The synthesis and biological evaluation of dopamine transporter inhibiting activity of substituted diphenylmethoxypiperidines

Article abstract of DOI:10.1016/j.bmcl.2005.08.028

The synthesis of potent 4-aryl methoxypiperidinol inhibitors of the dopamine transporter is described. Symmetrical para substituents of the benzene rings are important for high potency in binding to the dopamine transporter. 4-[Bis(4-fluorophenyl) methoxy]-1-methylpiperidine has an IC₅₀ of 22.1 ± 5.73 nM and increases locomotor activity in mice.

Full text of DOI:10.1016/j.bmcl.2005.08.028

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4915–4918
The synthesis and biological evaluation of dopamine transporter
inhibiting activity of substituted diphenylmethoxypiperidines
Gennady B. Lapa,a Gary D. Byrd,b Alla A. Lapa,a Evgeny A. Budygin,a
Steven R. Childers,a Sara R. Jonesa,* and Jill J. Harpc
aDepartment of Physiology and Pharmacology, Wake Forest University Health Sciences, Winston-Salem, NC 27157, USA
bHuman Studies Division, R.J. Reynolds Tobacco Co., Winston-Salem, NC 27157, USA
cDepartment of Life Sciences, Winston-Salem State University, Winston-Salem, NC 27157, USA
Received 16 May 2005; revised 11 July 2005; accepted 9 August 2005
Available online 13 September 2005
Abstract—The synthesis of potent 4-aryl methoxypiperidinol inhibitors of the dopamine transporter is described. Symmetrical para
substituents of the benzene rings are important for high potency in binding to the dopamine transporter. 4-[Bis(4-fluorophenyl)
methoxy]-1-methylpiperidine has an IC50 of 22.1 5.73 nM and increases locomotor activity in mice.
Ó 2005 Elsevier Ltd. All rights reserved.
The dopamine transporter (DAT) is an integral mem-
brane protein located in dopamine (DA) neuron termi-
nals. It is responsible for regulating the extracellular
concentrations of DA by taking up released DA into
the presynaptic terminal. It is also the biological target
for cocaine.1 The psychoactive effects of DAT inhibitors
like cocaine are associated with fast changes in the up-
take of DA.2 For example, within 4 s after intravenous
cocaine administration there is significant inhibition of
DA uptake that reaches a plateau in 20 s.3 The develop-
ment of potent long-acting inhibitors of the DAT has
provided a great deal of information about the cocaine
pharmacophore.4,5 Moreover, such compounds may be
useful as pharmacotherapeutic agents in the treatment
of cocaine addiction.2 Recent reports have shown that
potent DAT inhibitors such as benzotropine (BZT)
and GBR analogs have potential as pharmacotherapies
for treatment of cocaine abuse.4–6 Some of these analogs
exhibit a slow onset and long duration of stimulant
action that is evident as increased locomotor activity
in animal models.7 Both of these classes of DAT inhib-
itors have two common structural features: a diphenyl-
methoxy moiety and a nitrogen-based six-membered
ring (Fig. 1).
N
CH3
N
O
CH3
O
Benztropine
DPP
N
O
N
GBR12935
Figure 1. Chemical structures of known DAT inhibitors.
Diphenylpyraline (DPP), a well known H1-histamine
receptor antagonist, also has these structural features.8
A flexible piperidine ring and short ether bond allow
DPP to be considered a flexible analog of BZT and a
structurally constrained derivative of GBR, respectively.
Recently, we have shown that DPP is an effective DAT
inhibitor.9 This suggests that compounds structurally
similar to DPP could also be DAT inhibitors.
We have explored the structure–activity relationships
(SAR) of a series of piperidinol derivatives to develop
a more potent DAT inhibitor than DPP. In the cur-
rent study, we report the synthesis and in vitro biolog-
ical evaluation of a series of piperidinols structurally
Keywords: Dopamine transporter; Inhibitor; Piperidinol-4; Locomotor
activity.
*
Corresponding author. Tel.: +1 336 716 8533; fax: +1 336 716
8501; e-mail: srjones@wfubmc.edu
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.08.028

