Pharmacokinetics and metabolism studies on (3-tert-butyl-7-(5- methylisoxazol-3-yl)-2-(1-methyl-1H-1,2,4-triazol-5-ylmethoxy) pyrazolo[1,5-d][1,2,4]triazine, a functionally selective GABAA α5 inverse agonist for cognitive dysfunction

Article abstract of DOI:10.1016/j.bmcl.2005.11.012

(3-tert-Butyl-7-(5-methylisoxazol-3-yl)-2-(1-methyl-1H-1,2, 4-triazol-5-ylmethoxy)pyrazolo[1,5-d][1,2,4]triazine (1) was recently identified as a functionally selective, inverse agonist at the benzodiazepine site of GABA_A α5 receptors and enhan

Full text of DOI:10.1016/j.bmcl.2005.11.012

Bioorganic & Medicinal Chemistry Letters 16 (2006) 872–875
Pharmacokinetics and metabolism studies on
(3-tert-butyl-7-(5-methylisoxazol-3-yl)-2-(1-methyl-1H-1,2,4-
triazol-5-ylmethoxy) pyrazolo[1,5-d][1,2,4]triazine, a
functionally selective GABAA a5 inverse agonist
for cognitive dysfunction
Philip Jones,a,* John R. Atack,a Matthew P. Braun,b Brian P. Cato,b Mark S. Chambers,a
Desmond OꢀConnor,a Susan M. Cook,a Sarah C. Hobbs,a Robert Maxey,a
Helen J. Szekeres,a Nicola Szeto,a Keith A. Wafforda and Angus M. MacLeoda
aMerck Sharp and Dohme Research Laboratories, The Neuroscience Research Centre, Terlings Park,
Eastwick Road, Harlow, Essex, CM20 2QR, UK
bMerck Research Laboratories, 126 Lincoln Avenue, PO Box 2000, Rahway, NJ 07065-0900, USA
Received 13 October 2005; revised 2 November 2005; accepted 2 November 2005
Available online 21 November 2005
Abstract—(3-tert-Butyl-7-(5-methylisoxazol-3-yl)-2-(1-methyl-1H-1,2,4-triazol-5-ylmethoxy)pyrazolo[1,5-d][1,2,4]triazine (1) was
recently identified as a functionally selective, inverse agonist at the benzodiazepine site of GABAA a5 receptors and enhances per-
formance in animal models of cognition. The routes of metabolism of this compound in vivo in rat have been well characterised, the
identities of the major metabolites are confirmed by synthesis and their biological profiles were evaluated. An unusual oxidation of
the pyrazolo[1,5-d][1,2,4]triazine core to the corresponding pyrazolo[1,5-d][1,2,4]triazin-4(5H)-one scaffold by aldehyde oxidase has
been observed.
Ó 2005 Elsevier Ltd. All rights reserved.
c-Aminobutyric acid (GABA) is the major inhibitory
neurotransmitter in the central nervous system and
GABAA receptors are GABA-gated chloride ion chan-
nels which can be modulated by simultaneous binding
of chemical entities to allosteric sites on the pentameric
ion channel complex, the most important of which is the
benzodiazepine (BZ) site.1,2 The major GABAA recep-
tor subtypes which possess a BZ binding site present
in the brain are a1bc2, a2bc2, a3bc2 and a5bc2,3 and
BZ inverse agonists have been shown to decrease the
frequency of ion channel opening in the presence of
GABA, resulting in increased neuronal excitability.
Non-selective BZ inverse agonists have been shown to
improve cognitive function in animal models,4 but are
anxiogenic,5 convulsant,6 and proconvulsant.7 Recently,
(3-tert-butyl-7-(5-methylisoxazol-3-yl)-2-(1-methyl-1H-
1,2,4-triazol-5-ylmethoxy)pyrazolo[1,5-d][1,2,4]triazine
(1) has been identified as a functionally selective, inverse
agonist at the BZ site of GABAA a5 receptors which
enhances performance in animal models of cognition
and does not exhibit convulsant, proconvulsant or anx-
iogenic activity (see Fig. 1).8 This manuscript describes
metabolic studies carried out on this compound, the
identification of the routes of metabolism and the char-
acterisation of the major metabolites.
