Potent anti-muscarinic activity in a novel series of quinuclidine derivatives

Article abstract of DOI:10.1016/j.bmcl.2005.09.079

The synthesis and biological evaluation of a novel family of M₃ muscarinic antagonists are described. A systematic modification of the substituents to a novel alkyne-quinuclidine scaffold yielded original compounds displaying potent in vitro an

Full text of DOI:10.1016/j.bmcl.2005.09.079

Bioorganic & Medicinal Chemistry Letters 16 (2006) 373–377
Potent anti-muscarinic activity in a novel series
of quinuclidine derivatives
a
a
a
a
a
´ ´
Jean-Philippe Starck, Patrice Talaga, Luc Quere, Philippe Collart, Bernard Christophe,
Patrick Lo Brutto,a Sophie Jadot,a Dinesh Chimmanamada,b Matteo Zanda,b
Alain Wagner,b Charles Mioskowski,b Roy Massinghama and Michel Guyauxa,*
aUCB Pharma, Chemin du Foriest, 1420 Braine l’Alleud, Belgium
Universite´ Louis Pasteur de Strasbourg, UMR 7514 du CNRS, Laboratoire de Synthese Bioorganique,
b
`
Faculte´ de Pharmacie, 74 route du Rhin, 67401 Illkirch-Graffenstaden, France
Received 28 July 2005; revised 28 September 2005; accepted 28 September 2005
Available online 3 November 2005
Abstract—The synthesis and biological evaluation of a novel family of M3 muscarinic antagonists are described. A systematic mod-
ification of the substituents to a novel alkyne-quinuclidine scaffold yielded original compounds displaying potent in vitro anticho-
linergic properties.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Antimuscarinic agents are the most widely used therapy
for overactive bladder (OAB)1–3 despite exerting some
adverse effects like dry mouth, blurred vision, and
constipation.4,5
been reported to show interesting pharmacological
anti-muscarinic properties.8
The compounds were synthesized by means of standard
procedures as outlined in Scheme 1.
Although M2-muscarinic receptors (M2-R) are predom-
inant in the urinary bladder, it appears that the popula-
tion of M3-receptors (M3-R) is of great importance for
mediating direct contraction of the detrusor muscle.
Thus, M3-R antagonism represents the most clinically
effective treatment of overactive bladder related disor-
ders. Various anticholinergic agents (Fig. 1) have al-
ready proven to be effective in patients suffering from
OAB. Nevertheless, due to the side-effect profile dis-
played by these molecules, there is still a need to develop
new safer anti-muscarinic agents for the treatment of
OAB.
O
H
O
HO
N
O
N
O
N
Oxybutynin
Solifenacin
OH
O
O
N
NH2
N
As part of our continuous efforts to identify potent
antimuscarinic molecules displaying a good safety pro-
file,6 we wish to report preliminary results on a novel
quinuclidine containing series.7 Quinuclidine containing
derivatives such as Solifenacin (Fig. 1) have already
Tolterodine
MeO
Darifenacin
N
N
HO
22a
Keywords: Muscarinic receptor(s); Bladder; Urinary incontinence;
Quinuclidine; Muscarinic antagonist(s); Overactive bladder.
*
Corresponding author. Tel.: +32 2 386 2608; fax: +32 2 386 3133;
e-mail: michel.guyaux@ucb-group.com
Figure 1. Selected anti-muscarinic agents.
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.09.079

374
J.-P. Starck et al. / Bioorg. Med. Chem. Lett. 16 (2006) 373–377
Table 1. Affinity for human M3-R and M2-R for IV analogues
HO
MeO
a
b, c, d
R1
R2
Binding
affinity (Ki)
O
Compound
N
N
N
I
III
M3
M2
f
e
1*
2*
3
Ph
Ph
Ph
0.13
0.20
0.25
1.0
0.79
2.0
c-Hexyl
c-Pentyl
c-Butyl
n-Pentyl
n-Butyl
Bn
HO
MeO
R2
Ph
Ph
3.2
N
4
13
N
HO
R1
HO
R1
5
Ph
Ph
2.5
0.63
40
25
R2
IV
II
6
7
Ph
Bn
40
631
1.0
7.9
100
251
3161
20
Scheme 1. Reagents and conditions: (a) Li Acetylide.ed cpx, THF; (b)
BH3.THF, THF, ꢀ10 ꢁC; (c) NaH, NBu4I, MeI, THF, rt; (d) 5 M
HCl, acetone/Et2O, rt; (e) BuLi, ketone, THF, ꢀ78 ꢁC to rt; (f) 2 equiv
BuLi, ketone, THF, ꢀ78 ꢁC to rt.
