Synthesis and SAR of conformationally restricted inhibitors of soluble epoxide hydrolase

Article abstract of DOI:10.1016/j.bmcl.2006.07.009

A series of conformationally restricted inhibitors of human soluble epoxide hydrolase (sEH) has been developed. Inhibition potency of the described compounds ranges from 4.2 μM to 1.1 nM against recombinant sEH. N-(1-Acetylpiperidin-4-yl)-N′-(adamant-1-yl) urea (5a) was found to be a potent inhibitor (IC₅₀ = 7.0 nM) that was also orally bioavailable in canines.

Full text of DOI:10.1016/j.bmcl.2006.07.009

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5212–5216
Synthesis and SAR of conformationally restricted inhibitors
of soluble epoxide hydrolase
Paul D. Jones, Hsing-Ju Tsai, Zung N. Do, Christophe Morisseau
and Bruce D. Hammock*
Department of Entomology and Cancer Research Center, One Shields Avenue, University of California, Davis, CA 95616, USA
Received 22 May 2006; revised 28 June 2006; accepted 5 July 2006
Available online 25 July 2006
Abstract—A series of conformationally restricted inhibitors of human soluble epoxide hydrolase (sEH) has been developed. Inhi-
bition potency of the described compounds ranges from 4.2 lM to 1.1 nM against recombinant sEH. N-(1-Acetylpiperidin-4-yl)-
N0-(adamant-1-yl) urea (5a) was found to be a potent inhibitor (IC50 = 7.0 nM) that was also orally bioavailable in canines.
ꢀ 2006 Elsevier Ltd. All rights reserved.
The soluble epoxide hydrolase (sEH, E.C. 3.3.2.3) is a
member of the a/b-hydrolase fold family of enzymes.1
A major function of sEH is the conversion of epoxides
to vicinal diols through the catalytic addition of a water
molecule.2 The endogenous substrates for the sEH in-
clude the cytochrome P450 metabolites of arachidonic
acid (epoxyeicosatrienoic acids, EETs).3,4 EETs are
known modulators of blood pressure and inflammation,
and their conversion to dihydroxyeicosatrienoic acids
(DHETs) by sEH reduces the beneficial activity of
EETs. It has been shown that in vivo inhibition of
sEH with highly selective inhibitors results in an increase
in the concentration of EETs and is accompanied by a
reduction in blood pressure in rodent models, thereby
suggesting that sEH is a compelling target for the treat-
ment of hypertension.5–8
Through the course of our research we have found N,N0-
disubstituted ureas to be our most successful inhibitors
of sEH (Fig. 1).9–13 The inhibitors shown in Figure 1
can be divided into three basic categories: those that
Figure 1. Common inhibitors of sEH. IC50 is for in vitro inhibition
against recombinant human sEH.
are small, rigid, and contain non-polar groups
(e.g., DCU), those that contain both rigid and flexible
non-polar groups (e.g., CDU) and those that contain
both a rigid non-polar group and flexible polar chain
(e.g., 950, AUDA, and AUDA-BE).
We have found that while those compounds possessing a
flexible side chain do show biological effects when as-
sayed in vivo, they are rapidly metabolized and excreted,
limiting their utility. In contrast, compounds that lack a
flexible side chain (e.g., DCU) typically have physical
properties that are so poor that they show almost no
biological effects in vivo. Therefore, we have initiated
a study aimed at examining the utility of conformation-
ally restricted sEH inhibitors that also have a polar
secondary pharmacophore. Not only did we wish to
expand our current SAR by examining a broader range
Keywords: Soluble epoxide hydrolase; Enzyme inhibition;
Structure–activity relationships.
*
Corresponding author. Tel.: +1 530 752 7519; fax: +1 530 752
1537; e-mail: bdhammock@ucdavis.edu
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.07.009

