The putative endocannabinoid transport blocker LY2183240 is a potent inhibitor of FAAH and several other brain serine hydrolases

Article abstract of DOI:10.1021/ja062999h

How lipid transmitters move within and between cells to communicate signals remains an important and largely unanswered question. Integral membrane transporters, soluble lipid-binding proteins, and metabolic enzymes have all been proposed to collaborative

Full text of DOI:10.1021/ja062999h

Published on Web 07/06/2006
The Putative Endocannabinoid Transport Blocker LY2183240
Is a Potent Inhibitor of FAAH and Several Other Brain Serine
Hydrolases
Jessica P. Alexander and Benjamin F. Cravatt*
Contribution from the Skaggs Institute for Chemical Biology and Departments of
Cell Biology and Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road,
La Jolla, California 92037
Received April 29, 2006; E-mail: cravatt@scripps.edu
Abstract: How lipid transmitters move within and between cells to communicate signals remains an important
and largely unanswered question. Integral membrane transporters, soluble lipid-binding proteins, and
metabolic enzymes have all been proposed to collaboratively regulate lipid signaling dynamics in vivo.
Assignment of the relative contributions made by each of these classes of proteins requires selective
pharmacological agents to perturb their individual functions. Recently, LY2183240, a heterocyclic urea
inhibitor of the putative endocannabinoid (EC) transporter, was shown to disrupt the cellular uptake of the
lipid EC anandamide and promote analgesia in vivo. Here, we show that LY2183240 is a potent, covalent
inhibitor of the EC-degrading enzyme fatty acid amide hydrolase (FAAH). LY2183240 inactivates FAAH by
carbamylation of the enzyme’s serine nucleophile. More global screens using activity-based proteomic
probes identified several additional serine hydrolases that are also inhibited by LY2183240. These results
indicate that the blockade of anandamide uptake observed with LY2183240 may be due primarily to the
inactivation of FAAH, providing further evidence that this enzyme serves as a metabolic driving force that
promotes the diffusion of anandamide into cells. More generally, the proteome-wide target promiscuity of
LY2183240 designates the heterocyclic urea as a chemotype with potentially excessive protein reactivity
for drug design.
Introduction
role for FAAH in the control of fatty acid amide function in
vivo. For example, the inactivation of FAAH by genetic9 or
The fatty acid amide class of lipid transmitters modulates a
wide range of physiological and pathological processes.1,2
Prominent bioactive fatty acid amides include the endogenous
cannabinoid N-arachidonoyl ethanolamine (anandamide),3 the
sleep-inducing substance 9Z-octadecenamide (oleamide),4 and
the anti-inflammatory substance N-palmitoylethanolamine (PEA).5
In contrast to classical neurotransmitters, which are stored in
vesicles prior to release, lipid messengers such as the fatty acid
amides are thought to be enzymatically produced “on-demand”
(i.e., at the moment of their intended action).6 Termination of
fatty acid amide signaling also requires the action of enzymes
to degrade these molecules. Prominent among these catabolic
enzymes is the integral membrane protein fatty acid amide
hydrolase (FAAH).7,8 Several lines of evidence support a key
pharmacological10 techniques leads to highly elevated levels of
anandamide and related fatty acid amides in the nervous system
and concomitant CB1-dependent analgesic and anxiolytic
phenotypes.
The localization of FAAH to intracellular membrane com-
partments of neurons11,12 has raised a provocative question: how
are fatty acid amides such as anandamide delivered to the
enzyme for degradation? One model invokes the action of a
plasma membrane-associated transporter that promotes the
uptake of extracellular anandamide by facilitated diffusion.13
Evidence for the existence of this putative transporter includes
the saturability and temperature dependence of anandamide
uptake, as well as its disruption by pharmacological agents.
However, each of these criteria has also been subject to
alternative interpretations.14,15 For example, anandamide binds
(1) Di Marzo, V. Biochim. Biophys. Acta 1998, 1392, 153-175.
(2) Schmid, H. H.; Berdyshev, E. V. Prostaglandins Leukotrienes Essent. Fatty
Acids 2002, 66, 363-376.
(9) Cravatt, B. F.; Demarest, K.; Patricelli, M. P.; Bracey, M. H.; Giang, D.
K.; Martin, B. R.; Lichtman, A. H. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,
9371-9376.
(3) Devane, W. A.; Hanus, L.; Breuer, A.; Pertwee, R. G.; Stevenson, L. A.;
Griffin, G.; Gibson, D.; Mandelbaum, A.; Etinger, A.; Mechoulam, R.
Science 1992, 258, 1946-1949.
(10) Kathuria, S.; Gaetani, S.; Fegley, D.; Valino, F.; Duranti, A.; Tontini, A.;
Mor, M.; Tarzia, G.; La Rana, G.; Calignano, A.; Giustino, A.; Tattoli,
M.; Palmery, M.; Cuomo, V.; Piomelli, D. Nat. Med. 2003, 9, 76-81.
