Discovery, synthesis and characterization of a series of (1-alkyl-3-methyl-1H-pyrazol-5-yl)-2-(5-aryl-2H-tetrazol-2-yl)acetamides as novel GIRK1/2 potassium channel activators

Article abstract of DOI:10.1016/j.bmcl.2019.01.027

The present study describes the discovery and characterization of a series of 5-aryl-2H-tetrazol-3-ylacetamides as G protein-gated inwardly-rectifying potassium (GIRK) channels activators. Working from an initial hit discovered during a high-throughput screening campaign, we identified a tetrazole scaffold that shifts away from the previously reported urea-based scaffolds while remaining effective GIRK1/2 channel activators. In addition, we evaluated the compounds in Tier 1 DMPK assays and have identified a (3-methyl-1H-pyrazol-1-yl)tetrahydrothiophene-1,1-dioxide head group that imparts interesting and unexpected microsomal stability compared to previously-reported pyrazole head groups.

Full text of DOI:10.1016/j.bmcl.2019.01.027

Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery, synthesis and characterization of a series of (1-alkyl-3-methyl-
1H-pyrazol-5-yl)-2-(5-aryl-2H-tetrazol-2-yl)acetamides as novel GIRK1/2
potassium channel activators
Swagat Sharmaa, Krystian A. Kozekb,d, Kristopher K. Abneyc, Sushil Kumara, Nagsen Gautama,
⁎
Yazen Alnoutia, C. David Weaverb,d, Corey R. Hopkinsa,
a Department of Pharmaceutical Sciences, College of Pharmacy, University of Nebraska Medical Center, Omaha, NE 68198-6125, USA
b Department of Pharmacology, Vanderbilt University School of Medicine, Nashville, TN 37232, USA
c School of Graduate Studies and Research, Meharry Medical College, Nashville, TN 37208, USA
d Vanderbilt Institute of Chemical Biology, Vanderbilt University, Nashville, TN 37232, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
The present study describes the discovery and characterization of a series of 5-aryl-2H-tetrazol-3-ylacetamides as
G protein-gated inwardly-rectifying potassium (GIRK) channels activators. Working from an initial hit dis-
covered during a high-throughput screening campaign, we identified a tetrazole scaffold that shifts away from
the previously reported urea-based scaffolds while remaining effective GIRK1/2 channel activators. In addition,
we evaluated the compounds in Tier 1 DMPK assays and have identified a (3-methyl-1H-pyrazol-1-yl)tetra-
hydrothiophene-1,1-dioxide head group that imparts interesting and unexpected microsomal stability compared
to previously-reported pyrazole head groups.
GIRK
G protein-regulated inwardly-rectifying
potassium channel
Activator
Tetrazole
Introduction
activators (i.e., 1); however, the utility of this series of compounds was
limited by poor pharmacokinetic (PK) properties (low brain penetration
G protein-gated inwardly-rectifying potassium channels (GIRK),
also known as Kir3, are a family of inwardly-rectifying potassium
channels that are key effectors in GPCR signaling pathways that mod-
ulate excitability in cells.1,2 In mammals, four GIRK channel subunits
are expressed, GIRK1-4 (Kir3.1–3.4), which form either homo- or het-
erotetramers.3,4 GIRK1-3 are typically expressed in the brain, whereas
GIRK4 is found in heart atria, where it plays a key role in regulating
heart rate. Previous research supports roles for GIRK channels in a
number of normal and pathophysiological processes, including pain
perception,5–7 epilepsy,8–10 memory,11 and reward/addiction.12–14 Due
to our interest in discovering novel targets and tool compounds for
diseases of the brain, we have focused our efforts on the GIRK1/2
channel due to its localization in the CNS. However, due to a dearth of
subunit-selective GIRK activators, efforts to investigate the potential
benefits of GIRK channels in disease therapy have been hampered.
