Pyrazolopyridazine alpha-2-delta-1 ligands for the treatment of neuropathic pain

Article abstract of DOI:10.1016/j.bmcl.2010.05.026

Optimization of the novel alpha-2-delta-1 ligand 4 provided compounds 37 and 38 which have improved DMPK profiles, good in vivo analgesic activity and in vitro selectivity over alpha-2-delta-2. An in-house P-gp prediction programme and the MetaSite softwa

Full text of DOI:10.1016/j.bmcl.2010.05.026

Bioorganic & Medicinal Chemistry Letters 20 (2010) 4683–4688
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pyrazolopyridazine alpha-2-delta-1 ligands for the treatment
of neuropathic pain
*
James W. Myatt , Mark P. Healy, Gianpaolo S. Bravi, Andrew Billinton, Christopher N. Johnson,
Kim L. Matthews, Karamjit S. Jandu, Wenjing Meng, Anne Hersey, David G. Livermore, Clement B. Douault,
Jason Witherington, Rino A. Bit, James E. Rowedder, Jason D. Brown, Nick M. Clayton
Neurosciences Centre of Excellence for Drug Discovery, GlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Optimization of the novel alpha-2-delta-1 ligand 4 provided compounds 37 and 38 which have improved
DMPK profiles, good in vivo analgesic activity and in vitro selectivity over alpha-2-delta-2. An in-house
P-gp prediction programme and the MetaSiteÒ software package were used to help solve the specific
problems of high P-gp efflux and high in vivo clearance.
Received 14 April 2010
Revised 9 May 2010
Accepted 11 May 2010
Available online 31 May 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Neuropathic pain
Alpha-2-delta
Pyrazolopyridazine
P-glycoprotein
Metasite
Alpha-2-delta is an auxiliary subunit of voltage-gated calcium
channels that was originally reported to modify the activity of
the channel formed by the alpha-1 protein.1 Binding to alpha-2-
delta was suggested to inhibit the flux of calcium into the neuron2,3
and subsequently inhibit the evoked release of various neurotrans-
mitters.4–6 More recent data suggests that alpha-2-delta is impor-
tant in the trafficking of calcium channels from the endoplasmic
reticulum to the cell membrane.7 Alpha-2-delta is up-regulated
after nerve injury and is believed to play a significant role in estab-
lishing the sensitization which causes neuropathic pain,8 therefore
alpha-2-delta ligands have become attractive targets for the dis-
covery of neuropathic pain treatments.9 Indeed, this mechanism
is believed to be the mode-of-action of the currently marketed
compounds gabapentin and pregabalin (1 and 2, Fig. 1).10–13
Gabapentin and pregabalin are non-selective ligands of two of
the four known alpha-2-delta subtypes, alpha-2-delta-1 and al-
pha-2-delta-2.14 Studies on the distribution and differential ef-
fects9 of the two subtypes have suggested that the analgesia
produced by these compounds is due to interaction with alpha-
2-delta-1 and the major side effects of sedation, dizziness and
ataxia are caused by interaction with alpha-2-delta-2. Our aim
was to find novel alpha-2-delta-1 selective ligands to provide
patients with relief from neuropathic pain without the side
effects suffered during treatment with the current gold-standard
medicines.
A knowledge-based design approach (which will be the subject
of a future publication) starting from the known alpha-2-delta li-
gand 315–18 (Fig. 1) gave rise to the amino-substituted pyrazolopy-
CO2H
NH2
CO2H
N
NH2
N
2
1
O
N
N
3
Figure 1. Gabapentin (1), pregabalin (2) and alpha-2-delta-1 ligand (3).
N
Cl
i
N
N
N
N
N
N
N
O
N
O
4
5
* Corresponding author. Tel.: +44 7789308416.
E-mail address: jameswmyatt@yahoo.co.uk (J.W. Myatt).
