A functional proteomic strategy to discover inhibitors for uncharacterized hydrolases

Article abstract of DOI:10.1021/ja073650c

Hydrolytic enzymes constitute one of the largest and most diverse protein classes in Nature and play key roles in nearly all physiological and pathological processes. The mammalian serine hydrolase superfamily contains a remarkable number of uncharacterized members, with at least 40-50% of these enzymes lacking experimentally verified endogenous substrates and products. Assignment of metabolic and cellular functions to these enzymes requires the development of pharmacological tools to selectively perturb their activity. We describe herein a functional proteomic strategy to systematically develop potent and selective inhibitors for uncharacterized serine hydrolases and its application to the brain-enriched enzyme α/β-hydrolase-6. We anticipate that the methods described herein will facilitate the development of selective chemical probes to annotate the metabolic and (patho)physiological functions of many of the uncharacterized serine hydrolases that currently populate eukaryotic and prokaryotic proteomes. Copyright

Full text of DOI:10.1021/ja073650c

Published on Web 07/13/2007
A Functional Proteomic Strategy to Discover Inhibitors for Uncharacterized
Hydrolases
Weiwei Li, Jacqueline L. Blankman, and Benjamin F. Cravatt*
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 May 21, 2007; E-mail: cravatt@scripps.edu
Hydrolytic enzymes constitute one of the largest and most diverse
protein classes in Nature. Of the four major classes of hydrolytic
enzymes (aspartyl, cysteine, metallo, serine), serine hydrolases
(SHs) are a particularly expansive family in mammals, representing
∼1% of the predicted protein products of the human genome.1 SHs
Figure 1. Functional proteomic strategy to discover carbamate inhibitors
for uncharacterized SH enzymes. Rh, rhodamine.
play key roles in nearly all physiological and pathological processes
and are targeted by drugs that treat diseases such as diabetes,
Alzheimer’s disease, obesity, and blood clotting disorders.2 The
mammalian SH superfamily contains a remarkable number of
uncharacterized members, with at least 40-50% of these enzymes
lacking experimentally verified endogenous substrates and products.
Assignment of metabolic and cellular functions to these enzymes
requires the development of pharmacological tools to selectively
perturb their activity. We describe herein a functional proteomic
strategy to systematically develop potent and selective inhibitors
for uncharacterized SHs.
SHs are nearly universally susceptible to inactivation by fluo-
rophosph(on)ates (FPs), which covalently modify the catalytic serine
nucleophile in these enzymes.3 Reporter-tagged versions of FPs
(Figure 1) have been shown to serve as versatile activity-based
protein profiling (ABPP) probes for SHs,4 leading to the discovery
of enzymes dysregulated in aggressive cancer cells, activated
chondrocytes, and fatty livers.5 Beyond their application for SH
discovery in disease models, reporter-tagged FPs have formed the
basis of competitive assays to evaluate the specificity of SH
inhibitors in native proteomes.6 In select cases, inhibitors for
uncharacterized SHs have emerged from competitive ABPP experi-
ments and used, in combination with metabolomic methods,7 to
determine the endogenous metabolic function of the enzymes.8
Nonetheless, it remains unclear whether ABPP methods can be
systematically applied to engender potent and selective inhibitors
for uncharacterized SHs. We set out to test this premise by
performing competitive ABPP with a representative unannotated
SH, R/â-hydrolase domain 6 (ABHD6) and a library of SH-directed
inhibitors based on the carbamate reactive group (Figure 1).
Carbamates have proven to be a versatile class of SH inhibitors,
with multiple agents in active clinical use or development, including
those that target the SHs acetylcholinesterase (AChE) and FAAH,9
which regulate acetylcholine and endocannabinoid signaling in the
nervous system, respectively. ABHD6 represented an attractive
initial target for competitive ABPP, as this enzyme, like AChE and
FAAH, is highly expressed in mammalian brain (Supporting
Figure 1).
Figure 2. Discovery of lead inhibitors for ABHD6. (A) Screening of
representative carbamates for inhibition of FP-Rh labeling of ABHD6.
Active compounds shown: 5, 16, and 18 (see Supporting Figure 2 for
complete carbamate library screen). (B) Selectivity of lead inhibitors
(10 µM) was assessed by performing competitive ABPP with the mouse
brain membrane proteome. Representative SH targets identified in previous
studies6 are shown in bold. Arrowheads designate additional SHs sensitive
to carbamate 5. Fluorescent gels shown in grayscale.
analyzed by SDS-PAGE and in-gel fluorescence scanning. Multiple
carbamates blocked FP-Rh labeling of ABHD6, including com-
pounds 5, 16, and 18 (Figure 2A). These inhibitors were tested
over a concentration range of 0.01-50 µM to generate IC50 values
for ABHD6 inhibition of 2.3, 1.0, and 0.35 µM, respectively (see
Supporting Figure 2 for IC50 curves). Carbamates 5, 16, and 18
were next assessed for their selectivity in competitive ABPP assays
with the mouse brain membrane proteome. Marked differences in
selectivity were observed (Figure 2B). Carbamate 5 inhibited several
brain SHs, including FAAH, KIAA1363, and multiple enzymes with
masses between 20 and 40 kDa (arrowheads, Figure 2B). Carbamate
16 also cross-reacted with FAAH. In contrast, carbamate 18
selectively inhibited, albeit incompletely, a single 35 kDa FP-Rh-
reactive band provisionally assigned as ABHD6.
