Synthesis and NMDA-receptor affinity of 4-oxo-dexoxadrol derivatives

Article abstract of DOI:10.1016/j.bmc.2006.05.020

A synthesis of novel dexoxadrol analogues is described, which allows modifications of the piperidine substructure. The key step of the synthesis is a hetero Diels-Alder reaction of the imine 12 with Danishefsky's diene 6. After separation of the diastereo

Full text of DOI:10.1016/j.bmc.2006.05.020

Bioorganic & Medicinal Chemistry 14 (2006) 5955–5962
Synthesis and NMDA-receptor affinity
of 4-oxo-dexoxadrol derivatives
Michael Sax,a Kirstin Ebert,a Dirk Schepmann,a Birgit Wibbelingb and
Bernhard Wunscha,*
¨
aInstitut fu¨r Pharmazeutische und Medizinische Chemie, Westfa¨lische Wilhelms-Universita¨t Mu¨nster,
Hittorfstr. 58-62, D-48149 Mu¨nster, Germany
bInstitut fu¨r Organische Chemie, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, Corrensstr. 40, D-48149 Mu¨nster, Germany
Received 1 February 2006; revised 4 May 2006; accepted 15 May 2006
Available online 5 June 2006
Abstract—A synthesis of novel dexoxadrol analogues is described, which allows modifications of the piperidine substructure. The
key step of the synthesis is a hetero Diels–Alder reaction of the imine 12 with Danishefsky’s diene 6. After separation of the
diastereomeric piperidones 14a and 14b, the relative configuration of the unlike configured piperidone 15b was determined by
X-ray crystal structure analysis. In receptor binding studies the like configured secondary amine 15a (racemate) showed considerable
affinity toward the phencyclidine binding site of the NMDA receptor (Ki = 470 nM).
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
With regard to Alzheimer’s disease, NMDA receptor
antagonists have neuroprotective potential by protecting
neurons from excessive pathological calcium influx,
which leads to damage of neurons and finally to neuro-
nal cell death. Up to now, memantine (1a) is the only
NMDA receptor antagonist with moderate NMDA
receptor affinity (Ki = 1.2 lM) which has been intro-
duced for the treatment of severe Alzheimer’s disease.1,2
The particular advantage of memantine is the fast off-
rate kinetic preventing the drug from accumulating in
the ion channel. Therefore, memantine enters the open
channel preferentially when it is pathologically activated
for long periods of time. The physiological neurotrans-
mission is not disturbed by memantine resulting in min-
imal adverse effects.1 Parkinson’s disease is treated with
the NMDA receptor antagonists amantadine (1b) and
budipine (1-tert-butyl-4,4-diphenylpiperidine) display-
ing also moderate NMDA receptor affinity (Fig. 1).
Diseases with age as a major risk factor are becoming
more and more threatening for the older population.
Such is the case for neurodegenerative disorders like
Parkinson’s disease (PD) and Alzheimer’s disease
(AD), which are caused by the degeneration of dopami-
nergic and cholinergic neurons, respectively. Unfortu-
nately, the possibilities for the treatment of these
diseases are rather poor because up to now the therapeu-
tic gold standard is compensation of the decreased neu-
rotransmitter level. This is usually achieved by giving
levodopa as a dopamine precursor (PD) and acetylcho-
linesterase inhibitors (AD) to prevent degradation of the
neurotransmitter acetylcholine, which leads to higher
neurotransmitter concentrations in the synaptic cleft.
These treatments do lead to higher efficacy of the
remaining intact neurons, but the result is merely a
symptomatic therapy. Unfortunately, the treatments
do not prevent nor reduce the progression of neurode-
generation. Therefore, great efforts have been undertak-
en to develop drugs which meet this impairment.1
In contrast to memantine, the piperidine derivative dex-
oxadrol (2) binds with high affinity (Ki = 39 nM) at the
phencyclidine binding site within the NMDA receptor-
associated cation channel. Dexoxadrol (2) was synthe-
sized by Hardie et al. in the 1960s and revealed local
anesthetic, spasmolytic, and central nervous system
activity.3a Later, phencyclidine-like (3) dissociative anes-
thetic activities were found.4 Unfortunately, clinical
Keywords: NMDA receptor antagonists; 4-Oxo-dexoxadrol; Hetero
Diels–Alder reaction; X-ray crystal structure; Piperidines 2-
substituted.
*
Corresponding author. Tel.: +49 0251 83 33311; fax: +49 0251 83
32144; e-mail: wuensch@uni-muenster.de
trials of 2 had to be stopped because of the
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.05.020

5956
M. Sax et al. / Bioorg. Med. Chem. 14 (2006) 5955–5962
the acetalic center have been modified; the amino moiety
H
O
has been altered from a cyclic amine (piperidine) to ali-
phatic primary, secondary, and tertiary amines; the dis-
tance between the dioxolane ring and the amino moiety
has been changed; the dioxolane ring has been enlarged
to the corresponding dioxane ring; furthermore, the
piperidine ring has been replaced by a pyrrolidine ring
and the dioxolane ring by a cyclopentane ring.3 But in
no case has the piperidine substructure itself been
modified or substituted.
NH2
N
H
O
H
N
R
R
1
3
2
1a: R = CH3
1b: R = H
In this article, we wish to present a synthesis of dexoxa-
drol derivatives along a synthetic route that allows
manifold substitution of the piperidine moiety. Pharma-
cological properties of these new dexoxadrol analogues
are also included.
Figure 1. Structures of some important NMDA receptor antagonists:
1a, memantine; 1b, amantadine; 2, dexoxadrol; 3, phencyclidine (PCP).
psychotomimetic side effects. However, many deriva-
tives of 2 have been synthesized since then in order to
get more insight into the structure–affinity relationships
within this substance class.3 Since the severe side effects
of dexoxadrol are attributed to its high NMDA receptor
affinity, novel analogues with affinities between those of
dexoxadrol (Ki = 39 nM) and memantine (Ki = 1.2 lM)
should display an improved side-effect profile. The fol-
lowing modifications of the dexoxadrol structure have
already been described in the literature: the residues at
2. Chemistry
The key step of our synthetic strategy comprises the con-
struction of 2-substituted piperidines 7 from aldehydes 4
with acetalic substructures Racetalic. Condensation of
these aldehydes 4 with benzylamine should lead to the
corresponding imines 5, which subsequently react in a
OTMS
O
H3CO
Racetalic
O
4
6
5
6
3
H
Racetalic
N
2
Racetalic
1N
4
Bn
Bn
5
2
NH
Bn
7
Figure 2. Plan for the synthesis of dexoxadrol analogues substituted at the piperidine ring.
