SAR, species specificity, and cellular activity of cyclopentene dicarboxylic acid amides as DHODH inhibitors

Article abstract of DOI:10.1016/j.bmcl.2005.07.053

Novel DHODH inhibitors were developed based on a previously described series by introduction of heteroatoms into the cyclopentene ring and hydroxyl groups attached to it. Also, the hydrophobic biphenyl side chain was replaced with benzyloxy phenyl groups. Activities on human, rat, and mouse enzymes indicate a species specificity of these inhibitors.

Full text of DOI:10.1016/j.bmcl.2005.07.053

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4854–4857
SAR, species specificity, and cellular activity of
cyclopentene dicarboxylic acid amides as DHODH inhibitors
Johann Leban,* Martin Kralik, Jan Mies, Michael Gassen, Karin Tentschert and
Roland Baumgartner
4SC AG, Am Klopferspitz 19a, 82152 Martinsried, Germany
Received 30 May 2005; revised 5 July 2005; accepted 7 July 2005
Available online 6 September 2005
Abstract—Novel DHODH inhibitors were developed based on a previously described series by introduction of heteroatoms into the
cyclopentene ring and hydroxyl groups attached to it. Also, the hydrophobic biphenyl side chain was replaced with benzyloxy phen-
yl groups. Activities on human, rat, and mouse enzymes indicate a species specificity of these inhibitors.
Ó 2005 Elsevier Ltd. All rights reserved.
Dihydroorotate dehydrogenase (DHODH) is responsi-
ble for the conversion of dihydroorotate to orotate,
which is the rate-limiting step in pyrimidine biosynthe-
sis. Inhibitors of DHODH show immunosuppressant
and antiproliferative activities, which are most pro-
nounced on activated T-cells.1
boronic acids and halo anilines with Pd and KF in meth-
anol as described.3 Benzyloxy phenylanilines were
prepared, as shown in Figure 3.
The final compounds 1–22 were obtained by reacting the
appropriate dicarboxylic acid anhydrides with the
anilines in an inert solvent.8
Leflunomide, an inhibitor of DHODH, is currently used
to treat rheumatoid arthritis, and analogs are in clinic to
treat graft versus host disease and multiple sclerosis.2
Enzyme inhibition was measured in an in vitro enzyme
assay.4 For the assay, N-terminally truncated recombi-
nant human DHODH was used.5 The assay was based
on the coupling reaction of the enzyme to the redox
dye 2,6-dichlorophenolindophenol (DCIP) as described.
The reduction of DCIP can be monitored photometri-
cally by a decreasing absorption at 600 nm.
A novel series of DHODH inhibitors was developed by
us based on a lead that was discovered during a docking
procedure and medicinal chemistry exploration. The
activity of the initial lead was improved by a QSAR
method and yielded low nanomolar inhibitors.3
T-cell (PBMC) proliferation was measured at variance
to a published method and is described in reference.6
To evaluate the compounds further, we introduced
heteroatoms into the cyclopentene ring and replaced
the biphenyl moiety by the benzyloxy.
Previously, we have described the discovery of cyclopen-
tene dicarboxylic acid monoamides3 as potent DHODH
inhibitors. Here, we describe further data on the under-
standing of the SAR of these compounds.
Cyclopentene dicarboxylic anhydride was purchased
from Aldrich. Thio and oxocyclopentene carboxylic
acids were synthesized by the cyanhydrine synthesis, as
shown in Figures 1 and 2.
As can be seen on the lead compound 1 (Table 1), intro-
duction of sulfur into the petacyclic ring did not lower
the activity significantly in compound 2. A similar
behaviour was seen on all other carbon sulfur replace-
ments (6, 10, 13, 16, 17). Oxidation of the sulfide to
the sulfone led to less active compound 3. Inspection
of the binding mode in a X-ray structure of such an
Substituted biarylanilines were obtained by the Suzuki
cross-coupling procedure using commercial aromatic
*
Corresponding author. Fax: +49 89 700 763 28; e-mail:
leban@4sc.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.07.053