Herein, we report a new and distinct scaffold that shows promise as a
selective GIRK1/2 activator that we developed based on a molecule
discovered from our initial high-throughput screening (HTS) campaign
(See Fig 1).
and poor solubility).8,10 From the urea-series, we scaffold-hopped to the
phenylacetamide series, 2,15 which showed promise as a potent GIRK1/
2 activator. Although this series of compounds was able to improve
some pharmacokinetic properties (e.g., brain penetration), they suf-
fered from poor metabolic stability as measured in human and mouse
liver microsomes. Thus, we further investigated the compounds iden-
tified from a previous HTS campaign.8 We were interested in a new,
tetrazole containing scaffold, 3. This represented an interesting and
unique starting point for optimization because it did not contain the
privileged pyrazole head group. Further, the phenyl tetrazole was
curious because previous attempts to derivatize off the right side phenyl
groups of 1 or 2 did not prove productive.16,17 Herein, we report the
synthesis, biological characterization, and in vitro PK properties of this
new scaffold of GIRK1/2 activators.
Synthesis of the studied compounds started with the disubstituted
pyrazole head group (e.g., 7). For starting materials that were not
commercially available, the synthesis outlined in Scheme 1A and B was
followed. Namely, the appropriately-substituted ethyl ester, 4, was re-
acted with acetonitrile (n-BuLi, THF, −78 °C to rt) yielding 5.15,17 This
compound was then cyclized with the substituted hydrazine 6 under
We have previously reported on a series of urea-based GIRK1/2
⁎ Corresponding author.
E-mail address: corey.hopkins@unmc.edu (C.R. Hopkins).
https://doi.org/10.1016/j.bmcl.2019.01.027
Received 19 November 2018; Received in revised form 21 January 2019; Accepted 22 January 2019
0960-894X/©2019ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:Sharma,S.,Bioorganic&MedicinalChemistryLetters,https://doi.org/10.1016/j.bmcl.2019.01.027

S. Sharma et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Fig. 1. Previously reported GIRK1/2 activators, 1 and 2, and the new N-(bi-
cyclo[2.2.1]heptan-2-yl)-2-(5-phenyl-2H-tetrazol-2-yl) acetamide,
a starting
lead for GIRK1/2 activator optimization.
Scheme 2. Reagents and conditions: (a) NaN3, Et3N-HCI, toluene, 120 °C; (b)
bromoethyl acetate, NaOEt, EIOH; (c) NaOH, H2O, THF; (d) 2-chloroacethyl
chloride, Et3N, CH2Cl2; (e) NaOEt, EtOH, μW, 90 °C; (f) T3P, Et3N, CH2Cl2, rt.
Scheme 1. Reagents and conditions: (a) CH3CN, n-BuLi, THF, −78 °C to rt; (b)
AcOH, EtOH, reflux; (c) NaCNBH3, AcOH, MeOH, rt; (d) TFA, CH2Cl2, rt.
acidic conditions (AcOH, EtOH, reflux) to yield the desired dis-
ubstituted pyazoloamine, 7. For those substituted hydrazines that were
not commercially available, the synthetic procedure outlined in
Scheme 1B was followed. The Boc-hydrazine, 9, was reacted with the
carbonyl, 8, under reductive amination conditions (NaCNBH3, AcOH,
MeOH, rt) to yield 10, which was subjected to TFA in order to remove
the Boc protecting group and yield the desired hydrazine, 6.18
knowledge from previous work, we attached the 3-methyl-1-cyclohex-
ylpyrazole group to generate 15a, which we found to be a potent and
efficacious GIRK1/2 activator (EC50 = 96 nM; Efficacy = 92%). This
compound was also a GIRK1/4 activator, but it demonstrated an ap-
proximate 3-fold preference for GIRK1/2. Methylation of the amide
nitrogen led to an inactive molecule (data not shown). Moving from the
amide to the thioamide, 15b, produced an active molecule; however,
we observed an approximate 7-fold loss of potency (EC50 = 623 nM).
Further saturated 6-membered analogs were evaluated and the tetra-
hydropyran, 15c, lost activity compared to 15a. However, the 4,4-di-
fluorocyclohexane was equipotent (15d, EC50 = 84 nM), but potency
eroded with substituting dimethyl, 15e, for the difluoro. Further
branched alkyl analogs were less potent (15f–h), and the cyclopropyl
was also inferior, but some activity could be regained with the cyclo-
pentyl group (15j, EC50 = 163 nM). The 5- and 6-membered sulfone
derivatives (15k, l) were also less active; however, these were shown to
have superior in vitro PK properties (vide infra). Branched groups at the
3-position of the pyrazole were not productive analogs (15m–p). The 1-
cyclohexylmethyl group, as we have seen previously, was an active
analog (15q, EC50 = 176 nM); however, this molecule lost selectivity
between GIRK1/2 and GIRK1/4. Other substituents (15r), or 5-mem-
bered pyrazole replacements (15s, t) were not active, nor were pyridine
replacements for the pyrazole (15v).