8
Bn
9*
10*
11*
12*
13*
14*
4-F-Ph
4-Cl-Ph
4-Me-Ph
3-F-Ph
3-Me-Ph
3-CF3-Ph
4-F-Ph
4-Cl-Ph
4-Me-Ph
3-F-Ph
3-Me-Ph
3-CF3-Ph
126
501
0.79
7.9
5.0
50
#
Briefly, a mixture of 3-quinuclidinone (free base) and
lithium acetylide (ethylenediamine complex) in THF
yielded the acetylenic quinuclidinol I. The diol deriva-
tives II were obtained by the coupling of I with the cor-
responding ketones at low temperature. The synthesis
of III was accomplished via the protection of I as an
amino-borane complex, determinant for the following
methylation step.9 Thus, methylation of the hydroxyl
group of I using classical conditions (NaH, TBAI cata-
lytic, and MeI) only occurred after protection of the
nitrogen via the formation of this borane complex.
Addition of methyl iodide to the unprotected molecule
resulted in the exclusive formation of the corresponding
ammonium salt. The deprotection of the borane com-
plex in a mixture of 5 N HCl/acetone was followed by
the deprotonation of the alkyne using BuLi at low tem-
perature. The resulting alkynelide III was then reacted
with the appropriate ketone10 yielding compounds IV
which were isolated as mixtures of diastereomers or
as optically pure products.
1000
100
0.05
15*
631
16*
17*
18
2-Thienyl
3-Thienyl
Ph
2-Thienyl
3-Thienyl
4-Piperidyl
c-Pentyl
0.20
0.40
0.08
#
#
19
20
2-Pyridyl
2-Pyridyl
Ph
4.0
50
316
79
c-Butyl
2-Pyridyl
20
10
0.40
1.3
50
21
22
Ph
Ph
3-Pyridyl
4-Pyridyl
1.3
5.0
100
23
24*
25*
26
3-Pyridyl
2-Pyridyl
Ph
3-Pyridyl
2-Pyridyl
#
#
6-F-3-pyridyl
2-F-3-pyridyl
3-Pyridazinyl
5-Pyrimidinyl
0.32
0.79
7.9
1.0
3.2
1.3
1.3
4.0
3.2
40
27
28
Ph
Ph
29
Ph
2.0
4.0
Tolterodine13
Oxybutynin14
Darifenacin13
16
50
Results are expressed as Ki (nM) or # (when <50% inhibition of
radioligand specific binding by 10 lM of compounds). The compounds
have been tested as optically pure R isomers* or as a mixture of dia-
stereomers bearing the S/R, R/R configurations.
Methoxylation of the quinuclidine moiety, as shown in
IV, turned out to be of crucial importance. As a repre-
sentative example, IV (R1/R2 = Ph/cyclohexyl (2) in
Table 1) showed almost a 60-fold increase in affinity
for the M3-R (Ki: 0.2 nM) and the M2-R (Ki: 2.0 nM)
when compared to the corresponding diol analogue II
(Ki M3-R = 13 nM, Ki M2-R = 100 nM).
was tolerated, introduction of bulkier groups, such as
chlorine (10) and methyl (11), induced a decrease in
affinity. The same observation was made when the meta
position of the aromatic rings was considered. Thus,
fluorine (12) substitution was far better tolerated than
methyl (13) or trifluoromethyl (14) replacements. Fur-
thermore, rigidification of the hydrophobic part (R1,
R2) of the molecule, as exemplified by the suberenyl ana-
logue (15), caused almost a 1000-fold decrease in
affinity. These results suggest that strong binding to
the muscarinic receptors requires the molecules to have
their hydrophobic groups in orthogonal orientations.
Taking advantage of the presence of methoxy group in
IV, various chemical modulations were performed to
better characterize the hydrophobic part (R1, R2) of
the molecule (Table 1). The introduction of non-aromat-
ic groups in various anti-muscarinic series has been
reported to be well tolerated.11,12 Similarly, compounds
(2) and (3), respectively, bearing a cyclohexyl and a
cyclopentyl moiety had similar affinity to (1). Substitu-
tion of a single aryl moiety by a cyclobutyl (4) or short
linear alkyl groups such as n-pentyl (5) or n-butyl (6)
resulted in a slight decrease in affinity for the M2 mus-
carinic receptor subtype.