O
HO
O
(a)
O
+
HO
OH
OH
8
9
10
(b)
O
O
N
OTMS
NH2
O
6
N
H3CO
O
H
O
O
O
(d)
O
(c)
12
13a: like
11
13b: unlike
Scheme 1. Synthesis of dihydropyridones 13a/b (all compounds are racemic). Reagents and conditions: (a) TosOHÆH2O, toluene, reflux, 6 h, yield
73%; (b) oxalyl chloride, DMSO, CH2Cl2, À78 ꢁC, 30 min, then NEt3, rt, yield 72%; (c) benzylamine, trimethyl orthoformate, rt, 16 h; (d) Yb(OTf)3,
6, THF, 0 ꢁC (4 h) then 25 ꢁC (16 h), yield 70% from 11.

M. Sax et al. / Bioorg. Med. Chem. 14 (2006) 5955–5962
5957
hetero Diels–Alder reaction5 with Danishefsky’s diene
3. NMDA receptor affinity
66 to afford dihydropyridones 7. The dihydropyridones
7 allow the introduction of various substituents in posi-
tions 3, 4, 5, and 6 and, therefore, represent ideal precur-
sors for the synthesis of substituted piperidine
derivatives7 (Fig. 2).
The affinity of the four compounds 14a, 14b, 15a and
15b to the phencyclidine binding site of the NMDA
receptor was examined in competition experiments using
the radioligand [3H]-MK-801. The receptor material was
prepared by homogenization and fractional centrifuga-
tion of the grey substance of pig brain cortex. The
non-specific binding was determined in the presence of
The synthesis was started with acetalization of benzo-
phenone (9) with glycerol (8) to yield the (diphenyldiox-
olanyl)methanol 10. Swern oxidation8 of the primary
alcohol 10 afforded the aldehyde 11, which was con-
densed with benzylamine to give the imine 12.
Yb(OTf)3-catalyzed hetero Diels–Alder reaction of the
imine 12 with freshly9 prepared10 Danishefsky’s diene
6 yielded the 1,2-disubstituted dihydropyridone 13 as a
mixture of diastereomers. The diastereomeric ratio of
13a:13b (59:41) was determined by integration of char-
1
acteristic signals in the H NMR spectrum (Scheme 1).
Selective reduction of the double bond of 13 with
ꢀ
11
LiEt3BH in the presence of BF3 OEt2 afforded the
N-protected piperidin-4-ones 14a (like configuration)
and 14b (unlike configuration), which were separated
by flash chromatography. Finally, debenzylation with
H2 and Pd/C provided the desired 4-oxo-dioxadrols
15a and 15b12 (Scheme 2).
It was not possible to determine unequivocally the rela-
tive configuration of 15a and 15b by interpretation of
their NMR spectra. Therefore, 15b was carefully recrys-
tallized (Et2O) to obtain colorless crystals suitable for an
X-ray crystal structure analysis, which proved the unlike
configuration (SR/RS) of the centers of chirality (Fig. 3).
Figure 3. X-ray crystal structure of unlike configured oxo-dexoxadrol
derivative 15b (racemate).
O
O
H
H
N
(b)
N
O
H
O
H
O
H
O
14a*
15a*
(a)
13a,b
+
O
O
H
H
N
H
(b)
N
O
O
H
O
H
O
14b*
15b*
14a, 15a: like
14b, 15b: unlike
*only one enantiomer is shown
Scheme 2. Synthesis of 4-oxo-dexoxadrol derivatives (only one enantiomer is shown, all compounds are racemic). Reagents and conditions: (a)
BF3ÆOEt2, THF, À78 ꢁC, 30 min, then LiEt3BH, 1 h, À78 ꢁC, yields of 14a: 55%, 14b: 37%. (b) H2, balloon, Pd/C 10%, methanol, 2 h or 4 h, rt, yields
of 15a: 78%, 15b: 86%.

5958
M. Sax et al. / Bioorg. Med. Chem. 14 (2006) 5955–5962
Table 1. Affinities of 14 and 15 toward the phencyclidine binding site
of the NMDA receptor
in the formalin assay and the Irwin screen the pharma-
cological evaluation of 15a was terminated at this
point.
Compound
NMDA receptor affinity ([3H]-MK-801)
14a
14b
96%a
95%a
5. Conclusion
15a
15b
Ki = 470 173 nM (n = 3)
84%a
Ki = 39 10 nM (n = 3)
Dexoxadrol (2)
Herein, we have demonstrated that the Lewis acid-cata-
lyzed hetero Diels–Alder reaction of Danishefsky’s diene
6 with imines provides piperidine derivatives bearing
acetalic substituents in position 2 in good yields. In com-
parison with the lead compound dexoxadrol (2) the
NMDA receptor affinity of racemic oxo-dexoxadrol
15a (Ki = 470 nM) is considerably reduced indicating
the H-bond acceptor in position 4 of the piperidine ring
being disadvantageous for the NMDA receptor interac-
tion. However, the carbonyl moiety in position 4 of 15a
allows several further variations of the piperidine sub-
stituents, including non-polar residues and substituents
with H-bond donor properties.
a Specific binding of the radioligand at a concentration of 10 lM of the
test compound.
a large excess of cold MK-801.13–15 The Ki-values were
calculated according to the equation of Cheng and
Prussoff.16 The NMDA receptor binding data are
summarized in Table 1.