Scheme 1. Reagents and conditions: (i) pyrrolidine, NEt3, EtOH, 140 °C MW, 100%.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.05.026

4684
J. W. Myatt et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4683–4688
Table 1
and azepine 10 also had significantly lower potency than com-
pound 4. This indicated the smaller-ring pyrrolidine and azetine
groups were optimal in this position.
Effect of pyridazine amino group on alpha-2-delta-1 potency
R
DMPK profiling of 4 gave a moderate in vitro intrinsic clearance
in rat microsomes of 3.5 ml/min/g. However, this translated to a
high in vivo rat clearance of 84 ml/min/kg. To reduce this high
clearance, the lipophilicity of this template was lowered through
the introduction of polar groups into the amino-ring substituent
to give compounds 11–16, Table 2.
N
N
N
O
N
Compound
R
c log P
IC50 (nM)
a
2d1
a
2d2
Piperazine 11 and 1,4-diazepine 13 were less potent than the
piperidine 9 and azepine 10, respectively, but did show low intrin-
sic clearance (both <0.5 ml/min/g). This corresponded to a moder-
ate in vivo rat clearance of 47 ml/min/kg in the case of 13.
N-Methyl piperazine 12 gave a higher intrinsic clearance of
2.5 ml/min/g but a similar in vivo clearance of 57 ml/min/kg and
4-hydroxypiperidine 14 showed similar values to 13. The 3-(S)-
aminopyrrolidine 15 gave a lower in vivo clearance of 34 ml/min/
kg but the 3-(R)-hydroxypyrrolidine 16 was much less metaboli-
cally stable with a clearance of 78 ml/min/kg, despite the low
in vitro value (<0.5 ml/min/g). This showed that low in vitro clear-
ance was not always predictive of low in vivo clearance in this ser-
ies. The similar plasma protein binding data for these compounds
(15: 19% uBl, 4% uBr; 16: 14% uBl, 8% uBr) did not help to explain
the discrepancy. It is notable that alcohol 16 showed remarkably
higher selectivity over alpha-2-delta-2 (approximately 300-fold)
compared to amine 15 (fivefold) and that their opposite enantio-
mers were equipotent to 15 and 16.
Compound 14 was progressed the Complete Freund’s Adjuvant
(CFA)19 pharmacodynamic model of pain to establish whether this
novel series of alpha-2-delta-1 selective ligands is effective in vivo.
Pleasingly, 14 showed a dose-related reversal of induced hypersen-
sitivity with an ED50 of 5.9 mg/kg that was not statistically differ-
ent from pregabalin (dosed at 10 mg/kg) at the top dose of
30 mg/kg. No overt side effects were observed in the study. This
experiment also revealed a moderate brain:blood ratio of 0.17:1
despite in vitro Kbb measurement20 giving a ratio of 5:1. This dis-
crepancy could be explained by the high P-glycoprotein (P-gp) ef-
flux ratio21 of 9:1.
3
4
–NMe2
5.1
3.7
28
501
2
22
59
N
6
3.2
1
N
7
8
–NMe2
–NHMe
3.6
3.4
7
22
158
537
9
4.3
4.9
42
22
<1000
2131
N
N
10
O
O
i
ii
N
H2N
OR
N
OR
Cl
H
22
23
O
O
iii
N
N
O
HN
N
N
OR
N
OR
O
P-gp efflux ratios were measured for a sub-set of the com-
pounds shown in Table 2. The less polar parent compound 4
proved to not be a P-gp substrate (efflux ratio of 1.4:1) whereas
the more polar compounds 11, 13, 15 and 16 were all substrates,
giving ratios of 21:1, 14:1, 29:1 and 9:1, respectively, demonstrat-
ing that this liability had arisen due to the introduction of the polar
groups. In an effort to address this issue, prospective target com-
pounds were analysed by the in-house P-gp prediction programme
which uses the Abraham BetaH, Abraham Alpha, total charge,
c log D at pH 7.4 and c log P to calculate the probability of the test
compound being a substrate for P-gp. Table 3 shows that this mod-
el correctly predicted the high P-gp efflux ratios of compounds 14,
15 and 16 and the low P-gp efflux ratios of compounds 4 and 8.