The superior selectivity exhibited by carbamate 18 suggested
that this agent might serve as a useful lead for generating higher
affinity ABHD6 inhibitors. Approximately 20 derivatives of 18 were
synthesized with various modifications to the O-aryl group and
screened for activity against ABHD6 (Supporting Figure 3).
Modification of the terminal phenyl ring of 18 with a p-carboxamide
group improved potency ∼5-fold, generating an agent 70 that
inhibited ABHD6 with an IC50 value of 70 nM (Figure 3A,B). This
boost in potency was achieved with no discernible loss in selectivity
as judged by competitive ABPP with brain membrane proteome
(Supporting Figure 4). Curiously, however, even at concentrations
COS-7 cells were transiently transfected with the human ABHD6
cDNA, resulting in high expression levels of active, recombinant
enzyme in the membrane proteome (Figure 2A).10 This proteome
was treated with individual members of an initial library of 55
carbamates (50 µM; see Supporting Information for compound
structures) for 1 h, followed by the addition of the ABPP probe
FP-rhodamine (FP-Rh;4c 1 µM) for 1 h. Reactions were then
9
9594
J. AM. CHEM. SOC. 2007, 129, 9594-9595
10.1021/ja073650c CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
where the future application of 70, in conjunction with metabolomic
methods,7,8 may reveal endogenous biochemical and (patho)-
physiological functions for ABHD6. Projecting beyond ABHD6,
it is noteworthy to emphasize that the development of 70 required
screening fewer than 75 compounds. This achievement reinforces
the idea that carbamates may offer a privileged molecular scaffold
for the streamlined development of SH inhibitors that display an
excellent combination of potency and selectivity. Of course, our
competitive ABPP studies only speak to the specificity of carbam-
ates within the SH family; potential protein targets outside of this
enzyme class are not discriminated. Conversion of carbamates into
probes using bio-orthogonal reactions such as the Cu(I)-catalyzed
Huisgen’s azide-alkyne cycloaddition (click chemistry)14 provides
a complementary route to directly visualize protein targets of
carbamates.15 We anticipate that continued efforts to screen
structurally diverse libraries of carbamates against the daunting
number of uncharacterized SHs that populate eukaryotic and
prokaryotic proteomes will engender a suite of valuable pharma-
cological tools for annotating new biochemical pathways.
Figure 3. Development of carbamate 70, a potent and selective inhibitor
of ABHD6. (A) Structures of carbamates 18 and 70. (B) IC50 curves for
inhibition of FP labeling of recombinant ABHD6 by 18 and 70. Data
represent the average fluorescent signals ( standard error (SE) for three
independent competitive ABPP experiments. See Supporting Figure 5 for
gel images. (C) Blockade of FP labeling of endogenous ABHD6 in brain
membrane proteome by 70 (see Supporting Figure 4 for quantification of
signals). (D) ABPP-MudPIT analysis of the impact of 70 (10 µM, 1 h
preincubation) on the labeling of brain membrane SH activities by FP-
biotin (5 µM, 1 h). See Supporting Table 1 for full names of brain SHs.
Data represent the average spectral counts ( standard error (SE) for three
independent experiments. /, p < 0.01 for 70- versus DMSO-treated
proteomes.
Acknowledgment. We thank the Cravatt lab for critical reading
of the manuscript. This work was supported by the NIH (CA087660),
California Institute for Regenerative Medicine (T2-00001), and
Skaggs Institute for Chemical Biology.
Supporting Information Available: Synthesis and experimental
protocols; figures showing tissue distribution of ABHD6, IC50 curves
for 5 and 16, and SDS-PAGE analysis of competitive ABPP experi-
ments; table of brain membrane SH enzymes. This material is available
free of charge via the Internet at http://pubs.acs.org.
up to 10 µM, carbamate 70 failed to completely block FP labeling
of the 35 kDa brain protein band assigned as ABHD6 (Figure 3C
and Supporting Figure 4). We speculated that the residual 35 kDa
FP signal observed in competitive ABPP reactions with 70 (and
previously with 18) might represent an additional brain SH that
co-migrates with ABHD6. Consistent with this premise, analysis
of the brain soluble proteome identified a 35 kDa SH activity that
was insensitive to 70 (Supporting Figure 4).
References
(1) (a) Lander, E. S.; et al. Nature 2001, 409, 860-921. (b) Venter, J. C.; et
al. Science 2001, 291, 304-51.
(2) (a) Thornberry, N. A.; Weber, A. E. Curr. Top. Med. Chem. 2007, 7,
557-68. (b) Jann, M. W. Pharmacotherapy 2000, 20, 1-12. (c) Nelson,
R. H.; Miles, J. M. Exp. Opin. Pharmacother. 2005, 6, 2483-91. (d)
Srivastava, S.; Goswami, L. N.; Dikshit, D. K. Med. Res. ReV. 2005, 25,
66-92.