J. Leban et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4854–4857
4855
O
COOMe
HO
NC
COOMe
HOOC
COOH
NC
COOMe
a
b
c
O
O
O
O
Figure 1. Reagents and conditions: (a) NaCN, water/ether; (b) H2SO4, stirring 18 h at room temperature; (c) concd HCl, reflux, 3 h.
O
O
O
O
OH
HO
HO
COOMe
COOMe
b
OH
O
a
c
O
CN
S
S
S
S
Figure 2. Reagents and conditions: (a) KCN, methanol/water/acetic acid, 0 °C, 8 h; (b) concd HCl, acetic acid, reflux, 20 h; (c) acetic acid anhydride,
reflux, 4 h.
Figure 3. Reagents and conditions: (a) NaOH, water/ethanol, 80 °C, 3 h; (b) Fe, NH4Cl, water/acetic acid/ethanol, 75 °C, 3 h.
analog indicated a close proximity of the pentacycle to
the FMN cofactor in the enzyme, leaving little space
for the sulfone without disrupting a good binding mode.9
Compounds with hydroxyl groups attached to the pen-
tacyclic ring (21, 22) were synthesized by the Wohl–
Ziegler allylbromination with N-bromosuccinimide and
light from a 300 W lamp, in chloroform under refluxing
conditions (Fig. 4). Hydrolysis of the brominated com-
pound, without isolation, yielded two regioisomers that
were separated by HPLC (21, 22) in the ratio 7:3. The
more active compound 21 was found to be the isomer
with the hydroxygroup adjacent to the free carboxylic
acid, as shown in the X-ray structure. We did not sepa-
rate the enantiomers however, the X-ray structure
electrodensity of compound 21 accommodates equal
binding of both hydroxyl enantiomers. The hydroxylat-
ed compound 21 was 20-fold less active than the parent
compound 4.
Introduction of oxygen into the pentacyclic ring yielded,
in every case, compounds with lower activity (5, 9, 12,
and 19). A model derived from docking and from the
X-ray structure did not indicate any contradiction from
a good binding mode. Increased hydrophilicity of the
compound may prevent diffusion from into the active
side of this enzyme. The enzyme is anchored in a very
hydrophobic environment and the inhibitor has to dif-
fuse through a very hydrophobic environment. Such
an effect could explain the lower activity of hydrophilic
compounds in general. In the biphenyl series, there was
a clear trend toward better activity with the number of
fluor substitutions in the first aromatic ring (4, 7). We
initially attributed the better activity to increased hydro-
phobicity of the biphenyl ring system, since this pharma-
cophore binds to a hydrophobic environment in the
active site of the enzyme and such compounds diffuse
better into the active site.
Another suitable pharmacophore for binding into the
hydrophobic pocket of the active site was found to be
the benzyloxygroup 14. Although the parent compound
was less active than the unsubstituted biphenyl com-
pound 1, the introduction of halogen residues in ortho
and meta positions toward the benzylether linkage
greatly improved inhibitory activity (14–20).
The X-ray structure of these compounds showed a trend
toward a Brequinar-like binding mode with increasing
number of fluor substituents in the first aromatic ring.
The more Brequinar-like the binding the more active
the compounds.9
In this series, however, the alternative binding mode did
not seem to be of importance, but increased activity was
considered to be a consequence of lipophilicity, since
we did not observe any Brequinar-like binding mode.9
F
O
O
OH
HN
70
O
F
O
O
OH
a. b.
OH
HN
O
F
O
O
OH
HN
HO
30
O
Figure 4. Reagents and conditions: (a) 300 W lamp, NBS, chloroform, reflux; (b) water, RT.

4856
J. Leban et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4854–4857
Table 1. Inhibition on human DHODH
Table 1 (continued)
a
a
Compound
IC50 (lM)
Compound structure
Compound
IC50 (lM)
Compound structure
HO
O
HO
O
1
0.41
13
0.004
O
F
F
F
F
OCF3
O
S
HN
HN
HO
HO
O
O
O
O
14
15
16
17
18
19
20
21
22
2.0
2
0.667
S
HN
O
O
HN
Br
O
OH
HO
O
0.127
0.173
0.105
0.011
0.041
0.110
0.68
HN
O
3
4
3.8
O
S
O
Br
Br
HN
O
O
OH
HN
HO
O
O
O
O
F
F
OCH3
0.134
0.360
0.131
0.011
0.033
0.205
0.015
0.007
0.058
S
O
Br
Br
HN
F
O
OH
HN
HO
O
O
OCH3
Cl
5
S
O
Br
O
HN
HO
O
F
O
Cl
HO
O
HN
O
O
F
F
OCH3
OCH3
6
Cl
S
HN
HO
O
O
F
HO
O
Cl
O
O
HN
O
Cl
7
HN
F
Br
O
F
F
OH
HN
HO
HO
O
O
O
OCF3
OCF3
OCF3
OCF3
OCF3
Cl
O
Br
8
HN
HO
O
HO
F
F
O
F
OCH3
O
HN
O
9
O
HO
HN
O
O
F
OCH3
F
F
1.46
HO
O
HN
O
OH
10
11
12
S
a Values are means of three experiments, errors are usually around
10%.
HN
F
F
HO
O
O
F
We also tested some of the most active compounds on
the human enzyme, and on rat and mouse DHODH en-
zymes, and the results were compared. As a reference
compound, A771276, the active metabolite of Lefluno-
mide, shows a marked better activity on the mouse
versus rat versus human enzyme.
HN
F
F
F
F
HO
O
O
O
HN
Contrary to this result, all our compounds are better
inhibitors on the human versus rat and mouse enzymes.
F
F