This model also predicted that the putative methoxy derivatives
17 and 18 would be borderline P-gp substrates. These compounds
were duly prepared and tested, giving P-gp efflux ratios below the
accepted in-house threshold value of 2:1 and so were deemed to be
non-substrates.
25
24
iv, v
N
R = CF3, 19
R = CH2CF3, 20
N
N
N
N
OR
Scheme 2. Reagents and conditions: yields R = CF3/R = CH2CF3: (i) (a) isoamyl
nitrite, TFA, MeCN, 0 °C; (b) ethyl 2-chloro-3-oxobutanoate, NaOAc, ethanol, H2O,
53%/96%; (ii) 2,4-pentadione, NaH, THF, 81%/74%; (iii) N2H4ÁH2O, EtOH, 100 °C, 82%/
94%; (iv) POCl3, 120 °C, 46%/24%; (v) pyrrolidine, NEt3, 160 °C MW, 31%/53%.
ridazine compound 4, prepared from the commercially available
chloride 5 as shown in Scheme 1. This Letter, describes the subse-
quent optimization of this template to develop a selective alpha-2-
delta-1 ligand for further profiling.
Exploration of the SAR of the pyrazolopyridazine series com-
menced with the variation of the amino moiety on the pyridazine
ring. Compounds 6–10 (Table 1) were prepared analogously to 4
(Scheme 2), employing the requisite amines for the chloride
displacement.
Reducing the amino-ring size to give the azetidine 6 retained
potency and moderately improved selectivity. The acyclic dimeth-
ylamino derivative 7 had lower potency at both receptor subtypes
and the removal of one methyl group to give 8 reduced these
potencies further. Both the larger-ring derivatives piperidine 9
In addition to being non-P-gp substrates, compounds 17 and 18
showed good potency and greater than 10-fold selectivity over al-
pha-2-delta-2. Both compounds exhibited moderate to high in vivo
clearance in rat of 51 and 61 ml/min/kg, respectively, which was a
significant improvement compared with compound 4 but this was
still predicted to limit exposure in higher species. Further optimi-
zation was clearly required to reduce in vivo clearance to a more
acceptable level and to guide this effort an in silico analysis of com-
pound 4 using the MetaSiteÒ software package was undertaken.22
MetaSiteÒ uses computational methods to calculate the accessibil-

J. W. Myatt et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4683–4688
4685
Table 2
Effect of pyridazine amino group on in vitro and in vivo clearance
R
N
N
N
N
O
Compound
R
c log P
IC50 (nM)
CLi (ml/min/g)
CLb (ml/min/kg)
a
2d1
a
2d2
4
3.7
2.9
2
22
3.5
84
—
N
H
N
11
213
169
58
1000
1071
1000
<0.5
N
N
N
12
13
14
3.4
2.9
2.2
2.5
<0.5
<0.5
57
47
55
H
N
N
OH
32
398
N
NH2
15
16
2.5
2.4
2
5
10
0.8
34
78
N
N
OH
1585
<0.5
ity and reactivity of each atom of the test substrate towards the
reactive heme of a range of CYP450 enzymes. The product of these
two values is the probability of metabolism at that specific atom.
Figure 2 shows the resulting plot for compound 4 which clearly
identifies the CH2 of the ethoxy group as the most likely site of
metabolism by the CYP450 3A4 isoform.
To explore the potential for improving metabolic stability by
blocking this ethoxy group, the trifluoromethoxy, trifluoroethoxy
and isopropoxy analogues 19–21 were synthesized. Compounds
19and20 wereprepared bytheroute describedinScheme2, starting
from the trifluoromethoxy aniline and the trifluoroethoxy aniline,
respectively. Alkoxyaniline 22 underwent a Japp–Klingemann23
reaction with ethyl 2-chloroacetoacetate to give the chloro-hydra-
zone 23. Chloride displacement with acetoacetone and subsequent
ring-closure gave the pyrazole 24 which was treated with hydrazine
to form the pyridazine ring of intermediate 25.24 Chlorination and
subsequent displacement with pyrrolidine gave the pyrazolopyrid-
azines 19 and 20.