(11) Egertova, M.; Giang, D. K.; Cravatt, B. F.; Elphick, M. R. Proc. R. Soc.
London, Ser. B 1998, 265, 2081-2085.
(4) Cravatt, B. F.; Prospero-Garcia, O.; Siuzdak, G.; Gilula, N. B.; Henriksen,
S. J.; Boger, D. L.; Lerner, R. A. Science 1995, 268, 1506-1509.
(5) Lambert, D. M.; Vandevoorde, S.; Jonsson, K. O.; Fowler, C. J. Curr. Med.
Chem. 2002, 9, 663-674.
(6) Marsicano, G.; et al. Science 2003, 302, 84-88.
(7) Cravatt, B. F.; Giang, D. K.; Mayfield, S. P.; Boger, D. L.; Lerner, R. A.;
Gilula, N. B. Nature 1996, 384, 83-87.
(12) Tsou, K.; Nogueron, M. I.; Muthian, S.; Sanudo-Pena, M. C.; Hillard, C.
J.; Deutsch, D. G.; Walker, J. M. Neurosci. Lett. 1998, 254, 137-140.
(13) Beltramo, M.; Stella, N.; Calignano, A.; Lin, S. Y.; Makriyannis, A.;
Piomelli, D. Science 1997, 277, 1094-1097.
(8) McKinney, M. K.; Cravatt, B. F. Annu. ReV. Biochem. 2005, 74, 411-
432.
(14) Patricelli, M. P.; Cravatt, B. F. Vitam. Horm. 2001, 62, 95-131.
9
10.1021/ja062999h CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 9699-9704
9699

A R T I C L E S
Alexander and Cravatt
Figure 2. LY2183240 is a potent inhibitor of FAAH. (A) TLC plate
showing conversion of 14C-oleamide to 14C-oleic acid by FAAH(+/+) brain
membranes and its blockade by LY2183240 (1 µM). Control reactions with
brain tissue from FAAH (-/-) mice confirm that the observed activity is
due to FAAH. (B) Concentration-dependent inhibition of FAAH by
LY2183240. From this curve, an estimated IC50 value of 12.4 nM (8.3-
18.6 nM, 95% confidence limits) was measured.
Figure 1. Structures of the putative anandamide transport blocker
LY2183240 and the FAAH inhibitor URB597.
to culture dishes without cells, especially in the absence of serum
albumin, and this process is saturable, temperature-dependent,
and sensitive to chemical inhibitors.16,17 Likewise, most of the
transport inhibitors described to date also block FAAH,18-20
suggesting that this enzyme may create a metabolic driving force
for the cellular uptake of anandamide by simple diffusion.18,21
Consistent with this model, cells that express high levels of
FAAH show greatly accelerated rates of anandamide uptake.22-24
Other studies have indicated that the cellular uptake of free
anandamide can appear saturable due to the presence of an
unstirred water layer adjacent to the cell membrane that impedes
the uptake of this lipid.21,25
The molecular characterization of a putative anandamide
transporter would be facilitated by the creation of potent and
selective agents that perturb its function. Toward this end, a
novel blocker of anandamide uptake, LY2183240 (Figure 1),
was recently described that displays unprecedented potency (IC50
value of 270 pM in a cellular assay).24 LY2183240 also showed
impressive pharmacological activity in vivo, raising anandamide
levels and reducing pain responses in the formalin test.24 Unlike
most transport inhibitors, LY2183240 is not a structural analogue
of anandamide, suggesting that it might avoid interactions with
other natural protein partners of this lipid (e.g., cannabinoid
receptors, FAAH). On the other hand, LY2183240 bears some
structural resemblance to certain FAAH inhibitors, such as
URB597 (Figure 1), including a biraryl substituent and a
potentially reactive carbonyl. Here, we show that LY2183240
is, in fact, an extremely potent inhibitor of FAAH both in vitro
and in vivo. LY2183240 inactivates FAAH by covalently
labeling the enzyme’s serine nucleophile. Functional proteomic
screens identified several additional serine hydrolases that were
also inactivated by LY2183240, indicating that the compound
possesses rather promiscuous activity against this large and
diverse enzyme class.
Results
LY2183240 Is a Potent, Covalent Inhibitor of FAAH and
Several Other Brain Serine Hydrolases. In the original report
describing LY2183240 as a blocker of anandamide transport,
this compound was shown to bind to cells that lack FAAH,
suggesting that it may target a distinct protein(s).24 However,
direct interactions between LY2183240 and FAAH were not
investigated. To address this important question, we incubated
brain membranes with LY2183240 for 10 min and then tested
for residual FAAH activity using a 14C-substrate assay.
LY2183240 potently inhibited FAAH activity with an IC50 value
of 12.4 nM (Figure 2). This value approaches the estimated
concentration of FAAH protein present in the brain proteome,
as judged by quantitative labeling with a fluorophosphonate-
rhodamine (FP-rhodamine) active site-directed probe,26,27 sug-
gesting that LY2183240 may titrate the enzyme in these
reactions.
