Discovery of MK-3168: A PET tracer for imaging brain fatty acid amide hydrolase

Article abstract of DOI:10.1021/ml4000996

We report herein the discovery of a fatty acid amide hydrolase (FAAH) positron emission tomography (PET) tracer. Starting from a pyrazole lead, medicinal chemistry efforts directed toward reducing lipophilicity led to the synthesis of a series of imidazole analogues. Compound 6 was chosen for further profiling due to its appropriate physical chemical properties and excellent FAAH inhibition potency across species. [¹¹C]-6 (MK-3168) exhibited good brain uptake and FAAH-specific signal in rhesus monkeys and is a suitable PET tracer for imaging FAAH in the brain.

Full text of DOI:10.1021/ml4000996

Letter
pubs.acs.org/acsmedchemlett
Discovery of MK-3168: A PET Tracer for Imaging Brain Fatty Acid
Amide Hydrolase
Ping Liu,*,† Terence G. Hamill,¶ Marc Chioda,† Harry Chobanian,† Selena Fung,† Yan Guo,†
Linda Chang,† Raman Bakshi,† Qingmei Hong,† James Dellureficio,† Linus S. Lin,† Catherine Abbadie,‡
Jessica Alexander,‡ Hong Jin,‡ Suzanne Mandala,‡ Lin-Lin Shiao,‡ Wenping Li,¶ Sandra Sanabria,¶
David Williams,¶ Zhizhen Zeng,¶ Richard Hajdu,§ Nina Jochnowitz,§ Mark Rosenbach,§
Bindhu Karanam,∥ Maria Madeira,∥ Gino Salituro,∥ Joyce Powell,∥ Ling Xu,∥ Jenna L. Terebetski,⊥
Joseph F. Leone,# Patricia Miller,¶ Jacquelynn Cook,¶ Marie Holahan,¶ Aniket Joshi,¶ Stacey O’Malley,¶
Mona Purcell,¶ Diane Posavec,¶ Tsing-Bau Chen,¶ Kerry Riffel,¶ Mangay Williams,¶ Richard Hargreaves,¶
Kathleen A. Sullivan,‡ Ravi P. Nargund,† and Robert J. DeVita†
‡
§
∥
⊥
†Departments of Medicinal Chemistry, Immunology, Pharmacology, Preclinical DMPK, Basic Pharmaceutical Sciences, and
#Process Research, Merck Research Laboratories, Rahway, New Jersey 07065, United States
¶Department of Imaging Research, Merck Research Laboratories, West Point, Pennsylvania 19486, United States
S
* Supporting Information
ABSTRACT: We report herein the discovery of a fatty acid amide hydrolase (FAAH)
positron emission tomography (PET) tracer. Starting from a pyrazole lead, medicinal
chemistry efforts directed toward reducing lipophilicity led to the synthesis of a series of
imidazole analogues. Compound 6 was chosen for further profiling due to its appropriate
physical chemical properties and excellent FAAH inhibition potency across species.
[11C]-6 (MK-3168) exhibited good brain uptake and FAAH-specific signal in rhesus
monkeys and is a suitable PET tracer for imaging FAAH in the brain.
KEYWORDS: Fatty acid amide hydrolase, FAAH, positron emission tomography, carbon-11, PET tracer, target engagement,
biomarker
atty acid amide hydrolase (FAAH) is a serine hydrolase
effects typically observed with direct cannabinoid receptor
agonists.10,11 Thus, FAAH represents a potential therapeutic
target for the treatment of pain, inflammation, and other clinical
disorders. Indeed, a number of FAAH inhibitors have been
reported including covalent, irreversible inhibitors; covalent,
reversible inhibitors; and noncovalent, reversible inhibi-
tors.12−16 To better understand FAAH biology and assist in
the design and testing of promising drug candidates targeting
this enzyme, a suitable FAAH positron emission tomography
(PET) or single photon emission computed tomography
(SPECT) radioligand is highly desirable, which would allow
FAAH imaging studies in the living human and animal brain
under normal physiological and diseased conditions. In general,
a successful PET ligand should provide a good specific signal
(total/nonspecific ≥ 1.5:1) to allow quantitative mapping of
the target of interest. Though not an absolute, ideally, a brain
PET ligand should have a good binding affinity with Bmax/Kd ≥
10 and reasonable lipophilicity with log P or log D of 1−3.5 for
adequate brain penetration and optimum specific to nonspecific
ratio.17,18 Much effort has been devoted to the search for
suitable FAAH PET ligands. The majority of the reported
F
characterized by an unusual Ser−Ser−Lys catalytic triad
that cleaves amides and esters at similar rates.1,2 It is an integral
membrane enzyme responsible for the breakdown of several
fatty acid ethanolamide (FAE) signaling molecules (Figure 1),
Figure 1. Substrates of fatty acid amide hydrolase.