4. Discussion
Table 1 shows that the tertiary amines 14a and 14b do
not interact significantly with the phencyclidine bind-
ing site within the NMDA receptor. Comparison of
the various NMDA receptor affinities of the secondary
amines indicates that only the racemate 15a (like con-
figuration) displays considerable NMDA receptor
affinity, whereas its diastereomeric racemate 15b
(unlike configuration) does not bind significantly.
These results support the literature data that tertiary
amines generally reveal lower affinity than secondary
and primary amines and, furthermore, the stereoiso-
mers with (S,S)-(like) configuration display higher
affinities than their enantiomers and unlike configured
diastereomeres.17
6. Experimental
6.1. Chemistry, general
Thin-layer chromatography (TLC): silica gel 60 F254
plates (Merck). Flash chromatography (FC):21 Silica
gel 60, 40–63 lm (Merck). MS, MAT GCQ (Thermo-
Finnigan); EI, electron impact; IR, IR spectrophotome-
1
ter 480Plus FT-ATR-IR (Jasco). H NMR (400 MHz),
13C NMR (100 MHz): Unity m400 NMR spectrometer
(Varian); d in ppm related to tetramethylsilane, coupling
constants are given with 0.5 Hz resolution; the assign-
1
ments of 13C and H NMR signals were supported by
The NMDA receptor affinity of 15a (racemate) is about
12 times lower than the affinity of enantiomerically pure
dexoxadrol. However, the Ki-value of 470 nM is in the
desired affinity range (50–1000 nM). Therefore, further
properties of 15a were investigated.
2D NMR techniques. Elemental analysis: CHNO-Rapid
Analysator (Fons-Heraeus). THF was dried with sodi-
um and was freshly distilled before use. CHCl3 was dis-
˚
tilled over CaH2 and stored over molecular sieves 4 A.
Petroleum ether fraction bp 40–60 ꢁC. Yb(OTf)3 (Al-
drich) was stored in a desiccator over P4O10 in vacuo
at room temperature. Danishefsky’s diene 6 was stored
under N2 at À25 ꢁC.9
According to Lipinski’s ‘rule of 5’ the number of
H-bond acceptors (n < 10) and H-bond donors (n < 5),
the molecular weight (MW < 500), and the lipophilicity
(logP < 5) determine the pharmacokinetic of a drug, in
particular its crossing of the blood–brain barrier.18 Since
the first three parameters are in agreement with the ‘rule
of 5,’ the logP value of 15a was determined with an
HPLC method.19 This method led to a logP value of
2.0 indicating sufficient water solubility and good pene-
tration into the central nervous system. These results
prompted the in vivo evaluation of oxo-dexoxadrol
15a in animal models.
6.2. trans-1-Methoxy-3-(trimethylsilyloxy)-1,3-butadiene
(Danishefsky’s diene, 6)
Procedure modified according to Refs. 9 and 10. Anhy-
drous LiBr (4.32 g, 49.8 mmol) was filled into a 100 mL
Schlenk flask, which was immediately sealed by a rubber
septum, flushed with N2, and heated to approximately
400 ꢁC with a heat gun. The flask was allowed to cool
down to room temperature under gentle N2 flow, then
THF (25 mL) was added. It was stirred until LiBr had
completely dissolved. Afterwards the mixture was cooled
to À15 ꢁC, chlorotrimethylsilane (4.72 mL, 37.3 mmol)
and trans-4-methoxybut-3-en-2-one (2.50 mL, 25 mmol)
were slowly added, and the mixture was stirred for
15 min. Then NEt3 (5.18 mL, 37.3 mmol) was added,
the solution was stirred at À15 ꢁC for 1 h and another
24 h at 40 ꢁC. Then the reaction mixture was transferred
with 30 mL of cold (4 ꢁC) pentane into a separating funnel
loaded with ice (15 g), a cold saturated solution of NaH-
The analgesic potency was investigated in the mouse for-
malin assay, a model for neuropathic pain. However, a
dose of 0.3 mg/kg of 15a did not cause significant activ-
ity in this assay. The safety of 15a was evaluated by
observing mice behavior after intraperitoneal applica-
tion of the test compound (Irwin screen).20 At a dose
of 1.0 mg/kg of 15a the mice showed excitation, abnor-
mal gait, and stereotypes. Increasing the dose to
30 mg/kg additionally led to convulsions, tremor, and
the Skraup tail phenomenon. Due to these observations

M. Sax et al. / Bioorg. Med. Chem. 14 (2006) 5955–5962
5959
CO3 (15 mL), cold brine (15 mL), and cold pentane
(30 mL) (4 ꢁC each). The organic layer was separated
and the aqueous layer was extracted twice with cold pen-
tane (30 mL). The combined organic layers were washed
with cold brine (15 mL), cold water (5· 15 mL), dried
(MgSO4), filtered, and concentrated (400 mbar; 40 ꢁC),
and the resulting brownish oil was distilled in vacuo to
yield 3.19 g (74%) of a clear colorless liquid, bp10 Torr
66–67 ꢁC (Ref. 10, bp7 Torr 65–70 ꢁC.).
d 66.5 (C-5), 80.4 (C-4), 111.9 (C-2), 126.1 (2 Ph-C),
126.4 (2 Ph-C), 128.3 (Ph-C), 128.39 (2 Ph-C), 128.40
(2 Ph-C), 128.5 (Ph-C), 141.1/141.2 (2 quart Ph-C),
201.4 (CHO); MS (70 eV) m/e (rel int): 255 (MH+, 6),
225 (MÀCHO, 65), 177 (MÀPh, 36), 167 (PhCPh,
100), 105 (PhCO, 40); Anal. Calcd for C16H14O3: C,
75.57; H, 5.55. Found: C, 74.97; H, 5.60.
