the original ligand
between chains A and C with automatic ligand size. Initial Glide
docking experiments were run without any pharmacological con-
straints. Type I constrained docking was run with a tolerance of
0.1 Å, while in Type II constrained docking, the tolerance was set to
2 Å. Docking scores and MM/GBSA calculation results of reversible
complexes in kcal/mol are summarised in Table 2. The model was
LIGAND_SMARTS_PATTERN 1,[C] ¼ [C]-[C] ¼ [O]
RECEPTOR_SMARTS_PATTERN 2,[C]-[N;H1,-1]
CUSTOM_CHEMISTRY ("<1>",("charge",0,1))
CUSTOM_CHEMISTRY ("<1>j<2>",("bond",1,(1,2)))
CUSTOM_CHEMISTRY ("<2> =[C]-[C] ¼ [O]",("bond",1,(1,2)))
Pro1A is present in the conjugate acid form, and was manually
validated by docking 3-(3,4-dihydroxyphenyl)-7-hydroxy-4H-chromen- deprotonated before initiating covalent dockings. The same core

1360
J. GARAI ET AL.
Table 2. Glide XP docking scores and CovDock affinity scores of the compounds using type I and type II docking constraints.
Reversible
Covalent
Type I
docking score (XP)
Type II
Type I
Type II
MM/GBSA
dG Bind
MM/GBSA
dG Bind
MM/GBSA
dG Bind
MM/GBSA
dG Bind
docking score (XP)
cdock affinity
cdock affinity
1
2
3
4
5
6
7
8
ꢁ8.499
ꢁ8.132
–
ꢁ54.71
ꢁ50.83
–
ꢁ8.519
ꢁ7.819
ꢁ6.134
ꢁ7.034
ꢁ4.216
ꢁ7.225
ꢁ8.360
ꢁ6.560
ꢁ5.437
ꢁ6.788
ꢁ7.176
ꢁ6.785
ꢁ7.075
–
ꢁ57.99
ꢁ55.61
ꢁ28.08
ꢁ42.78
ꢁ46.44
ꢁ38.32
ꢁ51.57
ꢁ39.76
ꢁ36.13
ꢁ53,12
ꢁ20.02
ꢁ44.62
ꢁ42.91
–
ꢁ8.693
ꢁ8.233
ꢁ7.312
ꢁ8.188
ꢁ8.141
ꢁ8.269
ꢁ8.045
ꢁ8.046
ꢁ8.183
ꢁ8.626
ꢁ7.897
ꢁ8.186
ꢁ8.024
ꢁ8.301
ꢁ8.369
ꢁ7.457
ꢁ8.464
ꢁ8.197
ꢁ7.890
ꢁ8.308
ꢁ8.000
ꢁ7.823
ꢁ8.344
ꢁ8.504
ꢁ8.256
ꢁ8.601
ꢁ8.533
ꢁ8.837
ꢁ8.228
ꢁ8.667
ꢁ8.560
ꢁ8.626
ꢁ7.404
ꢁ8.134
–
ꢁ61.64
ꢁ34.65
ꢁ24.51
ꢁ35.80
ꢁ37.61
ꢁ45.84
ꢁ50.49
ꢁ36.68
ꢁ37.85
ꢁ41.35
ꢁ42.24
ꢁ40.35
ꢁ52.93
ꢁ40.16
ꢁ37.02
ꢁ45.26
ꢁ45.79
ꢁ46.33
ꢁ41.16
ꢁ34.01
ꢁ28.22
ꢁ39.63
ꢁ39.62
ꢁ42.35
ꢁ41.14
ꢁ40.48
ꢁ48.12
ꢁ48.69
ꢁ49.88
ꢁ50.57
ꢁ51.83
ꢁ45.56
ꢁ50.21
ꢁ44.49
–
ꢁ6.249
ꢁ6.685
ꢁ6.981
ꢁ6.140
ꢁ5.607
ꢁ6.819
ꢁ7.748
ꢁ6.685
ꢁ6.176
ꢁ6.855
ꢁ7.804
ꢁ6.525
ꢁ7.111
ꢁ5.758
ꢁ6.363
ꢁ6.289
ꢁ5.367
ꢁ6.542
ꢁ6.452
ꢁ6.735
ꢁ7.221
ꢁ6.360
ꢁ6.480
ꢁ7.072
ꢁ7.077
ꢁ6.242
ꢁ6.750
ꢁ6.608
ꢁ7.206
ꢁ5.632
ꢁ6.300
ꢁ6.887
ꢁ6.634
ꢁ7.360
–
ꢁ37.79
ꢁ36.06
36.11
ꢁ47.58
ꢁ24.69
ꢁ27.79
6.62
ꢁ43.60
ꢁ34.18
ꢁ34.79
46.15
ꢁ48.47
ꢁ32.16
ꢁ29.96
ꢁ47.81
ꢁ43.93
ꢁ44.99
ꢁ41.71
ꢁ44.63
ꢁ32.73
ꢁ34.68
ꢁ18.93
ꢁ35.48
ꢁ33.41
ꢁ35.23
ꢁ32.12
ꢁ34.59
ꢁ33.40
ꢁ28.65
ꢁ36.95
ꢁ34.00
9.84
ꢁ8.213
ꢁ8.210
ꢁ8.135
ꢁ8.309
ꢁ8.193
ꢁ8.052
ꢁ8.204
ꢁ8.236
ꢁ7.926
ꢁ8.278
ꢁ8.209
ꢁ8.221
ꢁ8.281
ꢁ8.369
ꢁ8.143
ꢁ7.935
ꢁ8.063
ꢁ8.019
ꢁ7.929
ꢁ8.200
ꢁ8.227
ꢁ8.093
ꢁ8.206
ꢁ8.425
ꢁ8.289
ꢁ8.196
ꢁ8.315
ꢁ8.132
ꢁ8.500
ꢁ8.341
ꢁ8.195
–
ꢁ52.36
ꢁ53.75
ꢁ52.37
ꢁ55.48
ꢁ51.5
ꢁ53.51
ꢁ50.87
ꢁ57.97
ꢁ52.70
ꢁ57.06
ꢁ51.36
ꢁ52.12
ꢁ51.63
ꢁ54.02
ꢁ50.91
ꢁ53.14
ꢁ54.70
ꢁ51.92
ꢁ54.38
ꢁ50.17
ꢁ55.40
ꢁ50.90
ꢁ50.79
ꢁ56.44
ꢁ58.79
ꢁ58.01
ꢁ58.44
ꢁ54.03
ꢁ61.06
ꢁ57.21
ꢁ54.00
–
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
ꢁ4.795
ꢁ6.784
–
ꢁ2.46
ꢁ41.76
–
ꢁ6.521
ꢁ7.953
ꢁ7.505
ꢁ7.732
ꢁ6.722
ꢁ6.152
ꢁ7.714
ꢁ6.606
ꢁ6.501
ꢁ6.836
ꢁ7.046
ꢁ6.252
ꢁ4.182
ꢁ4.606
ꢁ6.732
ꢁ6.865
ꢁ8.28
–
ꢁ41.23
ꢁ57.54
ꢁ54.38
ꢁ54.85
ꢁ49.94
ꢁ36.16
ꢁ60.28
ꢁ53.61
ꢁ54.63
ꢁ56.88
ꢁ44.12
ꢁ43.46
ꢁ47.12
ꢁ41.29
ꢁ15.06
ꢁ48.23
ꢁ43.08
–
ꢁ39.94
ꢁ27.69
–
ꢁ27.60
–32.26
–
ꢁ7.700
ꢁ37.08
ꢁ6.352
ꢁ48.78
ꢁ7.495
–5.623
–5.975
ꢁ25.07
–26.39
12.70
ꢁ6.765
–6.764
–
–
–
–
–
–
–
–
–
Calculated binding energies are provided for both, the reversible and covalent MIF-ligand complexes.
constraint settings were applied as to the Glide docking experi- simulation. The 1.2 ns simulation of (24) was also conducted with
ments. The tolerance was set to 2 Å in both instances. The pose
prediction docking mode was used. Docking scores and MM/GBSA
calculation results of covalent complexes in kcal/mol are summar-
ised in Table 2.
TIP3P water model to explore the nature of interactions on the ps
scale in a different MD setup.