P. D. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5212–5216
5213
Scheme 2. Reagents and conditions: (a) bromoalkane, KI, DMF,
K2CO3, 50 ꢁC, 12 h, 42–60%.
Figure 2. Piperidine-based scaffolds for rigid sEH inhibitors. Urea,
amide, and carbamate groups as the central pharmacophore have
yielded potent inhibitors with 1,3-substitution with aryl, alkyl,
cycloalkyl or as shown, adamantyl groups.9,10
Table 1. IC50 values for alkylpiperidine-based sEH inhibitors
n = 0
n = 1
O
of structures, but we also aimed to synthesize inhibitors
that were more ‘drug-like’ in structure.14–16 For the pur-
poses of this study, we utilized the scaffolds shown in
Figure 2 (compounds I and II) as platforms for inhibitor
development. These scaffolds were designed to not only
test the effectiveness of nitrogen containing secondary
pharmacophores, but also to compare the more rigid
scaffold I to the more flexible scaffold II.
N
N
Compound IC50 Compound IC50
(lM)a
n
N
H
H
R
(lM)a
R: H
I
0.30 II
4.2
3a
3.8
4a
3.9
3b
3c
3d
0.81 4b
2.6
The synthesis of scaffold I is illustrated in Scheme 1.
We began by reacting the commercially available
piperidine 1 and benzaldehyde to form the corre-
sponding Schiff base. Treatment with Boc anhydride
followed by acid hydrolysis afforded the protected
piperidine 2.17 Reaction of compound 2 with 1-ada-
mantylisocyanate followed by removal of the Boc
group with methanolic HCl afforded scaffold I with
a high overall yield. Scaffold II was synthesized in
an analogous manner.
1.2
4c
0.61
0.11
0.01 4d
a As determined via a kinetic fluorescent assay.22
We found that the piperidine nucleus was readily
alkylated as shown in Scheme 2 via reaction with an
alkylbromide in the presence of catalytic KI to give
the products in modest yield. Interestingly, we found
that the alkylpiperidines listed in Table 1 were, in gen-
eral, poor inhibitors. It is clear from these results that
a positively charged piperidine is not tolerated by the
enzyme. The exceptions to this trend are benzyl-substi-
tuted compounds 3d and 4d. These results suggest that
the destabilizing effects of a cation may be compensat-
ed for by aryl–aryl interactions within the active site.
It is also interesting to note that those compounds,
which are based on scaffold II show an SAR that var-
ies logically with substitution, whereas those which are
based on scaffold I do not.
diimide mediated coupling chemistry gave the desired
products in consistently higher yield (Scheme 3).18 As
can be seen in Table 2, conversion of the piperidine
nitrogen from an amine to an amide resulted in a
dramatic increase in potency. Inhibitors 5a–h show,
in general, better potency than inhibitors 6a–h. This
suggests that scaffold I may facilitate the formation
of a beneficial polar interaction between the piperidi-
nyl amide and a residue in the active site. In addi-
tion, the potency of inhibitors 5a–h and 6a–h seems
to rely more on the presence of the amide function-
ality rather than the actual identity of the amide
fragment.
Having firmly established the importance of the amide
functionality, we then investigated the effects of addi-
tional polar functionality on the potency of these inhib-
itors. Using the chemistry outlined in Scheme 3, a series
of methyl ester based inhibitors was generated (Table 3,
7a–e and 8a–e). As with the other amide-based inhibi-
tors described in this study, esters 7a–e and 8a–e do
not show a large variation in potency. There does seem
From the results described above, we postulated that
eliminating the possibility of nitrogen protonation
would result in an increase in potency. With this in
mind, we turned our attention to acylated piperidines.
We found that while acylation of scaffolds I and II
can be accomplished through the reaction of scaffolds
I or II with the requisite acid chloride, using carbo-
Scheme 1. Reagents and conditions: (a) toluene, benzaldehyde (1 equiv), reflux, Dean–Stark trap; (b) Boc anhydride, 0–25 ꢁC, 12 h; (c) KHSO4(aq)
(1 equiv); (d) adamant-1-yl isocyanate, THF, rt; (e) HCl/MeOH (4 equiv H+), 74% (overall).