including the endocannabinoid arachidonoyl ethanolamide
(anandamide, AEA), and the related lipids N-palmitoyl
ethanolamide (PEA) and N-oleoyl ethanolamide (OEA).3−5
Inhibition of FAAH leads to elevated levels of these
endogenous FAEs,6 which act on cannabinoid, vanilloid, and
other receptors to induce anti-inflammatory, antidepressant,
analgesic, and anxiolytic effects in preclinical animal models.7−9
In addition, these actions occur in the absence of the adverse
Received: March 11, 2013
Accepted: April 20, 2013
Published: April 20, 2013
© 2013 American Chemical Society
509
dx.doi.org/10.1021/ml4000996 | ACS Med. Chem. Lett. 2013, 4, 509−513

ACS Medicinal Chemistry Letters
Letter
radioligands are URB597 analogues19,20 or anandamide
analogues21,22 (Figure 2). Recent reports have appeared for
Scheme 2. SAR Efforts and Identification of Compound 6
Figure 2. Representative examples of FAAH PET tracers.
[18F]PF-981123 and [11C-carbonyl]O-arylcarbamates24 and the
characterization of these tracers in rodents. Herein, we wish to
report the discovery of a structurally distinct and reversible
FAAH PET tracer (MK-3168) that may be suitable for clinical
application.
The medicinal chemistry effort began with compound 2,25
a
potent FAAH inhibitor that had been identified from our lead
optimization program. Compound 2 has an HPLC log D value
of 4.2 and subnanomolar FAAH inhibition potency across
species. Although the lipophilicity of compound 2 is higher
than desired, we were encouraged by its excellent potency and
decided to radiolabel compound 2. 11CH3 was introduced
through precursor 1 to yield [11C]-2 (Scheme 1). However, in
Scheme 1. Initial Efforts for a FAAH PET Tracer
vivo PET studies of [11C]-2 in rhesus monkeys were
complicated by brain-penetrant and radiolabeled metabolites,
suggesting the need for a different scaffold for radiolabeling.
Our attention shifted to compound 3, another analogue from
the pyrazole series25 (Scheme 2). This compound has a
reasonable FAAH inhibition potency in human and rat, but
because rhesus monkey is our preferred preclinical species for
evaluation of CNS PET tracers, its rhesus potency and
lipophilicity required further optimization. More importantly,
a reliable route to introduce a radiolabel to the molecule was
needed since this was unable to be realized on compound 3.
Thus, our structure−activity relationship (SAR) objective was
to modify structure 3 to decrease lipophilicity, improve
potency, and locate a structural feature for efficient radio-
labeling.
Replacing pyrazole with imidazole is an effective way to
lower log D, provided other properties are maintained. In this
regard, compound 4 (R = cyclopropyl) was synthesized;
however, it was a much less potent compound (Scheme 2 and
its table).26 To our delight, when R was changed to methyl
(compound 5), the potency was improved by 3-fold; when R
was removed (R = H; compound 6), the potency was improved
significantly. However, compound 7, the imidazole regio
isomer, and compound 8, the uncapped imidazole analogue,
reduced potency by about 100-fold. With the log D of
compound 6 in an acceptable range, the methyl substituent on
the imidazole could serve as a handle to introduce 11C. To
further lower the log D, primary amide 9 and secondary amide
10 were synthesized. While compound 9 was less potent than
aFAAH inhibition IC50 data expressed as mean
independent experiments). nd = not determined. Assayed only once
except for compound 6.
SD (n ≥ 2
b
c
compound 6, compound 10 offered comparable potency.