6.5. ( )-1-Benzyl-2-(2,2-diphenyl-1,3-dioxolan-4-yl)-2,3-
dihydropyridin-4(1H)-one (13a and 13b)
6.3. ( )-(2,2-Diphenyl-1,3-dioxolan-4-yl)methanol (10)
The aldehyde 11 (773 mg, 3.0 mmol) was dissolved in
trimethyl orthoformate (7 mL), benzylamine (332 lL,
3.0 mmol) was added, and the solution was stirred over-
night at room temperature. The mixture was evaporated
to dryness, the obtained yellow oil (12) was dissolved in
THF (20 mL), and the resulting solution was cooled
down to 0 ꢁC. Then a solution of Yb(OTf)3 (337 mg,
0.61 mmol) in THF (4 mL) was added and it was stirred
at 0 ꢁC for 15 min. Afterwards, diene 6 (984 lL,
5.17 mmol) was added and it was stirred for another
4 h at 0 ꢁC before the reaction mixture was allowed to
warm to room temperature overnight. Then, water
(5 mL) and after 15 min Et2O (20 mL), water, a saturat-
ed solution of NaHCO3, and brine were added, the
organic layer was separated and the aqueous layer was
extracted with Et2O (3·). The organic layer was dried
(K2CO3), filtered and concentrated in vacuo. Flash
Glycerol 98% (8, 4.14 g, 44 mmol), benzophenone (9,
7.29 g, 40 mmol) and p-toluenesulfonic acid monohy-
drate (0.38 g, 2.0 mmol) were refluxed in toluene
(50 mL) at a water separator for 6 h. After cooling down
Et2O and a saturated solution of NaHCO3 were added.
The organic layer was separated and the aqueous layer
was extracted twice with Et2O. The combined organic
layers were dried (K2CO3), filtered, and concentrated
in vacuo. Flash chromatographic purification (n-hex-
ane/EtOAc = 7:3) of the residue afforded 7.55 g (73%)
of 10 as a colorless solid, mp 50.5–51.5 ꢁC; Rf = 0.22
(n-hexane/EtOAc = 7:3); IR (neat): 3421 (br m, –O–H),
1449 (s, –C–H deformation), 1204 (s)/1026 (s)/991 (s,
–C–O–C–); 1H NMR (CDCl3): d 1.84 (t, J = 6.4 Hz,
1H, OH), 3.66 (dd, J = 11.9/5.0 Hz, 1H, 5-H), 3.83
(dd, J = 11.8/3.5 Hz, 1H, 5-H), 4.00 (dd, J = 8.0/
6.1 Hz, 1H, CH2OH), 4.05 (dd, J = 8.0/7.0 Hz, 1H,
CH2OH), 4.34 (dddd, J = 7.1/5.8/4.9/3.6 Hz, 1H, 4-H),
7.26–7.42 (m, 6H, –Ph), 7.47–7.58 (m, 4H, –Ph); 13C
NMR (CDCl3): d 63.5 (C-5), 66.5 (CH2OH), 77.2 (C-
4), 110.3 (C-2), 126.1 (2 Ph-C), 126.4 (2 Ph-C), 128.3
(Ph-C), 128.39 (2 Ph-C), 128.40 (2 Ph-C), 128.46 (Ph-
C), 142.0/142.1 (2 quart Ph-C). Spectroscopic and ana-
lytical data of 10, which has been prepared from the cor-
responding chloromethyl derivative, are given in Ref.
22. However, these data do not match with the structure.
chromatographic
purification
(petroleum
ether/
EtOAc = 1:1) of the oily residue yielded 874 mg (70%)
of 13 as a yellow oil, Rf = 0.12 (petroleum ether/
EtOAc = 1:1); IR (neat): 1636 (m, C@O), 1577 (s,
C@C), 1205 (s)/1162 (m, characteristic dihydropyridone
1
fingerprint), 1073 (m, –C–O–C–); H NMR (CDCl3): d
1.91 (d, J = 16.8 Hz, 0.41H, 3x-H), 2.42 (dd, J = 16.8/
2.0 Hz, 0.59H, 3o-H), 2.64 (dd, J = 16.8/8.2 Hz, 0.59H,
3o-H), 2.69 (dd, J = 17.0/7.2 Hz, 0.41H, 3x-H), 3.39–
3.43 (m, 0.41 H, 2x-H), 3.63 (dd, J = 8.6/5.9 Hz, 0.41H,
OCHx2), 3.72–3.75 (m, 0.59H, 2o-H), 3.93 (dd, J = 8.2/
7.0 Hz, 0.59H, OCHo2), 3.96 (dd, J = 8.4/6.5 Hz, 0.41H,
OCHx2), 4.00 (dd, J = 8.2/6.3 Hz, 0.59H, OCHo2), 4.24
(td, J = 6.5/4.9 Hz, 0.59H, OCHo), 4.27 (d,
J = 15.3 Hz, 0.59H, PhCHo2), 4.42 (d, J = 15.3 Hz,
0.59H, PhCHo2), 4.43 (d, J = 14.3 Hz, 0.41H, PhCHx2),
4.65 (d, J = 14.9 Hz, 0.41H, PhCHx2), 4.75 (dt, J = 9.9/
6.3 Hz, 0.41H, OCHx), 4.89 (d, J = 7.4 Hz, 1H, 0.41
5x-H + 0.59 5o-H), 6.99–7.11 (m, 3H, 2· –Ph + 0.41
6x-H + 0.59 6o-H), 7.18–7.31 (m, 9H, –Ph), 7.39–7.44
(m, 4H, –Ph); 13C NMR (CDCl3): d 37.0 (C-3o), 37.6
(C-3x), 56.6 (C-2o), 58.8 (C-2x), 59.5 (PhCoH2), 60.2
(PhCxH2), 67.0 (OCoH2), 67.2 (OCxH2), 74.4 (OCxH),
77.4 (OCoH), 97.4 (C-5x), 98.6 (C-5o), 109.8 (OCoO),
110.9 (OCxO), 125.9/126.1/126.2/126.4/127.6/127.8/128.4/
128.5/128.6/129.2/129.3/136.7/137.1/142.0/142.1/142.2
(Ph-C), 153.2 (C-6x), 153.8 (C-6o), 189.1 (C-4x), 189.9
(C-4o); MS (70 eV) m/e (rel int): 411 (M, 2), 229
(MÀPh2CO, 38), 186 (M-diphenyldioxolanyl, 39), 91
(PhCH2, 100).