As seen with compounds (16) and (17), the 2-thienyl or
3-thienyl moiety could successfully be used as a phenyl
bioisostere. On the contrary, replacement of one aro-
matic group by basic moieties such as 4-piperidyl (18)
led to a total loss of affinity.
The introduction of a benzyl group at the R2 position (7)
or the replacement of both phenyl groups by benzyl
moieties (8) induced a 100- to 1000-fold decrease in
affinity. Along the same lines, while substitution of the
para position of both aromatic rings by fluorine (9)
The 2-pyridyl derivatives (19, 20, and 21) displayed a
30- to 100-fold decreased muscarinic affinity when

J.-P. Starck et al. / Bioorg. Med. Chem. Lett. 16 (2006) 373–377
Table 3. Functional antagonism in isolated organs
375
compared to their corresponding cycloalkyl (5, 6) and
phenyl analogues (1). Interestingly, compounds bearing
a 3-pyridyl (22) or a 4-pyridyl (23) moiety as potential
phenyl isosteres were highly active on both subtypes of
muscarinic receptors. However, the bis-pyridyl deriva-
tives (24, 25) were less potent. Similar to what was ob-
served with the phenyl analogues, fluorination at the
para position (26) or the ortho (27) position of the 3-pyr-
idyl moeity is well tolerated. Among the various nitro-
gen containing heterocycles that have been investigated
(pyridyl, 21–23 and 26–27; pyridazinyl, 28; pyrimidinyl,
29), those bearing the Ph/3-pyridyl moieties in R1/R2
showed the highest affinity for the M3-R and M2-R. It
should be noted that compound 22 (R1/R2 = Ph/3-pyr-
idyl) has almost the same affinity as the one displayed
by 1, highlighting the fact that the nitrogen located at
the meta position of the pyridyl ring does not affect
the affinity for either muscarinic receptor subtype.
Compound
R1/R2
Guinea pig (pA2)
Bladder (M2) Atria (M3)
1
Ph/Ph
Ph/c-hexyl
Ph/c-butyl
9.3
9.1
8.5
9.1
8.8
8.0
8.6
7.5
8.9
8.4
8.6
7.7
7.1
7.3
2
4
16
2-Thienyl/2-thienyl 9.3
21
22
Ph/2-pyridyl
Ph/3-pyridyl
Ph/4-pyridyl
Ph/2-F-3-pyridyl
7.5
9.1
8.2
8.7
7.8
7.4
7.8
23
27
Tolterodine17
Oxybutynin18
Darifenacin14
Values are means of at least three experiments. Antagonism was
measured by the ability of compounds to shift to the right (pA2)19 the
concentration-response curve to carbachol on isolated guinea-pig uri-
nary bladder and left atria.13,20
In order to characterize the biologically active isomers,
enantiomerically pure compounds were prepared15 and
their absolute configuration was established via X-ray
crystallography.
M3-R antagonists and bladder anticontractile agent
in vitro.
During evaluation of the aqueous chemical stability of
this novel series, we demonstrated that compounds bear-
ing two aromatic rings (e.g., 1) or two thienyl groups
(e.g., 16) were unstable at acidic pH while remaining sta-
ble at neutral pH. We succeeded in isolating and charac-
terizing the rearrangement product of 1 under acidic
conditions as being the corresponding a,b-unsaturated
ketone, formed via a classical Meyer–Schuster reaction
(Fig. 3).21
As shown in Table 2, the (R)-enantiomer 1a displayed a
greater affinity than that of the corresponding (S)-iso-
mer 1b. The same trend was observed when testing the
diastereomers 2a/2b (R1/R2 = Ph/cyclohexyl). Thus, the
molecules bearing the (R,R) configuration exhibited
affinity for muscarinic receptors higher than those in
the (S,R) configuration. These data are in line with those
reported for other structurally close muscarinic antago-
nists described in the literature (e.g., trihexylphenidyl,
oxybutynin, and p-fluoro-HHSiD).16
On the contrary, the 2-pyridyl compound 21 was shown
to be stable at pH 1, whereas it was surprisingly unstable
at pH 7.4. Interestingly, the 3-pyridyl (22) and the 4-
pyridyl (23) derivatives were stable under acidic as well
as neutral conditions.