Furthermore, to consider the option of introducing 18F to the
molecule, fluoroethyl analogues 11−13 were synthesized.
Compounds 11 and 12 could be candidates for a 18F PET
tracer, but analogue 13 was not an option due to its poor
potency in rhesus, which is our preferred preclinical species for
imaging studies. Finally, to complete the SAR studies, the
chlorophenyl analogues 14−18 were also synthesized. These
chlorophenyl analogues offered better potency (for example, 14
vs 6) but increased lipophilicity.
Compound 6 was chosen for further profiling. In the whole
blood FAAH inhibition assay, compound 6 exhibited excellent
potency in human (IC90 = 5.5 nM) and rhesus (IC90 = 29 nM).
An off-target screen of this compound was performed against a
panel of 168 receptors, ion channels, and enzymes; only two
off-target activities, acetyl cholinesterase (IC50 = 1.63 μM) and
phosphodiesterase PDE4 (IC50 = 9.75 μM), were identified.
Compound 6 was very selective over the two key ion channels,
hERG (IC50 = 37 μM) and DLZ (IC50 = 16 μM); CYP 3A4
(IC50 > 50 μM); as well as the two cannabinoid receptors, CB1
(IC50 = 30 μM) and CB2 (IC50 = 150 μM). This compound
was not a substrate for rat or human P-glycoprotein in vitro and
rapidly penetrated into the brain and achieved a brain-to-
510
dx.doi.org/10.1021/ml4000996 | ACS Med. Chem. Lett. 2013, 4, 509−513

ACS Medicinal Chemistry Letters
Letter
a
Table 1. Tissue homogenate binding studies with [3H]-6
plasma concentration ratio of 7:1 at 2 h following a 2 mg/kg
oral dose to rats. On the basis of its overall profile, we believed
compound 6 would provide a high probability of success to
develop as a 11C PET tracer.
tissue
Kd (nM)
Bmax (nM)
Bmax/Kd
rat cortex
0.6 0.1
1.4 0.2
0.8 0.2
43
21
15
4
3
1
73 11
The synthesis of [3H]-6 and [11C]-6 is outlined in Scheme 3
and began with a Negishi-type coupling27 between chiral
rhesus cortex
human cortex
15
19
1
6
a
Data expressed as mean SD (n = 3 independent experiments).
a
Scheme 3. Synthesis of [3H]-6 and [11C]-6
Scatchard plot of transformed data showed that [3H]-6 bound
to a single, saturable site in these species.31 More importantly,
the comparable Bmax/Kd ratio for rhesus monkey and human
suggested that the magnitude of an observed in vivo specific
signal of [11C]-6 in rhesus monkey could be similar to that in
human.
The pharmaceutical properties of compound 6 were also
assessed and the crystalline free base was characterized as its
solid state form. This form of compound 6 showed excellent 24
h pH dependent solution stability, and solubility values
obtained in 10 mM saline buffer at pH 3−8, with 10% ethanol,
should provide adequate solubility to support clinical iv dosing.
[11C]-6 was evaluated in isoflurane-anesthetized rhesus
monkeys as a potential PET ligand. Baseline scans after a
single iv (5 mCi) administration revealed that [11C]-6 rapidly
penetrated the blood/brain barrier and accumulated in the
frontal cortex, striatum, and hippocampus regions (FAAH
enriched area),32 reaching a maximum signal in ∼45 min. A
blockade experiment was also carried out with compound 2, a
potent FAAH inhibitor in rhesus (Figure 3). The total to
a
Reagents and conditions: (a) 19, EtMgBr, ZnCl2 in THF, then 20
and Pd(PPh3)4, reflux, 72%; (b) 1 N HCl, MeOH, reflux, 66%; (c)
NIS, dichloromethane, rt, 53%; (d) CuI, 1,10-phenanthroline, DMSO,
100 °C, 26%; (e) KOH, MeCN/water, 80 °C; (f) dimethylamine,
HOBT, EDC, diisopropylethylamine, DMF, 72% for 2 steps; (g) CT3I,
Cs2CO3, DMF, rt; (h) [11C]-MeI, Cs2CO3, DMF, 65 °C.
Figure 3. In vivo PET images of [11C]-6 in rhesus monkey brain.