6.4. ( )-2,2-Diphenyl-1,3-dioxolane-4-carbaldehyde (11)
Under N2 CH2Cl2 (5 mL) and oxalyl chloride (103 lL,
1.2 mmol) were cooled down to À78 ꢁC. Then a solution
of dry DMSO (171 lL, 2.4 mmol) in CH2Cl2 (1.5 mL)
was added dropwise and the mixture was stirred for
10 min at À78 ꢁC. Afterwards, a solution of 10
(256 mg, 1.0 mmol) in CH2Cl2 (1.5 mL) was added drop-
wise and the solution was stirred for another 30 min at
À78 ꢁC before NEt3 (707 lL, 5.0 mmol) was added.
After warming to room temperature, n-hexane was add-
ed (10 mL), the mixture was filtered, and the precipitate
was washed with Et2O (10 mL). The organic layer was
concentrated (600 mbar, 40 ꢁC) and the filtration proce-
dure was repeated once. Then the solvent was evaporated
and the residue was purified by flash chromatography
(petroleum ether/EtOAc = 8:2) to afford 182 mg (72%)
of 11 as a highly hygroscopic colorless oil, Rf = 0.17
(petroleum ether/EtOAc = 8:2); IR (neat): 1734 (s,
1
C@O), 1206 (s)/1072 (s)/991 (s, –C–O–C–); H NMR
(CDCl3): d 4.08 (t, J = 8.4 Hz, 1H, 5-H), 4.14 (dd,
J = 8.7/4.8 Hz, 1H, 5-H), 4.41 (ddd, J = 8.0/4.8/1.9 Hz,
1H, 4-H), 7.20–7.30 (m, 6H, –Ph), 7.44–7.47 (m, 4H,
–Ph), 9.65 (d, J = 1.9 Hz, 1H, CHO); 13C NMR (CDCl3):
1
o = index for major diastereomer 13a (59%; H NMR
integral for 2o-H); x = index for minor diastereomer
1
13b (41%; H NMR integral for 2x-H).

5960
M. Sax et al. / Bioorg. Med. Chem. 14 (2006) 5955–5962
6.6. ( )-(2RS)-1-Benzyl-2-[(4RS)-2,2-diphenyl-1,3-dioxo-
lan-4-yl]piperidin- 4-one (14a) and ( )-(2RS)-1-benzyl-2-
[(4SR)-2,2-diphenyl-1,3-dioxolan-4-yl]piperidin-4-one
(14b)
(PhCH2, 100); Anal. Calcd for C27H27NO3: C, 78.42;
H, 6.58; N, 3.39. Found: C, 78.14; H, 6.52; N, 3.30.
6.7. ( )-(2RS)-2-[(4RS)-2,2-Diphenyl-1,3-dioxolan-4-
yl]piperidin-4-one (15a) (4-oxo-a-dioxadrol,12 15a)
Under N213 (340 mg, 0.83 mmol) was dissolved in THF
(8.3 mL) and the solution was cooled down to À78 ꢁC.
Then BF3 Æ OEt2 (115 lL, 0.91 mmol) was added drop-
wise and it was stirred for 30 min at À78 ꢁC. Afterwards,
a solution of 1 M LiEt3BH in THF (0.91 mL, 0.91 mmol)
was added very slowly and the solution was stirred for
another 60 min at À78 ꢁC. After addition of a saturated
solution of NaHCO3 (200 lL) and EtOAc (20 mL), the
reaction mixture was warmed to room temperature. The
solution was washed with a saturated solution of NaH-
CO3 and brine, the organic layer was dried (K2CO3), fil-
tered, and concentrated in vacuo, and the residue was
purified by flash chromatography (petroleum ether/
EtOAc = 8:2) to afford 125 mg (37%) of 14b (Rf = 0.22)
as a colorless oil and 187 mg (55%) of 14a (Rf = 0.16) as
a colorless oil.
The N-benzylpiperidone 14a (170 mg, 0.41 mmol) was
dissolved in MeOH (6 mL), Pd/C 10% (70 mg) was add-
ed, and the suspension was vigorously stirred under H2-
atmosphere (balloon) at room temperature for 4 h. The
reaction mixture was filtered through Celite, which was
rinsed with MeOH, and the filtrate was evaporated to
dryness. Flash chromatographic purification (EtOAc/
MeOH/NH3cc = 94:5:1) of the oily residue yielded
104 mg (78%) of 15a as a colorless solid, mp = 116.6–
117.6 ꢁC; Rf = 0.44 (EtOAc/MeOH/NH3cc = 94:5:1); IR
(neat): 3329 (w)/3305 (w, –N–H), 1712 (s, C@O), 1449
(s, –C–H deformation), 1204 (s)/1065 (s, –C–O–C–);
1H NMR (CDCl3): d 2.17 (ddd, J = 14.1/11.8/0.8 Hz,
1H, 3-H), 2.31–2.36 (m, 1H, 5-H), 2.40 (dddd, J = 14.4/
11.6/6.5/0.9 Hz, 1H, 5-H), 2.47 (ddd, 1H, J = 14.1/3.1/
1.9 Hz, 3-H), 2.86 (ddd, J = 12.4/11.6/4.1 Hz, 1H, 6-H),
3.10 (ddd, J = 11.8/4.1/3.1 Hz, 1H, 2-H), 3.38 (ddd,
J = 12.5/6.5/2.5 Hz, 1H, 6-H), 3.98 (dd, J = 7.8/7.0 Hz,
1H, OCH2), 4.13 (dd, J = 7.8/5.9 Hz, 1H, OCH2), 4.19
(ddd, J = 7.1/5.8/4.2 Hz, 1H, OCH), 7.26–7.37 (m, 6H,
–Ph), 7.47–7.54 (m, 4H, –Ph), the NH-signal was not vis-
ible; 13C NMR (CDCl3): d 43.1 (C-5), 44.8 (C-3), 45.6 (C-
6), 59.1 (C-2), 66.2 (OCH2), 78.8 (OCH), 110.4 (OCO),
126.2 (2 Ph-C), 126.4 (2 Ph-C), 128.41 (Ph-C), 128.42
(2 Ph-C), 128.5 (3 Ph-C), 141.8/142.0 (2 quart benzophe-
none-C), 208.7 (C-4); MS (70 eV) m/e (rel int): 182
(Ph2CO, 57), 105 (PhCO, 68), 77 (–Ph, 100); Anal. Calcd
for C20H21NO3: C, 74.28; H, 6.55; N, 4.33. Found: C,
74.01; H, 6.49; N, 4.36.