The synthesis of the final tetrazole or heterocyclic compounds was
completed as outlined in Scheme 2A–C. The tetrazole-containing com-
pounds were synthesized by reacting the appropriately-substituted aryl
cyanide with NaN3 (Et3N∙HCl, toluene, 120 °C) to yield 12.19 Next, 12
was reacted with bromoethyl acetate (NaOEt, EtOH)20 and followed by
saponification of the ester (NaOH, H2O, THF) to give the acid coupling
partner, 13. In an analogous fashion to the final targets, the dis-
ubstituted pyrazoloamine, 7, was coupled with 2-chloroacetyl chloride
(Et3N, CH2Cl2) affording the α-chloroamide, 14. This compound could
then be reacted with a heterocyclic partner under basic conditions to
yield the final targets, 15. Also, the acid coupling partner (e.g., 13 or
16) could be reacted with 7 under standard amide coupling procedures
(T3P, Et3N, CH2Cl2) to yield the final targets, 17.21
The initial structure-activity relationship (SAR), which was centered
around the left-hand amide portion, is detailed in Table 1. The initial
HTS hit molecule was resynthesized and tested on HEK293 cells ex-
pressing GIRK channels using thallium flux assays, as previously de-
scribed.8 This molecule demonstrated weak activity against GIRK1/2
(3, EC50 = 1980 nM, Efficacy = 13% of a maximally effective con-
centration of VU0466551), which agreed with previous data showing
non-pyrazole “head groups” to be weak GIRK1/2 activators. Using our
We next set out to evaluate the right-hand phenyl group (Table 2).
Replacement of the phenyl with either 3-pyridyl (16a) or 2-pyridyl
(16b) led to a > 20-fold loss of activity. Addition of halogens (either
mono- or di-halogens) maintained activity. The 3,4-dichloro and 3,4-
difluoro were notable exceptions, with > 10-fold loss of activity. In
2

S. Sharma et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Table 1
Table 1 (continued)
SAR of the amide portion of compound, 3.
Cmpd
Structure
GIRK1/2
GIRK1/4
SEMa;
(nM
%
SEMa;
Cmpd
Structure
GIRK1/2
GIRK1/4
(nM
%
SEMb)
SEMb)
(nM
%
SEMa;
(nM
%
SEMa;
SEMb)
SEMb)
15l
> 10,000;
> 25
Inactive
3
1983
13
546;
2761
7
753;
1
1
15a
96
92
7;
2
259
63
25;
2
15m
843
78
139;
> 10,000;
> 19
4
15b
15c
623
80
72;
1256
36
92;
3
1
15n
15o
15p
> 10,000;
> 10
Inactive
Inactive
435
115
22;
2
2,037
109
114;
3
Inactive
15d
15e
84
68
9;
2
383
52
36;
2
271
45
17;
16;
659
15
57;
> 221;
> 19
Inactive
1
0
15q
15r
176
77
283
58
26;
15f
556
107
55;
3
> 4000;
> 96
2
3
2
1
15g
> 5000;
> 69
> 10,000;
> 20
3019
66
517;
4093
25
528;
15h
> 6000;
> 105
> 10,000;
> 97
15s
15t
15u
Inactive
Inactive
495
Inactive
Inactive
848
15i
15j
> 5000;
> 78
> 10,000;
> 39
163
105
7;
612
84
32;
1
2
75;
121;
74
2
48
2
15k
1034
93
66;
2168
75
172;
1
2
15v
Inactive
Inactive
a
Potency values were obtained from triplicate determinations; values are
average of n = 3.
b
Reported efficacy values shown are obtained from triplicate determina-
tions; values are average of n = 3 and are normalized to standard compound,
VU0466551.
3

S. Sharma et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Table 2
Table 3
SAR of the tetrazole substituent.
Aryl and heterocyclic replacements.