Compound 1 displays a short half-life in rats t1/2 = 0.5 h,
oral bioavailability F = 52% and modest clearance
Cl = 27 11 mL/min/kg, whereas in dog t1/2 = 0.5 h,
F = 8% and Cl = 39 10 mL/min/kg and rhesus
t1/2 = 0.3 h, F < 2% and Cl = 33 2 mL/min/kg. Initial
analysis of plasma samples from the rat in vivo studies
by mass spectroscopy showed the presence of three
M + 16 metabolites. The same metabolites were detected
in incubations with rat hepatocytes. Analysis of mass
Keywords: GABA; GABAA a5; Inverse agonist; Cognition; Metabo-
lism; Aldehyde oxidase; Radiolabel.
*
Corresponding author. Tel.: +39 0691093559; e-mail: philip_jones@
merckcom
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.11.012

P. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 872–875
873
OHC
N N
N
N
O
NH
N
CH3
N
N
N
N
OH
O
O
EtO2C
EtO2C
H2NHNOC
ii, 90 %
5
OTs
Cl
O
N
2
i, 74 %
OH
iii, 22 %
OH
OH
N
3
4
CH3
N
N
N N
N
O
N N
N
N
O
Figure 1. (3-tert-Butyl-7-(5-methylisoxazol-3-yl)-2-(1-methyl-1H-1,2,4-
triazol-5-ylmethoxy)pyrazolo[1,5-d][1,2,4]triazine (1).
iv, 86 %
N
OH
v, 67 %
Cl
N
OH
O
N
OTs
N N
N
N N
N
N
N
O
O
CH3
CH3
N
N
N
N
N
OH
HO
6
7
8
CH3
CH3
N
N
N N
N
O
Scheme 1. Reagents and conditions: (i) Et3N; (ii) NH2NH2 Æ H2O; (iii)
5, xylene, reflux, then Dowtherm A, 200 °C; (iv) NaOH; (v) 7, K2CO3.
O
N
O
N
N
CH3
N
CH3
CH3
N
N
N
N
O
OH
O
O
O
O
CH3
O
O
NH
N
N
N
N
A
B
C
i, 79 %
O
ii, 35 %
O
9
CO2CH3
OH
10
11
O
A and B
C
SEM
N
Plasma
Microsomes
Hepatocytes
Rat, Dog, Rhesus Rat, Rhesus
Rat, Dog, Rhesus Not formed
Rat, Dog, Rhesus Rat, Rhesus
O
N
O
NH
N
iii, 66 %
iv, 99 %
v, 35 %
vi, 65 %
vii, 72 %
viii, 99 %
O
Figure 2. Potential metabolites.
OTs
CH3
N
N
12
13
spec fragmentation patterns tentatively assigned these as
shown in Figure 2: A, oxidation, of the 5-methylisoxaz-
ole; B, oxidation of the tert-butyl group; and C, oxida-
tion of the pyrazolotriazine core. Metabolite C is
observed in rat and rhesus plasma and following incuba-
tions with rat hepatocytes, but not microsomes. Howev-
er, it is seen in rat liver cytosol and S9 incubations,
where formation does not require NADPH. It is not ob-
served in the corresponding dog samples. These species
differences, the subcellular localisation and cofactor
requirements, and the nature of the biotransformation
(adjacent to a nitrogen on an electron-poor aromatic
ring), point to the involvement of a relatively unstudied
enzyme, aldehyde oxidase.9
N
N N
OH
O
N
O
N
O
H2NHNOC
NH
N
N
CH3
15
CH3
N
ix, 20 %
O
O
N
OH
CH3
CH3
N
N
N
14
16
N
N
Scheme 2. Reagents and conditions: (i) Ph3P@CHCO2CH3; (ii)
NH2NH2 Æ H2O, AcOH; (iii) TsCl, Et3N; (iv) SEMCl, NaH; (v)
NaOH; (vi) 7, K2CO3, 50 °C; (vii) DIBAL, À78 °C; (viii) 40% TFA/
DCE; (ix) 15, xylene, 145 °C.