The Japp–Klingemann reaction failed with isoproxy aniline so
analogue 21 was provided by the alternative route outlined in
Scheme 3. 2,4-Pentanedione 26 was treated with oxalyl chloride
and magnesium chloride to give the 4-acetyl-5-methyl-2,3-furan-
dione 27.25 Pyrazole 28 was then formed by treatment of 27 with
4-isoproxyphenylhydrazine. The final three steps of pyridazine for-
mation with hydrazine, chlorination and displacement proceeded
as described above to provide compound 21.
All three analogues 19, 20 and 21 showed reduced in vitro clear-
ance in rat microsomes compared with compound 4, but this
improvement was at the expense of an approximate 10-fold drop
in potency (Table 4). The trifluoroethoxy analogue 20 was taken for-
ward to an in vivo rat PK study, giving a CLb of 49 ml/min/kg, again a
significant improvementcompared with 4 (84 ml/min/kg). These re-
sults correlate well with the outcome of the MetaSiteÒ analysis, sug-
gesting the ethoxy group is a major liability for metabolism. Further
analogues 30 and 31 that omitted the 4-ethoxyphenyl group in
question were prepared by the route described in Scheme 2. The
2,4-dichlorophenyl derivative 30 and the 4-chloro-2-trifluorometh-
oxyphenyl derivative 31 both had reduced in vitro rat clearance
while retaining acceptable levels of alpha-2-delta-1 potency. Com-
pound30 was progressedto an in vivo rat PK study and gavea similar
moderate CLb value to compound 20 (51 ml/min/kg).
Returning to the Metasite results, the C-3 and C-4 methylene
groups on the pyrrolidine ring were also identified as potential
sites of metabolism.26 Substitutions on the 3-position of the pyrrol-
idine had already been explored by the compounds described in
Tables 2 and 3 which identified the 3-(R)-methoxypyrrolidine
and the 3-methoxyazetidine R fragments as advantageous for met-
abolic stability. In addition to these, the 3-(R)-3-fluoro-and 3,3-di-
fluoro-pyrrolidine analogues 32 and 33 were prepared (Table 5).
Compound 32 gave a rat microsomal clearance of 2.4 ml/min/g
and a reduced rat in vivo clearance of 56 ml/min/kg (the (S) enan-
tiomer of 32 gave a higher CLi of 6.7 ml/min/g). Interestingly, the

4686
J. W. Myatt et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4683–4688
Table 3
Predicted P-gp score and measured P-gp efflux ratio
R
N
N
N
N
O
Compound
R
c log P
IC50 (nM)
Predicted P-gp scorea
P-gp efflux ratiob
a
2d1
a2d2
4
8
3.7
3.6
2
22
0.38
0.06
1.4:1
1.4:1
N
–NHMe
22
537
OH
14
2.2
32
398
À0.40
9.1:1
N
NH2
15
16
17
18
2.5
2.4
3.2
3.4
2
5
6
5
10
À0.68
À0.36
À0.01
À0.03
29:1
9:1
N
N
N
OH
1585
79
OMe
1.8:1
1.8:1
OMe
N
126
a
P-gp score calculated by an unpublished in-house method, <À0.20 predicts a substrate with high confidence, À0.20 to 0.00 predicts a borderline substrate, 0.00–0.20
predicts a borderline non-substrate and >0.20 predicts a non-substrate with high confidence.
b
P-gp efflux ratio determined by measuring the permeation through a monolayer of hMDR-MDCK (type II) cells in the presence and absence of a known P-gp ligand.21
O
O
O
O
N
O
HO
O
i
ii
N
O
O
O
26
28
27
O
N
iii
iv, v
HN
N
N
O
29
N
N
N
N
N
O
21
Figure 2. Metasite output showing the most likely sites of metabolism of
compound 4 by CYP450 3A4. Each atom on the test compound is ranked in the
histogram (and by shading on the structure) according to the calculated probability
of metabolism. Higher score (darker shading) = higher probability of metabolism.