Considering that LY2183240 possesses a potentially reactive
heterocyclic urea group, we next tested if this agent might
inactivate FAAH by a covalent mechanism. Purified, recom-
binant rat FAAH protein (∼23 µM) was incubated with or
without LY2183240 (35 µM, ∼1.5 equiv) for 60 min, after
which the reaction mixtures were subjected to tryptic digestion
and MALDI-TOF mass spectrometry (MS) analysis. A new
mass signal was observed for the LY2183240-treated sample
that corresponded to the mass of the FAAH tryptic peptide
amino acids (AA) 213-243, which contains the FAAH nucleo-
phile S241,28,29 modified by the C(O)N-dimethyl substituent of
the inhibitor (Figure 3). These results indicate that LY2183240
covalently inhibits FAAH by carbamylation of the enzyme’s
serine nucleophile, analogous to the mechanism of action of
URB597 and other carbamate FAAH inhibitors.30 Consistent
with an irreversible mechanism of inhibition, LY2183240-treated
(15) Glaser, S. T.; Kaczocha, M.; Deutsch, D. G. Life Sci. 2005, 77, 1584-
1604.
(16) Ortega-Gutierrez, S.; Hawkins, E. G.; Viso, A.; Lopez-Rodriguez, M. L.;
Cravatt, B. F. Biochemistry 2004, 43, 8184-8190.
(17) Karlsson, M.; Pahlsson, C.; Fowler, C. J. Eur. J. Pharm. Sci. 2004, 22,
181-189.
(18) Glaser, S. T.; Abumrad, N.; Fatade, F.; Kaczocha, M.; Studholme, K. M.;
Deutsch, D. G. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 4269-4274.
(19) Vandevoorde, S.; Fowler, C. J. Br. J. Pharmacol. 2005, 145, 885-893.
(20) Fowler, C. J.; Tiger, G.; Ligresti, A.; Lopez-Rodriguez, M. L.; Di Marzo,
V. Eur. J. Pharmacol. 2004, 492, 1-11.
(21) Kaczocha, M.; Hermann, A.; Glaser, S. T.; Bojesen, I. N.; Deutsch, D. G.
J. Biol. Chem. 2006, 281, 9066-9075.
(22) Deutsch, D. G.; Glaser, S. T.; Howell, J. M.; Kunz, J. S.; Puffenbarger, R.
A.; Hillard, C. J.; Abumrad, N. J. Biol. Chem. 2001, 276, 6967-6973.
(23) Day, T. A.; Rakhshan, F.; Deutsch, D. G.; Barker, E. L. Mol. Pharmacol.
2001, 59, 1369-1375.
(26) Patricelli, M. P.; Giang, D. K.; Stamp, L. M.; Burbaum, J. J. Proteomics
2001, 1, 1067-1071.
(27) Jessani, N.; Liu, Y.; Humphrey, M.; Cravatt, B. F. Proc. Natl. Acad. Sci.
U.S.A. 2002, 99, 10335-10340.
(24) Moore, S. A.; Nomikos, G. G.; Dickason-Chesterfield, A. K.; Schober, D.
A.; Schaus, J. M.; Ying, B. P.; Xu, Y. C.; Phebus, L.; Simmons, R. M.; Li,
D.; Iyengar, S.; Felder, C. C. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
17852-17857.
(28) Patricelli, M. P.; Lovato, M. A.; Cravatt, B. F. Biochemistry 1999, 38,
9804-9812.
(29) Bracey, M. H.; Hanson, M. A.; Masuda, K. R.; Stevens, R. C.; Cravatt, B.
F. Science 2002, 298, 1793-1796.
(25) Bojesen, I. N.; Hansen, H. S. J. Lipid Res. 2006, 47, 561-570.
(30) Alexander, J. P.; Cravatt, B. F. Chem. Biol. 2005, 12, 1179-1187.
9
9700 J. AM. CHEM. SOC. VOL. 128, NO. 30, 2006

LY2183240 Is a Covalent Inhibitor of FAAH
A R T I C L E S
Figure 3. LY2183240 is a covalent inhibitor of FAAH. (A) Predicted mode of irreversible inactivation of FAAH by LY2183240, involving carbamylation
of the enzyme’s serine nucleophile (S241). (B) MALDI-MS mapping of a tryptic digest of purified recombinant FAAH, highlighting the tryptic peptide that
contains S241 (AA213-243). (C) MALDI-MS mapping of a tryptic digest of FAAH pretreated with LY2183240, highlighting a new mass that corresponds
to the AA213-243 peptide modified by one C(O)N-dimethyl molecule.
FAAH samples showed no evidence of recovery of activity for
up to 24 h following gel filtration to remove free inhibitor (data
not shown). Attempts to measure a k2/Ki value for LY2183240
were complicated by the extremely rapid rate of inactivation
(complete inactivation of FAAH with a 30-s preincubation),
even at low inhibitor/FAAH ratios (10:1).
labeling of several brain serine hydrolases with potencies
comparable to that observed for FAAH (Figure 4A). These
enzymes included the uncharacterized hydrolase KIAA1363,33
as well as multiple unidentified enzymes with molecular masses
between 25 and 35 kDa. We next pursued the characterization
of enzymes targeted by LY2183240 in vivo.