J. Leban et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4854–4857
Table 2. Comparison of selected compounds on human, rat, and
mouse DHODH
4857
References and notes
a
1. Fairbanks, L. D.; Bofill, M.; Ruckemann, K.; Simmonds,
H. A. J. Biol. Chem. 1995, 270, 29682.
2. Allred, A.; Emery, P. Exp. Opin. Pharmacother. 2001, 2,
125.
3. Leban, J.; Saeb, W.; Garcia, G.; Baumgartner, R.; Kramer,
B. Bioorg. Med. Chem. Lett. 2004, 14, 55.
4. Davis, J. P.; Cain, G. A.; Pitts, W. J.; Magolda, R. L.;
Copeland, R. A. Biochemistry 1996, 35, 1270.
Compound
IC50 (lM)
Human
Rat
Murine
1
0.41
1.5
4.0
2.0
8
11
0.033
0.007
0.041
0.42
0.504
0.078
1.2
0.422
1.8
0.156
19
A771726
0.009
5. Recombinant human, rat and murine DHODH were
obtained from Prof. Dr. M. Lo¨ffler, University of Marburg,
Karl-von-Frisch-Str.1 35043 Marburg.
a Values are means of three experiments, errors being usually around
10%.
6. Neidhardt, E. A.; Punreddy, S. R.; McLean, J. E.;
Hedstrom, L.; Grossman, T. H. J. Mol. Microbiol.
Biotechnol. 1999, 1, 183.
Table 3. Inhibition of PBMC proliferation
7. Magaud, J.-P.; Sargent, I.; Mason, D. Y. J. Immunol.
Methods 1988, 106, 95, Cells (Mononuclear cells) were
isolated from human peripheral blood by AccuspinTM
System-HistopaqueÒ-1077 (Sigma, Germany). After wash-
ing, the cells were diluted to approximately 100,000–200,000
cells/well in a sterile 96-well flat-bottomed MP (Corning,
Netherlands). T-lymphocytes were stimulated by addition of
20 lg/ml Phytoha¨magglutinin-L (Roche, Germany). Incu-
bation at 37 °C, 5% CO2, 90% humidity was carried out in
the presence of different concentrations of the compounds.
All cells were incubated for 48 h at 37 °C, 5% CO2, 90%
relative humidity over a concentration range of 0.4–50 lM
compound solutions with a final volume per well of 100 ll.
After the initial incubation period, 10 lM BrdU (Roche,
Germany) was added for an additional 4 h incubation. The
culture medium employed was RPMI 1640, which contained
10% heat-inactivated fetal bovine serum, 2 mM L-gluta-
mine, 100 U/ml penicillin G, and 100 lg/ml streptomycin
sulfate. After 48 h incubation, the cells were labeled by
adding 10 ll BrdU-Solution (Roche, Germany) and reincu-
bated for further 4 h. Following incubation, the media plus
BrdU and drug were removed, the cells were fixed, and the
DNA was denatured in a single step using FixDenat (Roche,
Germany) The anti-BrdU-POD (Roche, Germany) binds to
the BrdU incorporated into a newly synthesized, cellular
DNA. The immune complexes were detected by a subse-
quent substrate reaction. The reaction product was quan-
tified by measuring the absorbance at the respective
wavelength using an ELISA reader. The EC50 values were
determined using a fitting function.
Compound
IC50 (lM)
13
15
0.175–0.327
1.0
108
4
8
11
19
A771726
To show the immunosuppressive effects of compounds,
we studied the inhibition of proliferation of our com-
pounds human peripheral blood mononuclear cells
(PBMCs) stimulated with Phytohemagglutinin-L.7
Data in Table 3 show that this series of DHODH inhib-
itors was significantly more active than A77127, which is
the active metabolite of Leflunomide (Tables 2 and 3).
Further studies on PK and animal models should iden-
tify clinical candidates for the development of drugs to
treat autoimmune diseases. Such diseases are rheuma-
toid arthritis, multiple sclerosis, lupus erythematosus,
ulcerative colitis, and neurodermitis.
Acknowledgments
We thank Matthias Dormeyer for helpful discussions,
and Gerhard Keilhauer and Daniel Vitt for support.
We also thank Prof. Dr. V. Bakulev, Urals State Univer-
sity, Jekatarinenburg, for the synthesis of intermediates.
8. All compounds were characterized by HPLC- MS and
NMR (300 MHz), and exhibited satisfactory properties.
9. R. Baumgarter et al., J. Med. Chem., submitted.