4916
G. B. Lapa et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4915–4918
similar to DPP. In all recent reviews, the group of
very potent DAT inhibitors has bis-para-halogen
substituted diphenylmethoxy moieties like BZT,
GBR, and their piperidine analogs.4,5,10 These com-
pounds are more lipophilic than their unsubstituted
counterparts. In addition, one of the most reliable
methods in medicinal chemistry for improving in vitro
activity is to incorporate a properly positioned lipo-
philic group11 (e.g., by increasing the length of N-alkyl
groups). As a result, we synthesized known8,12 DPP-
related structures and new compounds. This study
also evaluates the in vivo effects of novel inhibitors
with high DAT affinity on locomotor activity in mice.
DAT inhibitors are known to elevate striatal extracel-
lular DA levels and this effect is associated with
behavioral hyperactivity.7
and final mixtures were purified by flash chromatogra-
phy.15 The yields ranged from 60 to 100% based on
recovered starting material. Another described method
resulted in poor yields.12
The in vitro IC50 values (Table 1) of compounds 3a–h at
DAT were determined by displacement of [125I]RTI-55
binding in rat striatal membranes, as described previous-
ly.16 The most potent compounds, 3a–c,g, were tested
in vivo for effects on locomotor activity in mice as
described previously.9 Figure 2 shows the time course
effects of these compounds on locomotor activity over
the entire 3 h session in 5 min bins. Table 1 reports loco-
motor activity over the entire 3 h sessions. Statistical
analyses were carried out using a two-way ANOVA.
We used the ‘‘rule of 5’’ to select compounds for in vivo
experiments.11 ‘‘Molinspiration’’ is an on-line software
tool that allows us to rapidly calculate all parameters
of this rule and have an Nv descriptor.17 The Nv-molec-
ular violating parameter is an integral number which re-
flects the importance of each parameter of the rule of 5
The compounds were synthesized as previously de-
scribed for the series of tropane compounds13,14
(Scheme 1). Briefly, N-substituted piperidin-4-ols 2g,h
were prepared by alkylation with alkyl halides in N,N-
dimethylformamide in the presence of anhydrous potas-
sium carbonate13 in yields of approximately 90%. Com-
mercially available substituted benzophenones were
reduced to the benzhydrols using sodium borohydride
in isopropanol in nearly quantitative yield. The ethers
were synthesized by condensation of benzhydrols with
25% molar excess of N-substituted piperidin-4-ols or
N-methylpiperidin-4-ol in benzene with a Dean–Stark
trap.14 para-Toluenesulfonic acid was used as a catalyst.
It was removed by extraction with sodium hydroxide
1800
3b
GBR12909
Cocaine
Saline
1500
1200
900
600
300
0
R3
HO
R3
R4
0
50
100
150
R4
N
time min
R2
1
R
R1
N
O
OH
b
Figure 2. Effects of 3b, GBR12909, cocaine and saline on locomotor
activity. Mice were injected intraperitoneally with equivalent doses
(17.8 lmol/kg) of each drug. Horizontal activity was monitored for 3 h
as described previously,9 and data are expressed as cm/5 min. All
compounds significantly increased locomotor activity levels compared
with saline, P < 0.001. Fifty minutes after injection, there were
significant differences in locomotor activity produced by each com-
pound compared with the others, P < 0.001.
R2
3a-h
1
1 R1 = H
2g R1 = Bu, 2h R2 = 4-F-C6H4-CH -
a
2
Scheme 1. Reagents and conditions: (a) Hal-R, K2CO3, DMF, 40 °C;
(b) p-TSA, benzene/DMF (25:1), reflux.
Table 1. Binding and horizontal locomotor activity of 4-arylmethoxypiperidinols
R1
R2
R3 R4
Calculated logP IC50 (DAT) nMc Locomotor activity (m/3 h)c
GBR 12909
—
—
—
—
—
—
Ph
—
—
H
—
—
H
H
H
H
H
—
—
0.43 0.15b
104 49b
26.3 4.4
16.7 2.8
12.2 2.1
5.18 0.87
34.8 5.2
5.89 1.01
—
Cocaine
DPPa
3a
C19H23NO
C19H21Cl2NO
CH3
CH3
CH3
CH3
CH3
CH3
CH3
C4H9
3.95
5.20
3.99
4.58
3.97
420
9
4-Cl–C6H4– Cl
F
44.0 10.9
22.1 5.7
155 10
277 10
293 10
3b
3c
C
C
19H21F2NO
19H22ClNO
4-F–C6H4–
Ph
Cl
F
3d
3e
C19H22FNO
20H25NO
C14H20FNO
22H27Cl2NO
Ph
Ph
C
H
F
CH3 4.27
—
—
3f
3g
CH3
H
H
H
3.11
6.55
6.53
264
9
C
4-Cl–C6H4– Cl
12.5 7.5
50.6 2.8
3.87 0.64
—
3h
C25H24Cl2FNO 4-F–C6H4–CH2– 4-Cl–C6H4– Cl
a Purchased from Aldrich.
b See Ref. 16a.
c The data are means; SEM; 0.9% NaCl was used as a vehicle. Locomotor activity produced by vehicle was À3.92 m/3 h.