Molecular dynamics simulations ranging between 1.2 and
175 ns were applied to a selected group of the compounds with-
out using constraints. All processes were initiated using Desmond
in Maestro. Output structures of a docking experiment with Type I
core constraints were used as starting points. SPC solvent model
was used and box boundaries were adjusted to a distance of 10 Å
from the protein–ligand complexes in each axis. Net positive
charge of the protein was neutralised by chloride ions. 0.15 M
NaCl was added to the systems in each instance. Simulations were
carried out using NPT conditions, at a constant temperature of
Cell culture and treatments
In cell culture experiments, we used RAW264.7 mouse monocyte/
macrophage cell line (ECACC, Salisbury, UK; passage number:
8–15) and RAW-BlueTM cells (Invivogen, Toulouse France, passage
number: 10–14). RAW264.7 and RAW-BlueTM cells were cultured in
5% CO2 at 37 ꢀC in endotoxin-tested Dulbecco’s Modified Eagle’s
Medium (high glucose, 4.5 g/L, 2 mM L-glutamine; Corning, Corning
Incorporated Costar, Corning, NY) supplemented with 10% FBS. For
culturing RAW-BlueTM cells, we also added 100 mg/ml normocin and
200 mg/ml zeocin to the medium. The day before the experiments,
cells were plated onto 24- or 96-well plates and cultured overnight.
Then medium was replaced to fresh one and cells were induced by
0.1 mg/ml or 1 mg/ml lipopolysaccharide (LPS; E. coli, 0127:B8; Sigma-
Aldrich, Budapest, Hungary). The freshly synthesised tetralone deriva-
tives were dissolved in DMSO and applied in 20 mM concentration as
a pre-treatment, added 30 min before LPS induction. To exclude the
effects of vehicle, both CTRL and LPS-treated cells received the same
ꢀ
300 K and pressure of 1 atm, with Nose–Hoover chain thermostat
(1 ps relaxation time) and MTK barostat (2 ps relaxation time)
methods. The cut-off radius for coulombic interactions was set to
9 Å. The systems were relaxed before simulation using the default
relaxation process of Desmond57. The stability of (24) in the active
site was explored through a 175 ns simulation with 25 ps record-
ing intervals, and the Thermal MMGBSA script available in
€
Schrodinger was used to obtain multiple binding free energy val-
ues for the complex, sampling every 100th frame of the amount of DMSO.

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
1361
ROS and nitrite production
chamber (model MIDI F230S; PL Maschine Ltd., Tarnok, Hungary)
set to 31 ꢀC. Each mouse was implanted intraperitoneally with a
miniature biotelemetry transmitter (G2 E-Mitter series; Starr Life
Sciences Corp., Oakmont, PA) to record abdominal temperature
(Tab) and general locomotor activity. The latter has been shown
to play an important thermoregulatory role in mice58–60. The
transmitter was inserted into the abdominal cavity through a
small midline incision and was fixed to the left side of the abdom-
inal wall with suture. The surgical wound was closed and the mice
were returned to their home cages. The mice were allowed a
2–3 weeks post-surgical recovery period before an experiment.
Telemetry receivers (model ER-4000; Starr Life Sciences Corp.,
Oakmont, PA, USA) were positioned in a temperature-controlled
room, and the home cages of mice were placed on top of the
receivers, as described earlier59. The ambient temperature of the
room was set to 26ꢀC, which is near the lower end of the thermo-
neutral zone for mice61. On the day of an experiment, compound
(24) at 1.6 mg/kg or its vehicle (10% DMSO in saline) was adminis-
tered intraperitoneally in bolus (3.3ml/kg). Thirty minutes later,
either LPS (5.0mg/kg) from Escherichia coli 0111:B4 (Sigma-Aldrich,
St. Louis, MO) or saline was intraperitoneally injected to the mice.
Thermoregulatory responses were compared by two-way
ANOVA followed by Fisher LSD post hoc test, as appropriate, using
Sigma Plot version 11.0 (Systat Software, San Jose, CA, USA) with
the level of significance set at p < 0.05. The data are reported
as means SEM.
To measure the amount of reactive oxygen species we seeded
RAW264.7 cells onto 96-well plates in a density of 105cells/well.
Cells were treated with the tetralones as a pre-treatment and they
were induced with 1 mg/ml LPS for 24 h. Then 2 mM dihydrorhod-
amine123 (DHR123; Life Technologies, Carlsbad, CA) fluorescent
dye was added and incubated for an additional 2 h. Fluorescent
intensity of the dye (excitation 490 nm/emission 510–570 nm) was
R
measured with Glomax Multi Detection System (PromegaV,
Madison, WI, USA).
For nitrite measurements, we applied the same culturing condi-
tions, treatments, and equipment as described above. After 24 h of
incubation, 50 ml medium was removed and added to equal amount
of Griess reagent (Sigma-Aldrich, St. Louis, MO, USA) in 96-well plate.
The optical density was measured at 550 nm wavelength.
NF-jB activation
The detection of NF-jB activation was accomplished by using
RAW-BlueTM cells. RAW-BlueTM cells are genetically modified
RAW264.7 macrophages, which are able to secrete embryonic
alkaline phosphatase (EAP) upon LPS induction and following NF-
jB activation. The levels of EAP can be examined using QUANTI-
BlueTM detection medium.
RAW-BlueTM cells were seeded into 96-well plate at a density
of 105cells/well. Cells were treated with tetralones (pre-treatment,
20 mM) and with 1 mg/ml LPS. 24 h after LPS treatment 20 ml
medium was removed and incubated with 100 ml QUANTI-BlueTM
detection medium (Invivogen; Toulouse, France) in 96-well plates.
The optical density was measured at 600 nm.
Results
Tautomerase assay
First the inhibitory effect on the ketonase reaction is discussed. As
regards the structure–activity relationship, the size of the cyclic
ketone ring has an impact on the inhibitory effect (see Table 1).
Cytokine production
For cytokine concentration measurements, RAW264.7 cells were From the phenyl substituted compounds, the five-membered (1)
5
ꢂ
cultured in 24-well plates at a density of 5 10 cells/well and
treated with tetralones (pre-treatment, 20 mM) and with 1 mg/ml
LPS for 24 h. TNF-a, IL-6 and CCL-2 levels were determined from
the culturing media by using Ready-Set-Go ELISA kits (eBioscience,
San Diego, CA, USA) for TNF-a and IL-6 detection and mouse CCL-
2 uncoated ELISA-kit (Invitrogen, Vienna, Austria) for CCL-2 deter-
mination. ELISA-kits were applied as provided by the protocol of
the manufacturer, the optical density was measured at 450 nm.
and the seven-membered (3) showed almost identical inhibitory
effect (IC50¼82.8 and 84.0 mM, respectively), while the six-mem-
bered (2) exerted a lower activity (IC50¼127 mM). The substituted
derivatives of (2) containing both homoaromatic and hetero-
aromatic side chain were examined.
Summarising the results of the measurements studying the
possible inhibitors of ketonase reaction, the best inhibitors are the
compounds with a nonpolar aromatic side chain, such as the
methyl derivative (4), the ortho,para-dichloro compound (13) and
the polar 2-pyridyl derivative (24). Analysis of the structure–activ-
ity relationship is presented later on in more detail.
Mouse thermometry studies
The inhibition of the enolase activity was also investigated. In
the group of the compounds with a homocyclic skeleton, some
good inhibitors can be found, such as five-membered phenyl
derivative (1) (IC50¼20.2 mM). Compounds with a heteroaryl side
chain produce the highest bioactivity. From the five-membered
derivatives the indolyl- (23, IC50¼2.89 mM), the furyl derivative (19,
IC50¼25.4 mM) and the N-metylpyrrolyl compound (22,
IC50¼27.8 mM) are the best inhibitors. As for the six-membered
ones, two pyridyl derivatives (24 and 25, IC50¼28.6 mM for both)
exerted the best activity, while the 4-pyridyl derivative (26) proved
to be a weaker enolase inhibitor. Comparing the enolase and
ketonase activity and focussing on the selectivity, some of our
compounds acted as good ketonase inhibitors having high select-
Thermophysiological experiments were performed in 24 C56BL/6
adult male mice. Experimental procedures were approved by the
ꢀ
Animal Research Review Committee of the University of Pecs,
Medical School (Permit number: BA02/2000–6/2018). At the time
of the experiments, the mice weighed 28 4 g. Animals were
housed in temperature-controlled rooms on a 12/12 h light/dark
cycle. Standard rodent chow and tap water were available ad libi-
tum. All experimental procedures were carried out according to
the European Communities Council Directive of 2010/63/EU under
protocols approved by the Institutional Animal Use and Care
ꢀ
Committee of the University of Pecs. Mice were anaesthetised
with an intraperitoneal (i.p.) injection of ketamine-xylazine (81.7
and 9.3 mg/kg, respectively) and received antibiotic prophylaxis
intramuscularly (gentamycin, 6 mg/kg). During the surgery, mice
were kept on a heating pad (PECO Services Ltd., Brough, UK), and
ivity, such as (4) (IC50enolase/IC50ketonase
¼
13.9), (13)
(IC50enolase/IC50ketonase ¼18), (26) (IC50enolase/IC50ketonase
then to prevent postsurgical hypothermia, the animals were ¼10), while the substance (23) (IC50ketonase/IC50enolase ¼20)
allowed to recover from anaesthesia in a temperature-controlled behaved as a strong and selective enolase inhibitor.