Scheme 3. Reagents and conditions: (i) oxalyl chloride, MgCl2, Et2O, À5 °C, 64%; (ii)
4-iso-propoxyphenylhydrazine, PhMe, 60 °C, 89%; (iii) N2H4ÁH2O, EtOH, 100 °C, 87%;
(iv) POCl3, 120 °C, 54%; (v) pyrrolidine, NEt3, 120 °C MW, 37%.
difluoro analogue 33 gave a much higher rat microsomal clearance
of 10.8 ml/min/g so was not progressed to an in vivo study.
Unfortunately, the combination of these amine groups with the
promising phenyl groups was not always successful (Table 6). The
3-(R)-3-fluoropyrrolidine amino group in combination with 4-tri-
fluoroethoxyphenyl fragment gave compound 34 which had low
rat microsomal clearance but was not progressed due to potent
hERG inhibition (316 nM). The 3-methoxyazetidine amino group
gave a large drop in potency when combined with the 4-trifluoro-
ethoxyphenyl (35) and 2,4-dichlorophenyl (36) groups, but
pleasingly the 4-chloro-2-trifluoromethoxyphenyl derivative 37
retained the high potency of the parent compound. The in vitro

J. W. Myatt et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4683–4688
4687
Table 4
PK profiles of substituted phenyl analogues
N
N
N
N
N
R
Compound
R
c log P
IC50 (nM)
CLi(r) (ml/min/g)
CLb(r) (ml/min/kg)
a2d1
a2d2
4
O
3.7
4.3
2
22
3.5
0.8
84
—
OCF3
19
79
1995
398
O
20
21
4.0
4.0
20
32
<0.5
0.8
49
—
CF3
O
316
631
Cl
30
31
4.5
5.1
25
3
1.0
0.6
51
—
Cl
Cl
794
F3CO
Table 5
Profiles of fluoro-pyrrolidine analogues
R1
R2
N
N
N
N
O
N
Compound
R1
R2
c log P
IC50 (nM)
CLi(r) (ml/min/g)
CLb(r) (ml/min/kg)
a2d1
a2d2
4
32
33
H
F
F
H
H
F
3.7
3.7
4.0
2
3
25
22
85
794
3.5
2.4
10.8
84
56
—
rat clearance of this analogue was below 0.5 ml/min/g and this
translated to an in vivo rat clearance of 52 ml/min/kg. This was still
higher than desired but lower than the value for compound 18
(61 ml/min/kg), showing this phenyl substitution has had an addi-
tional clearance lowering effect over that provided by the 3-meth-
oxyazetidine group. This positive effect was probably tempered by
the increase in lipophilicity going from compound 18 (c log P = 3.4)
to compound 37 (c log P = 4.7). This particular phenyl substitution
also had the advantage of increasing the selectivity over alpha-2-
delta-2. This appeared to be a general trend with the 2-trifluoro-
methoxy-, difluoromethoxy- and methoxy-phenyl analogues (data
not shown).