G. B. Lapa et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4915–4918
4917
for the structure of interest.17 On the basis of experimen-
tal IC50 values, we developed a simple quantitative
structure–activity relationship (QSAR) model to predict
IC50 values for new compounds.17 We chose a linear
type of regression equation which allowed us to reach
a reasonable value for r2. Our equation shows that more
active compounds should be more lipophilic than DPP
in in vitro assays.17 This requirement may be because
our target protein, the DAT, is an integral membrane
protein, and the binding site may be on the transmem-
brane domains. We predicted the IC50 values for 3g
and 3h to be 17.55 and 26.06 nM, respectively. Their
actual IC50 values were 12.5 and 50.6 nM, respectively.
inhibitors only 3b had a greater effect than DPP, both
in binding and on locomotor activity in mice. This com-
pound is a good candidate for further QSAR study. We
can conclude that the short inflexible link of the diphe-
nylmethyloxy fragment in this series of piperidines can
still adopt the bioactive conformation for high affinity
for DAT in vitro. However, DPP and 3b were less
long-lasting than BZT or GBR12909 but more long-
lasting than cocaine in live animal locomotor tests.
Acknowledgments
This work was supported by Wake Forest University
School of Medicine Venture Funds, National Center
for Minority Health and Health Disparities Grant
MD00232 and National Institutes of Health Grant
AA014091.
All of our compounds are more potent than DPP. Com-
pound 3e is more lipophilic than DPP and has increased
affinity for the DAT. The results of the binding assay
show that replacing the aromatic ring of DPP with a
methyl group, in order to reduce lipophilicity, leads to
a higher affinity of 3f than DPP for the DAT. Perhaps
a substituent on the aromatic ring is important for bind-
ing, in addition to lipophilicity. Symmetrically disubsti-
tuted halogen compounds 3a and 3b were more active
than their monosubstituted counterparts 3c and 3d,
respectively. Chlorosubstituted compound 3a was a two-
fold less potent inhibitor of DAT than fluorosubstituted
3b. The most lipophilic compound 3g exhibited higher
potency in the binding assay, as was predicted. Perhaps
an increase in the molecular volume of substituents on
the nitrogen atom of the piperidine ring led to the stabil-
ization of the biologically active conformer. Aliphatic
substitution on the nitrogen of 3g led to a more potent
DAT inhibitor than with the fluorobenzyl fragment in
3h. This is in accord with the previous statement con-
cerning a properly positioned lipophilic group.11 Our
binding results agree with the trends reported for halo-
genated benztropine analogs.18
References and notes
1. Torres, G. E.; Gainetdinov, R. R.; Caron, M. G.
Neuroscience 2003, 4, 13.
2. Volkow, N. D.; Wang, G. J.; Fowler, J. S.; Learned-
Coughlin, S.; Yang, J.; Logan, J.; Schlyer, D.; Gatley, J.
S.; Wong, C.; Zhu, W.; Pappas, N.; Schueller, M.; Jayne,
M.; Carter, P.; Warner, D.; Ding, Y. S.; Shea, C.; Xu, Y.
Biol. Psychiatry 2005, 57(6), 640.
3. Mateo, Y.; Budygin, E. A.; Morgan, D.; Roberts, D. C. S.;
Jones, S. R. Eur. J. Neurosci. 2004, 20, 2838.
4. Singh, S. Chem. Rev. 2000, 100(3), 925.
5. Carroll, F. I.; Howell, L. L.; Kuhar, M. J. J. Med. Chem.
1999, 42, 2721.
6. Dessai, R. I.; Kopajtic, T. A.; Koffarnus, M.; Newman, A.
H.; Katz, J. L. J. Neurosci. 2005, 23, 1889.
7. Budygin, E. A.; Kilpatrick, M. R.; Gainetdinov, R. R.;
Wightman, R. M. Neurosci. Lett. 2000, 281, 9.
8. Idson, B. Chem. Rev. 1951, 307.
9. Lapa, G. B.; Mathews, T. A.; Harp, J. J.; Budygin, E. A.;
Jones, S. R. Eur. J. Pharmacol. 2005, 506, 237.
10. Dutta, A. K.; Zhang, S.; Kolhatkar, R.; Reith, M. E. A.
Eur. J. Pharmacol. 2003, 479, 93.
11. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P.
J. Adv. Drug Del. Rev. 2001, 46, 26.
12. Brit. Patent 702067, 1954; Chem. Abstr. 1955, 4023i.
13. Robarge, M. J.; Agoston, G. E.; Izenwasser, S.; Kopajtic,
T.; George, C.; Katz, J. L.; Newman, A. H. J. Med. Chem.
2000, 43, 1085.
14. Meltzer, P. C.; Liang, A. Y.; Madras, B. K. J. Med. Chem.
1994, 37, 2001.
In our previous research, we found that DPP was fairly
effective at increasing locomotor activity at a dose of
5.0 mg/kg (17.8 lmol/kg).9 All compounds were tested
at this dose. Compound 3b was most active in this assay
(see Table 1 and Fig. 2). All synthesized compounds had
a longer duration of effect on locomotor activity
(approximately 3 h) than cocaine, but shorter than
GBR (more than 3 h). DPP and 3b are more lipophilic
than cocaine but less than GBR or BZT.17 This fact
may explain the slow onset and long duration of effects
of these compounds on locomotor activity in mice. We
made a prediction of the bioavailability of our com-
pounds using the rule of 5 to better understand the
extremely low activity of 3a, 3c, and 3g in vivo.11 Calcu-
lations of parameters of the rule of 5 showed that com-
pounds 3a and 3g with heavy substituents like chlorine
atoms should have poor bioavailability (see the value
of log P in Table 1). These compounds violate LipinskiÕs
rule of 5.11,17 Compounds 3a, 3c, and 3g have a lower
solubility in the assay buffers. Thus, it is likely that they
have poor penetration through the blood–brain barrier.
15. For all compounds the mass spectra contained weak
(MÀH)+ ions characteristic of piperidines and the typical
isotope patterns for one or more halogen atoms in the
structures. Characteristic ions that occurred for com-
pounds 3a–3f were m/z 99 (methylpiperidine fragment) as
the base peak and m/z 114 (methylpiperidinol fragment) at
approximately 50% R.A. in each mass spectrum. Signif-
icant diarylmethane fragments also occurred for all
compounds except 3f. For compounds with very weak
(MÀH)+ ions in the EI mass spectrum, electrospray
ionization mass spectra from a methanol solution were
taken to confirm the molecular weight as the (M+H)+ ion.
The relative response of all compounds was from 95.8%
1
We established that the para-substituents of the benzene
rings of 3a–c, 3g, and 3h are important for more effective
binding to the DAT. Among our synthesized DAT
up to 99.1%. The H NMR (300 MHz, CDCl3) spectra of
N-methylpiperidines showed typical signals: 1.30–1.35 (m,
2H, H-2,6 ax), 1.85–1.90 (m, 2H, H-2,6 equiv.), 2.00–2.05

