Design, synthesis, and activity of novel cis- and trans-3,6-disubstituted pyran biomimetics of 3,6-disubstituted piperidine as potential ligands for the dopamine transporter

Article abstract of DOI:10.1016/S0960-894X(03)00169-0

In our effort to develop novel molecules for the dopamine transporter, we converted our previously designed dopamine transporter specific 3,6-disubstituted piperidine template into corresponding pyran derivatives. cis-Pyran derivative 7b, like their piper

Full text of DOI:10.1016/S0960-894X(03)00169-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1591–1595
Design, Synthesis, and Activity of Novel cis- and trans-3,6-
Disubstituted Pyran Biomimetics of 3,6-Disubstituted Piperidine
as Potential Ligands for the Dopamine Transporter
Shijun Zhang,a Maarteen E. A. Reithb and Aloke K. Duttaa,*
aDepartment of Pharmaceutical Sciences, Wayne State University, Detroit, MI 48202, USA
bUniversity of Illinois, College of Medicine, Department of Biomedical and Therapeutic Sciences, Peoria, IL 61605, USA
Received 4 November 2002; accepted 16 December 2002
Abstract—In our effort to develop novel molecules for the dopamine transporter, we converted our previously designed dopamine
transporter specific 3,6-disubstituted piperidine template into corresponding pyran derivatives. cis-Pyran derivative 7b, like their
piperidine counterparts, exhibited greater activity for the dopamine transporter compared to the trans-isomer. Further molecular
modifications of the cis derivative led to the development of potent analogues which indicated successful bioisosteric replacement of
the piperidine ring by a pyran moiety in these 3,6-disubstituted derivatives.
# 2003 Elsevier Science Ltd. All rights reserved.
Cocaine is a potent drug of abuse which poses a great
burden to our society due to its impact on the economy,
crime and spread of HIV.1 Extensive studies have been
conducted so far to understand the mechanism of action
of cocaine which might eventually lead to development
of a much needed medication for cocaine dependence.
Cocaine binds to all three monoamine transporter sys-
tems in the brain but its central reinforcing action is
thought to be derived mainly from binding to the
dopamine transporter (DAT).2,3 This does not rule out
the involvement of nondopaminergic systems in cocaine
reward pathways (Fig. 1).4
structurally constrained 3,6-disubstituted piperidine deri-
vatives (Fig. 2). Our study demonstrated that the cis-
structure in this novel template exhibited preferential
interaction with the DAT compared to the trans-struc-
ture.10 Our initial study was later followed by a more
extensive structure–activity study that produced potent
optically active 3,6-disubstituted compounds, thus, con-
firming the specificity of this novel template for the DAT.11
In our current study, we wanted to extend these findings by
replacing the piperidine ring by a pyran moiety.
Many efforts have been directed so far to develop
molecules targeting DAT. These have resulted in the
generation of a database with many molecules exhibit-
ing high selectivity and potency for the DAT. Two
comprehensive review articles have been published so
far which describe this work in detail.5,6 Our effort to
develop molecules targeting DAT started with piper-
idine analogues of GBR 12909. Our structure–activity
relationship study led to development of many potent
and highly selective molecules for the DAT and some of
them exhibited interesting in vivo properties.7ꢀ9 In our
recent effort to design structurally constrained molecule
of flexible piperidine analogues of GBR derivatives, we
converted one of our lead piperidine analogues into
*Corresponding author. Tel.: +1-313-577-1064; fax: +1-313-577-
2033; e-mail: adutta@wayne.edu
Figure 1. Structure of dopamine transporter blockers.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00169-0

1592
S. Zhanget al. / Bioorg. Med. Chem. Lett. 13 (2003) 1591–1595
Figure 2. Structural transformation of I into more conformationally constrained molecule.
Transformation of certain DAT-selective 3-aryltropane
analogues into oxy-3-aryltropane derivative was carried
out earlier and was shown to have minimal influence on
activity for the DAT compared to its parent bioisosteric
N-analogues.12 On the other hand, benztropine mole-
cules that underwent such transformation to oxy-benz-
tropine analogues, lost their potency completely for the
DAT.13 These results are interesting since both oxy-3-
aryltropane and oxy-benztropine analogues contain
constrained pyran moiety albeit oxy-3-aryltropane ana-
logues contain additional substitutions; yet they exhib-
ited different pharmacological properties. The results
also point to the N-atom in benzopyran as a critical
requirement for binding to the DAT whereas the
N-atom in 3-aryltropane analogues may be not so
critical, reflecting a possible involvement of different
binding modes between these two series of compounds.