A search of the scientific and patent literature identified
multiple reports of heterocyclic ureas as inhibitors of serine
hydrolases, including peptidases31 and hormone-sensitive li-
pase.32 These previous studies led us to consider whether
LY2183240 is a selective FAAH inhibitor or, alternatively, a
more general inactivator of enzymes in the serine hydrolase
superfamily. This question was addressed by treating brain
proteomes with increasing concentrations of LY2183240 fol-
lowed by the activity-based serine hydrolase probe FP-rho-
damine.26,27 In this competitive activity-based protein profiling
(ABPP) screen, the selectivity of an inhibitor for members of
the serine hydrolase class is determined by measuring reductions
in FP-rhodamine labeling intensity for many enzymes in
parallel.33 Remarkably, LY2183240 blocked the FP-rhodamine
LY2183240 Is an Inhibitor of FAAH and Several Other
Serine Hydrolases in Vivo. LY2183240 has been reported at
10 mg/kg (intraperitoneal, ip) to produce analgesic effects in
the formalin test of noxious pain, a phenotype that was correlated
with a significant elevation in brain anandamide levels.24 These
molecular and behavioral effects are similar to those observed
in FAAH-disrupted animals,9,34 suggesting that pharmacologi-
cally active doses of LY2183240 may inhibit FAAH in vivo.
We tested this premise by treating mice with LY2183240 (10
mg/kg, ip) for 90 min and then sacrificing these animals and
characterizing their brain serine hydrolase activity profiles by
labeling with FP-rhodamine. Parallel studies were conducted
with URB597 (Figure 1, 10 mg/kg, ip), a covalent carbamate
inhibitor of FAAH that shows good selectivity for this enzyme
(31) Maezaki, H.; Ichikawa, T.; Ikedo, K. Jpn. Kokai Tokkyo Koho, Patent JP
2003327532, 2003.
(32) Ebdrup, S.; Sorensen, L. G.; Olsen, O. H.; Jacobsen, P. J. Med. Chem.
2004, 47, 400-410.
(33) Leung, D.; Hardouin, C.; Boger, D. L.; Cravatt, B. F. Nat. Biotechnol.
2003, 21, 687-691.
(34) Lichtman, A. H.; Shelton, C. C.; Advani, T.; Cravatt, B. F. Pain 2004,
109, 319-327.
9
J. AM. CHEM. SOC. VOL. 128, NO. 30, 2006 9701

A R T I C L E S
Alexander and Cravatt
Figure 4. LY2183240 inhibits several brain serine hydrolases in vitro and in vivo. (A) Competitive ABPP profiles of brain membranes treated with LY2183240
(0.001-10 µM, 10 min) in vitro, followed by the serine hydrolase activity-based probe FP-rhodamine (0.1 µM, 10 min). The FP-labeling of several enzymes
was blocked in a concentration-dependent manner by LY2183240, including FAAH, the uncharacterized hydrolase KIAA1363,33 and multiple proteins
between the molecular masses of 25-35 kDa (arrows). (B) Competitive ABPP profiles of brains isolated from mice treated with LY2183240 or URB597
(10 mg/kg, ip, 90 min). The FP-labeling of several enzymes was blocked by in vivo treatment with LY2183240, including FAAH, whose identity was
confirmed by comparison with in vivo treatment of FAAH (-/-) mice, and a set of 25-35 kDa proteins (arrows). In contrast, in vivo treatment with
URB597 selectively blocked the FP-labeling of FAAH.
Table 1. Levels of Representative Serine Hydrolase Activities in
Brains from Mice Treated with LY2183240 (10 mg/kg, ip), URB597
in brain.30 In vivo administration of LY2183240 was found to
inactivate FAAH and multiple other serine hydrolases that reside
in the molecular mass range of 25-35 kDa (Figure 4B, arrows).
In contrast, URB597 selectively inactivated FAAH in brain
tissue.
(10 mg/kg, ip), or Vehicle (Saline with 5% Ethanol, 5% Emulphor)
spectral countsa
control (C)
URB597
ave SE
LY2183240 (L)
relative activity
L/C
ave
SE
ave
SE
Identification and Characterization of Serine Hydrolase
Targets of LY2183240. To identify the enzymes inactivated
by LY2183240 in vivo, we analyzed tissues from vehicle- and
inhibitor-treated mice using an advanced functional proteomic
platform termed ABPP-multidimensional protein identification
technology (ABPP-MudPIT).35 In ABPP-MudPIT, proteomes
are treated with a biotinylated activity-based probe (in this case,
FP-biotin36), and probe-labeled enzymes are enriched by avidin
chromatography and subjected to on-bead trypsin digestion and
multidimensional LC-MS/MS analysis. Spectral counting
methods are then used to quantify the relative levels of enzyme
activities in different proteomic samples. ABPP-MudPIT analy-
sis identified 40 serine hydrolases in mouse brain (Supporting
Information Table 1). A number of these enzymes were partially
or completely inhibited by LY2183240 in vivo, as judged by a
reduction in spectral counts of greater than 2.5-fold relative to
vehicle-treated controls (Table 1). These LY2183240-inhibited
enzymes included FAAH (100% inhibited), R/â-hydrolase 6
(Abh6, >90% inhibited), and monoacylglycerol lipase (MAG
lipase, >60% inhibited). In contrast, URB597 selectively
inactivated FAAH in brain (Table 1).