1362
J. GARAI ET AL.
Figure 2. Type I (left) and Type II (right) binding modes of tetralones in the active site of MIF, identified through reversible docking experiments. Carbons of (24) are
coloured orange (Type I) and yellow (Type II). Residues are coloured according to their type of interaction with the ligand: yellow residues form H-bonds, blue residues
form p–p stacking and green form cation-p interactions, red residues are hydrophobic, and the catalytic Pro1 is coloured cyan. The grey area is the inner surface of
the binding site. Red dashed lines indicate steric clashes.
Molecular modelling
Docking experiments
The majority of MIF inhibitors target the active site where enzym-
atic reactions take place. The protein is known to form homo-
trimers, in which binding sites are formed by each of two
adjacent chains. As represented in the Protein Data Bank entry
3IJJ, 4-hydroxyphenylpyruvate contacts with Pro1, Lys32 and Ile64
of monomer A, and Tyr95 and Asn97 of monomer C. Available
structures of MIF in complex with various inhibitors reveal the
importance of these residues in ligand binding62. Like the series
of compounds presented in this paper, many MIF inhibitors con-
tain more than one aromatic ring63, suggesting the extensive
involvement of p-interactions.
Reversible docking
Docking experiments revealed two possible types of binding
mode (Figure 2). In Type I complexes, the A and B rings occupy
the binding pocket. The ketone oxygen forms a hydrogen bond
with the peptide NH of Ile64A. The aromatic ring interacts with
Tyr95C through p–p stacking. The catalytic Pro1A and Lys32A are
in the proximity of the methylene bridge. The aryl or heteroaryl
group bonds with Lys32A, Tyr36A and Phe113A through hydro-
phobic and p-interactions. Such binding mode can be observed in
the PDB entry 3L5R of a 3-phenylchromen-4-one compound64. In
type II binding mode, where the ligand is horizontally flipped, the
C-ring interacts with Tyr95C, the ketone oxygen with Ile64A and
Lys32A, and the A ring with Tyr36A. A similar binding mode of a
structurally related covalent inhibitor can be observed in the PDB
entry 4Z1U65. After visual analysis and consideration of docking
scores, we propose Type I binding mode to be more favourable.
Figure 3. Compound 24 covalently bound to Pro1A of MIF in Type I binding
mode. The H-bond with Ile64A is represented by the yellow, and the p–p stacking
with Tyr95C by the blue dashed line. The docking output structure was minimised
to avoid clashes due to the rigid receptor model. Active site residues involved in
reversible binding are shown. Some atoms are hidden for clearer view of the
covalent adduct.
Both Type I and Type II constraints were used for covalent
docking to assess whether any of the two determined binding
poses would be favourable for a covalent reaction to occur
between Pro1A and the methylene carbon. Covalent adduct for-
mation was successful for most molecules, with the exception of
(35) and (38) due to steric clashes. Covalent docking affinity and
MM/GBSA scores indicate that the reaction is more likely to hap-
pen in Type I binding mode (Table 2). Even though an increased
number of steric clashes with Asn97C, Tyr95C and Pro1A are vis-
ible, minimisation of the output structure results in a reasonable
covalent adduct (Figure 3).
Covalent adduct formation
In the reversible docking models, the position of the electrophilic
methylene bridge relative to Pro1A hints at the possibility of a
covalent mechanism. Such catalytic proline residues reportedly
possess the ability to mediate enzymatic conversions through
covalent catalysis66. NAPQI has been shown to bind covalently to
Pro1, while exhibiting non-covalent inhibitory activity67. The cova-
lent mechanism of a series of MIF inhibitors has also been verified
by X-ray crystallography and enzymatic assays65, where both cova-
Molecular dynamics
We conducted MD simulations to reveal binding characteristics
obscured by using a rigid protein structure for docking. The H-
bond between Ile64A and the ketone oxygen is a crucial contact.
The p–p stacking between the A ring and Tyr95C is a T-type stack-
lent and non-covalent tautomerase inhibitory activity have been ing interaction. Weak hydrophobic contacts with Val106A and
attributed to “compound 12”, visible in PDB entry 4Z1U65.
Met2A are present. The arylmethylene substituent is exposed to

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
1363
Free binding energies for this phase were calculated at an average
of ꢁ54.82 kcal/mol (max¼ ꢁ46.95, min=-63.18; SD ¼ 3.88). At
138.79 ns Trp108A disrupts binding as it intrudes the active site,
which results in cessation of most H-bond and p interactions, and
the distancing of the methylene carbon from Pro1A as well. As a
result, (24) undergoes a major positional shift, and only p–p stack-
ing interactions of the C-ring with Phe113A and Trp108A are vis-
ible with 41% and 27% frequencies. Weak hydrophobic contacts
with Phe49C and Tyr95C also develop. The complex remains
unchanged until the end point of the 175 ns simulation (see
Supplementary material Figures S4–S7 for RMSD plot and active
site residue contributions).
In simulation of (24) using TIP3P water model, the phenolic
oxygen of Tyr95C acts as the H-acceptor instead of Tyr36A; such
model describing an enolate intermediate was published previ-
ously68. We have also identified a potential chloride ion binding
motif at the active site. Chloride binding in MIF may be a remnant
of a lost dehalogenase function. Such enzymatic activity is exhib-
ited by structurally related bacterial tautomerases CaaD and cis-
CaaD, of which the former also possesses a phenylpyruvate tauto-
merase activity similar to that of MIF6.
Figure 4. Interactions between the active site residues and compound 24 as it
maintains its binding conformation. Both an active site-bound water molecule
and a chloride ion are visible. Hydrogen bonds are represented by yellow, p–p
stackings by blue, cation-p interactions by green and ionic bonds by purple
dashed lines. The grey area represents the inner surface of the binding site.
Some atoms of active site residues are not shown for clearer view on
interactions.
Structure–activity relationship
Both aromatic rings interact with active site residues. The ideal size
for the B ring is either 5 or 6 atoms without any substituents; larger
ring systems are less favourable. The 1-tetralone ring is preferred
over 4-chromanones, and thiochromanone analogues have even
weaker effect. The ortho-substitution of the C ring, as well as
switching an ortho-carbon to nitrogen, is favourable for a water
bridge interaction with Pro1A to occur. The heterocyclic nitrogen of
(24) and, to a lesser extent, the 2-chloro substituent of (13), are
able to bind a water molecule. For the 3-pyridyl and 4-pyridyl ana-
logues (25) and (26), or the 3-bromo substituted (9), no persistent
water bridge interaction is observed. Certain para-substituents are
the solvent (see ligand RMSF in Supplementary material, Figure
S2) and forms a sandwich-type p–p stacking with Tyr36A. Rotation
of the C-ring is hindered by constant interactions with Lys32A,
Tyr36A and Phe113A (Supplementary material, Figure S3). A steric
clash may occur for 3-substituted phenylmethylene-tetralones
such as the inactive analogue (9). The protonated amino group of
Lys32A is capable of forming a cation-p bond with the C-ring,
especially of heteroaryl derivatives (22–26). The interaction is less
pronounced for phenyl compounds, such as (4) and (13), and vir-
tually absent from models of (9). All three interactions can be pre-
sent concurrently. A water molecule can also be involved in partially solvated, or can interact with the phenolic hydroxyl of
ligand binding, where the catalytic Pro1A and the adjacent Tyr36A Tyr36A. Some may also stabilise the molecule and prevent changes
in binding pose. Meta-substituents seem to prevent tighter binding,
possibly due to steric clashes with proximal residues. Changing in
the phenyl group meta- or para-carbon to nitrogen, also enhances
inhibitory activity, although to a lesser extent compared to the 2-
pyridyl analogue. Changing the phenyl ring to a 5-membered het-
erocycle mostly increases activity, although not for unsubstituted
pyrrolyl derivative (21), that shows somewhat weaker effect com-
pared to the phenyl analogue. Electron rich derivatives such as (23)
are more likely to interact with the active site lysines, while elec-
tron-deficient rings are less likely to do so.
interacts with the ligand through a water bridge. The hydrogen of
the water molecule occasionally engages in close contact with the
methylene carbon or interacts with heteroatoms and halogen sub-
stituents of the C ring.
