A thiocarbamate inhibitor of endothelial lipase raises HDL cholesterol levels in mice

Article abstract of DOI:10.1016/j.bmcl.2013.02.113

By screening directed libraries of serine hydrolase inhibitors using the cell surface form of endothelial lipase (EL), we identified a series of carbamate-derived (EL) inhibitors. Compound 3 raised plasma HDL-C levels in the mouse, and a correlation was found between HDL-C and plasma compound levels. Spectroscopic and kinetic studies support a covalent mechanism of inhibition. Our findings represent the first report of EL inhibition as an effective means for increasing HDL-C in an in vivo model.

Full text of DOI:10.1016/j.bmcl.2013.02.113

Bioorganic & Medicinal Chemistry Letters 23 (2013) 2595–2597
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
A thiocarbamate inhibitor of endothelial lipase raises HDL
cholesterol levels in mice
⇑
M. N. Greco , M. A. Connelly, G. C. Leo, M. W. Olson, E. Powell, Z. Huang, M. Hawkins, C. Smith,
C. Schalk-Hihi, A. L. Darrow, H. Xin, W. Lang, B. P. Damiano, D. J. Hlasta
Janssen Research and Development, LLC, Welsh and McKean Roads, PO Box 776, Spring House, PA 19477, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
By screening directed libraries of serine hydrolase inhibitors using the cell surface form of endothelial
lipase (EL), we identified a series of carbamate-derived (EL) inhibitors. Compound 3 raised plasma
HDL-C levels in the mouse, and a correlation was found between HDL-C and plasma compound levels.
Spectroscopic and kinetic studies support a covalent mechanism of inhibition. Our findings represent
the first report of EL inhibition as an effective means for increasing HDL-C in an in vivo model.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 3 January 2013
Revised 15 February 2013
Accepted 25 February 2013
Available online 7 March 2013
Keywords:
Endothelial lipase
HDL cholesterol
Carbamate
Thiocarbamate
EL is a serine phospholipase that functions in the metabolism of
lipoproteins.1 Although EL is a member of the triglyceride lipase
family, it differs from other triglyceride lipases, such as lipoprotein
lipase (LPL) and hepatic lipase (HL), in its substrate preference for
phospholipids rather than triglycerides.1 Studies in mice have
shown that ELÀ/À (knockout) mice have a significant elevation in
HDL cholesterol relative to wild type mice, thus establishing that
EL plays a role in regulating HDL cholesterol in mice.2 Epidemiolog-
ical studies have shown that plasma HDL-C levels and cardiovascu-
lar disease are inversely related.3 In addition, EL is upregulated
under high inflammatory conditions and has been shown to have
negative effects at the vessel wall that further contribute to athero-
sclerosis.4 Therefore, EL inhibitors may represent potential thera-
peutic agents for the treatment of cardiovascular disease.5
A key strategic component of our drug discovery approach was
the identification of a tool compound suitable for assessment of
in vivo target validation, prior to commencement of lead
optimization.
(mouse EL IC50 = 0.68 0.45 lM) displayed sufficient potency and
stability to serve as a tool compound for in vivo evaluation.
Although it was not potent or selective enough to serve as a lead
compound (human HL IC50 = 1.36 1.26
lM; human LPL
IC50 = 1.69 1.86 M), we used 3 as a tool to get an early read on
l
the potential efficacy of an EL inhibitor and evaluated its effect
on plasma HDL-C levels in an in vivo model. Vehicle or 3 was
administered to C57BL/6J mice by the intraperitoneal route (five
doses over three days) and plasma HDL-C levels measured 1 h after
the last dose. Compound 3 at 100 mg/kg produced an increase in
plasma HDL-C of 37% (from 27 2 to 37 1 mg/dL) compared to
vehicle controls (Fig. 2). Because the HDL raising effect was not ro-
bust, we sought to identify a different animal model that might
produce a more robust HDL response with an EL inhibitor.