FAAHb
23
55
11
2
3
3
4
6
0
0
0
3
2
3
6
0
2
1
3
6
8
0
Abh6c
62 10
9
20
32 10
0.06
0.18
0.25
0.32
0.37
proteasome â1d
2
5
PLA2 group VIIe 12
Wbscr21f
19
73
MAG lipaseg
14 120 15
27
a Serine hydrolase activities were measured by ABPP-MudPIT using
spectral counting, as described previously.35 Only those hydrolases with at
least 10 spectral counts in control proteomes were considered candidates
for inhibition by LY2183240 or URB597. Ave ) average of 3 independent
experiments/group. SE ) standard error. For a full list of serine hydrolases
identified by ABPP-MudPIT, see Supporting Information Table 1. b Fatty
acid amide hydrolase. c R/â-Hydrolase domain containing 6. d Proteasome
subunit â-type 1. e Phospholipase A2 group VII. f Williams-Beuren syn-
drome critical region protein 21. g Monoacylglycerol lipase.
Figure 5. Inactivation of recombinantly expressed serine hydrolases by
LY2183240. Two in vivo targets (Abh6 and MAG lipase) and one in vitro
target (KIAA1363) of LY2183240 were recombinantly expressed in COS-7
cells and shown by competitive ABPP to be inactivated by LY2183240
(see Table 2 for a list of calculated IC50 values). For MAG lipase and
KIAA1363, 5000-5 nM LY2183240 (represented in the figure by a
descending slope) and 0.6 and 0.2 mg/mL of transfected cell proteome were
used in the assays; for Abh6, which proved to be the most potently inhibited
target, 50-0.1 nM LY2183240 and 0.01 mg/mL of transfected cell proteome
were used to avoid enzyme titration.
Recombinant expression of Abh6 and MAG lipase in COS-7
cells confirmed that these enzymes were targets of LY2183240
(Figure 5), which inhibited their FP-rhodamine labeling with
IC50 values of 0.09 and 5.3 nM, respectively (Table 2). Both of
these enzymes migrated as 30-35 kDa proteins by SDS-PAGE,
suggesting that they corresponded to the LY2183240 targets
observed in this molecular mass range by gel-based ABPP
(Figure 4). In contrast to these enzymes, which were inactivated
by LY2183240 both in vitro and in vivo, KIAA1363 was only
inhibited by LY2183240 in vitro (compare parts A and B of
Figure 4). Why KIAA1363 shows resistance to LY2183240 in
vivo remains unclear, especially considering that LY2183240
(35) Jessani, N.; Niessen, S.; Wei, B. Q.; Nicolau, M.; Humphrey, M.; Ji, Y.;
Han, W.; Noh, D. Y.; Yates, J. R., 3rd; Jeffrey, S. S.; Cravatt, B. F. Nat.
Methods 2005, 2, 691-697.
(36) Liu, Y.; Patricelli, M. P.; Cravatt, B. F. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 14694-14699.
9
9702 J. AM. CHEM. SOC. VOL. 128, NO. 30, 2006

LY2183240 Is a Covalent Inhibitor of FAAH
A R T I C L E S
Table 2. IC50 Values for the Inhibition of Serine Hydrolases by
of FAAH. Our results argue that LY2183240 does not qualify
as such an agent, as this compound inhibits FAAH in vitro and
in vivo at pharmacologically active doses.
LY2183240 as Measured by Competitive ABPPa
enzyme
source
IC50 (nM)
95% confidence
FAAH
brain membrane
13
5.3
0.09
8.2
8.0-20
3.9-7.3
0.07-0.10
6.1-11
LY2183240 proved to be a remarkably potent inactivator of
FAAH (IC50 ) 13 nM), initially suggesting that it might serve
as a useful pharmacological tool for studying this enzyme.
However, competitive ABPP studies revealed that LY2183240
also inhibited several other brain serine hydrolases with IC50
values in the low nanomolar range. Additional in vivo targets
of LY2183240 included lipid-metabolizing enzymes, such as
MAG lipase, which has been proposed to regulate the endocan-
nabinoid 2-arachidonoyl glycerol,40 and uncharacterized hydro-
lases such as Abh6. Assuming that LY2183240 inactivates each
of these enzymes by the same mechanism as FAAH (i.e.,
carbamylation of the serine nucleophile), these data suggest that
the heterocyclic urea group may display excessive inherent
reactivity for use in the design of selective pharmacological
agents.