Hence, a structurally constrained pyran moiety requires
more molecular specificity for exhibiting activity at
DAT compared to a structurally constrained piperidine
motif. In our study, we decided to explore this specifi-
city further by converting our newly developed 3,6-di-
substituted piperidine template into corresponding
pyran analogues. Several changes may occur in mole-
cular interaction with the receptor with such modifi-
cation. Conceivably, the mode of interaction between
DAT and pyran derivatives may differs from their
bioisosteric piperidine counterparts. Moreover, CNS-
active pyran derivatives are relatively rare and it will be
interesting to examine how pyran derivatives behave in
in vivo experiments compared to their piperidine
counterparts.
Scheme 1 delineates the preparation of the key inter-
mediate 5. Diphenylacetaldehyde 1 reacted with a
Grignard reagent, prepared in situ by reacting magne-
sium with 4-bromo-1-butene, to furnish 1,1-diphenyl-
hex-5-en-2-ol 2 in 91% yield.14 Vinyllation of com-
pound 2 with ethyl vinyl ether in presence of mercury
trifluoroacetate produced 1,1-diphenyl-2-(1-ethenoxy)-
hex-5-ene 3 in 66% yield.15 Ring closing metathesis of 3
with Grubb’s catalyst in refluxing benzene afforded
cyclic olefin 4 in 92.6% yield.16 Hydroboration of 4 with
9-BBN in THF, followed by oxidation with NaOH and
H2O2, regioselectively produced hydroxylated com-
pound 5. The compound trans-6-benzhydryl-tetra-
hydropyran-3-ol 5 was obtained in major amount
(93.5% yield).
As described in Scheme 2, compound 5 was subjected
to Swern oxidation condition, which produced the keto
product 6, 6-benzhydryl-dihydro-pyran-3-one, in 91%
yield. This was followed by reductive amination10 with
4-fluoro-benzylamine to produce the target compound
7a, trans-(6-benzhydryl-tetrahydropyran-3-yl)-(4-fluoro-
benzyl)-amine, as a major product in 50% yield.
Scheme 3 shows the synthesis of cis-target compounds.
Compound 5 was converted into amine 10, cis-(6-benz-
hydryl-tetrahydropyran-3-yl)-amine, by three steps in
overall 85% yield. Thus, mesylation with methane-
sulfonyl chloride and triethylamine in dichloromethane
produced 8 which on displacement with sodium azide in
DMF produced azide 9. Catalytic hydrogenation in
presence of 10% Pd/C as a catalyst in methanol yielded
amine 10 in good yield. Reductive amination of com-
pound 10 with various aldehydes furnished target cis
compounds 7b–e in good yield.
The synthesis of our target compounds is shown in
Schemes 1–3.
Scheme 1. (a) 4-Bromo-1-butene, Mg, Et2O; (b) ethyl vinyl ether, Hg(OCOCF3)2; (c) Grubb’s catalyst, benzene; (d) 9-BBN/THF NaOH, H2O2.

S. Zhanget al. / Bioorg. Med. Chem. Lett. 13 (2003) 1591–1595
1593
Scheme 2. (a) DMSO, oxalyl chloride, Et3N; (b) 4-fluorobenzylamine, AcOH, ClCH2CH2Cl/NaCNBH3 MeOH.
The assignment of the structures of target compounds
was determined by 1-D and 2-D NMR (300 MHz)
experiments.17,18 The assumption that the diphe-
nylmethyl moiety would assume the equatorial position
was confirmed by the NMR data.
tion of H-2ax resulted in an enhancement of H-6 and
H-2eq signal. Therefore, diphenylmethyl moiety must
be equatorially oriented in order to have this interac-
tion.
Our earlier SAR results delineated certain differences
between flexible and constrained piperidine analogues
which reflected different requirement of molecular
determinants for their activity for DAT. One of our
goals in this endeavor is to gain an understanding of
bioactive conformations of molecules through this
molecular rigidification. As a part of extension of our
studies from piperidine derivatives, in this current study,
we wanted to examine biological properties of corre-
sponding bioisosteric pyran derivatives.