ent synthetic approach as shown in Scheme 2 and dihy-
dro-4,4-dimethyl-2,3-furandione (9) was chosen as a
suitable hydroxyl-tert-butyl synthon. Wittig olefination
of 9 gave diester 10, which upon treatment with
NH2NH2 Æ H2O in AcOH resulted in 1,4-addition to
the least sterically encumbered end of the alkene, concur-
rent ring opening of the lactone, followed by cyclisation
of the hydroxyl onto the ethyl ester to furnish the pyraz-
ololactone 11 as described by Hackler.12 Further protec-
tion group manipulation,13 alkylation with 7 and
controlled DIBAL reduction of the lactone gave the crit-
ical cyclisation precursor, lactol 14. Addition of the isox-
azole hydrazide 1514 triggered a cascade sequence:
opening of the lactol; condensation with the hydrazide;
and thermal cyclisation to yield compound 16.
Synthetic chemistry was then deployed to synthesise
these compounds with a view to confirming the metabol-
ic routes and ensuring the biological activity observed
was not the result of an active metabolite.
The intermediate for the preparation of the hydroxylated
methylisoxazole A was readily obtained utilising a [3 + 2]
dipolar cycloaddition of an in situ generated nitrile oxide
and propargyl alcohol (Scheme 1). Conversion of the
resulting ester 3 to the hydrazide 4 allowed subsequent
cyclisation to the pyrazolo[1,5-d][1,2,4]triazine 6, upon
treatment of 4 with the pyrazoloaldehyde 58 and cyclisa-
tion in Dowtherm A at 200 °C.10 After unmasking the
pyrazolone, alkylation with 5-chloromethyl-1-methyl-
1H-1,2,4-triazole (7)11 furnished compound 8.
Formation of the putative aldehyde oxidase metabolite
C necessitated an entirely different cyclisation protocol,
Scheme 3. Oxidation of the pyrazoloaldehyde 5 to the
Introduction of a hydroxyl group onto the tert-butyl as
in hypothetical metabolite B required an entirely differ-

874
P. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 872–875
Table 1. Affinity and efficacy of metabolites
Compound Ki (a5) Efficacy in L(tkÀ) cells (%) Occ.
O
N
O
HO2C
HN NH
CH3
NH
N
nM
(3 mpk ip)
O
a1
a2
+6
À1 À15 À52 12%
a3
a5
5
i, 74 %
ii, 57 %
NH
N
1
8
1.4
1.1
À16
À11
À9 À55 > 95%
OTs
OTs
17
18
16
22
23
24
25
9.4
À41 À23 À32 À40 21%
CH3
O
N
700
370
21
1
ND
0
À8 À47
ND ND
ND
ND
ND
HN N
N
O
O
À45 À31 À39 À44 8%
ND ND ND ND ND
N
N
N
+
CH3
>666
O
iv, v
iii
N
82 % (2 steps)
NH
N
OTs
binding affinity;17 efficacy values were determined using
whole cell patch-clamp recordings from mouse fibro-
blast L(tkÀ) cells stably expressing the human GABAA
receptor subtypes relative to the GABA-evoked current
using a submaximal (EC20) GABA concentration;18 and
receptor occupancy was measured for those compounds
showing high affinity as determined by inhibition of
in vivo binding of [3H]Ro 15-1788, a non-selective BZ
receptor ligand.8 All six potential metabolites showed
either weak affinity or poor receptor occupancy at 3
mpk ip, at which dose 1 gave complete occupancy, it
was concluded that the activity of 1 was not due to cir-
culating metabolites.