Further SAR exploration of the phenyl ring led to the discovery
of the ortho-fluoro-para-ethoxyphenyl group which in combina-
tion with the 3-methoxyazetidine amino group gave compound
38. This analogue had good potency, acceptable selectivity and a
low to moderate in vivo rat clearance of 37 ml/min/kg. Further pro-
filing of compounds 37 and 38 showed no problematic activity at
hypersensitivity, 37 giving full reversal when dosed at 30 mg/kg
and 38 giving full reversal at 10 mg/kg—the same level of in vivo
potency as the gold-standard alpha-2-delta-1 ligand pregabalin
(Fig. 3). No side effects were observed in the study and both com-
pounds showed improved blood:brain ratios compared to com-
pound 14 (0.94:1 for 37 and for 38 0.92:1 compared with 0.17:1
for 14). Analysis of the blood and brain samples from this study
showed that the unbound concentrations in blood and brain are
similar for 37 and 38 and so do not help to explain the difference
between their in vivo potencies.27
Theoverallprofiles of37and 38 showthey are promising potential
development candidates that could test the hypothesis that alpha-2-
delta-1 selective ligands can provide effective pain relief without the
common side effects of the currently marketed treatments.28
In summary, the novel pyrazolopyridazine alpha-2-delta-1
ligand 4 has been optimized by variation of the amino group and
the phenyl substituents, with guidance from in silico tools, to
provide compounds 37 and 38. These compounds have improved
DMPK profiles, in vivo activity in the CFA pain model and in vitro
selectivity for alpha-2-delta-1 over alpha-2-delta-2.
the hERG channel (IC50 of 4 and 6
lM, respectively) and that 37
was a non-P-gp substrate. Both compounds reversed CFA induced

4688
J. W. Myatt et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4683–4688
3. McClelland, D.; Evans, R. M.; Barkworth, L.; Martin, D. J.; Scott, R. H. BMC
Table 6
Pharmacol. 2004, 4, 14.
Profiles of analogues 34–38
4. Dooley, D. J.; Mieske, C. A.; Borosky, S. A. Neurosci. Lett. 2000, 280, 107.
5. Maneuf, Y. P.; Hughes, J.; McKnight, A. T. Pain 2001, 93, 191.
6. Dooley, D. J.; Donovan, C. M.; Pugsley, T. A. J. Pharmacol. Exp. Ther. 2000, 295,
1086.
7. Bauer, C. S.; Nieto-Rostro, M.; Rahman, W.; Tran-Van-Minh, A.; Ferron, L.;
Douglas, L.; Kadurin, I.; Sri Ranjan, Y.; Fernandez-Alacid, L.; Millar, N. S.;
Dickenson, A. H.; Lujan, R.; Dolphin, A. C. J. Neurosci. 2009, 29, 4076.
8. Li, C.-Y.; Song, Y.-H.; Higuera, E. S.; Luo, Z. D. J. Neurosci. 2004, 24, 8494.
9. Field, M. J.; Li, Z.; Schwarz, J. B. J. Med. Chem. 2007, 50, 2569.
10. Baillie, J. K.; Power, I. Curr. Opin. Invest. Drugs 2006, 7, 33.
11. Sills, G. J. Curr. Opin. Pharmacol. 2006, 6, 108.
OMe
N
F
N
N
N
N
N
R2
N
N
N
O
N
CF3
R1
35-38
IC50 (nM)
34
12. Cheng, J.-K.; Chiou, L.-C. J. Pharmacol. Sci. 2006, 100, 471.
13. Dooley, D. J.; Taylor, C. P.; Donevan, S.; Feltner, D. Trends Pharmacol. Sci. 2007,
28, 75.
Compound R1
R2
c log P
CLb(r) (ml/min/kg)
a
2d1
a2d2
14. Taylor, C. P.; Angelotti, T.; Frauman, E. Epilepsy Res. 2009, 73, 137.
15. Stearns, B. A.; Anker, N.; Arruda, J. M.; Campbell, B. T.; Chen, C.; Cramer, M.; Hu,
T.; Jiang, X.; Park, K.; Ren, K. K.; Sablad, M.; Santini, A.; Schaffhauser, H.; Urban,