SEMa;%
Cmpd
Structure
GIRK1/2 (nM
SEMa;
GIRK1/4 (nM
SEMa;
Cmpd
Structure
GIRK1/2
GIRK1/4 (nM
SEMb)
(nM
%
SEMa;
SEMb)
%
SEMb)
%
SEMb)
16a
16b
16c
1931
96
89;
1
4908
80
165;
17a
17b
116
87
3;
1
412
76
26;
1
1
14,571
1379;
> 33,000;
375
87
33;
2
1027
49
85;
1
81
3
> 41
300
29;
662
66
75;
89
2
2
1
2
1
2
16d
16e
1217
84
91;
2201
70
130;
17c
17d
631
72
154;
5
1132
27
79;
1
> 11,000;
> 69
3979
22
547;
980
82
68;
2122
51
331;
1
17e
17f
17g
511
88
29;
998
49
105;
15;
2
2
1
16f
180
92
11;
634
74
45;
65
72
6;
2
348
52
1
2
2
16g
81
7;
264
22;
96
2
80
1326
44
227;
> 3000;
> 18
3
16h
187
88
22;
447
60
42;
3
2
17h
17i
> 3000;
> 60
> 10,000;
> 21
16i
16j
16k
116
93
6;
392
79
23;
370
72
95;
> 4000;
> 51
1
2
4
5
441
95
57;
1,337
68
181;
4
17j
17k
17l
1657
69
297;
115;
318;
1825
20
424;
6
2
> 3000;
> 83
> 10,000;
> 38
1520
36
Inactive
16l
111
86
9;
408
64
29;
1
2
2
1
6
16m
129
15;
435
113;
93
3
63
4471
49
4352
326;
11
1
16n
16o
175
90
13;
9;
779
65
78;
18;
1
2
1
1
117
299
76
40
a
Potency values were obtained from triplicate determinations; values are
average of n = 3.
16p
16q
> 3000;
> 68
> 10,000;
> 25
b
Reported efficacy values shown are obtained from triplicate determina-
241
11;
448
39;
tions; values are average of n = 3 and are normalized to standard compound,
VU0466551.
94
1
2
50
1
4
16r
16s
16t
183
86
17;
18;
25;
850
57
200;
compounds were potent and efficacious against GIRK1/2, they did not
impart selectivity versus GIRK1/4 (only approximately 3–5-fold selec-
tive). The 2-fluoro-4-pyridyl analog maintained potency (16j,
EC50 = 441 nM); however, the 2-fluoro-6-pyridyl analog was much less
potent (16k, EC50 = > 3000 nM). Other substituents that were well
tolerated and generated active compounds included alkyl, methoxy,
trifluoromethyl, and cyano (16l–s). The cycloalkyl substituents appear
to be poorly tolerated (16p); however, we only investigated a small
sample size.
183
80
619
55
80;
89;
2
1
3
1
712
89
1510
51
a
Potency values were obtained from triplicate determinations; values are
average of n = 3.
b
Reported efficacy values shown are obtained from triplicate determina-
tions; values are average of n = 3 and are normalized to standard compound,
VU0466551.
Finally, we explored aryl and heterocyclic replacements for the
tetrazole moiety (Table 3). Replacing the tetrazole with the triazole led
to two regioisomers (17a, b). The 4-substituted triazole was more active
(17a, EC50 = 116 nM); however, the 5-substituted analog did retain
some potency (17b, EC50 = 375 nM). The 5-phenyloxazol-2-ylacetic
acid derivative (17c, EC50 = 631 nM) was much more potent than the
addition, the fluoro substituted compounds were more active than the
chloro substitutions (e.g., 16g, EC50 = 81 nM, vs. 16c, EC50 = 300 nM,
and 16i, EC50 = 116 nM, vs. 16d, EC50 = 1218 nM). Although these
4

S. Sharma et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
when compared to 1. The %Emax has been normalized to the standard
compound, 1. Data are the average of three independent determina-
tions. Error bars represent the standard error of the mean (SEM).
Having established a robust SAR for this new scaffold of GIRK1/2
activators, we next evaluated select compounds in a panel of Tier 1 in
vitro DMPK assays (Table 4, Q2 Solutions, Indianapolis, IN).22,23 Un-
fortunately, all of the compounds tested were unstable in both human
and mouse liver microsomes, with the singular exception of 15k. This
compound was unique in that it possessed the cyclic five-membered
sulfone moiety on the pyrazolo head group. Compound 15k was stable
in both human and mouse liver microsomes and was stable in human S9
fractions.24,25 In addition to displaying excellent stability in liver mi-
crosomes, 15k, also showed increased free fraction in human plasma (%
fu = 3.7). Unfortunately, this improvement in stability did not include
an increase in potency as 15k was much less potent (EC50 = 1034 nM).