Treatment of C57BL/6 HLÀ/À (knockout) mice with an EL-specific
antibody was reported to produce a larger increase in plasma HDL-
C than was observed in C57BL/6 mice.6 Therefore, C57BL/6J HLÀ/À
mice were treated intraperitoneally with 3 or vehicle for three days
and the levels of plasma lipoproteins measured 1 hour after the
last dose. In HLÀ/À mice, 3 produced a robust increase in HDL-C
levels at the highest dose (Fig. 3A). Plasma HDL-C increased 23%
and 77% over vehicle controls at the 30 and 100 mg/kg dose,
respectively, although the changes at the 10 and 30 mg/kg doses
were not statistically significant. The plasma compound levels in-
creased in a dose related fashion (Fig. 3B). Further analysis of this
experiment and repeat experiments suggested that the plasma
Carbamate 1 (IC50 = 0.47 0.70 lM) was identified in a directed
high-throughput, cell-based screen of a serine hydrolase inhibitor li-
brary using cell surface bound mouse EL (Fig. 1). We addressed the
solution instability of 1 by focusing on analogs containing replace-
ments for the labile phenylcarbamate moiety. Although the corre-
sponding benzylcarbamate 2 was inactive, thiobenzylcarbamate 3
compound concentration needed to be greater than 3.0
der to obtain a statistically significant increase in HDL-C. In
lM in or-
⇑
Corresponding author. Tel.: +1 215 872 6695.
E-mail address: grecom97@gmail.com (M.N. Greco).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.02.113

2596
M. N. Greco et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2595–2597
Ph (CH2)6 NH
O
Ph (CH2)6 NH
O
F
Ph (CH2)6 NH
O
F
O
F
O
S
1
2
3
IC50 > 33 µM
IC50 = 0.68 µM
IC50 = 0.47 µM
HTS "hit" from hydrolase library
•unstable in pH 7.4 buffer
•stable in simulated gastric fluid, pH 1.1
Figure 1. Progression of a screening hit 1 to a lead 3.
of inhibition (Fig. S1, Supplementary data). 8 However, recovery of
enzyme activity did not occur via a rapid-reversible mechanism
and a koff-value could not be determined (Fig. S2, Supplementary
data). 9 In the course of optimizing 3, we identified a series of oxa-
diazole derivatives represented by 4, which proved to be a potent
EL inhibitor (IC50 = 0.023 0.006 lM). Because 4 was more potent
than 3, we thought it might have a higher affinity for the enzyme
and serve as a better tool for unequivocally determining the bind-
ing mode of this series. Therefore we conducted additional spectro-
scopic studies using 4.
We prepared the 13C labeled carbonyl analog 5 and examined
its NMR spectrum in the presence of EL.
Figure 2. Effects of compound 3 on HDL-C in C57BL/6 mice after intraperitoneal
administration. Bars represent means SD (n = 8 per group; ⁄⁄p <0.01 relative to
vehicle group).
Ph
O
H
N
N
F
O
S
N
(CH2)5
O
individual animals, plasma levels of compound 3 correlated with
elevated plasma HDL-C and increased plasma phospholipids
(Fig. 3C and D), both of which are pharmacodynamic markers of
EL inhibition in vivo. Because 3 inhibited both EL and LPL, triglyc-
eride levels also increased significantly at the 30 and 100 mg/kg
doses (data not shown). Given that we used HLÀ/À mice, and inhi-
bition of LPL would most likely result in a reduction in HDL-C,7 we
were fairly confident that the resulting increase in HDL-C in these
studies was primarily due to inhibition of EL. Therefore, these
in vivo results were promising enough to prompt us to continue
optimizing this series in search of a more drug-like molecule.
We next examined the mechanism of EL inhibition by 3. The
kinetics of the reaction were consistent with time-dependent onset
F
4
EL IC50 = 23 nM
Ph
H
O
N
F
O
N
(CH2)5
S
13C
N
O
F
5
A stable tetrahedral enzyme–inhibitor complex such as 6
(Fig. 4) would be expected to cause a large upfield shift of the
Figure 3. Effects of compound 3 in C57BL/6 HLÀ/À mice after intraperitoneal administration, (A) plasma HDL-C, (B) plasma compound levels, (C) correlation between plasma
HDL-C and compound levels and (D) correlation between plasma phospholipids and compound levels. Bars represent means SD (n = 8 per group; ⁄⁄⁄p <0.001 relative to
vehicle group).