The importance of the interaction is most pronounced in the
case of (24), where the 2-pyridyl nitrogen is adequately positioned
to maintain
a constant water bridge interaction (Figure 4).
Consequently, a stable complex forms between (24) and MIF,
where no major conformational change is observed for 65 ns. The
presence of interactions in percentage of frames are: Ile64A(H-bond)
¼ 65%; Tyr95C(p–p) ¼ 64%, Lys32A(cation–p) ¼ 69%, Pro1A(water
¼ 91%, Tyr36A(water bridge) ¼ 83%. Additionally, p–p interac-
bridge)
Macrophage activation
tions with Tyr36A and Phe113A were observed in 18% of the
frames for both residues. The most prominent weak hydrophobic
contacts are formed with Met2A, Tyr36A, Val106A and Phe113A
and are present throughout the entirety of the simulation. The
averaged free binding energy was calculated at ꢁ59.32 kcal/mol
(max¼ ꢁ51.19, min¼ ꢁ68.75; SD ¼ 3.40) for this phase. Minor pro-
tein and ligand movements at 65 ns result in (24) assuming a min-
imally altered, but stable position until 138.79 ns. In this phase of
roughly 73 ns in length, H-bond interaction with Ile64A diminish
to 21% as it forms weak hydrophobic contacts instead, and p–p
interaction with Tyr95C falls to 2%. The water bridge interaction
becomes even more pronounced with 94% and 90% presence
rates for Pro1A and Tyr36A. Lys32A and Phe113A continue to
Tetralone derivatives reduce ROS and NO production in LPS-
induced macrophages
MIF has been involved for long in the pathomechanism of inflam-
mation including macrophage activation. One of the most import-
ant signs of inflammatory macrophage induction is the production
of reactive oxygen and nitrogen species, like superoxide (Oꢃ ꢁ) and
2
nitrogen monoxide (NOꢃ)69.ꢁTherefore, we measured ROS and nitrite
(NO rapidly oxidises to NO2 in the medium of cells) production of
LPS-induced and tetralone derivatives-treated RAW264.7 macro-
phage cells. 24h after LPS treatment we detected a significant,
approximately ꢄ3-fold increase in ROS production compared to the
basal level of vehicle-treated control cells (Figure 5(A)). Three out of
interact with the p –electron system of the C ring (60% and 32%). the five selected compounds slightly inhibited ROS production. To

1364
J. GARAI ET AL.
Figure 6. The effect of tetralone derivatives on NF-jB production. NF-jB activa-
tion in macrophages was evaluated by using RAW-BlueTM cells. Macrophages
were pre-treated with tetralones for 30 min and they were induced by 1 lg/ml
LPS for 24 h. The activity of secreted embryonic alkaline phosphatase was meas-
ured by using QUANTI-BlueTM detection medium. The optical density was meas-
ured at 600nm. Data are presented as means SD in percentage of LPS-treated
group; n ¼ 3 ꢅ 6 (pooled data of three independent experiments with six paral-
ꢂꢂꢂ
lels); Student’s t test, p values <0.05 were considered significant.
n.s.¼non-significant.
p < 0.001;
macrophages upon tetralone treatments. LPS induced a powerful
increase (ꢄ5–20-fold) in the concentrations of all investigated
cytokines and chemokines (Figure 7). (24), (26) and (32) reduced
TNF-a production, but (23) and (32) failed to do so (Figure 7(A)).
In case of IL-6 all of the five selected tetralone compounds could
diminish IL-6 production, but (24), (26) and (32) had the most
dominant effect (Figure 7(B)). CCL2 levels were also negatively
modified by all of the investigated tetralones, which showed com-
parable inhibitory efficacy (Figure 7(C)).
Figure 5. Tetralone derivatives reduce ROS and nitrite production. RAW264.7 cells
were pre-treated with tetralones for 30 min and they were induced by 1 lg/ml
LPS for 24h. (A) ROS production was measured by using 2 lM dihydrorhodamine
123 fluorescent dye (fluorescent intensity was measured at 490 nm [excitation]/
510–570 nm [emission] wavelengths). (B) Nitrite production was evaluated by
applying Griess reagent (optical density was measured at 550 nm). Data are pre-
sented as means SD in percentage of LPS-treated group; n ¼ 3 ꢅ 6 (pooled data
of 3 independent experiments with 6 parallels); Student’s t test, p values < 0.05
ꢂ
ꢂꢂ
ꢂꢂꢂ
p < 0.01; p < 0.001; n.s. ¼ non-
were considered significant. p < 0.05;
significant.
Thermophysiology
The administration of compound (24) or its vehicle on its own did
not cause any change in the abdominal temperature (Tab) and
locomotor activity of the mice. On the contrary, the applied high
dose (5.0 mg/kg i.p.) of LPS in a near subthermoneutral environ-
ment caused marked hypothermia and hypokinesis in vehicle-pre-
treated mice (Figure 8), as expected74. Compared to saline, LPS
caused a significant drop in Tab of vehicle-pre-treated mice from
8 to 22 h post-administration (p < 0.05), with the biggest inter-
group difference of 2.3 ꢀC at 13–14 h (p < 0.001). The LPS-induced
hypothermia was brought about, at least in part, by suppressed
locomotor activity, which was lower in LPS-treated than in saline-
treated mice throughout the experiment and was significantly dif-
ferent between the groups at 11 h (p < 0.05) and from 13 to 22 h
post-administration (p < 0.001) (Figure 8). Pre-treatment of the
mice with compound (24) exaggerated and prolonged the LPS-
induced decrease in Tab. In compound (24)-pre-treated mice the
LPS hypothermia was significant from 7 to 24 h post-administra-
tion, compared to saline-treated mice, with the biggest intergroup
difference of 4.1 ꢀC at 14 h post-administration (p < 0.001).
Importantly, the decrease of Tab in response to LPS was signifi-
cantly more pronounced in mice pre-treated with compound (24)
than in vehicle-pre-treated mice between 10 and 24 h post-LPS
administration (p < 0.05). The LPS-induced decrease in locomotor
activity could be also observed throughout the experiment after
pre-treatment with compound (24), and did not differ from what
was seen in vehicle-pre-treated mice. However, it should be noted
that the locomotor activity was reduced to near-zero level in both
LPS-treated groups, thus an exaggeration of the response (i.e. a
be more precise, we detected at about ꢄ20% decrease in ROS con-
centrations of (24)-, (26)- and (32)-treated, LPS-activated cells. In
contrast to these (4) failed to modify ROS level and surprisingly,
(23) induced a slight increase in it. Nitrite concentrations were
found to be considerably increased (ꢄ20-fold) by LPS treatment
(Figure 5(B)), however in the media of (24)-, (26)- and (32)-treated
active macrophages, we detected markedly lower nitrite concentra-
tions (ꢄ50%). Interestingly, compounds (4) and (23) could also
inhibit nitrite production, but to a lower extent.
Majority of tetralone derivatives diminish NF-jB activation in
macrophages
NF-jB is an inflammatory transcription factor, which can be acti-
vated by PAMPs, like LPS via TLR470 or even by oxidative stress71
and induces proinflammatory protein expression. LPS induced a
dramatic increase (ꢄ40-fold) in the transcriptional activity of NF-
jB compared to DMSO-control (Figure 6). We found slight, but
statistically significant inhibitory effect in the case of (24), (26),
(23) and (4) treatments, in relation to the LPS-activated cells.
Compound (32) could not show any reducing effect on NF-
jB activation.
Tetralone derivatives modulate cytokine and chemokine produc-
tion in RAW264.7 cells
NF-jB participates in the protein expression of many inflammatory
cytokines, like TNF-a, IL-672 and chemokines like, CCL273. Levels of
TNF-a, IL-6 and CCL2 were measured in the media of activated further decrease) to compound (24) was hardly possible.

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
1365
Figure 8. The effect of compound 24 on thermoregulation and locomotor activ-
ity in mice. The effects of compound 24 (or vehicle) pre-treatment at ꢁ30min
on the abdominal temperature (top) and locomotor activity (bottom) of mice
treated with lipopolysaccharide (LPS) or saline at time zero (doses indicated). Pre-
treatment with compound 24 significantly exaggerates the hypothermic response
to LPS, while it does not further reduce the already near-zero level of locomotor
activity in LPS-treated mice. The number of mice was 8 (n ¼ 8) in each group.