Color scale indicates standard uptake value (SUV) units (nCi/cc/
mCi/kg).
bromide 2028 and the zinc species derived from the
commercially available 19. This C−C bond formation provided
compound 21 in 72% yield. After the trityl group was removed
with HCl, imidazole 22 was converted by NIS to iodoimidazole
23 that was ready for the C−S bond formation. This C−S
coupling proved challenging, most likely due to the interference
of the free imidazole N−H.29 In the end, the CuI-mediated
coupling conditions were used for the reaction between
iodoimidazole 23 and thiol 2428 that afforded sulfide 25 in
26% yield. With intermediate 25 in hand, the remaining
synthesis was straightforward. Ester hydrolysis and amide
formation gave 8, a precursor to both [3H]-6 and [11C]-6.
Reaction of 8 with CT3I provided [3H]-6. Finally, 8 was reacted
with [11C]-CH3I completing the synthesis of PET tracer [11C]-
6.30
nonspecific signal ratio was ∼2:1, exceeding the targeted criteria
1.5:1. Table 2 compares the profile of [11C]-6 against our target
Table 2. Comparison of Target Criteria and Profile of [11C]-
6
target criteria for a FAAH PET tracer
[11C]-6 profile
19
human Bmax/Kd > 10
human Kd < 2 nM
0.8
moderate log D (1−3.5)
non- or weak substrate for hPGP (MDR)
3.3
0.8
a
Papp > 20 × 10−6 cm/s
27 × 10−6 cm/s
[11C]-MeI
reliable route for C-11 or F-18 labeling
good brain uptake in monkey
b
>1 SUV
specific signal ≥ 1.5:1 (total/nonspecific)
∼2:1
[3H]-6 was used in tissue homogenate binding studies. As
shown in Table 1, compound 6 exhibited excellent binding
affinity to the cortex of rat, rhesus, and human. In addition,
a
b
Measurement of cell permeability. SUV: nCi/cc/mCi/kg body
weight.
511
dx.doi.org/10.1021/ml4000996 | ACS Med. Chem. Lett. 2013, 4, 509−513

ACS Medicinal Chemistry Letters
Letter
criteria for a FAAH PET tracer. We believe [11C]-6 has the
criteria of a successful FAAH PET tracer and should find wide
use for in vivo FAAH PET studies.
blockade of anandamide hydrolysis. Proc. Natl. Acad. Sci. U.S.A. 2005,
102, 18620−18625.
(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. Modulation of
anxiety through blockade of anandamide hydrolysis. Nat. Med. 2003, 9,
76−81.
In summary, a high quality FAAH PET tracer [11C]-6 (MK-
3168) was developed through optimization of lipophilicity and
potency for FAAH inhibition of the initial lead compound. This
PET tracer exhibited good brain uptake and FAAH-specific
signal in rhesus monkey PET studies and therefore was a
suitable PET tracer for imaging FAAH in the brain. These
results led to the decision to investigate [11C]-6 in the clinic,
which will be the subject of another publication.
(11) Piomelli, D. The molecular logic of endocannabinoid signalling.
Nat. Rev. Neurosci. 2003, 4, 873−884.
(12) Ahn, K.; Johnson, D. S.; Cravatt, B. F. Fatty acid amide
hydrolase as a potential therapeutic target for the treatment of pain
and CNS disorders. . Expert Opin. Drug Discovery 2009, 4, 763−784.
(13) Minkkila, A.; Saario, S.; Nevalainen, T. Discovery and
development of endocannabinoid-hydrolyzing enzyme inhibitors.
Curr. Top. Med. Chem. 2010, 10, 828−858.
ASSOCIATED CONTENT
* Supporting Information
■
S
(14) Seierstad, M.; Breitenbucher, J. G. Discovery and development
of fatty acid amide hydrolase (FAAH) inhibitors. J. Med. Chem. 2008,
51, 7327−7343.
(15) Otrubova, K.; Ezzili, C.; Boger, D. L. The discovery and
development of inhibitors of fatty acid amide hydrolase (FAAH).
Bioorg. Med. Chem. Lett. 2011, 21, 4674−4685.