Compound 14a: IR (neat): 1710 (s, C@O), 1450 (m,
–C–H deformation), 1204 (s)/1069 (s, –C–O–C–); 1H
NMR (CDCl3): d 2.31 (dtd, J = 15.1/4.5/1.4 Hz, 1H,
5-H), 2.50 (dddd, J = 15.2/9.3/5.9/0.7 Hz, 1H, 5-H),
2.64 (ddd, J = 14.9/5.8/0.9 Hz, 1H, 3-H), 2.70 (ddd,
J = 14.9/4.3/1.5 Hz, 1H, 3-H), 2.95 (dddd, J = 13.5/
5.9/4.8/1.0 Hz, 1 H, 6-H), 3.12–3.20 (m, 2H, 2-H + 6-
H), 3.87 (s, 2H, PhCH2), 3.88 (dd, J = 8.2/6.7 Hz,
1H, OCH2), 4.08 (dd, J = 8.2/7.0 Hz, 1H, OCH2),
4.29 (‘q,’ J = 6.8 Hz, 1H, OCH), 7.22–7.35 (m, 11H,
–Ph), 7.38–7.41 (m, 2H, –Ph), 7.48–7.51 (m, 2H,
–Ph); 13C NMR (CDCl3): d 38.1 (C-5), 39.6 (C-3),
47.8 (C-6), 57.1 (PhCH2), 62.4 (C-2), 68.6 (OCH2),
76.4 (OCH), 110.7 (OCO), 126.2 (2 Ph-C), 126.3 (2
Ph-C), 127.7 (Ph-C), 128.25 (2 Ph-C), 128.28 (Ph-C),
128.4 (Ph-C), 128.5 (2 Ph-C), 128.77 (2 Ph-C), 128.79
(2 Ph-C), 138.7 (quart Bn-C), 142.0/142.4 (2 quart benzo-
phenone-C), 208.7 (C-4); MS (70 eV) m/e (rel int): 413 (M,
0.37), 336 (MÀPh, 1.5), 188 (M-diphenyldioxolanyl = N-
benzylpiperidinone, 47), 91 (PhCH2, 100); Anal. Calcd
for C27H27NO3: C, 78.42; H, 6.58; N, 3.39. Found: C,
77.94; H, 6.13; N, 3.11.
6.8. ( )-(2RS)-2-[(4SR)-2,2-Diphenyl-1,3-dioxolan-4-
yl]piperidin-4-one (4-oxo-b-dioxadrol,12 15b)
The N-benzylpiperidone 14b (106 mg, 0.26 mmol) was
dissolved in MeOH (5 mL), Pd/C 10% (43 mg) was add-
ed, and the suspension was vigorously stirred under H2-
atmosphere (balloon) at room temperature for 2 h. The
reaction mixture was filtered through Celite, which was
rinsed with MeOH, and the filtrate was evaporated to
dryness. Flash chromatographic purification (EtOAc/
MeOH/NH3cc = 94:5:1) of the oily residue yielded
71 mg (86%) of 15b as a colorless solid. Recrystallization
for X-ray crystal structure analysis was performed with
Et2O, colorless solid, mp = 96.7–97.4 ꢁC; Rf = 0.50
(EtOAc/MeOH/NH3cc = 94:5:1); IR (neat): 3337 (w,
–N–H), 1712 (s, C@O), 1205 (m)/1068 (m, –C–O–C–);
1H NMR (CDCl3): d 2.13 (br s, 1H, –NH), 2.25 (d,
J = 7.8 Hz, 2H, 3-H), 2.35 (br d, J = 14.5 Hz, 1H, 5-
H), 2.44 (ddd, J = 14.3/11.9/6.6 Hz, 1H, 5-H), 2.85 (td,
J = 12.0/3.7 Hz, 1H, 6-H), 2.93 (‘q,’ J = 7.2 Hz, 1H, 2-
H), 3.40 (ddd, J = 12.2/6.6/2.3 Hz, 1H, 6-H), 3.93 (dd,
J = 8.2/6.3 Hz, 1H, OCH2), 4.03 (dd, J = 7.8/7.0 Hz,
1H, OCH2), 4.11 (‘q,’ J = 6.7 Hz, 1H, OCH), 7.26–7.36
(m, 6H, –Ph), 7.46–7.53 (m, 4H, –Ph); 13C NMR
(CDCl3): d 42.7 (C-5), 45.26 (C-3), 45.36 (C-6), 60.0
(C-2), 67.0 (OCH2), 79.3 (OCH), 110.4 (OCO), 126.3
(2 Ph-C), 126.4 (2 Ph-C), 128.43 (2 Ph-C), 128.48
Compound 14b: IR (neat): 1714 (s, C@O), 1450 (m, –C–
H deformation), 1205 (s)/1070 (s, –C–O–C–); H NMR
1
(CDCl3): d 2.21 (ddd, J = 14.7/5.1/1.6 Hz, 1H, 3-H), 2.31
(dddd, J = 15.0/5.9/4.6/1.4 Hz, 1H, 5-H), 2.40–2.48 (m,
1H, 5-H), 2.66 (ddd, J = 14.7/5.8/1.2 Hz, 1H, 3-H),
2.82 (dt, J = 12.8/6.1 Hz, 1H, 6-H), 3.16 (ddd,
J = 12.8/8.1/4.7 Hz, 1H, 6-H), 3.31 (dt, J = 7.5/5.4 Hz,
1H, 2-H), 3.91 (d, J = 14.2 Hz, 1H, PhCH2), 3.93 (t,
J = 7.6 Hz, 1H, OCH2), 4.04 (dd, J = 7.8/6.7 Hz, 1H,
OCH2), 4.11 (d, J = 13.8 Hz, 1H, PhCH2), 4.36 (‘q,’
J = 7.2 Hz, 1H, OCH), 7.25–7.39 (m, 11H, –Ph), 7.46–
7.53 (m, 4H, –Ph); 13C NMR (CDCl3): d 37.8 (C-5),
40.1 (C-3), 45.7 (C-6), 56.6 (PhCH2), 59.7 (C-2), 65.1
(OCH2), 74.7 (OCH), 108.3 (OCO), 124.0/124.1/125.4/
126.2/126.4/126.5/126.6 (15 Ph-C), 137.3 (quart Bn-C),
140.1/140.3 (2 quart benzophenone-C), 206.9 (C-4);
MS (70 eV) m/e (rel int): 336 (MÀPh, 1.2), 188
(M-diphenyldioxolanyl = N-benzylpiperidinone, 55), 91

M. Sax et al. / Bioorg. Med. Chem. 14 (2006) 5955–5962
5961
(Ph-C), 128.52 (2 Ph-C), 128.6 (Ph-C), 141.9/142.1 (2
7.2. Preparation of the tissue
quart benzophenone-C), 208.2 (C-4); MS (70 eV) m/e
(rel int): 246 (MÀPh, 3.4), 167 (PhCPh, 28), 105 (PhCO,
28), 98 (M-diphenyldioxolane = piperidinone, 100);