Data are presented as means SD.
Figure 7. Cytokine production of activated macrophages: (A) TNF-a; (B) IL-6; (C)
CCL-2. LPS-activated RAW264.7 cells were pre-treated with tetralones for 30 min
and they were induced by 1 lg/ml LPS for 24 h. Cytokine concentrations were
measured with ELISA-kits, optical density was measured at 450 nm. Data are pre-
sented as means SD in percentage of LPS-treated group; n ¼ 6; Student’s t test;
piperidine78. The corroboration of the proposed mechanism is
under progress.
Concerning potential anti-inflammatory activity of the tetralone
analogues reported here, it is noteworthy that certain herbs used
against inflammatory conditions are rich in tetralone
ꢂꢂ
ꢂꢂꢂ
p < 0.001; n.s. ¼
p values < 0.05 were considered significant. p < 0.01;
non-significant.
substances79,80
.
Next, we intended to give a picture of the biological efficacy of
five preferred tetralone derivatives, namely (4), (23), (24), (26) and
(32). These compounds were selected based on their excellent
tautomerase inhibitor activity, i.e. they are potent MIF inhibitors
(see Table 1). The most efficient compounds, namely (24):
(IC50¼5.63 mM for ketonase activity) or (32): (IC50¼4.31 mM for
ketonase activity) are even more powerful pharmacological inhibi-
tors, than the well-known and widely accepted ISO-1 (14.41 mM)23.
Since a strong association between tautomerase inhibition and
attenuation of inflammatory macrophage activation exists81, and
other MIF inhibitors were previously found to possess anti-inflam-
matory effects in LPS-induced RAW264.7 macrophages82, we
Discussion
As regards the mechanism of action of the test compounds, we
suppose, similarly to other authors, the importance of the N-ter-
minal proline. The role of this basic amino acid in the mechanism
of MIF was discussed by some researchers75. Mc Lean et al.
studied hydroxyquinoline derivatives as efficient MIF inhibitors76.
Applying crystallographic methods they were able to detect the
covalent adduct formed in the reaction of Pro1 and an intermedi-
ate unsaturated ketone (“quinone methide”). Similar observations
were made by Orita et al. also using hydroxyquinolines25.
Our test compounds, the cyclic a,b-unsaturated ketones are
well known alkylators reacting with nucleophiles (amines, thiols) decided to use this model in our investigations as well.
All of the chosen tetralone derivatives were found to have
in the b-position in a Michael addition. They show a preferential
affinity towards the thiols77. Thus, we suggest a Michael addition anti-inflammatory potencies in activated macrophages, but two of
with Pro1 for our test compounds. This type of synthetic product them (24 and 26) showed remarkably powerful effects. These two
has been described by Shankar et al. when reacting enones with compounds inhibited ROS and NO production, NF-jB activation,

1366
J. GARAI ET AL.
and expression of the investigated cytokines (TNF-a, IL-6, CCL-2). It inhibitor of the ketonase was the five membered indanone deriva-
is important to note, that these molecules have similar structures; tive (1). As regards the tetralone derivatives with homoaromatic
they are isomeric compounds where only the position of the rings the p-methyl derivative (4) showed the highest activity. As
nitrogen atom differs. Compound (4), however, with a quite simi- for the heteroaromatic C-rings, the five-membered derivatives as
lar methyl substituted structure was found to have only moderate
effects compared to (24) and (26). The third best compound (32)
had very similar effects on RONS and cytokine production, like
(24) and (26), but surprisingly it could not inhibit NF-jB activation.
Compound (32) is a methoxy substituted compound without
nitrogen in the six membered C-ring and it has oxygen in its chro-
manone ring system compared to (24) and (26). Compound (23),
the least effective tetralone in the investigated group also lacks a
pyridyl ring, since it is an indolyl-compound. These findings might
underline the importance of the pyridyl moiety in the biological
activity of the investigated MIF inhibitors.
In thermophysiological experiments, we studied the effect of
compound (24), a potent, preferential ketonase inhibitor (Table 1),
on LPS-induced hypothermia in freely moving mice. This experi-
mental model resembles the manifestation of severe forms of sys-
temic inflammation (e.g. septic shock)74. We found that pre-
treatment of the mice with compound (24) augmented the hypo-
thermic response to high-dose LPS in mice, suggesting that the
enzymatic, mostly ketonase, activity of MIF plays a limiting role in
the thermoregulatory changes associated with severe forms of sys-
temic inflammation.
the indolyl- (23), the furyl derivative (9) and N-metylpyrrolyl com-
pound (22) proved to be the most efficient. From the derivatives
with six membered heteroaromatic C-ring, the 2-pyridyl substance
(24) was the most efficient. The inhibitory effect of the enolase
was much weaker. Some test substances, such as (4), p-methyl
derivative possessed high selectivity. The possible mechanism of
action is probably based on a Michael addition between the
enone groups of the test compounds and the Pro1 of MIF. The
selected five compounds (4), (23), (24), (26) and (32) inhibited
LPS-induced macrophage activation, but to different extent. The
pyridyl-derivatives (24) and (26) showed the highest potency,
which might underline the importance of six membered hetero-
aromatic C-ring in the biological efficacy of our MIF inhibitors. We
have determined a binding model for the compounds through
docking experiments, and generated a custom covalent docking
protocol, which we used to provide the structure of the hypothes-
ised covalent adduct with Pro1. Through MD simulations, we have
demonstrated that the determined reversible binding mode is sta-
ble, and covalent reaction between Pro1 and the electrophilic
methylene carbon may be possible. Our work provide useful infor-
mation for further understanding the ligand binding of MIF, and
could be of aid in designing more potent inhibitors.
Circulating inflammatory cytokines are known to be involved in
the mediation of endotoxin-induced fever and hypothermia74,81,83
,
but the key cryogenic substances responsible for triggering the
hypothermic response in severe systemic inflammation are still
not entirely identified. LPS binds to toll-like receptor 4, which
results in the activation of the NF-jB, thereby leading to the
expression of proinflammatory cytokines such as tumour necrosis
factor (TNF)-a, interleukin (IL)-1ß, and IL-6. At moderate-to-high
concentrations TNF-a causes hypothermia, and, at present, it is
considered to be the most important cryogenic mediator, whereas
IL-6 is the most important pyrogenic cytokine74. MIF, as a potent
proinflammatory cytokine, also plays an important role in systemic
inflammation84, however, to our knowledge; it has not been clari-
fied whether it also contributes to the thermoregulatory manifes-
tations of the disease. In the present study, we demonstrated, for
the first time, that LPS hypothermia is more pronounced when
the ketonase (and possibly, to lesser extent, the enolase) activity
of MIF is inhibited. These findings suggest that the enzymatic
activity of MIF limits the effects of the endogenous, thermally
active mediators, e.g. IL-6 and TNF-a, in severe systemic inflamma-
tion. In accordance, MIF tautomerase inhibitors blocked the pro-
duction of TNF-a and to a bigger extent that of IL-6 in infected
mice in vivo85, as well as, in LPS-activated macrophages in vitro in
the present study (Figure 7). Compound (24) inhibits the ketonase
activity of MIF with a five-fold higher potency than its enolase
activity, but it cannot be ruled out that the enolase activity was
also inhibited by the applied dose of compound (24) to some
extent, therefore, that the enolase activity of MIF also plays a role
in the control of the systemic inflammation-associated hypother-
mic response. The contribution of MIF and its enzymatic activities
to the thermoregulatory manifestation of mild and moderate
forms of systemic inflammation, which are typically accompanied
by fever, also remain subject of future studies.
Acknowledgements
The authors thank Krisztina Sajti for her efficient tech-
nical assistance.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
ꢀ
Andras Garami acknowledges the Janos Bolyai Research
Scholarship of the Hungarian Academy of Sciences (BO/00670/18).