Synthetic procedures and characterization data of selected
compounds, conditions for the biological assays, reversibility
study of compound 6, and protocol for rhesus PET imaging.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(16) Scott, C. W.; Tian, G.; Yu, X. H.; Paschetto, K. A.; Wilkins, D.
E.; Meury, L.; Cao, C. Q.; Varnes, J.; Edwards, F. D. Biochemical
characterization and in vitro activity of AZ513, a noncovalent,
reversible, and noncompetitive inhibitor of fatty acid amide hydrolase.
Eur. J. Pharmacol. 2011, 667, 74−79.
(17) Eckelman, W. C.; Gibson, R. E.; Rzeszotarski, W. J.; Vieras, F.;
Mazaitis, J. K.; Francis, B.; Reba, W. C. The Design of Receptor
Binding Radiotracers. In Principles of Radiopharmacology; Colombetti,
L., Ed.; CRC Press: New York, 1979; Vol. 1, pp 251−274.
(18) Waterhouse, R. N. Determination of lipophilicity and its use as a
predictor of blood-brain barrier penetration of molecular imaging
agents. Mol. Imaging Biol. 2003, 5, 376−389.
(19) Wyffels, L.; Muccioli, G. G.; Kapanda, C. N.; Labar, G.; De
Bruyne, S.; De Vos, F.; Lambert, D. M. PET imaging of fatty acid
amide hydrolase in the brain: synthesis and biological evaluation of an
11C-labelled URB597 analogue. Nucl. Med. Biol. 2010, 37, 665−675.
(20) Wilson, A. A.; Garcia, A.; Parkes, J.; Houle, S.; Tong, J.; Vasdev,
N. [11C]CURB: Evaluation of a novel radiotracer for imaging fatty acid
amide hydrolase by positron emission tomography. Nucl. Med. Biol.
2011, 38, 247−253.
(21) Wyffels, L.; Muccioli, G. G.; De Bruyne, S.; Moerman, L.;
Sambre, J.; Lambert, D. M.; De Vos, F. Synthesis, in vitro and in vivo
evaluation, and radiolabeling of aryl anandamide analogues as
candidate radioligands for in vivo imaging of fatty acid amide
hydrolase in the brain. J. Med. Chem. 2009, 52, 4613−4622.
(22) Wyffels, L.; De Bruyne, S.; Blanckaert, P.; Lambert, D. M.; De
Vos, F. Radiosynthesis, in vitro and in vivo evaluation of 123I-labeled
anandamide analogues for mapping brain FAAH. Bioorg. Med. Chem.
2009, 17, 49−56.
AUTHOR INFORMATION
Corresponding Author
*(P.L.) Phone: 732-594-0321. Fax: 732-594-9556. E-mail:
ping_liu2@merck.com.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank the department of Laboratory Animal Resources for
their assistance in animal dosing and sampling.
■
REFERENCES
■
(1) Ahn, K.; McKinney, M. K.; Cravatt, B. F. Enzymatic pathways
that regulate endocannabinoid signaling in the nervous system. Chem.
Rev. 2008, 108, 1687−1707.
(2) McKinney, M. K.; Cravatt, B. F. Structure and function of fatty
acid amide hydrolase. Annu. Rev. Biochem. 2005, 74, 411−432.
(3) Lambert, D. M.; Vandevoorde, S.; Jonsson, K. O.; Fowler, C. J.
The palmitoylethanolamide family: A new class of anti-inflammatory
agents? Curr. Med. Chem. 2002, 9, 663−674.
(4) Devane, W. A.; Hanus, L.; Breuer, A.; Pertwee, R. G.; Stevenson,
L. A.; Griffin, G.; Gibson, D.; Mandelbaum, A.; Etinger, A.;
Mechoulam, R. Isolation and structure of a brain constituent that
binds to the cannabinoid receptor. Science 1992, 258, 1946−1949.
(5) Rodriguez de Fonseca, F.; Navarro, M.; Gomez, R.; Escuredo, L.;
Nava, F.; Fu, J.; Murillo-Rodriguez, E.; Giuffrida, A.; LoVerme, J.;
Gaetani, S.; Kathuria, S.; Gall, C.; Piomelli, D. An anorexic lipid
mediator regulated by feeding. Nature 2001, 414, 209−212.