In the trans compound 7a (Table 1), the signal attrib-
uted to H-6, d 3.99, is a doublet of triplets (3J=2.1 and
3
9 Hz) where J5ax,6=3J6,CH(Ph)2. The large coupling
constant between H-6 and H-5ax is consistent with H-6’s
axial orientation. The triplet of triplets signal at d 2.68
(3J=4.2Hz and 10.5Hz) attributed to H-3 is 39Hz broad
À
Á
3
3
3
3
composed of S J2ax;3 þ J3;4ax þ J2eq;3 þ J3;4eq . The
axial orientation of H-3 can be confirmed by having two
3
3
smalls Jax,eq and two large Jax,ax coupling. This orien-
tation also can be confirmed by the signal at d 3.11 ppm
attributed to H-2ax which appears as triplet (2J2ax,2eq
and J2ax,3) with J=3J=10.5 Hz.
Following our synthesis, novel molecules were char-
acterized for their potency in binding to the mono-
amine transporter systems by monitoring inhibition of
binding of [3H]WIN 35, 428 to DAT, [3H]citalopram
to SERT, and [3H]nisoxetine to NET in rat brain
tissue (Table 2). The current trans- and cis-derivatives,
7a and 7b correspond to our initial 3,6-disubstituted
piperidine derivatives. The objective was to test
whether the same cis isomeric preference for the DAT
exhibited by the piperidine version is also maintained
in these pyran derivatives. Binding results confirmed
this outcome by showing higher activity in the expec-
ted cis-compound 7b albeit less potent than the corre-
3
2
In the cis compound 7b (Table 1), the signal from H-6, d
4.08, appears as doublet of triplet with J=2.4 Hz and
3
9 Hz, same as in the trans compound. The multiplet
signal at d 2.64 with width of 12 Hz, attributed to H-3.
This small S3J is consistent with H-3 equatorial orien-
tation, which has same vicinal coupling from adjacent
four protons (3J2ax,3+3J2eq,3+3J4ax,3+3J4eq,3). This
equatorial orientation also can be confirmed by the sig-
3
nals from H-2, two doublet of doublets with J=1.8 Hz
2
and J=11.4 Hz. the large coupling is from the geminal
coupling of the two protons in C-2, the small one is
from vicinal coupling with H-3.
sponding piperidine version II. Interestingly,
a
difference in activity was found for their interaction
with the NET as the cis-pyran derivative 7b was
4-fold more active at the NET compared to the
piperidine version II.
In addition, the axial orientation of H-6 was also con-
firmed by the nuclear Overhauser experiment. Irradia-
Scheme 3. (a) Ch3SO2Cl, Et3N/CH2Cl2; (b) NaN3/DMF; (c) H2/Pd/C, MeOH; (d) aldehyde, AcOH, ClCH2CH2Cl/NaCNBH3, MeOH.

1594
S. Zhanget al. / Bioorg. Med. Chem. Lett. 13 (2003) 1591–1595
Table 1. Selected NMR data of 7a and 7b
Compd
Chemical shift
H2ax, 3.11
H3ax, 2.68
Multiplicity
Coupling constants
7a
t
tt
J2ax,2eq=J2ax,3eq=10.5 Hz
J2eq,3ax=J4eq,3ax=4.2Hz
J2ax,3ax=J4ax,3ax=10.5 Hz
J5eq,6ax=2.1 Hz
H6ax, 3.99
dt
JCH(Ph)2,6ax=J5ax,6ax=9 Hz
J2ax,3eq=1.8 Hz, J2eq,2ax=11.4 Hz
Peak width=12 Hz
7b
H2ax, 3.57
H3eq, 2.64
H6ax,4.08
dd
m
dt
J5eq,6ax=2.4 Hz
JCH(Ph)2,6ax=J5ax,6ax=9 Hz
Table 2. Affinity of drugs at the dopamine, serotonin and norepinephrine transporters in rat striatum and in inhibition of dopamine (DA) reuptake
Compd
DAT binding, IC50 (nM),
[3H]WIN 35, 428a
SERT binding, IC50 (nM),
[3H]citaloprama
NET binding, IC50 (nM),
[3H]nisoxetinea
DAT uptake, IC50 (nM),
[3H]DAa
Cocaine
GBR 12909
I
(ꢁ)-II
(ꢁ)-7a
(ꢁ)-7b
(ꢁ)-7c
(ꢁ)-7d
(ꢁ)-7e
266ꢁ37
10.6ꢁ1.9
19.7ꢁ1.4
32.5ꢁ12.6
313ꢁ71
737ꢁ160
132ꢁ0
3530ꢁ554