19, 16 % 20, 16 %
OTs
HN N
N
O
O
N
N
N N
N
N
CH3
O
SEMO
CH3 vi, vii
61 % (2 steps)
O
N
OH
CH3
N
N
21
22
N
Scheme 3. Reagents and condition: (i) AgNO3, KOH; (ii) 15, BOPCl,
Et3N; (iii) mesitylene, 165 °C; (iv) SEMCl, NaH; (v) NaOH; (vi) 7,
K2CO3; (vii) TFA, then K2CO3.
corresponding carboxylic acid 17 was achieved with
AgNO3/KOH, and coupling to the isoxazole hydrazide
15 gave the required cyclisation precursor 18. However,
all attempts to prepare the pyrazolotriazinone skeleton
20 were accompanied with formation of the 1,3,4-
oxadiazole 19 as the major product (4:1 upon acid catal-
ysed cyclisation). Even the optimised procedure in
refluxing mesitylene yielded only 16% of the required
pyrazolo[1,5-d][1,2,4]triazin-4(5H)-one 20, together with
16% of 19.15 Subsequent protecting group manipulation
and alkylation with 7 gave compound 22, which was
identical to the material isolated from biological
sources.16
In order to assist ADME studies and to ensure that no
metabolites had been missed radiolabelled 1 was pre-
pared, introducing a 14C-isotope at the C-5 position of
the isoxazole, through a palladium-catalysed cross-cou-
pling of the iodoisoxazole 26 with an appropriately
radiolabelled methyl organometallic, Scheme 4.19 Lithi-
ation of [14C]methyl iodide with n-BuLi gave 14CH3Li,
which was then reacted with chlorostannatrane 27 to
give the required radiolabelled methyl organometallic
28.20 Coupling to the iodoisoxazole using catalytic
Pd(PPh3)4 in DMF at 100 °C occurred successfully to
give [14C]-1.21
Excretion studies with [14C]-1 in rat showed that the
dose was distributed equally between urine and faeces,
40 8% and 45 12%, respectively, the majority being
cleared in 24 h. Analysis of urine following dosing with
[14C]-1 showed one major component, the hydroxyl-
tert-butyl metabolite 16, accounting for 40% of the urine
extract. Other minor metabolites included 8, hydroxyl-
ation of the 5-methylisoxazole, and 24 a M + 32 metab-
olite (see Fig. 4).
Utilisation of combinations of the above synthetic
approaches enabled the preparation of the three possible
M + 32 metabolites, Figure 3. Compound 23 was pre-
pared from 4 + 17; 24 from 14 + 4; whereas 25 was pre-
pared using a modification of Scheme 2 using 13,
hydrazine and 5-methylisoxazole-3-carbonyl chloride.
With all the potential metabolites of 1 on hand, their
biological profiles were determined (Table 1): in vitro
14CH3
14CH3Li
+ BuLi
I
HN N
N
HN N
N
N
N
O
O
i
O
O
27
ii
CH3
N N
N
N N
N
N
Sn Cl
N
O
N
O
N
N
OH
OH
N N
N
N
O
14CH3
I
O
N
O
N
OH
OH
Sn 14CH3
28
iii, 37 %
N
N
N
N
CH3
CH3
O
N
O
N
N
N
N
N
O
N
CH3
N
N
CH3
N
N
[14C]-1
CH3
26
N
N
23
24
25
Scheme 4. Reagents and conditions: (i) hexanes, À78 °C; (ii) 27, THF,
À20 °C; (iii) 28, cat. Pd(PPh3)4, DMF, 100 °C.
Figure 3. Other potential metabolites.

P. Jones et al. / Bioorg. Med. Chem. Lett. 16 (2006) 872–875
875
Acknowledgment
The authors thank Laure Hitzel and Paul Scott-Stevens
for their support of this project.
References and notes
1. Barnard, E. A.; Skolnick, P.; Olsen, R. W.; Mohler, H.;
Sieghart, W.; Biggio, G.; Braestrup, C.; Bateson, A. N.;
Langer, S. Z. Pharmacol. Rev. 1998, 50, 291.
2. Bonnert, T. P.; McKernan, R. M.; Farrar, S.; Le Bour-
`
delles, B.; Heavens, R. P.; Smith, D. W.; Hewson, L.;
Rigby, M. R.; Sirinathsinghji, D. J.; Brown, N.; Wafford,
K. A.; Whiting, P. J. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 9891.
Figure 4. Radiochemical analysis of rat urine upon dosing of [14C]-1.
3. McKernan, R. M.; Whiting, P. J. Trends Neurosci. 1996,
19, 139.