(6) Cravatt, B. F.; Demarest, K.; Patricelli, M. P.; Bracey, M. H.;
Giang, D. K.; Martin, B. R.; Lichtman, A. H. Supersensitivity to
anandamide and enhanced endogenous cannabinoid signaling in mice
lacking fatty acid amide hydrolase. Proc. Natl. Acad. Sci. U.S.A. 2001,
98, 9371−9376.
(7) Lichtman, A. H.; Shelton, C. C.; Advani, T.; Cravatt, B. F. Mice
lacking fatty acid amide hydrolase exhibit a cannabinoid receptorme-
diated phenotypic hypoalgesia. Pain 2004, 109, 319−327.
(8) Gaetani, S.; Dipasquale, P.; Romano, A.; Righetti, L.; Cassano, T.;
Piomelli, D.; Cuomo, V. The endocannabinoid system as a target for
novel anxiolytic and antidepressant drugs. Int. Rev. Neurobiol. 2009, 85,
57−72.
(23) Skaddan, M. B.; Zhang, L.; Johnson, D. S.; Zhu, A.; Zasadny, K.
R.; Coelho, R. V.; Kuszpit, K.; Currier, G.; Fan, K.-H.; Beck, E. M.;
Chen, L.; Drozda, S. E.; Balan, G.; Niphakis, M.; Cravatt, B. F.; Ahn,
K.; Bocan, T.; Villalobos, A. The synthesis and in vivo evaluation of
[18F]PF-9811: a novel PET ligand for imagingbrain fatty acid amide
hydrolase (FAAH). Nucl. Med. Biol. 2012, 39, 1058−1067.
(24) Wilson, A. A.; Hicks, J. W.; Sadovski, O.; Parkes, J.; Tong, J.;
Houle, S.; Fowler, C. J.; Vasdev, N. Radiosynthesis and evaluation of
[11C-carbonyl]-labeled carbamatesas fatty acid amide hydrolase
radiotracers for positron emission tomography. J. Med. Chem. 2013,
56, 201−209.
(25) The medicinal chemistry efforts on the pyrazole series is the
subject of another publication.
(26) The potency drop was primarily due to the core change. There
were two other changes from 3 to 4: chlorophenyl changed to
chloropyridyl; primary amide changed to N,N-dimethyl tertiary amide.
The former slightly decreased potency, and the latter slightly increased
potency. They should cancel each other. Chloropyridyl was used here
to further lower Log D.
(9) Gobbi, G.; Bambico, F. R.; Mangieri, R.; Bortolato, M.;
Campolongo, P.; Solinas, M.; Cassano, T.; Morgese, M. G.;
Debonnel, G.; Duranti, A.; Tontini, A.; Tarzia, G.; Mor, M.; Trezza,
V.; Goldberg, S. R.; Cuomo, V.; Piomelli, D. Antidepressant-like
activity and modulation of brain monoaminergic transmission by
512
dx.doi.org/10.1021/ml4000996 | ACS Med. Chem. Lett. 2013, 4, 509−513

ACS Medicinal Chemistry Letters
Letter
(27) Negishi, E.-I.; Hu, Q.; Huang, Z.; Wang, G.; Yin, N. Palladium-
or Nickel-Catalyzed Cross-Coupling Reactions with Organozincs and
Related Organometals. In Chemistry of Organozinc Compounds; Wiley:
New York, 2006; Part 1, pp 457−553.
(28) For its synthesis, see Supporting Information.
(29) During the synthesis of the cold 6, the similar C−S coupling
between methyl capped 23 and thiol 24 was high yielding (see
Supporting Information).
(30) The methylation of imidazole 8 to product 6 was highly regio-
selective where the regioisomer 7 was unable to detect from the crude
product by NMR. In fact, compound 7 had to be prepared by a
different route (see Supporting Information).
(31) Dahlquist, F. W. The meaning of Scatchard and Hill plots.
Methods Enzymol. 1978, 48, 270−299.
(32) Ueda, N.; Puffenbarger, R. A.; Yamamoto, S.; Deutsch, D. G.
The fatty acid amide hydrolyse (FAAH). Chem. Phys. Lipids 2000, 108,
107−121.
513
dx.doi.org/10.1021/ml4000996 | ACS Med. Chem. Lett. 2013, 4, 509−513