ꢀ
Zoltan Rumbus acknowledges the New National Excellence
Program of the Ministry for Innovation and Technology from the
source of the National Research, Development and Innovation
Fund (UNKP-20-3-IIPTE-877). Balazs Radnai acknowledges the
ꢀ
ꢀ
Janos Bolyai Research Scholarship of the Hungarian Academy of
ꢀ
Sciences (BO/00855/18/5); the UNKP-19–4 (UNKP-19-4-PTE-405)
ꢀ
and UNKP-20–5 (UNKP-20-5-PTE-762) New National Excellence
Program of the Ministry for Innovation and Technology from the
source of the National Research, Development and Innovation
Fund. This work was supported by the National Research,
Development and Innovation Office (FK 124483 to AG); the
Medical School, University of Pecs (KA-2019-27 to AG); the Higher
Education Institutional Excellence Program of the Ministry of
Human Capacities in Hungary (20765-3/2018/FEKUTSTRAT to AG)
and the European Union, co-financed by the European Social
Fund (EFOP-3.6.1-16-2016-00004 to AG); Ministry of Finance,
Hungary GINOP-2.3.3-15-2016-00025, GINOP-2.3.2-15-2016-00049.
ꢀ
ꢀ
Andras Garami acknowledges the Janos Bolyai Research
In conclusion, we have selected two families of compounds, E-
2-arylmethylene-1-tetralones and their heteroanalogues as poten- Scholarship of the Hungarian Academy of Sciences (BO/00670/18).
ꢀ
tial MIF inhibitors. The type and the size of the B-ring and the Zoltan Rumbus acknowledges the New National Excellence
substitution pattern of the C-ring have a high impact on the bio- Program of the Ministry for Innovation and Technology from the
activity. From the compounds with unsubstituted C-ring the best source of the National Research, Development and Innovation

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
1367
ꢀ
Fund (UNKP-20-3-IIPTE-877). Balazs Radnai also acknowledges the 12. Rosengren E, Bucala R, Aman P, et al. The immunoregulatory
ꢀ
Janos Bolyai Research Scholarship of the Hungarian Academy of
mediator macrophage migration inhibitory factor (MIF) cata-
lyzes a tautomerization reaction. Mol Med 1996;2:143–9.
13. Rosengren E, Aman P, Thelin S, et al. The macrophage
migration inhibitory factor MIF is a phenylpyruvate tauto-
merase. FEBS Lett 1997;417:85–8.
Sciences (BO/00855/18/5); the New National Excellence Program
of the Ministry for Innovation and Technology from the source of
the National Research, Development and Innovation Fund (UNKP-
19-4-PTE-405 and UNKP-20-5-PTE-762).
14. Swope M, Sun HW, Blake PR, Lolis E. Direct link between
cytokine activity and a catalytic site for macrophage migra-
tion inhibitory factor. Embo J 1998;17:3534–41.
15. Al-Abed Y, Dabideen D, Aljabari B, et al. ISO-1 binding to
the tautomerase active site of MIF inhibits its pro-inflamma-
tory activity and increases survival in severe sepsis. J Biol
Chem 2005;280:36541–4.
ORCID
ꢀ
Janos Garai
http://orcid.org/0000-0002-3335-4169
http://orcid.org/0000-0002-4972-972X
http://orcid.org/0000-0001-6149-2385
ꢀ
Marcell Kreko
}
ꢀ
ꢀ
Laszlo Orfi
16. Al-Abed Y, VanPatten S. MIF as a disease target: ISO-1 as a
proof-of-concept therapeutic. Future Med Chem 2011;3:45–63.
17. Stosic-Grujicic S, Stojanovic I, Maksimovic-Ivanic D, et al.
Macrophage migration inhibitory factor (MIF) is necessary
for progression of autoimmune diabetes mellitus. J Cell
Physiol 2008;215:665–75.
18. Zhang Y, Zeng X, Chen S, et al. Characterization, epitope
identification and mechanisms of the anti-septic capacity of
monoclonal antibodies against macrophage migration
inhibitory factor. Int Immunopharmacol 2011;11:1333–40.
19. Garai J, Lorand T. Macrophage migration inhibitory factor
(MIF) tautomerase inhibitors as potential novel anti-inflam-
matory agents: current developments. Curr Med Chem 2009;
16:1091–114.
20. Kang I, Bucala R. The immunobiology of MIF: function, gen-
etics and prospects for precision medicine. Nat Rev
Rheumatol 2019;15:427–37.
21. Cournia Z, Leng L, Gandavadi S, et al. Discovery of human
macrophage migration inhibitory factor (MIF)-CD74 antago-
nists via virtual screening. J Med Chem 2009;52:416–24.
22. Tsai TL, Lin TH. Virtual screening of some active human
ꢀ
ꢀ
Peter Balazs Jakus
http://orcid.org/0000-0002-1494-9837
http://orcid.org/0000-0003-2898-6628
http://orcid.org/0000-0002-9171-6834
http://orcid.org/0000-0003-2493-0571
http://orcid.org/0000-0003-0622-442X
http://orcid.org/0000-0002-8904-3976
ꢀ
Zoltan Rumbus
Patrik Keringer
Andras Garami
Eszter Vamos
ꢀ
ꢀ
ꢀ
ꢀ
Dominika Kovacs
ꢀ ꢀ
ꢀ
Viola Bagone Vantus
http://orcid.org/0000-0001-7557-3004
http://orcid.org/0000-0002-7147-0080
http://orcid.org/0000-0002-0599-0758
ꢀ
Balazs Radnai
Tamas Lorand
ꢀ
ꢀ ꢀ
References
1. Bloom BR, Bennett B. Mechanism of a reaction in vitro asso-
ciated with delayed-type hypersensitivity. Science 1966;153:
80–2.
2. David JR. Delayed hypersensitivity in vitro: its mediation by
cell-free substances formed by lymphoid cell-antigen inter-
action. Proc Natl Acad Sci USA 1966;56:72–7.
3. Bucala
R (ed.) The Mif Handbook. Singapore: World
Scientific; 2012.
4. Bucala R, Donnelly SC. Macrophage migration inhibitory fac-
tor: a probable link between inflammation and cancer.
Immunity 2007;26:281–5.
5. Sugimoto H, Taniguchi M, Nakagawa A, et al. Crystallization
and preliminary X-ray analysis of human D-dopachrome tau-
tomerase. J Struct Biol 1997;120:105–8.
macrophage migration inhibitory factor antagonists.
J
Biomol Screen 2014;19:1116–23.
23. Xu L, Zhang Y, Zheng L, et al. Discovery of novel inhibitors
targeting the macrophage migration inhibitory factor via
structure-based virtual screening and bioassays.
J Med
Chem 2014;57:3737–45.
6. Poelarends GJ, Johnson WH, Serrano H, Whitman CP.
Phenylpyruvate tautomerase activity of trans-3-chloroacrylic
acid dehalogenase: evidence for an enol intermediate in the
dehalogenase reaction? Biochemistry 2007;46:9596–604.
7. Whitman CP. The 4-oxalocrotonate tautomerase family of
enzymes: how nature makes new enzymes using a beta-
alpha-beta structural motif. Arch Biochem Biophys 2002;402:
1–13.
24. Dios A, Mitchell RA, Aljabari B, et al. Inhibition of MIF bio-
activity by rational design of pharmacological inhibitors of
MIF tautomerase activity. J Med Chem 2002;45:2410–6.
25. Orita M, Yamamoto S, Katayama N, Fujita S. Macrophage
migration inhibitory factor and the discovery of tautomerase
inhibitors. Curr Pharm Des 2002;8:1297–317.
26. Sarkar S, Siddiqui AA, Mazumder S, et al. Ellagic acid, a diet-
ary polyphenol, inhibits tautomerase activity of human
macrophage migration inhibitory factor and its pro-inflam-
matory responses in human peripheral blood mononuclear
cells. J Agric Food Chem 2015;63:4988–98.
8. Sun HW, Bernhagen J, Bucala R, Lolis E. Crystal structure at
2.6-A resolution of human macrophage migration inhibitory
factor. Proc Natl Acad Sci USA 1996;93:5191–6.
9. Muhlhahn P, Bernhagen J, Czisch M, et al. NMR characteriza-
tion of structure, backbone dynamics, and glutathione bind-
ing of the human macrophage migration inhibitory factor
(MIF). Protein Sci 1996;5:2095–103.
27. Senter PD, Al-Abed Y, Metz CN, et al. Inhibition of macro-
phage migration inhibitory factor (MIF) tautomerase and
biological activities by acetaminophen metabolites. Proc
Natl Acad Sci USA 2002;99:144–9.
10. Cherepkova OA, Gurvits BY. Macrophage migration inhibi-
tory factor: identification of the 30-kDa MIF-related protein
in bovine brain. Neurochem Res 2004;29:1399–404.