MAG lipase
Abh6
COS-7 transfection
COS-7 transfection
COS-7 transfection
(brain membrane)
KIAA1363b
(12)
(7.4-20)
a Note that because LY2183240 appears to act as an irreversible inhibitor,
the reported IC50 values are dependent on the pre-incubation time of the
assay (10 min). b IC50 values are shown for KIAA1363 in both transfected
and native (brain membrane) proteomes.
inactivated this enzyme in vitro with an IC50 value similar to
those observed for MAG lipase and FAAH (Table 2). Regard-
less, these data underscore the value of profiling inhibitors in
living systems, where potential issues such as the cellular and
subcellular distribution of these pharmacological agents, as well
as their metabolism, are taken into account.
The broad target selectivity of LY2183240 might also explain
the previous finding that this agent binds to cells that lack
FAAH.24 Perhaps other serine hydrolases sensitive to LY2183240
inactivation are responsible for these FAAH-independent bind-
ing events. More generally, our competitive ABPP studies of
LY2183240 underscore the importance of defining proteome-
wide selectivity patterns for bioactive small molecules to
illuminate potentially unanticipated targets of these agents.
Indeed, none of the additional enzymes targeted by LY2183240
display any discernible sequence homology with FAAH, sug-
gesting that their common inhibitor sensitivity profiles derive
from a higher-order relatedness in active-site structure and/or
reactivity. These data thus provide another salient example of
enzymes that share “active-site homology” in the absence of
significant sequence similarity.33,41,42
Discussion
The recent description of LY2183240 as a highly potent
small-molecule inhibitor of anandamide transport has sparked
renewed interest in elucidating the protein(s) responsible for
this process.24,37 Here, we show that LY2183240 is a covalent
inhibitor of FAAH, thereby adding to a growing body of
evidence that ascribes a primary role for this degradative enzyme
in the regulation of anandamide uptake. FAAH’s contribution
to the cellular uptake of anandamide was originally delineated
by Deutsch and colleagues, who showed that FAAH inhibitors
reduce this process.22 More recent studies have revealed that
most, if not all, well-characterized anandamide transport block-
ers inhibit FAAH to a degree that correlates with their impact
on anandamide uptake.19,21 Additionally, the genetic disruption
of FAAH from neurons slows anandamide uptake,16 and
conversely, the introduction of FAAH into cells lacking this
enzyme increases the rate of this process.22-24 These findings
can be explained by a model in which the rapid intracellular
hydrolysis of anandamide by FAAH establishes a concentration
gradient that drives the uptake of this lipid.22,21
The participation of FAAH in anandamide uptake could also
rationalize, at least in part, the saturability of this process,
especially at later time points where catabolism might make its
greatest contribution. Indeed, careful characterization of the
uptake process for anandamide at early time points (less than 1
min) indicates that it is neither saturable18,21,38 nor sensitive to
chemical inhibition.18,21 These results suggest that the FAAH-
independent portion of anandamide uptake into cells may occur
largely by passive diffusion. Still, none of these previous studies
can exclude the participation of additional proteins, including
a putative plasma membrane and/or intracellular transporter, in
the cellular uptake and distribution of anandamide, which is
supported by some evidence (e.g., partial pharmacological
blockade of anandamide uptake in cells lacking FAAH16,23,39).
They do, however, emphasize that the discovery of such an
“anandamide transporter” will require chemical probes that
selectively disrupt its function without impacting the activity
Experimental Procedures
Synthesis of LY2183240. To a round-bottom flask fitted with a
magnetic stirrer were added TBAF‚3H2O (315 mg, 1 mmol, 1 equiv),
4-biphenylacetonitrile (193 mg, 1 mmol, 1 equiv), and trimethylsilyl
azide (175 mg, 1.5 mmol, 1.5 equiv). The mixture was heated at 85 °C
for 18 h and then cooled to room temperature. The mixture was then
diluted with EtOAc (20 mL) and washed with 1 M HCl (3 × 5 mL).
The organic layer was then dried with sodium sulfate and concentrated
under reduced pressure to afford 5-biphenyl-1-H-tetrazole as a white
solid (156 mg, 66% yield). The 5-biphenyl-1-H-tetrazole (23.6 mg, 0.1
mmol, 1 equiv) was then combined with dimethylcarbamyl chloride
(21.4 mg, 0.2 mmol, 2 equiv) and 1,4-diazabicyclo[2.2.2]octane (33.6
mg, 0.3 mmol, 3 equiv) in DMF. After 14 h, the reaction mixture was
dried and purified by silica gel chromatography (50% EtOAc in
1
hexanes) to afford LY2183240 (24 mg, 77%). H NMR (400 MHz,
CDCl3) 7.58-7.54 (m, 4H), 7.44-7.34 (m, 5H), 4.48 and 4.37 (s, 2H),
3.25 and 3.06 (s, 3H), 3.11 and 2.7 (s, 3H); HRMS m/z calculated for
C17H17N5O [M + Na]+ 330.1325, found 330.1331, 1.8 ppm.