The anti-muscarinic potency of several representative
molecules was next evaluated in isolated guinea-pig
bladder (M3-R) and atria (M2-R) (Table 3).
Furthermore, as illustrated in Figure 4, within that series
of molecules, a positive correlation between in vitro met-
abolic c0learance (human liver microsomes) and lipophil-
icity ðKIAMÞ has been established.
On scaffold IV (Scheme 1) R1 = indicated groups,
R2 = Ph. Exception: when R2 = 2-thiophen then
R1 = 2-thiophen. Parent drug consumption was mea-
sured following incubation (up to 90 min) of test com-
pounds (0.25 lM) with NADPH-fortified human liver
microsomes (0.1 mg/ml). Metabolic clearance (Clint)
was calculated with 0the following equation:
Clint = dose/AUC.22 LogKIAM was measured using ace-
tonitrile/phosphate buffer mobile phase.23
All these compounds behaved as competitive antago-
nists of the carbachol-induced response. Their pA2
values were consistently similar in both organs confirm-
ing the compounds were non-selective M3-R and M2-R
antagonists. The pA2 values as well as the pKi values
of the most potent compounds were superior to those
documented for darifenacin, tolterodine, and oxybuty-
nin, indicating that these compounds are very potent
Table 2. In vitro M3/M2 binding affinity of optically pure compounds
HO
OMe
N
3
1
R1
R2
Overall, there was a trend toward improving the meta-
bolic clearance profile when a less hydrophobic group
Compound
R1/R2
Configuration
Binding
(Ki, nM)
M3
63
M2
HO
Ph
OMe
N
OMe
N
O
H3O+
1a
1b
2a
2b
Ph/Ph
Ph/Ph
(3R)
(3S)
0.13
0.79
251
2.0
20
Ph
Ph
Ph
Ph/c-hexyl
Ph/c-hexyl
(1R,3R)
(1S,3R)
0.20
3.2
Figure 3. Rearrangement of 1 under acidic conditions.

376
J.-P. Starck et al. / Bioorg. Med. Chem. Lett. 16 (2006) 373–377
Table 5. In vitro properties of diastereomers 23a and 23b
500
450
400
350
300
250
200
150
100
50
CH3
b
Compounds
Ki (nM)a
M2
0.79
50
pA2
CYP enzymec
M3
13
Bladder
3A4
2D6
22a (1R,3R)
22b (1S,3R)
0.25
9.1
7.4
51%
18%
2%
4%
a Results are expressed as Ki (nM).
b pA2 (Antagonism of carbachol induced guinea-pig bladder
contraction13).
CH3
S
N
c Percentage of inhibition of midazolam 10-hydroxylation (CYP3A4)
and of dextromethorphan O-demethylation (CYP2D6) as measured
using NADPH-fortified human liver microsomes24 in the presence of
50 lM of the test compound.
0
1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9
log K'IAM
Figure 4. Correlation between metabolic clearance and lipophilicity.
ter in vitro CYP interaction profile than other com-
pounds within this series.
(e.g., pyridyl moieties) was introduced in0to the core
structure of IV. Thus, the 2-pyridyl ðLogKIAM ¼ 1:87Þ,
3-pyridyl ðLogK0IAM ¼ 1:40Þ, and 4-pyridyl ðLog
K0IAM ¼ 1:41Þ derivatives were less rapidly metabolized
ðLogK0IAM ¼ 2:17Þ.
Chiral chromatography separation of this compound led
to the two optically pure diastereomers 22a (R/R) and
22b (S/R). As shown in Table 5, 22a did not affect
CYP2D6 activity (<5%), displaying slightly less favor-
able profile toward CYP3A4 inhibition than 22b but
was almost ꢁ50-fold more potent than 22b in terms of
M3-R affinity and bladder anticontractile potency
in vitro.
than
the
corresponding
Ph/Ph
analogue
1
We also investigated in vitro (human microsomes) the
capacity of the compounds to inhibit marker activities
for CYP3A4 and CYP2D6 (Table 4).
In conclusion, these studies describe the structure–
activity relationships regarding the affinity for the M3-
R and M2-R of a new alkyne-quinuclidine scaffold.