Curiously, all of the compounds tested had poor recovery in mouse
plasma, leading us to theorize that the compounds were unstable due to
the amidases present in plasma.26 Several compounds were made to test
this hypothesis (thioamide, 15b, amide N-methylation, and α-carbon
methylation). However, all of these compounds also showed poor re-
covery in mouse plasma (data not shown).
Fig. 2. Activation CRC’s for VU0466551, 1 and 16g.
2-phenyloxazol-4-ylacetic acid analog (17d, EC50 = > 11,000 nM).
Although, the oxazole was much less active than the tetrazole or tria-
zole. Further exploration of the regio-chemistry of the nitrogens and
determination of whether these atoms were necessary led us to the
pyrazole derivatives (17e, f). Interestingly, we identified a distinct re-
gioisomeric requirement for potency. The 3-phenyl-1H-pyrazole, 17e,
was ∼8-fold less potent than the 4-phenyl-1H-pyrazole, 17f
(EC50 = 511 nM vs. EC50 = 65 nM). The nitrogen placement in the 4-
phenyl-1H-pyrazole maintained the nitrogen in a similar arrangement
as the triazole (17a) and the original tetrazole (15a). Thiazole, 17g,
and thiadiazole, 17h, did not maintain potency, and this can be ex-
plained by the position of the nitrogen/sulfur in the thiazole and the
increased ring size in the thiadiazole (or due to the presence of the
sulfur atom, in general). The meta-phenyl substituted analog possessed
moderate activity (17i, EC50 = 370 nM); however, all of the para-sub-
stituted analogs were less active (17j–l).
In conclusion, we have identified a novel series of 5-aryl-2H-tet-
razol-3-ylacetamides as GIRK1/2 channel activators. This series was
born out of an HTS hit molecule. SAR around the initial head group
found that the pyrazolo privileged group was optimal. Further SAR
identified the optimal tetrazole substituent as well as heterocyclic re-
placements. These compounds were found to be unstable in liver mi-
crosomes as well as mouse plasma, with the one exception being the
cyclic five-membered sulfone derivative. Although the in vitro DMPK
profile of these compounds are less than ideal, our discovery of these
compounds increases the number of potent, efficacious, and modestly-
selective GIRK1/2 channel activators from a novel scaffold, and these
compounds will add to the armament of GIRK channel researchers.
Further work and evaluation of lead molecules is ongoing and will be
reported in due course.
Acknowledgments
The thallium flux data resulting from the treatment of GIRK1/2 or
GIRK1/4 expressing cells with an activator (1 or 16g) is shown in Fig. 2.
As can be seen, both 1 and 16g activate GIRK1/2 preferentially over
GIRK1/4; however, 16g is more selective for GIRK1/2 vs. GIRK1/4
The authors would like to thank Q2 solutions for the in vitro DMPK
experiments. The study was supported by a grant from the NIH (NIMH:
R01MH107399) to C.R.H. and C.D.W. C.D.W. is an owner of WaveFront
Table 4
In vitro Tier 1 DMPK of select compounds.
Cmpd
Intrinsic Clearance (mL/min/kg)a,b
Plasma unbound (%fu)a,c
hCLINT
hCLHEP
mCLINT
mCLHEP
Human
Mouse
*
*
*
*
*
*
*
*
*
15a
15b
15d
15k
15q
16d
16g
17a
17b
186.5
346.5
77.3
18.9
18.9
15.9
< 10.7
20.4
17.4
17.8
14.6
18.3
2917.5
> 5930
722.7
87.0
1.1
> 88.7
80.0
< 0.06
1.5
< 23.1
753.1
99.9
< 49.5
> 5930
489.1
< 31.9
88.7
3.7
0.3
76.0
< 0.3
1.0
162.4
54.6
2710.6
596.0
87.1
78.2
3.0
209.1
3487.8
87.7
0.4
Stability in Human Liver S9 fractions
T1/2 (min)
CLINT (mL/min/kg)
2.6
CLHEP (mL/min/kg)
< 1.0
ERc
15k
647.5
< 0.05
a
b
c
In vitro DMPK studies performed at Q2 Solutions, Indianapolis, IN
Intrinsic and predicted hepatic clearance based on experiments in liver microsomes. b%fu = % fraction unbound.