Preparation of Mouse Tissue Proteomes. Mouse tissues were
Dounce-homogenized in 50 mM Tris‚HCl, pH 8.0, for assays with 14C-
oleamide or in 10 mM sodium/potassium phosphate buffer (pH 8) (PB)
for ABPP reactions followed by a low-speed spin (1400g, 3 min) to
remove debris. The supernatant was then subjected to centrifugation
(40) Dinh, T. P.; Freund, T. F.; Piomelli, D. Chem. Phys. Lipids 2002, 121,
149-158.
(41) Lichtman, A. H.; Leung, D.; Shelton, C.; Saghatelian, A.; Hardouin, C.;
Boger, D.; Cravatt, B. F. J. Pharmacol. Exp. Ther. 2004, 311, 441-448.
(42) Saghatelian, A.; Jessani, N.; Joseph, A.; Humphrey, M.; Cravatt, B. F. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101, 10000-10005.
(37) Mechoulam, R.; Deutsch, D. G. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
17541-17542.
(38) Sandberg, A.; Fowler, C. J. Chem. Phys. Lipids 2005, 134, 131-139.
(39) Fegley, D.; Kathuria, S.; Mercier, R.; Li, C.; Goutopoulos, A.; Makriyannis,
A.; Piomelli, D. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 8756-8761.
9
J. AM. CHEM. SOC. VOL. 128, NO. 30, 2006 9703

A R T I C L E S
Alexander and Cravatt
at 145000g for 45 min to provide the cytosolic fraction in the
supernatant and the membrane fraction as a pellet that was washed
and resuspended in Tris or PB buffer by sonication. Total protein
concentration in each fraction was determined using a protein assay
kit (Bio-Rad). Samples were stored at -80 °C until use.
labeled with 5 µM FP-biotin probe for 2 h at room temperature and
prepared for MudPIT analysis as previously described except that the
Lys-C digestion step was omitted.34 MudPIT analysis of eluted peptides
was carried out as previously described35,44 on a coupled Agilent 1100
LC-ThermoFinnigan LTQ MS system. All data sets were searched
against the mouse IPI database using the SEQUEST search algorithm,45
and results were filtered and grouped with DTASELECT.46 Peptides
with cross-correlation scores greater than 1.8 (+1), 2.5 (+2), 3.5 (+3)
and delta CN scores greater than 0.08 were included in the spectral
counting analysis. Only proteins for which 10 or more peptides were
identified on average in the control samples were considered for
analysis. Specifically, probe-labeled proteins were further identified by
their presence in FP-treated samples with a spectral number at least
5-fold or greater than that observed in “no probe” control runs
(experiments performed as described above, but without inclusion of
biotinylated FP). Spectral counts are corrected for the background
observed in the no-probe control and reported as the average of three
samples with the standard error (SE) of the mean.
Inactivation of Recombinant Serine Hydrolases by LY2183240.
Candidate serine hydrolase targets of LY2183240 were recombinantly
expressed in COS-7 cells by transient transfection. COS-7 cells were
grown in 100-mm dishes in complete medium (DMEM with
L-glutamine, nonessential amino acids, sodium pyruvate, and FBS). The
cells were transiently transfected using the appropriate cDNA in the
eukaryotic expression vector pcDNA3 (Invitrogen) and the FuGENE
6 transfection system (Roche) following the manufacturer’s protocol.
Control COS-7 cells were prepared in the same way except that an
empty pcDNA3 vector was used for transfection. Cells were washed
three times with PBS, collected by scraping, brought up in 400 µL of
PBS, and homogenized by sonication, and the membrane fractions were
isolated by centrifugation at 145000g. Membrane proteomes of
KIAA1363-transfected cells and MAG lipase-transfected cells were
diluted to 0.2 and 0.6 mg/mL, respectively, in PBS (∼1 nM active
enzyme as estimated by ABPP as previously described27), and the
membrane proteome from Abh6-transfected cells was diluted to 0.01
mg/mL (∼0.1 nM active enzyme). IC50 values were obtained by
competitive ABPP with FP-rhodamine as described above for FAAH
in tissue samples.