M. N. Greco et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2595–2597
2597
H2N
F
O
(CH2)5
R =
F
Ser149
O
Enz
Enz
Ph
Ph
N
Ph
N
EL
O
O
O
O
O
N
S
HS
+
N
S
13C
N
13C
R
13C
O
N
R
R
O
OH
4
6
7
8
Figure 4. Proposed pathway for EL inhibition.
Acknowledgments
The authors would like to thank Sharon L. Burke and Wendy
Sun for their technical expertise. We thank Dr. George Furst for
assistance in using the Bruker 500 with the carbon detect cryo-
probe which is part of the University of Pennsylvania Chemistry
Department’s NMR facility. We also thank Bruce Maryanoff and
Matthew Todd for helpful discussions.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.02.
113.
References and notes
1. (a) Duffy, D.; Rader, D. J. Nat. Rev. Cardiol. 2009, 6, 455; (b) Yasuda, T.; Ishida, T.;
Rader, D. J. Circ. J. 2010, 74, 2263.
Figure 5. Left: 125 MHz 13C NMR spectra of EL with and without 5 obtained using a
carbon detect cryoprobe. Right: 470 MHz 19F NMR of bound EL:5 complex and free
5.
2. (a) Jaye, M.; Lynch, K. J.; Krawiec, J.; Marchadier, D.; Maugeais, C.; Doan, K.;
South, V.; Amin, D.; Perronwe, M.; Rader, D. J. Nat. Genet. 1999, 21, 424; (b)
Ishida, T.; Choi, S.; Kundu, R. K.; Hirata, K.; Rubin, E. M.; Cooper, A. D.;
Quertermous, T. J. Clin. Invest. 2003, 111, 347; (c) Edmondson, A. C.; Brown, R. J.;
Kathiresan, S.; Cupples, L. A.; Demissie, S.; Manning, A. K.; Jensen, M. K.; Rimm,
E. B.; Wang, J.; Rodrigues, A.; Bamba, V.; Khetarpal, S. A.; Wolfe, M. L.;
Derohannessian, S.; Li, M.; Reilly, M. P.; Aberle, J.; Evans, D.; Hegele, R. A.;
Rader, D. J. J. Clin. Invest. 2009, 1042, 119; (d) Edmondson, A. C.; Braund, P. S.;
Stylianou, I. M.; Khera, A. V.; Nelson, C. P.; Wolfe, M. L.; Derohannessian, S. L.;
Keating, B. J.; Qu, L.; He, J.; Tobin, M. D.; Tomaszewski, M.; Baumert, J.; Klopp,
N.; Doring, A.; Thorand, B.; Li, M.; Reilly, M. P.; Koenig, W.; Samani, N. J.; Rader,
D. J. Circ. Cardiovasc. Genet. 2011, 4, 145.
13C labeled carbon relative to unbound 5. The 13C NMR of a solution
of 5 and EL (2:1 ratio of 5 to EL) showed a sharp signal at
160.5 ppm corresponding to the unbound inhibitor and a line
broadened signal at 160.2 ppm corresponding to the bound ligand.