496ꢁ22
6.63ꢁ0.43
49.6ꢁ7.2
ꢁ5.1
137ꢁ46
1110ꢁ120
1020ꢁ7245.7
12,700ꢁ3,180
232ꢁ46
2220ꢁ590
8410ꢁ163
1860ꢁ22
863ꢁ521580
1180ꢁ269
738ꢁ164
163ꢁ29
156ꢁ36
58.6ꢁ13.2
ꢁ11.6
52.6ꢁ5.9
84.0ꢁ6.5
38.3ꢁ3.9
ꢁ89
1550ꢁ68259.5
968ꢁ98
102ꢁ7
Results are average ꢁSEM of three independent experiments assayed in triplicate
aThe DAT was labeled with [3H]WIN 35,428, the SERT with [3H]citalopram and the NET with [3H] nisoxetine (see ref 8 for details).
Effects of both electron withdrawing and electron
donating substituents were tried on the phenyl ring of
the N-benzyl group. Thus, the electron donating
methoxy substituent in compound 7d made it more
potent at DAT than 7b (Table 2). On the other
hand, the strong electron withdrawing substituents
cyano and nitro in compounds 7c and 7e resulted in the
development of greater potencies in these compound. In
this regard, the racemic cyano compound 7c was about
one and half time less potent at DAT than the corres-
ponding piperidine version analogue. Affinity of these
compounds for the SERT and NET were weak demon-
strating their preferential interactions with the DAT.
products 7b–e was controlled by selective displacement
of a good leaving group in compound 8. Our current
results indicate that bioisosteric replacement of the
N-atom by an oxygen atom did not interfere with
activity at the DAT, although it is not known at this
time whether the two groups of compounds bind in an
identical manner. We are currently expanding our
initial findings with expanded SAR studies to explore
this aspect in greater detail. In addition, our future in
vivo studies with these compounds will tell whether
3,6-pyran derivatives, despite the fact that their bind-
ing activities at DAT are similar to those of the cor-
responding piperidine analogues, exert differential
behavioral activities.
Evaluation of inhibition of dopamine uptake activity
was carried out with the new derivatives. No differential
activity was observed between binding and uptake
potency except in compound 7e. Compound 7e was
almost three times more potent in inhibiting binding
than in inhibiting uptake.
Acknowledgements
This work was supported by the National Institute on
Drug Abuse, Grant No. DA 12449 (AKD). We thank
Mrs. Janet L. Berfield and Li Juan Wang for perform-
ing the binding and uptake assays.
In conclusion, we have synthesized novel 3,6-di-
substituted-pyran derivatives, which showed pre-
ferential affinity for the DAT. The synthesis of these
compounds was carried out with a key intermediate
step, which involved ring formation via a ring closing
metathesis process by using Grubb’s catalyst to generate
the key intermediate 5. Stereochemistry of the target
References and Notes
1. (a) Substance Abuse and Mental Health Services Adminis-
tration National Household Survey on Drug Abuse. Main

S. Zhanget al. / Bioorg. Med. Chem. Lett. 13 (2003) 1591–1595
1595
Findings 1997; Department of Health and Human Services:
Washington, DC, 1998. (b) Mathias, R. NIDA Notes 2000, 15,
4.
Rossi, M.; Invidiata, F. P.; Merighi, S.; Varani, K.; Gessi, S.;
Borea, P.; Marino, S.; Cavallini, S.; Bianchi, C.; Siniscalchi, A.
Bioorg. Med. Chem. Lett. 2001, 11, 823.
2. Wilcox, K. M.; Paul, I. A.; Woolverton, W. L. Eur. J.
Pharmacol. 1999, 367, 175.
14. Holladay, M. W.; Salituro, F. G.; Rich, D. H. J. Med.
Chem. 1987, 30, 374.