M. O.; Munoz, B. Bioorg. Med. Chem. Lett. 2004, 14, 1295.
16. Hu, T.; Stearns, B. A.; Campbell, B. T.; Arruda, J. M.; Chen, C.; Aiyar, J.;
Bezverkov, R. E.; Santini, A.; Schaffhauser, H.; Liu, W.; Venkatraman, S.; Munoz,
B. Bioorg. Med. Chem. Lett. 2004, 14, 2031.
4
—
—
H
Cl
—
—
3.7
4.0
2
22 84
34
35
36
37
38
32
126
200
8
631
1995
6310
—
—
—
–OCH2CF3 3.7
Cl
4.2
4.7
3.6
–OCF3 Cl
OEt
3981 52
200 37
F
8
17. Chen, C.; Stearns, B.; Hu, T.; Anker, N.; Santini, A.; Arruda, J. M.; Campbell, B. T.;
Datta, P.; Aiyar, J.; Munoz, B. Bioorg. Med. Chem. Lett. 2006, 16, 746.
18. Anker, N. B.; Arruda, J. M.; Campbell, B. T.; Munoz, B.; Prasit, P.; Stearns, B. A.
PCT Int. Appl. WO2004006836 A2 20040122, 2004.
19. (a) Iadarola, M. J.; Douglass, J.; Civelli, O.; Naranjo, J. R. Brain Res. 1988, 455,
205; (b) Hay, C. H.; Trevethick, M. A.; Wheeldon, A.; Browers, J. S.; De
Belleroche, J. S. Neuroscience 1997, 78, 843.
100
20. Summerfield, S. G.; Stevens, A. J.; Cutler, L.; Osuna, M. d. C.; Hammond, B.;
Tang, S.-P.; Hersey, A.; Spalding, D. J.; Jeffrey, P. J. Pharmacol. Exp. Ther. 2006,
316, 1282.
21. Wang, Q.; Rager, J. D.; Weinstein, K.; Kardos, P. S.; Dobson, G. L.; Li, J.; Hidalgo, I.
J. Int. J. Pharm. 2005, 288, 349.
22. Cruciani, G.; Carosati, E.; De Boeck, B.; Ethirajulu, K.; Mackie, C.; Howe, T.;
Vianello, R. J. Med. Chem. 2005, 48, 6970.
23. Japp, F. R.; Klingemann, F. Ann 1888, 247, 190.
24. Patel, H. V.; Fernandez, P. S.; Vyas, K. A. Indian J. Chem., Sect B 1990, 29, 966.
25. Saalfrank, R. W.; Lutz, T. Angew. Chem. 1990, 102, 1064.
26. An in vivo metabolism study of compound 4 largely confirmed the MetasiteÒ
predictions. The major metabolites found in urine were the result of O-
dealkylation, oxidation on the ethyl chain and oxidation of the pyrrolidine ring
and those found in bile were predominantly associated with glucuronidation of
the desethyl metabolite and again oxidation of the pyrrolidine ring.
80
60
40
20
0
-20
27. Compound 37: 30 mg/kg: [uBl] = 0.84
[uBl] = 0.46 M.
Compound 38: 30 mg/kg: [uBl] = 0.91
[uBl] = 0.48 M.
l
M, [uBr] = 0.20
l
M; 37 10 mg/kg:
l
l
M, [uBr] = 0.23
lM; 38 10 mg/kg:
Figure 3. CFA data for compounds 37, 38 and pregabalin.
l
28. Preliminary in vivo selectivity data was provided by the rotarod model of
sedation which showed compound 14 had no effect up to a dose of 300 mg/kg.
This correlates to at least a threefold window between pain relief and sedation
once the non-linear relationship between exposures and dose at the higher
concentrations of 30 and 300 mg/kg is taken into account. Compound 14 had
an in vitro selectivity for alpha-2-delta-1 over alpha-2-delta-2 of
approximately 10-fold.
References and notes
1. Felix, R. Receptors Channels 1999, 6, 351.
2. Fink, K.; Dooley, D. J.; Meder, W. P.; Suman-Chauhan, N.; Duffy, S.; Clusmann,
H.; Gothert, M. Neuropharmacology 2002, 42, 229.