However, a carbon signal indicative of a stable tetrahedral en-
zyme–inhibitor complex such as 6 was not observed (Fig. 5). Sim-
ilarly, the 19F NMR of this solution displayed a line broadened
signal at À110 ppm consistent with the EL bound inhibitor 5,
and a sharp signal for the unbound inhibitor at À120 ppm
(Fig. 5); only one adduct was detected. Thus, NMR analysis does
support formation of a covalent enzyme–inhibitor complex. Fur-
thermore, LCMS–MS analysis of this sample detected the thiol
leaving group 7, which is the expected by-product of formation
of the covalent acyl-enzyme complex 8 (Fig. S3, Supplementary
data). These results support a proposed pathway (Fig. 4) in which
8 is the most likely covalent adduct formed. Overall, our findings
are consistent with the covalent mode of inhibition observed for
carbamate-based inhibitors of fatty acid amide hydrolase (FAAH).10
In summary, we have identified a potent series of EL inhibitors
based on a thiocarbamate scaffold. Kinetic and spectroscopic stud-
ies indicate a covalent mechanism of inhibition for the series. The
tool compound 3 raised HDL-C plasma levels in both C57Bl/6J and
C57Bl/6 HLÀ/À mice. This work is the first report to demonstrate
that an EL inhibitor raises HDL-C and phospholipid plasma levels
in vivo. The results of our studies support the potential for EL
inhibitors as effective treatment for raising plasma HDL-C in the
treatment of atherosclerosis.
3. (a) Wilson, P. W.; Abbott, R. D.; Castelli, W. P. Arteriosclerosis 1988, 8, 737; (b)
Abbott, R. D.; Wilson, P. W.; Kannel, W. B.; Castelli, W. P. Arteriosclerosis 1988, 8,
207; (c) Boden, W. E. Am. J. Cardiol. 2000, 86, 19L.
4. (a) Ishida, T.; Choi, S. Y.; Kundu, R. K.; Spin, J.; Yamashita, T.; Hirata, K.; Kojima,
Y.; Yokoyama, M.; Cooper, A. D.; Quertermou, T. J. Biol. Chem. 2004, 279, 45085;
(b) Yu, K. C.; David, C.; Kadambi, S.; Stahl, A.; Hirata, K.; Ishida, T.; Quertermous,
T.; Cooper, A. D.; Choi, S. Y. J. Lipid Res. 2004, 45, 1614; (c) Hirata, K.; Ishida, T.;
Matsushita, H.; Tsao, P. S.; Quertermous, T. Biochem. Biophys. Res. Commun.
2000, 272, 90; (d) Broedl, U. C.; Jin, W.; Rader, D. J. Trends Cardiovasc. Med. 2004,
14, 202; (e) Badellino, K. O.; Wolfe, M. L.; Reilly, M. P.; Rader, D. J. PLoS Med.
2006, 3, e22; (f) Badellino, K. O.; Wolfe, M. L.; Reilly, M. P.; Rader, D. J.
Circulation 2008, 117, 678; (g) Brown, R. J.; Rader, D. J. Curr. Drug Targets 2007,
8, 1307; (h) Huang, J.; Qian, H. Y.; Li, Z. Z.; Zhang, J. M.; Wang, S.; Tao, Y.; Gao, Y.
L.; Yin, C. Q.; Que, B.; Sun, T.; Zhao, Z. Y.; Li, Z. Transl. Res. 2010, 156, 1.
5. DeSantis, P.; Coleman, T.; Schiekofer, S.; Nawroth, P. P.; Schlimmer, P.;
Schneider, J. G. Mini-Rev. Med. Chem. 2008, 8, 619.
6. Jin, W.; Millar, J. S.; Broedl, U.; Glick, J. M.; Rader, D. J. J. Clin. Invest. 2003, 111,
357.
7. Weinstock, P. H.; Bisgaier, C. L.; Aalto-Setälä, K.; Radner, H.; Ramakrishnan, R.;
Levak-Frank, S.; Essenburg, A. D.; Zechner, R.; Breslow, J. L. J. Clin. Invest. 1995,
96, 2555.
8. Morrison, J. F.; Walsh, C. T. Adv. Enzymol. Relat. Areas Mol. Biol. 1988, 61, 201.
9. Glick, B. R.; Brubacher, J. L.; Leggett, D. J. Can. J. Biochem. 1978, 56, 1055.
10. For a thorough review of carbamate-based FAAH inhibitors, see: Otrubova, K.;
Ezzili, C.; Boger, D. L. Bioorg. Med. Chem. Lett. 2011, 21, 4674.