(3) Walsh, C. T. Enzymatic Reaction Mechanisms; W.H. Freeman and
Company: New York, 1979.
(4) (a) Liu, Y.; Patricelli, M. P.; Cravatt, B. F. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 14694-9. (b) Kidd, D.; Liu, Y.; Cravatt, B. F. Biochemistry
2001, 40, 4005-15. (c) Patricelli, M. P.; Giang, D. K.; Stamp, L. M.;
Burbaum, J. J. Proteomics 2001, 1, 1067-71.
To more thoroughly characterize the target profile of carbamate
70, we performed competitive ABPP experiments with mouse brain
membrane proteome and FP-biotin4a and analyzed the reactions
using a shotgun LC-MS method termed ABPP-MudPIT.11 The
relative levels of SH activities in 70-treated versus DMSO-treated
(control) proteomes were quantified by spectral counting.11 For these
studies, we selected a concentration of 70 (10 µM) that was well
above the calculated IC50 value (70 nM), so as to discern the degree
of selectivity exhibited by this agent for ABHD6 relative to other
brain SHs. Under these conditions, carbamate 70 was found to block
greater than 90% of the activity of ABHD6 (Figure 3D). In contrast,
none of the other 27 SH activities identified in the brain membrane
proteome were significantly altered by 70, with the exception of
ABHD10, which showed a modest (30%) but significant reduction
in activity in 70-treated proteomes.12 Follow-up studies, however,
failed to reveal any inhibition of recombinant ABHD10 by 70 at
concentrations up to 100 µM, indicating that this enzyme is unlikely
to be a true target of 70. Collectively, these data indicate that 70 is
a highly selective inhibitor of ABHD6 compared to other brain
SHs. From a technical perspective, the results also highlight the
superior resolution afforded by LC-MS compared to gel-based
methods for broadly assessing the protein targets of chemical probes.
In summary, we have described herein a functional proteomic
strategy to develop inhibitors for uncharacterized SHs, and its
application to create a potent and selective inhibitor of ABHD6.
The enriched expression of ABHD6 in brain tissue suggests a role
for this enzyme in nervous system metabolism and/or signaling.
ABHD6 is also highly elevated in Epstein-Barr virus-transformed
B cells,13 indicating that it may contribute to cancer pathogenesis.
These expression patterns thus designate cell and organ targets
(5) (a) Jessani, N.; Liu, Y.; Humphrey, M.; Cravatt, B. F. Proc. Natl. Acad.
Sci. U.S.A. 2002, 99, 10335-40. (b) Milner, J. M.; Kevorkian, L.; Young,
D. A.; Jones, D.; Wait, R.; Donell, S. T.; Barksby, E.; Patterson, A. M.;
Middleton, J.; Cravatt, B. F.; Clark, I. M.; Rowan, A. D.; Cawston, T. E.
Arthritis Res. Ther. 2006, 8, R23. (c) Barglow, K. T.; Cravatt, B. F. Chem.
Biol. 2004, 11, 1523-31.
(6) Leung, D.; Hardouin, C.; Boger, D. L.; Cravatt, B. F. Nat. Biotechnol.
2003, 21, 687-91.
(7) Saghatelian, A.; Trauger, S. A.; Want, E. J.; Hawkins, E. G.; Siuzdak,
G.; Cravatt, B. F. Biochemistry 2004, 43, 14332-9.
(8) Chiang, K. P.; Niessen, S.; Saghatelian, A.; Cravatt, B. F. Chem. Biol.
2006, 13, 1041-50.
(9) (a) Desai, A.; Grossberg, G. Exp. Opin. Pharmacother. 2001, 2, 653-66.
(b) Piomelli, D.; Tarzia, G.; Duranti, A.; Tontini, A.; Mor, M.; Compton,
T. R.; Dasse, O.; Monaghan, E. P.; Parrott, J. A.; Putman, D. CNS Drug
ReV. 2006, 12, 21-38.
(10) The selective distribution of ABHD6 to the membrane proteome is
consistent with hydropathy plots, which predict that the enzyme contains
a single transmembrane domain from amino acids 8-24.
(11) 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, 691-7.
(12) A handful of other SHs showed 30-40% lower spectral counts in 70-
treated samples (e.g., FAAH, BAT5, AChE), but none of these differences
were statistically significant. Recombinant forms of these enzymes were
not inhibited by up to 100 µM of 70 (Supporting Figure 5).
(13) Maier, S.; Staffler, G.; Hartmann, A.; Hock, J.; Henning, K.; Grabusic,
K.; Mailhammer, R.; Hoffmann, R.; Wilmanns, M.; Lang, R.; Mages, J.;
Kempkes, B. J. Virol. 2006, 80, 9761-71.
(14) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2001,
40, 2004-21.
(15) Alexander, J. P.; Cravatt, B. F. Chem. Biol. 2005, 12, 1179-87.
JA073650C
9
J. AM. CHEM. SOC. VOL. 129, NO. 31, 2007 9595