Anal. Calcd for C20H21NO3: C, 74.28; H, 6.55; N,
4.33. Found: C, 73.93; H, 6.45; N, 4.24.
Modified according to Refs. 13–15: Fresh pig brain cor-
tex was homogenized with a potter (500–800 rpm, 10 up-
and-down strokes) in 6 volumes of cold 0.32 M sucrose.
The suspension was centrifuged at 1200g for 10 min at
4 ꢁC. The supernatant was separated, and centrifuged
at 23,500g for 20 min at 4 ꢁC. The pellet was resuspend-
ed in buffer (5 mM Tris–acetate with 1 mM EDTA, pH
7.5) and centrifuged again at 23,500g (20 min, 4 ꢁC).
This procedure was repeated twice. The final pellet
was resuspended in buffer, the protein concentration
was determined according to the method of Bradford29
using bovine serum albumin as standard, and subse-
quently the preparation was frozen (À83 ꢁC) in 1.5 mL
portions containing about 0.8 mg protein/mL.
6.9. X-ray crystal structure analysis of 15b
Formula C20H21NO3, M = 323.38, colorless crystal
˚
˚
0.25 · 0.20 · 0.15 mm, a = 6.032(1) A, b = 34.556(1) A,
3
˚
˚
c = 8.136(1) A, b = 101.23(1) ꢁ, V = 1636.4(3) A ,
q
calcd = 1.291 g cmÀ3, l = 0.87 cmÀ1, empirical absorp-
tion correction (0.979 6 T 6 0.987), Z = 4, monoclinic,
˚
space group P21/n (no. 14), k = 0.71073 A, T = 198(2)
K, x and / scans, 5990 reflections collected ( h,
À1
˚
k,
l), [(sinh)/kmax] = 0.66 A , 3698 independent
(Rint = 0.023) and 2693 observed reflections [I P 2
r(I)], 220 refined parameters, R = 0.046, wR2 · 0.125,
7.3. Performance of the assay
max residual electron density 0.27 (À0.20) e AÀ3, hydro-
Modified according to Refs. 13–15: The test was per-
formed with the radioligand [3H]-(+)-MK-801
(9.25 MBq; Biotrend). The thawed membrane prepara-
tion (about 100 lg of the protein) was incubated with
various concentrations of test compounds, 2 nM [3H]-
(+)-MK-801, and buffer (5 mM Tris–acetate, 1 mM
EDTA, pH 7.5) in a total volume of 200 lL for
120 min at 25 ꢁC. The incubation was terminated by ra-
pid filtration through the presoaked filtermats using a
cell harvester. After washing five times with 300 ll of
cold water, the filtermats were dried at 95 ꢁC. Subse-
quently, the solid scintillator was placed on the filtermat
and melted at 95 ꢁC. After 5 min, the solid scintillator
was allowed to solidify at room temperature. The bound
radioactivity trapped on the filters was counted in a scin-
tillation analyzer. Non-specific binding was determined
with 10 lM (+)-MK-801.
˚
gens calculated and refined as riding atoms. The data set
was collected with a Nonius KappaCCD diffractometer.
Programs used: data collection COLLECT,23 data
reduction Denzo-SMN,24 absorption correction Den-
zo,25 structure solution SHELXS-97,26 structure refine-
ment SHELXL-97,27 graphics SCHAKAL.28
CCDC-296453 contains the supplementary crystallo-
graphic data for this paper. These data can be obtained
free of charge at www.ccdc.cam.ac.uk/conts/retriev-
ing.html [or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax:
(internat.) +44(1223)336-033, E-mail: deposit@ccdc.
cam.ac.uk].
7. Receptor binding studies
7.1. General information
Acknowledgments
Homogenizer: Elvehjem Potter (B. Braun Biotech Inter-
national). Centrifuge: High-speed cooling centrifuge
model Sorvall RC-5C plus (Thermo Finnigan). Filter:
Printed Filtermat Type A (Perkin Elmer), presoaked in
0.5% aqueous polyethylenimine for 2 h at room temper-
ature before use. The filtration was carried out with a
MicroBeta FilterMate-96 Harvester (Perkin Elmer).
The scintillation analysis was performed using Meltilex
(Type A) solid scintillator (Perkin Elmer). The solid
scintillator was melted on the filtermat at a temperature
of 95 ꢁC for 5 min. After solidification of the scintillator
at room temperature, the scintillation was measured
We thank Dr. Jo¨rg Fabian for determination of the
logP-value. Thanks are also due to Degussa A. G., Jans-
sen-Cilag GmbH, and Bristol-Myers Squibb for dona-
tion of chemicals and reference compounds. The
performance of the behavioral studies with 15a by
Schwarz Pharma, Monheim, is gratefully acknowledged.