ER = extraction ratio and defines the % of drug amount eliminated by the liver from the incoming blood supply in every pass through the liver (0 not eliminated
by liver, 100 completely eliminated by liver).
* Poor recovery.
5

S. Sharma et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Biosciences, maker of the Panoptic kinetic imaging plate reader used in
these studies.
2014;111:10755.
11. Sánchez-Rodríguez I, Temprano-Carazo S, Nájera A, et al. Sci Rep. 2017;7:14658.
12. Fernandez de Velasco EM, McCall N, Wickman K. Int Rev Neurobiol. 2015;123:201.
13. Rifkin RA, Moss SJ, Slesinger PA. Trends Pharmacol Sci. 2017;38:378.
14. Sugaya N, Kobayashi T, Ikeda KJ. Drug Alc Res. 2013;2:235823.
15. Wieting JM, Vadukoot AK, Sharma S, et al. ACS Chem Neurosci. 2017:8.
16. Ramos-Hunter SJ, Engers DW, Kaufmann K, et al. Bioorg Med Chem Lett.
2013;23:5195.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2019.01.027.
17. Wen W, Wu W, Weaver CD, Lindsley CW. Bioorg Med Chem Lett. 2014;24:5102.
18. Akkari, R.; Alvey, L. J.; Bock, X. M.; Brown, B. S.; Claes, P. I. R.; Cowart, M. D.; De
Lemos, E.; Desroy, N.; Duthion, B.; Gfesser, G. A.; Gosmini, R. L. M.; Housseman, C.
G.; Jansen, K. K.; Ji, J.; Kym, P. R.; Lefrancois, J.-M.; Mamoliti, O.; Menet, C. J. M.;
Merayo, N. M.; Newsome, G. J. R.; Palisse, A. M. E.; Patel, S. V.; Pizzonero, M. R.;
Shrestha, A.; Swift, E. C.; Van der Plas, S. E.; Wang, X.; De Blieck, A. US2017/
0101405, 2017.
References
1. Dascal N. Cell Sig. 1997;9:551.
2. Yamada M, Inanobe A, Kurachi Y. Pharmacol Rev. 1998;50:723.
3. Aryal P, Dvir H, Choe S, Slesinger PA. Nat Neurosci. 2009;12:988.
4. Hibino H, Inanobe A, Furutani K, Murakami S, Findlay I, Kurachi Y. Physiol Rev.
2010;90:291.
19. Koguro K, Oga T, Mitsui S, Orita R. Synthesis. 1998:910.
20. Maxwell JR, Wasdahl DA, Wolfson A, Stenberg VI. J Med Chem. 1984;27:1565.
21. Vishwanatha BTM, Panguluri NR, Sureshbabu VV. Synthesis. 2013;45:1569.
22. Chang G, Steyn SJ, Umland JP, Scott DO. ACS Med Chem Lett. 2010;1:50.
23. Liu X, Wright M, Hop CECA. J Med Chem. 2014;57:8238.
5. Lyu C, Mulder J, Barde S, et al. Mol Pain. 2015;11:44.
6. Bruehl S, Denton JS, Lonergan DF, et al. Pain. 2013;154:2853.
7. Nishizawa D, Fukuda K, Kasai S, et al. J Pharmacol Sci. 2014;126:253.
8. Kaufmann K, Romaine I, Days E, et al. ACS Chem Neurosci. 2013;4:1278.
9. Luján R, Fernandez de Velasco EM, Aguado C, Wickman K. Trends Pharmacol Sci.
2014;37:20.
24. Richardson SJ, Bai A, Kulkami AA, Moghaddam MF. Drug Metab Lett. 2016;10:83.
25. Gautam N, Bathena SP, Chen Q, Natarajan A, Alnouti Y. Biomed Chromatogr.
2013;27:900.
26. Teffera Y, Berry LM, Brake RL, et al. Drug Metab Dispos. 2013;41:238.
10. Wydeven N, Fernandez de Velasco EM, Du Y, et al. Proc Natl Acad Sci USA.
6