5214
P. D. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5212–5216
Scheme 3. Reagents and conditions: (a) RCOCl, TEA, DCM, 6 h, 20–73%; (b) RCOOH, TEA, DMAP, EDCI, dichloromethane, 0 ꢁC to rt, 12 h,
75–99%.
Table 2. IC50 values for acylpiperidine-based sEH inhibitors
destabilizing any interactions that would support bind-
ing in the active site.
n = 0
n = 1
O
Compound IC50 Compound IC50
N
H
N
O
We were intrigued by the observation that trifluoroace-
tamides 5d and 6d showed a dramatic increase in poten-
cy over acetamides 5a and 6a. Using the published
crystal structure of human sEH with a bound urea-
based ligand (N-(4-iodophenyl)-N0-cyclohexyl urea,
CIU, PDB accession number 1VJ5),19 inhibitor 5d was
manually docked into the active site in order to further
understand the observed benefit of the trifluoroaceta-
mide functionality. As can be seen in Figure 3, com-
pound 5d is bound primarily through interactions with
Tyr381, Tyr465, and Asp333 with the urea pharmaco-
phore. In addition, the trifluoroacetamide functionality
of compound 5d can hydrogen bond with Gln382 via
the carbonyl and one of the fluorine atoms. It is reason-
able to postulate that the observed increase in potency
of trifluoroacetamides 5d and 6d over acetamides 5a
and 6a is mainly due to the presence of the additional
interactions between the –CF3 and Gln382. Docking
inhibitor 5d into the active site in the opposite orienta-
tion resulted in unfavorable steric interactions between
the adamantane and Met337, and removed any opportu-
nity for the trifluoroacetamide to participate in produc-
tive hydrogen bonding.
(nM)a
7.0
(nM)a
n
N
H
R
CH3
R:
5a
5b
5c
5d
5e
5f
6a
6b
6c
6d
6e
6f
5.0
8.7
6.7
1.8
3.2
7.6
5.4
7.3
3.2
2.6
CF3
1.1
1.3
N
1.2
N
5g
5h
1.7
6g
6h
N
At this point, we selected a small number of com-
pounds and screened them for oral bioavailability in
dogs.20 As can be seen in Table 4, not only do these
compounds have appreciable blood levels, but com-
pound 5a shows an almost 10-fold increase in AUC
as compared to AUDA. In addition, the observation
that blood level is clearly dependent on inhibitor
structure indicates that we may be able to optimize
a subset of the compounds reported herein for im-
proved oral availability.
2.1
a As determined via a kinetic fluorescent assay.22
to be a small advantage in having an aryl linker instead
of an alkyl one. Curiously, there is very little difference
in whether the methyl ester is in the ortho-, meta-, or
para-position, indicating that the active site is quite
tolerant of increased steric bulk.
In conclusion, we have reported a series of sEH
inhibitors that use a piperidine moiety to rigidify
their structure. A preliminary screen of inhibitor
potency against recombinant sEH reveals that simple
amide-based inhibitors are well tolerated. In contrast,
acid functionalized inhibitors show a distinct SAR,
which is consistently less potent than the correspond-
ing esters across both scaffolds. The data presented
clearly indicate the potential value of these and other
heterocyclic compounds as effective in vivo inhibitors
of sEH. We currently have detailed experiments
underway with the aim of determining the efficacy
and pharmacokinetic properties of these compounds.
The results of these studies will be reported in due
course.
Esters 7a–e and 8a–e were smoothly converted to the
corresponding acids (7f–j and 8f–j, respectively) by
reaction with methanolic KOH. The potency of the
resultant acids showed a remarkable dependence on
the position of the carboxylate, as shown by data in
Table 3. Conformational analysis (using CONFLEX,
as implemented in CAChe Workstation Pro 6.1, Fujit-
su Inc.) suggested that the carboxylate in compounds
7f could hydrogen bond to the urea in an intramolec-
ular fashion (data not shown). This implies that the
observed SAR trend for compound 7f–j and 8f–j
may be due, in part, to the ability of the inhibitor
to form intramolecular hydrogen bonds, thereby

P. D. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5212–5216
5215
Table 3. IC50 values for piperidine-based sEH inhibitors containing ester and acid functionalized amides
n = 0
n = 1
O
Compound
IC50
(nM)a
Compound
IC50
(nM)a
N
N
H
O
n
N
H
R
O
R:
7a
7b
7c
9.0
2.7
1.7
8a
8b
8c
6.2
3.4
1.8
O
O
O
O
O
O
O
7d
7e
7f
1.1
8d
8e
8f
4.1
1.5
O
O
1.1
O
2.5 · 102
1.7 · 102
OH
O
7g
7h
72
8g
8h
41
OH
O
HO
1.6 · 102
4.0 · 102
O
OH
O
7i
7j
10
8i
8j
43
OH
3.3
11.8
a As determined via a kinetic fluorescent assay.22
Table 4. Pharmacokinetic profile data for selected compounds as
obtained via oral dosing in a canine model
Compound
AUCa (·104 nM min)
5a
6a
3.7
0.55
0.65
0.25
0.061
0.033
0.33
0.47
0.31
5b
5c
5e
6e
5d
5f
AUDA
a Area under the curve, estimated from a plot of inhibitor plasma
concentration (nM) versus time (minutes) following an oral dose of
0.3 mg/kg of the indicated compounds in tristerate.20
Acknowledgments
The authors thank Dr. James Sanborn for many helpful
discussions. We also thank Dr. William Jewel and the
staff of the UC Davis Mass Spectrometry Center for
Figure 3. Compound 5d docked into the active site of human sEH.
Hydrogen bonds are indicated by the yellow lines. Tyr465 has been
removed for the sake of clarity.21