References and notes
1. Lipton, S. A. Nat. Rev. Drug Disc. 2006, 5, 160–170.
2. Stark, H.; Reichert, U.; Graßmann, S. Pharmazie in
unserer Zeit 2000, 4, 228.
using
a
MicroBeta Trilux scintillation analyzer
3. (a) Hardie, W. R.; Hidalgo, J.; Halverstadt, I. F.; Allen, R.
E. J. Med. Chem. 1966, 9, 127; (b) Hardie, W. R.; Aaron,
J. E. GB Patent 1 199 110, 1968; (c) Ornstein, P. L.;
Zimmermann, D. M.; Leander, J. D.; Mendelson, L.;
Reel, J. K.; Evrard In Sigma and Phencyclidine-like
Compounds as Molecular Probes in Biology; Kamenka,
J.-M., Domino, E. F., Eds.; NPP Books: Ann Arbor, MI,
1988; pp 19–25; (d) Thurkauf, A.; Mattson, M. V.;
Richardson, S.; Mirsadeghi, S.; Ornstein, P. L.; Harrison,
E. A., Jr.; Rice, K. C.; Jacobson, A. E.; Monn, J. A.
(Perkin-Elmer). The counting efficiency was 40%. All
experiments were carried out in triplicate using standard
96-well-multiplates (Diagonal). The IC50-values were
determined in competition experiments with six concen-
trations of the test compounds and were calculated with
the program GraphPad Prismꢂ 3.0 (GraphPad Soft-
ware) by non-linear regression analysis. The Ki-values
were calculated according to Cheng and Prusoff.16 The
Ki-values are given as mean values SEM from three
independent assays.
J. Med. Chem. 1992, 35, 1323; (e) Aepkers, M.; Wunsch,
¨

5962
M. Sax et al. / Bioorg. Med. Chem. 14 (2006) 5955–5962
B. Arch. Pharm. Pharm. Med. Chem. 2004, 337, 67; (f)
Utech, T. PhD thesis, University of Freiburg, Germany,
2003.
14. Ho¨fner, G.; Wanner, K. T. J. Rec. Sign. Trans. Res. 1996,
16, 297–313.
15. Aepkers, M.; Wunsch, B. Arch. Pharm. Pharm. Med.
¨
Chem. 2004, 337, 67–75.
4. Jacobson, A. E.; Harrison, E. A.; Mattson, M. V.;
Rafferty, M. F.; Rice, K. C.; Woods, J. H.; Winger, G.;
Solomon, R. E.; Lessor, R. A.; Silverton, J. V. J.
Pharmacol. Exp. Ther. 1987, 243, 110.
5. For reviews concerning hetero Diels–Alder reactions, see:
(a) Jørgensen, K. A. Angew. Chem. 2000, 112, 3702; (b)
Buonora, P.; Olsen, J.-C.; Oh, T. Tetrahedron 2001, 57,
6099.
6. Danishefsky, S. Acc. Chem. Res. 1981, 14, 400.
7. (a) Brown, J. D.; Foley, M. A.; Comins, D. L. J. Am.
Chem. Soc. 1988, 110, 7445; (b) Heintzelmann, G. R.;
Weinreb, S. M. J. Org. Chem. 1996, 61, 4594; (c)
Weymann, M.; Pfrengle, W.; Schollmeyer, D.; Kunz, H.
Synthesis 1997, 1151; (d) Kranke, B.; Hebrault, D.;
Schultz-Kukula, M.; Kunz, H. Synlett 2004, 671; (e)
Shintani, R.; Tokunaga, N.; Doi, H.; Hayashi, T. J. Am.
Chem. Soc. 2004, 126, 6240.
16. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22,
3099–3108.
17. Thurkauf, A.; Zenk, P. C.; Balster, R. L.; May, E. L.;
George, C.; Carroll, F. I.; Mascarella, S. W.; Rice, K. C.;
Jacobson, A. E.; Mattson, M. V. J. Med. Chem. 1988, 31,
2257.
18. Lipinski, C. A. Adv. Drug Deliv. Rev. 1997, 23, 3.
19. OECD (Organization for Economic Co-operation and
Development) 1989. OECD Guideline for testing of
Chemicals. 117. Partition coefficient (n-octanol/water),
HPLC-Method. OECD. Paris, France. Download under:
http://www.oecd.org/dataoecd/17/36/1948177.pdf.
20. (a) Irwin, S. Psychopharmacologia 1968, 13, 222–257; (b)
Campbell, D. E. S.; Richter, W. Acta Pharmacol. Toxicol.
1967, 25, 345–365.
21. Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43,
2923–2925.
22. Miyazawa, T.; Kurita, S.; Sakamoto, H.; Otomatsu, T.;
Katsutoshi, H.; Yamada, T. Biotechnol. Lett. 1999, 21, 447–
450.
23. B.V. Nonius, Delft, The Netherlands, 1998.
24. Otwinowski, Z.; Minor, W. Methods in Enzymology 1997,
276, 307–326.
25. Otwinowski, Z.; Borek, D.; Majewski, W.; Minor, W.
Acta Crystallogr. 2003, A59, 228–234.
8. Manusco, J.; Swern, D. Synthesis 1981, 165–185.
9. Diene 6 decomposes rapidly under standard storing
conditions (4 ꢁC, N2) but is stable for at least 21 days
when stored under N2 at À25 ꢁC in the darkness (proved
1
by H NMR spectroscopy).
10. Hansson, L.; Carlson, R. Acta Chem. Scand. 1989, 43, 188.
11. Comins, D.; LaMunyon, D. H. Tetrahedron Lett. 1989,
30, 5053–5056.
12. The racemate consisting of dexoxadrol (2) and its enan-
tiomer levoxadrol has been termed a-dioxadrol, the
diastereomeric racemate b-dioxadrol.
26. Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467–473.
27. Sheldrick, G. M. Universita¨t Go¨ttingen, 1997.
28. Keller, E. Universita¨t Freiburg, 1997.
13. McKernan, R. M.; Castro, S.; Poat, J. A.; Wong, E. H. F.
J. Neurochem. 1989, 52, 777–785.
29. Bradford, M. M. Anal. Biochem. 1976, 72, 248–254.