The core structure of the most active compounds bears
functional groups (hydrogen-acceptor site, basic nitro-
gen, and two lipophilic moieties) similar to those found
in other muscarinic M3-R antagonists. The methoxy
group attached to the quinuclidine moeity constitutes
an additional and original interaction point with the
muscarinic receptor. Chemical modulations around this
scaffold gave compounds displaying higher affinities for
M3-R and M2-R as well as more potent in vitro antago-
nism in isolated tissues than the known reference drugs.
Lead optimization involving the introduction of nitro-
gen heteroatoms in the lipophilic part of the scaffold
led to compounds with lower lipophilicity, less metabo-
lism, advantageous drug interaction profiles whilst at the
same time solving the problem of aqueous chemical
instability.
It is known that oxybutynin25 is metabolized by
CYP3A4, and tolterodine and darifenacin also by
CYP2D626,27 and are inhibitors of CYP 3A4 or/and
CY2D6 isoenzymes in vitro.28,29 The 2- and 3-pyridyl
derivatives (21, 22, and 24) exhibited poor to moderate
inhibition of CYP3A4 and CYP2D6 activities contrarily
to the Ph/cyclohexyl (2) and the Ph/4-pyridyl analogues
(23) which inhibit CYP2D6 activity.
Moreover, representative compounds 1, 2, 6, 16, 21, 22,
and 23 did not compete (results not shown) with ligands,
selective for adenosine A1, A2, adrenergic a1, a2, and b,
serotoninergic 5-HT1–7, dopaminergic D1, D2, and hista-
minergic H1, H2, and H3 receptors at 10 lM. This preli-
minary counterscreen suggests that these compounds
might possess high selectivity for muscarinic receptors.
Taken together, these results emphasize that the 3-pyr-
idyl compound 22 was the most interesting. This
molecule is chemically stable, has high affinity for the
M3-R and M2-R and is less metabolized, and has a bet-
Acknowledgments
The authors thank C. Huart, B. Carlier, L. Ellens,
V. Pinilla, A. Fauconier, E. Cavoy, C. Audouin, M.
Lechien, and M. Gillard for the technical support they
provided to the realization of the present work.
Table 4. CYP interaction studies
Compound
R1/R2
CYP enzyme
3A4
2D6
2
21
22
23
24
Ph/c-hexyl
74%
6%
28%
8%
Ph/2-pyridyl
Ph/3-pyridyl
Ph/4-pyridyl
Bis(3-pyridyl)
References and notes
41%
58%
18%
6%
71%
4%
1. Andersson, K.-E. Lancet Neurol. 2004, 3, 46.
2. Hegde, S. S.; Mammen, M.; Jasper, J. R. Curr. Opin.
Investig. Drugs 2004, 5, 40.
3. Eglen, R. M.; Choppin, A.; Watson, N. Trends Pharmacol.
Sci. 2001, 22, 409.
4. Feinberg, M. Drugs Aging. 1993, 3, 335.
Specific substrates were incubated with NADPH-fortified human liver
microsomes in the presence of 50 lM of the test compound. Results are
expressed as percentage inhibition of midazolam 10-hydroxylation
(CYP3A4) and of dextromethorphan O-demethylation (CYP2D6).24

J.-P. Starck et al. / Bioorg. Med. Chem. Lett. 16 (2006) 373–377
377
5. Scarpero, H. M.; Dmochowski, R. R. Curr. Urol. Rep.
2003, 4, 421.
14. Gilberg, P. G.; Sundquist, S.; Nivelbrant, L. Eur. J.
Pharmacol. 1998, 349, 285.
´
6. Diouf, O.; Gadeau, S.; Chelle, F.; Gelbcke, M.; Talaga, P.;
15. Separation of the diastereomers has been performed via
chiral chromatography (Chiralpak AD column).
16. Baker, R.; Saunders, J. Annu. Rep. Med. Chem. 1989, 24, 31.
17. Newgreen, D. T.; Naylor, A. M. Br. J. Pharmacol. 1996,
117, 107P.
18. Wallis, R. M.; Napier, C. M. Life Sci. 1999, 64, 395.
19. Van Rossum, J. M.; Hurkmans, J. A. T. M.; Wolters, C. J.
J. Arch. Int. Pharmacodyn. Ther. 1963, 143, 299.
20. Lambrecht, G.; Feifel, R.; Wagner-Ro¨der, M.; Strohman,
C.; Zilch, H.; Tacke, R.; Waelbroeck, M.; Christophe, J.;
Boddeke,H.;Mutschler,E. Eur.J.Pharmacol. 1989,168,71.