In Vitro Analysis of Inhibitor Potency. Inhibitor analysis was
carried out in vitro using two different assays. In the first assay, FAAH
activity was measured in mouse brain membranes in the presence of
varying concentrations of inhibitor by following the conversion of
14C-oleamide, synthesized as previously described,7 to oleic acid by
thin-layer chromatography (TLC).43 Brain membranes (1 mg/mL) from
FAAH (-/-) mice were used as a control. Briefly, varying concentra-
tions of LY2183240 (1 µL, 100× stock in DMSO added to provide
0.001-10 µM final concentration) were preincubated with brain
membranes (1 mg/mL in 50 mM Tris, pH 8, 94 µL) for 10 min. 14C-
Oleamide (5 µL, 2 mM stock in DMSO, 100 µM final concentration)
was added, and 31 µL of each reaction was removed and quenched in
300 µL of 0.1 M HCl at 30, 45, and 60 min. The solution was then
extracted with 400 µL of ethyl acetate. The organic layer was removed
and dried under a stream of gaseous N2, solubilized in 25 µL of ethyl
acetate, and separated by TLC (60% ethyl acetate in hexanes). The
radioactive compounds were quantified using a Cyclone Phosphorim-
ager (PerkinElmer Life Sciences). Inhibitor potency was also examined
using competitive ABPP as described previously.33 Briefly, mouse brain
membrane proteomes were diluted to 1 mg/mL in PBS and preincubated
with varying concentrations of inhibitors (1 nM to 10 µM, 50× DMSO
stocks) for 10 min at room temperature prior to the addition of a
rhodamine-tagged fluorophosphonate (FP-rhodamine,26,27 50× DMSO
stock). The ABPP probe was used at a final concentration of 100 nM
in a 50 µL of total reaction volume. Reactions were quenched after 10
min with 2× SDS-PAGE loading buffer, subjected to SDS-PAGE, and
visualized in-gel using a flatbed fluorescence scanner. Dose response
curves obtained from both methods from three trials at each inhibitor
concentration were fit using Prism software (GraphPad) to obtain IC50
values with 95% confidence intervals.
MS Analysis of LY2183240 Covalent Labeling of FAAH. Inhibitor
labeling reactions were conducted at room temperature for 1 h with 23
µM recombinant purified FAAH (∼74 µg) prepared as described
previously,43 and 35 µM LY2183240 or DMSO for the control reaction
in 125 mM Tris‚HCl, pH 9, in a total volume of 50 µL. Any noncovalent
interactions between inhibitor and enzyme were disrupted by gel
filtration through a NAP 5 column (GE Healthcare), the desired fractions
were concentrated to dryness by speed vacuum centrifugation, and
reconstituted in 0.5% trifluoroacetic acid in H2O in a final volume of
50 µL. Samples were prepared for MALDI-TOF Reflectron mass
spectrometry analysis as previously described with 1:1 R-cyano-4-
hydroxycinnamic acid/sample (2 µL of final volume).30
Acknowledgment. We thank A. Saghatelian for the synthesis
of LY2183240, K. Chiang for assistance with the in vivo mouse
studies, S. Niessen for assistance with tandem MS analysis, M.
McKinney and K. Masuda for purified recombinant ∆TM-
FAAH, G. Siuzdak for access to the MALDI-TOF Reflectron
mass spectrometer, and the Cravatt lab for helpful discussions.
This work was supported by the National Institutes of Health
Grants DA015197 and DA017259 (B.F.C.), the Ruth L. Kir-
schstein NRSA Postdoctoral Fellowship DA019347 (J.P.A.), the
Skaggs Institute for Chemical Biology, and the Helen L. Dorris
Institute for the Study of Neurological and Psychiatric Disorders
of Children and Adolescents.
In Vivo Labeling of Proteins by LY2183240 in Mice. Mice were
male wild type or FAAH (-/-)9 animals on a C57Bl/6 background,
weighed between 20 and 30 g, and were between the ages of 8 and 9
weeks. Mice were given intraperitoneal injections of 10 mg/kg URB597
(in vehicle 18:1:1 saline/emulphor/EtOH), 10 mg/kg LY2183240 (in
vehicle), or vehicle alone. After 90 min, the mice were sacrificed by
CO2 asphyxiation and relevant tissues were flash frozen (liquid N2)
immediately upon removal. Tissues were processed as described above
to isolate membrane and cytosolic proteomes. The FP-rhodamine
labeling was carried out as described above, and reactions were analyzed
by SDS-PAGE. See the next paragraph for ABPP-MudPIT analysis of
inhibited enzymes in the proteome. The average of three independent
samples at each concentration was reported.
Supporting Information Available: Complete ref 6 and
Supporting Information Table 1. This material is available free
of charge via the Internet at http://pubs.acs.org
JA062999H
(43) Patricelli, M. P.; Lashuel, H. A.; Giang, D. K.; Kelly, J. W.; Cravatt, B. F.
Biochemistry 1998, 37, 15177-15187.
(44) Washburn, M. P.; Wolters, D.; Yates, J. R. Nat. Biotechnol. 2001, 19, 242-
247.
(45) Eng, J.; McCormack, A. L.; Yates, J. R. J. Am. Soc. Mass Spectrom. 1994,
ABPP-MudPIT Analysis of Proteins Labeled by LY2183240 in
Vivo. Brain membrane proteomes (1 mL, 1 mg/mL in PB) from mice
treated with LY2183240, URB597, or vehicle (as described above) were
5, 976-979.
(46) Tabb, D. L.; McDonald, W. H.; Yates, J. R. J. Proteome Res. 2002, 1,
21-26.
9
9704 J. AM. CHEM. SOC. VOL. 128, NO. 30, 2006