5216
P. D. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5212–5216
9. Morisseau, C.; Goodrow, M. H.; Dowdy, D.; Zheng, J.;
Greene, J. F.; Sanborn, J. R.; Hammock, B. D. Proc. Natl.
Acad. Sci. U.S.A. 1999, 96, 8849.
10. McElroy, N. R.; Jurs, P. C.; Morisseau, C.; Hammock, B.
D. J. Med. Chem. 2003, 46, 1066.
11. Kim, I. H.; Morisseau, C.; Watanabe, T.; Hammock, B.
D. J. Med. Chem. 2004, 47, 2110.
12. Kim, I. H.; Heirtzler, F. R.; Morisseau, C.; Nishi, K.;
Tsai, H. J.; Hammock, B. D. J. Med. Chem. 2005, 48,
3621.
the use of their mass spectrometer. P.D.J. was supported
by an NIH/NHLBI Ruth L. Kirchstein NRSA Grant
(F32 HL078096). This work was supported in part by
an NIEHS Grant (ES02710), an NIEHS Superfund
Grant (P42 ES04699), an NIH/NHLBI grant (R01
HL59699-06A1), and a Translational Technology Grant
from the UC Davis Medical Center.
13. Schmelzer, K. R.; Kubala, L.; Newman, J. W.; Kim, I. H.;
Eiserich, J. P.; Hammock, B. D. Proc. Natl. Acad. Sci.
U.S.A. 2005, 102, 9772.
14. Navia, M. A.; Chaturvedi, P. R. Drug Discovery Today
1996, 1, 179.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2006.07.009.
15. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P.
J. Adv. Drug Delivery Rev. 2001, 46, 3.
16. Veber, D. F.; Johnson, S. R.; Cheng, H. Y.; Smith, B. R.;
Ward, K. W.; Kopple, K. D. J. Med. Chem. 2002, 45,
2615.
References and notes
17. Sonda, S.; Kawahara, T.; Murozono, T.; Sato, N.; Asano,
K.; Haga, K. Bioorg. Med. Chem. 2003, 11, 4225.
18. Dhaon, M. K.; Olsen, R. K.; Ramasamy, K. J. Org.
Chem. 1982, 47, 1962.
1. Arand, M.; Grant, D. F.; Beetham, J. K.; Friedberg, T.;
Oesch, F.; Hammock, B. D. FEBS Lett. 1994, 338, 251.
2. Oesch, F. Xenobiotica 1973, 3, 305.
3. Zeldin, D. C.; Kobayashi, J.; Falck, J. R.; Winder, B. S.;
Hammock, B. D.; Snapper, J. R.; Capdevila, J. H. J. Biol.
Chem. 1993, 268, 6402.
4. Spector, A. A.; Fang, X.; Snyder, G. D.; Weintraub, N. L.
Prog. Lipid Res. 2004, 43, 55.
5. Yu, Z.; Xu, F.; Huse, L. M.; Morisseau, C.; Draper, A. J.;
Newman, J. W.; Parker, C.; Graham, L.; Engler, M. M.;
Hammock, B. D.; Zeldin, D. C.; Kroetz, D. L. Circ. Res.
2000, 87, 992.
6. Imig, J. D.; Zhao, X.; Capdevila, J. H.; Morisseau, C.;
Hammock, B. D. Hypertension 2002, 39, 690.
7. Imig, J. D.; Zhao, X.; Zaharis, C. Z.; Olearczyk, J. J.;
Pollock, D. M.; Newman, J. W.; Kim, I. H.; Watanabe,
T.; Hammock, B. D. Hypertension 2005, 46, 975.
8. Spiecker, M.; Liao, J. K. Arch. Biochem. Biophys. 2005,
433, 420.
19. Gomez, G. A.; Morisseau, C.; Hammock, B. D.; Chris-
tianson, D. W. Biochemistry 2004, 43, 4716.
20. Selected soluble epoxide hydrolase inhibitors were
prepared at 6 mg/mL in triglycerides. Each dog was
dosed orally with three different inhibitors per experi-
ment (dose = 0.3 mg/kg per inhibitor). Plasma samples
were collected at 0, 15, 30, 60, 120, 180, 240, 300, 360,
480, and 1440 min, and analyzed by LC/MS/MS
(Quattro Premier; Micromass, MA). The pharmacoki-
netic parameters were calculated with WinNonlin 5.0
(Pharsight, CA).
21. Humphrey, W.; Dalk, A.; Schulten, K. J. Mol. Graph.
1996, 14.1, 33.
22. Jones, P. D.; Wolf, N. M.; Morisseau, C.; Whetstone, P.;
Hock, B.; Hammock, B. D. Anal. Biochem. 2005, 343, 66.