3. Kuhar, M. J.; Ritz, M. C.; Boja, J. W. Trends. Neurosci.
1991, 14, 299.
15. Mukund, K.; Gurjar, L.; Krishna, M.; Reddy, B. S.;
Chorghade, M. S. Synthesis 2000, 4, 557.
4. Rocha, B. A.; Fumagalli, F.; Gainetdinov, R. R.; Jones, S.;
Ator, R.; Giros, B.; Miller, G. W.; Caron, M. Nat. Neurosci.
1998, 1, 132.
16. Sturino, C. F.; Wong, J. C. Y. Tetrahedron Lett. 1998, 3,
9623.
17. Racemic 7a H NMR (300MHz, CDCl3) d 1.18–1.44 (m,
1
5. Carrol, F. I.; Lewin, A. H.; Mascarella, S. W. In Neuro-
transmitter Transporters: Structure and Function; 2nd ed.;
Reith, M. E. A., Ed.; Human: Totowa, 2002; p 381.
6. Singh, S. Chem. Rev. 2000, 100, 925.
7. Dutta, A. K.; Coffey, L. L.; Reith, M. E. A. J. Med. Chem.
1997, 40, 35.
8. Dutta, A. K.; Fei, X.-S.; Beardsley, P.; Newman, J.; Reith,
M. E. A. J. Med. Chem. 2001, 44, 937.
9. Ghorai, S.; Cook, C. D.; Davis, M.; Venkataraman, S.;
Beardsley, P.; Reith, M. E. A.; Dutta, A. K. J. Med. Chem. in
press.
10. Dutta, A. K.; Davis, M. C.; Reith, M. E. A. Bioorg. Med.
Chem. Lett. 2001, 11, 2337.
11. Kolhatkar, R.; Ghorai, S.; Reith, M. E. A.; Dutta, A. K.
Submitted for publication.
12. Meltzer, P. C.; Liang, A. Y.; Blundell, P.; Gonzalez,
M. D.; Chen, Z.; George, C.; Madras, B. K. J. Med. Chem.
1997, 40, 2661.
2H, H-5), 1.55 (m, 1H, H-4), 1.75 (bm, NH), 2.02 (m, 1H, H-4),
2.68 (tt, J=4.2Hz, 10.5 Hz, 1H, H-3), 3.11 (t, J=10.8 Hz, 1H,
H-2ax), 3.76 (s, 2H, (F)–Ph–CH2), 3.86 (d, J=9 Hz, 1H,
Ph2CH), 3.99 (dt, J=3Hz, 8.7 Hz, 1H, H-6), 4.06 (m, 1H, H-
2eq), 6.9–7.38 (m, 14H, aromatic–CH); 13C NMR (300MHz,
CDCl3) d (ppm) 29.93, 31.46, 50.81, 53.37, 57.78, 73.00, 79.42,
126.50, 126.73, 128.56, 128.65, 128.80, 129.77, 129.88. Free base
was converted into oxalate C27H28NO5F 0.65H2O. Anal. theory
C 67.96H 6.19 N 2.94, found C 67.93H 6.02 N 3.02.
1
18. Racemic 7b H NMR (300 MHz, CDCl3) 1.33 (m, 1H, H-
5), 1.46–1.72 (m, 2H, H-5, H-4), 1.94 (m, 1H, H-4), 2.03 (bm,
1H, NH), 2.64 (m, 1H, H-3eq), 3.57 (dd, J=1.8 Hz, 11.4 Hz,
1H, H-2ax), 3.75 (m, 2H, (F)Ph–CH2), 3.95–4.14 (m, 3H, H-6,
H-2eq, Ph2CH), 4.08 (dt, J=2.4 Hz 9 Hz, 1H, H69x), 6.9–7.38
(m, 14H, aromatic–CH); 13C NMR (300 MHz, CDCl3) d
(ppm) 21.50, 29.95, 32.00, 55.72, 65.62, 76.05, 101.50, 126.88,
127.09, 128.60, 128.68, 128.90, 128.97, 141.36, 141.62. Free
base was converted into oxalate C27H28NO5F. Anal. theory C
69.66H 6.06 N 3.01, found C 69.60H 6.09 N 2.97.
13. Simoni, D.; Roberti, M.; Rondanin, R.; Baruchello, R.;