4. McNamara, R. K.; Skelton, R. W. Psychobiology 1993,
21, 101.
5. Dorow, R.; Horowski, R.; Paschelke, G.; Amin, M.;
Braestrup, C. Lancet 1983, 322, 98.
6. Petersen, E. N. Eur. J. Pharmacol. 1983, 94, 117.
7. Little, H. J.; Nutt, D. J.; Taylor, S. C. Br. J. Pharmacol.
1984, 83, 951.
8. Chambers, M. S.; Atack, J. R.; Carling, R. W.; Collinson,
N.; Cook, S. M.; Dawson, G. R.; Ferris, P.; Hobbs, S. C.;
OꢀConnor, D.; Marshall, G.; Rycroft, W.; MacLeod, A.
M. J. Med. Chem. 2004, 47, 5829.
9. See Ghafourian, T.; Rashidi, M. R. Chem. Pharm. Bull.
2001, 49, 1066, further details will be published shortly.
10. Bryant, H. J.; Carling, W. R.; Chambers, M. S.; Hobbs, S.
C.; Jones, P.; MacLeod, A. M. PCT Int. Appl. WO 2001/
038331.
11. Carling, W. R.; Mitchinson, A.; Moore, K. W.; Russell,
M. G. N.; Scott, G.; Street, L. J. PCT Int. Appl. WO 2000/
023449.
12. Hackler, R. E.; Dreikorn, B. A.; Johnson, G. W.; Varie,
D. L. J. Org. Chem. 1988, 53, 5704.
13. For clarity only one SEM-isomer shown, both N-1 and N-
2 alkylation occurred.
14. Sung, K.; Lee, A.-R. J. Heterocycl. Chem. 1992, 29, 1101.
15. For a related cyclisation, see Lancelot, J.-C.; Maume, D.;
Robba, M. J. Heterocycl. Chem. 1981, 18, 1319.
16. This material was isolated from rat liver cytosol
incubations.
17. Hadingham, K. L.; Wingrove, P.; Le Bourdelles, B.;
Palmer, K. J.; Ragan, C. I.; Whiting, P. J. Mol. Pharma-
col. 1993, 43, 970.
18. Horne, A. L.; Hadingham, K. L.; Macaulay, A. J.;
Whiting, P.; Kemp, J. A. Br. J. Pharmacol. 1992, 107, 732.
19. The 5-iodoisoxazole 26 was prepared in a similar manner
to 8 using TMSCCH in place of propargyl alcohol and
performing a Sn-I exchange with I2/Et2O after the
cycloaddition reaction.
20. For related reactions, see (a) Forngren, T.; Samuelsson,
L.; Langstroem, B. J. Label. Compd. Radiopharm. 2004,
47, 71; (b) Braun, M. P.; Dean, D. C.; Melillo, D. G. J.
Label. Compd. Radiopharm. 1999, 42, 469.
Figure 5. Radiochemical analysis of rat bile upon dosing of [14C]-1.
Meanwhile, analysis of bile revealed that the major peak
(40%) did not correspond to any of the synthetic metab-
olites. Mass spectroscopy showed that the component
had a mass of 481, corresponding to a sulfate conjugate
of a M + 32 metabolite. Hydrolysis of this conjugate lib-
erated 23, the product of oxidation of the methylisoxaz-
ole and also by aldehyde oxidase (see Fig. 5).
In summary, through synthesis of potential metabolites
of 1 and radiolabel [14C]-1 the fate of the compound
in vivo has been fully characterised and profiling of
the authentic samples has revealed that they probably
have little relevance to the pharmacological activity seen
with 1. There is no single dominant metabolic pathway,
the two major routes being renal excretion of 16 and bil-
iary excretion of a sulfate conjugate of 23, each account-
ing for 15% of the dose. Direct excretion of 1 accounts
for <2% dose. Although the metabolite 22 arising from
oxidation of the pyrazolo[1,5-d][1,2,4]triazine core by
aldehyde oxidase is not a significant excreted metabolite,
oxidation by this enzyme contributes to the formation
of 23.
21. Full details will be published shortly.