11. Meyer-Siegler KL, Iczkowski KA, Vera PL. Macrophage migra-
tion inhibitory factor is increased in the urine of patients
with urinary tract infection: macrophage migration inhibitory
factor-protein complexes in human urine. J Urol 2006;175:
1523–8.
28. Molnar V, Garai J. Plant-derived anti-inflammatory com-
pounds
affect
MIF
tautomerase
activity.
Int
Immunopharmacol 2005;5:849–56.
29. Duvoix A, Blasius R, Delhalle S, et al. Chemopreventive and
therapeutic effects of curcumin. Cancer Lett 2005;223:181–90.
30. Garai J, Molnar V, Eros D, et al. MIF tautomerase inhibitor
potency of alpha,beta-unsaturated cyclic ketones. Int
Immunopharmacol 2007;7:1741–6.

1368
J. GARAI ET AL.
31. Dickerhof N, Magon NJ, Tyndall JD, et al. Potent inhibition
of macrophage migration inhibitory factor (MIF) by myelo-
peroxidase-dependent oxidation of epicatechins. Biochem J
2014;462:303–14.
absorbents. (Kao Corp., Japan). Japan. Kokai Tokkyo Koho;
1992. 10 pp. CODEN: JKXXAF JP 04134043 A 19920507
Heisei. Patent written in Japanese. Application: JP 90-254738
19900925. Priority: CAN 117:157422 AN 1992:557422.
32. Carletti G, Nervo G, Cattivelli L. Flavonoids and Melanins: a 49. Perjesi P, Das U, De Clercq E, et al. Design, synthesis and
common strategy across two kingdoms. Int J Biol Sci 2014;
10:1159–70.
33. Matsunaga J, Sinha D, Solano F, et al. Macrophage migration
antiproliferative activity of some 3-benzylidene-2,3-dihydro-
1-benzopyran-4-ones which display selective toxicity for
malignant cells. Eur J Med Chem 2008;43:839–45.
inhibitory factor (MIF)–its role in catecholamine metabolism. 50. Perjesi P, Nusser T, Tarczay G, Sohar P. E-2-benzylideneben-
Cell Mol Biol (Noisy-le-Grand) 1999;45:1035–40.
zocycloalkanones. Stereostructure and NMR spectroscopic
34. Al-Nakib TM, Lorand T, Foldesi A, Varghese R. The in vitro
investigation. J Mol Struct 1999;479:13–9.
antimycotic activity and acute toxicity of third-generation 51. Perjesi P, Perjessy A, Kolehmainen E, et al. E-2-benzylidene-
bezylidenetetralones and heteroarylidenetetralones. Med
Prin Pract 2001;10:191–6.
35. A GB1255080 19711124; Nishio M, Ito T, Koeda T, Shibata U;
Meiji Seika Kaisha Ltd. 1971 11 24. 1-Oxy- 3-benzal-thiochro-
manone derivatives and process for the production thereof.
Great Britain, GB1255080A.
benzocycloalkanones III. Studies on transmission of substitu-
ent effects on IR carbonyl stretching frequencies and C-13
NMR chemical shifts of E-2-(X-benzylidene)-1-indanones.
Comparison with the IR data of E-2-(X-benzylidene)-1-inda-
nones, -tetralones, and -benzosuberones. J Mol Struct 2004;
697:41–7.
36. Adam W, Golsch D, Hadjiarapoglou L, et al. Dioxirane oxida- 52. Toth G, Szollosy A, Levai A, et al. Stereochemistry of 3-aryli-
tion of (Z)-1-thioaurones, (E)-3-arylidene-1-thiochroman-4-
ones and (E)-3-arylidene-1-thioflavan-4-Ones. Tetrahedron
1994;50:13113–20.
deneflavanones and their thio analogs by the use of 3j(Ch)
couplings. Magn Reson Chem 1991;29:801–4.
53. Wachter GA, Hartmann RW, Sergejew T, et al.
Tetrahydronaphthalenes: influence of heterocyclic substitu-
ents on inhibition of steroidogenic enzymes P450 arom and
P450 17. J Med Chem 1996;39:834–41.
37. Adam W, Halasz J, Jambor Z, et al. Synthesis and stereo-
chemistry of the epoxides of 2-arylmethylidene-1-tetralones.
J Chem Soc Perk T 1996;1:395–400.
38. AlAnsari IAZ. Ground- and excited-state properties of some 54. Hallgas B, Dobos Z, Osz E, et al. Characterization of lipophi-
3,4-dihydro-1-(2-p-substituted benzylidene)naphthalenones:
substituent and environmental effects. J Phys Org Chem
1997;10:687–96.
licity and antiproliferative activity of E-2-arylmethylene-1-tet-
ralones and their heteroanalogues. J Chromatogr B 2005;
819:283–91.
39. Alnakib T, Bezjak V, Meegan MJ, Chandy R. Synthesis and 55. Taylor AB, Johnson WH Jr, Czerwinski RM, et al. Crystal struc-
antifungal activity of some 3-benzylidenechroman-4-ones, 3-
benzylidenethiochroman-4-ones and 2-benzylidene-1-tetra-
lones. Eur J Med Chem 1990;25:455–62.
ture of macrophage migration inhibitory factor complexed
with (E)-2-fluoro-p-hydroxycinnamate at 1.8 A resolution:
implications for enzymatic catalysis and inhibition.
Biochemistry 1999;38:7444–52.
40. Azzolina O, Desimoni G, Ditoro V, et al. Influence of config-
uration and ring strain on transmission of electronic effect 56. Cheng Y, Prusoff WH. Relationship between the inhibition
constant (K1) and the concentration of inhibitor which
causes 50 per cent inhibition (I50) of an enzymatic reaction.
Biochem Pharmacol 1973;22:3099–108.
in Alpha,Beta-unsaturated carbonyl derivatives. Gazz Chim
Ital 1975;105:971–83.
41. Basavaiah D, Bakthadoss M, Pandiaraju S. A new protocol for
the syntheses of (E)-3-benzylidenechroman-4-ones: a simple 57. Bowers KJ, Chow E, Xu H, et al. Scalable algorithms for
molecular dynamics simulations on commodity clusters.
Proceedings of the ACM/IEEE Conference on
Supercomputing (SC06), 2006, Tampa, Florida, USA.
synthesis of the methyl ether of bonducellin. Chem
Commun 1998;697:1639–40.
42. Dimmock JR, Zello GA, Oloo EO, et al. Correlations between
cytotoxicity and topography of some 2-arylidenebenzocy-
cloalkanones determined by X-ray crystallography. J Med
Chem 2002;45:3103–11.
€
Schrodinger, LLC. Desmond User Manual, Chapter 3. 222
Third Street, Suite 2230, Cambridge, MA 02142, USA. D. E.
Shaw Research.
43. Duddeck H, Levai A. Natural abundance O-17 Nmr-spectra 58. Banki E, Pakai E, Gaszner B, et al. Characterization of the
of thio analogs of chromanones, chromones, flavanones and
flavones. Magn Reson Chem 1992;30:65–7.
44. Eiden F, Felbermeir G. 7-Phenyl[1]benzopyrano(or 7-thiopyr-
thermoregulatory response to pituitary adenylate cyclase-
activating polypeptide in rodents. J Mol Neurosci 2014;54:
543–54.
ano)pyranes from 3-benzylidenechromanones (or 3-thiochro- 59. Garami A, Pakai E, Oliveira DL, et al. Thermoregulatory
manones). Arch Pharm 1984;317:861–7.
phenotype of the Trpv1 knockout mouse: thermoeffector
45. Elrayyes NR, Aljawhary A. Heterocycles.8. Synthesis of new
dysbalance with hyperkinesis. J Neurosci 2011;31:1721–33.
substituted Benz[G]Indazoles. J Heterocycl Chem 1986;23: 60. Pozsgai G, Payrits M, Saghy E, et al. Analgesic effect of
135–40.
dimethyl trisulfide in mice is mediated by TRPA1 and sst4
46. Elrayyes NR, Ramadan HM. Heterocycles.10. Synthesis of
receptors. Nitric Oxide 2017;65:10–21.
new pyrimidine systems. J Heterocycl Chem 1987;24:589–96. 61. Kanizsai P, Garami A, Solymar M, et al. Energetics of fasting
47. Bowen PJ, Robinson TP, Ehlers T, Goldsmith D, Arbiser J;
University of Georgia Research Foundation and Emory
University, USA, 2001 06 28, Chalcone and its analogs as
agents for the inhibition of angiogenesis and related disease
states. United States Patent, WO2001046110A2.
heterothermia in TRPV1-KO and wild type mice. Physiol
Behav 2009;96:149–54.