21. Swaminathan, S.; Narayanan, K. V. Chem. Rev. 1971, 71,
429.
Christophe, B.; Gillard, M.; Massingham, R.; Guyaux, M.
Bioorg. Med. Chem. Lett. 2002, 12, 2535.
7. Guyaux, M.; Dinesh, C. U.; Mioskowski, C.; Quere, L.;
Starck, J.-P.; Talaga, P.; Wagner, A.; Zanda, M. WO 03/
033495, 2003.
8. Mealy, N.; Castaner, J. Drugs Future 1999, 24, 871.
9. Stotter, P. L.; Friedman, M. D.; Dorsey, G. O.; Shiely, R.
W.; Williams, R. F.; Minter, D. E. Heterocycles 1987, 25,
251.
10. Typical procedure: A 1.6 N solution of BuLi in hexane
(41.7 ml, 66.7 mmol) was added dropwise to a solution of
(3R)-3-ethynyl-1-azabicyclo[2.2.2]oct-3-yl methyl ether 2
(11 g, 66.7 mmol) in THF (100 ml) cooled at ꢀ5 ꢁC. After
stirring for 10 min, the medium was cooled at ꢀ15 ꢁC, and
a solution of benzophenone (12.8 g, 70.2 mmol) in THF
(50 ml) was slowly added. The cooling bath was removed
and the reaction was allowed to gradually warm to room
temperature during 2 h. The reaction was quenched by the
addition of saturated aqueous NH4Cl solution. The
reaction mixture was washed with brine, dried over
Na2SO4, and evaporated. The yellowish oil was purified
by chromatography over silica gel (CH2Cl2/EtOH/
NH4OH 94/6/1). 3-[(3R)-3-Methoxy-1-azabicyclo[2.2.2]
oct-3-yl]-1,1-diphenyl-2-propyn-1-ol 3a was obtained as a
white solid (20.9 g, 83%).
11. Mitsuya, M.; Mase, T.; Tsuchiya, Y.; Kawakami, K.;
Hattori, H.; Kobayashi, K.; Ogino, Y.; Fujikawa, T.;
Satoh, A.; Kimura, T.; Noguchi, K.; Ohtake, N.; Tomim-
oto, K. Bioorg. Med. Chem. 1999, 7, 2555.
12. Naito, R.; Takeuchi, M.; Morihira, K.; Hayakawa, M.;
Ikeda, K.; Shibanuma, T.; Isomura, Y. Chem. Pharm.
Bull. 1998, 46, 1274.
22. Carlile, D. J.; Stevens, A. J.; Ashforth, E. I. L.; Waghela,
D.; Houston, J. B. Drug Metab. Dispos. 1998, 26, 216.
23. Capacity factors (k0IAM) were established on
a
IAM.PC.DD
2
Drug-Discovery HPLC column 30 ·
4.6 mm (Regis Tech Inc., Morton Grove, IL, USA). The
mobile phase consisted of different mixtures of phosphate-
buffered saline (pH 7.4) and acetonitrile as co-solvent. The
published data are the extrapolated values at 0% in
CH3CN. See also: Ong, S.; Liu, H.; Pidgeon, C. J.
Chromatogr. A, 1996, 728, 113.
24. Yuan, R.; Madani, S.; Wei, X.; Reynolds, K.; Huang, S.
Drug Metab. Dispos. 2002, 30, 1311.
25. Yaich, M.; Popon, M.; Medard, Y.; Aigrain, E. J.
Pharmacogenetics 1998, 8, 449.
26. Postlind, H.; Danielson, A.; Lindgren, A.; Andersson, S.
H. G. Drug Metab. Dispos. 1998, 26, 289.
27. Beaumont, K. C.; Cussans, N. J.; Nichols, D. J.; Smith, D.
A. Xenobiotica 1998, 28, 63.
28. Novartis Pharmaceutical Corp. New Drug Application.
2004, NDA-021513.
13. Posangi, J.; Zar, M. A.; Harris, J. B. Br. J. Pharmacol.
1992, 106, 307.
29. Pharmacia and Upjohn. New Drug Application. 1998,
NDA-020771.