4918
G. B. Lapa et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4915–4918
(m, 2H, H-3,5 ax), 2.40–2.45 (s, 3H, N-CH3), 2.90–3.00
17. (a) Ertl, P.; Rohde, B.; Selzer, P. J. Med. Chem. 2000, 43,
3714; (b) Free online service: http://www.molinspira-
tion.com; (c) QSAR model was developed on the basis of
calculation of the molecular properties of compounds DPP,
3a–3h. The equation of correlation was log (IC50) =
À0.3403(logP + Na + pKa) + 7.5292 (where logP—calcu-
lated octanol/water partition coefficient, Na—the number of
hydrogen bond acceptors, pKa— calculated dissociation
constant of bases in water). R2 value was 0.7533, r2 = 0.8699
and q2 = 0.7855. The prediction of bioavailability as an
Nv—molecular violating parameter was performed on the
basis of Ôrule of 5Õ on the same web site. The compounds
might have poor bioavailability if Nv is more then 0.
18. Newman, A. H.; Allen, A. C.; Izenwasser, S.; Katz, J. L.
J. Med. Chem. 1994, 37, 2258.
(m, 2H, H-3,5 equiv.), 3.50–3.55 (m, 1H, H-4 ax), 5.35–
5.45 (s, 1H, Ar-CH-Ar), 7.20–7.60 (m, aromatic C–H).
16. (a) Letchworth, S. R.; Smith, H. R.; Porrino, L. J.;
Bennett, B. A.; Davies, H. M. L.; Sexton, T.; Childers, S.
R. JPET 2000, 293, 686; (b) Potencies were calculated
from displacement curves using 6–8 concentrations of
unlabeled compounds. All data are mean values SEM of
at least three separate experiments, each of which was
conducted in triplicate. Potencies of all unlabeled com-
pounds in displacing [125I]RTI-55 binding are expressed as
IC50 values because the biphasic nature of [125I]RTI-55
binding to striatal membranes makes determination of
accurate Ki values difficult. For a detailed description, see
the publication 16a.