62. (ai) Crichlow GV, Cho Y, Lolis EJ. Ternary complex of macro-
phage Migration Inhibitory Factor (MIF) bound both to 4-
hydroxyphenylpyruvate and to the allosteric inhibitor
AV1013 (R-stereoisomer). 2009. doi:10.2210/pdb3IJJ/pdb
(aii) Cho Y, Crichlow GV, Vermeire JJ, et al . Allosteric
48. Kawamata A, Jokura H, Imokawa G. Cosmetics containing
alfa-benzylidene ketone derivatives and ultraviolet

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
1369
inhibition of macrophage migration inhibitory factor 69. Zhang Z, Zhang R, Li L, et al. Macrophage migration inhibi-
tory factor (MIF) inhibitor, Z-590 suppresses cartilage
destruction in adjuvant-induced arthritis via inhibition of
macrophage inflammatory activation. Immunopharmacol
Immunotoxicol 2018;40:149–57.
revealed by ibudilast. Proc Natl Acad Sci
2010;107:11313–11318. doi:10.1073/pnas.1002716107
U
S A
(bi) Mclean LR, Zhang Y. Crystal structure of macrophage
migration inhibitory factor (mif) with hydroxyquinoline
inhibitor 707 at 1.58a resolution. 2009. doi:10.2210/
pdb3JSG/pdb
(bii) McLean LR, Zhang Y, Li H, et al. Discovery of covalent
inhibitors for MIF tautomerase via cocrystal structures with
phantom hits from virtual screening. Bioorg Med Chem Lett
2009;19:6717–6720. doi:10.1016/j.bmcl.2009.09.106
70. Zhang G, Ghosh S. Molecular mechanisms of NF-kappaB
activation induced by bacterial lipopolysaccharide through
Toll-like receptors. J Endotoxin Res 2000;6:453–7.
71. Wardyn JD, Ponsford AH, Sanderson CM. Dissecting molecu-
lar cross-talk between Nrf2 and NF-kappaB response path-
ways. Biochem Soc Trans 2015;43:621–6.
72. Schow SR, Joly A. N-Acetyl-leucinyl-leucinyl-norleucinal
inhibits lipopolysaccharide-induced NF-kappaB activation
and prevents TNF and IL-6 synthesis in vivo. Cell Immunol
1997;175:199–202.
73. Akhter N, Hasan A, Shenouda S, et al. TLR4/MyD88 -medi-
ated CCL2 production by lipopolysaccharide (endotoxin):
implications for metabolic inflammation. J Diabetes Metab
Disord 2018;17:77–84.
74. Garami A, Steiner AA, Romanovsky AA. Fever and hypother-
mia in systemic inflammation. Handb Clin Neurol 2018;157:
565–97.
75. Golubkov PA, Johnson WH Jr, Czerwinski RM, et al.
Inactivation of the phenylpyruvate tautomerase activity of
macrophage migration inhibitory factor by 2-oxo-4-phenyl-
3-butynoate. Bioorg Chem 2006;34:183–99.
76. McLean LR, Zhang Y, Li H, et al. Discovery of covalent inhibi-
tors for MIF tautomerase via cocrystal structures with phan-
tom hits from virtual screening. Bioorg Med Chem Lett
2009;19:6717–20.
77. Dimmock JR, Raghavan SK, Logan BM, Bigam GE. Anti-leu-
kemic evaluation of some mannich-mases derived from 2-
arylidene-1,3-diketones. Eur J Med Chem 1983;18:248–54.
78. Shankar R, Jha AK, Singh US, Hajela K. An efficient and
improved synthesis of 1,5-diketones: versatile conjugate
addition of nucleophiles to alpha,beta-unsaturated enones
and alkynones. Tetrahedron Lett 2006;47:3077–9.
79. Liu Q, Zhao P, Li XC, et al. New alpha-tetralone galloylgluco-
sides from the fresh pericarps of Juglans sigillata. Helv Chim
Acta 2010;93:265–71.
80. Upadhyay HC, Verma RK, Srivastava SK. Quantitative deter-
mination of bioactive 4-hydroxy-alpha-tetralone, tetralone-4-
O-beta-D-glucopyranoside and ellagic acid in Ammannia
baccifera (Linn.) by reversed-phase high-performance liquid
chromatography. J Chromatogr Sci 2013;51:21–5.
(ci) Oki H, Igaki S, Moriya Y, et al. Crystal structure of macro-
phage migration inhibitory factor in complex with BTZO-14.
2016. doi:10.2210/pdb5B4O/pdb
(cii) Kimura H, Sato Y, Tajima Y, et al . BTZO-1, a cardiopro-
tective agent, reveals that macrophage migration inhibitory
factor regulates ARE-mediated gene expression. Chem Biol
2010;17:1282–1294. doi:10.1016/j.chembiol.2010.10.011
(di) Robertson MJ, Jorgensen WL. Crystal structure of
Macrophage Migration Inhibitory Factor (MIF) with a Potent
Inhibitor (NVS-2). 2016. doi:10.2210/pdb5HVT/pdb
(dii) Cisneros JA, Robertson MJ, Valhondo M, Jorgensen WL.
A fluorescence polarization assay for binding to macrophage
migration inhibitory factor and crystal structures for com-
plexes of two potent inhibitors.
J
Am Chem Soc
2016;138:8630–8638. doi:10.1021/jacs.6b04910
(ei) Krimmer SG, Robertson MJ, Jorgensen WL. Macrophage
migration inhibitory factor in complex with
a
Naphthyridinone Inhibitor (3a), 2017. doi:10.2210/pdb6B1K/
pdb.
(eii) Dawson TK, Dziedzic P, Robertson MJ, et al. Adding a
hydrogen bond may not help: naphthyridinone vs quinoline
inhibitors of macrophage migration inhibitory factor. ACS
Med Chem Lett 2017;8:1287–1291. doi:10.1021/acsmed-
chemlett.7b00384
63. (ai) Dziedzic P, Cisneros JA, Robertson MJ, et al . Design,
synthesis, and protein crystallography of biaryltriazoles as
potent tautomerase inhibitors of macrophage migration
inhibitory factor.
J Am Chem Soc 2015;137:2996–3003.
doi:10.1021/ja512112j
(bi) Trivedi-Parmar V, Robertson MJ, Cisneros JA, et al.
Optimization of pyrazoles as phenol surrogates to yield
potent inhibitors of macrophage migration inhibitory factor.
ChemMedChem
201800158
2018;13:1092–1097.
doi:10.1002/cmdc.
81. Romanovsky AA, Almeida MC, Aronoff DM, et al. Fever and
hypothermia in systemic inflammation: recent discoveries
and revisions. Front Biosci 2005;10:2193–216.
82. Zhang Y, Xu L, Zhang Z, et al. Structure-activity relationships
and anti-inflammatory activities of N-carbamothioylforma-
mide analogues as MIF tautomerase inhibitors. J Chem Inf
Model 2015;55:1994–2004.
64. McLean LR, Zhang Y, Li H, et al. Fragment screening of
inhibitors for MIF tautomerase reveals a cryptic surface bind-
ing site. Bioorg Med Chem Lett 2010;20:1821–4.
65. Singh AK, Pantouris G, Borosch S, et al. Structural basis for
decreased induction of class IB PI3-kinases expression by
MIF inhibitors. J Cell Mol Med 2017;21:142–53.
66. Huddleston JP, Johnson WH Jr, Schroeder GK, Whitman CP.
The accidental assignment of function in the tautomerase
superfamily. Perspect Sci 2015;4:38–45.
67. Crichlow GV, Lubetsky JB, Leng L, et al. Structural and kin-
etic analyses of macrophage migration inhibitory factor
active site interactions. Biochemistry 2009;48:132–9.
68. Sean E, ed. Computational toxicology. Risk assessment for
pharmaceutical and environmental chemicals. Vol. 2, 2nd ed.
Boca Raton (FL): CRC Press; 2018:59.
83. Leon LR. Hypothermia in systemic inflammation: role of
cytokines. Front Biosci 2004;9:1877–88.
84. Larson DF, Horak K. Macrophage migration inhibitory factor:
controller of systemic inflammation. Crit Care 2006;10:
138–40.
85. Chuang CC, Chuang YC, Chang WT, et al. Macrophage
migration inhibitory factor regulates interleukin-6 produc-
tion by facilitating nuclear factor-kappa B activation during
Vibrio vulnificus infection. Bmc Immunol 2010;11:50.