Synthesis, crystal structure analysis, and pharmacological characterization of desmethoxy-sila-venlafaxine, a derivative of the serotonin/noradrenaline reuptake inhibitor sila-venlafaxine

Article abstract of DOI:10.1016/j.jorganchem.2006.04.015

rac-Desmethoxy-sila-venlafaxine (rac-3) is a derivative of the noradrenaline-selective serotonin/noradrenaline reuptake inhibitor rac-sila-venlafaxine (rac-1b), a silicon analogue of the serotonin-selective serotonin/noradrenaline reuptake inhibitor rac-venlafaxine (rac-1a) (rac-1a, rac-1-[2-(dimethylamino)-1-(4-methoxyphenyl)ethyl]cyclohexan-1-ol; rac-1b, rac-1-[2-(dimethylamino)-1-(4-methoxyphenyl)ethyl]-1-silacyclohexan-1-ol ). The synthesis and crystal structure analyses of rac-3 and rac-3 · HCl are reported, and the pharmacological properties of rac-1a, rac-1b, rac-2 (a sila-venlafaxine derivative with a silacyclopentanol skeleton instead of a silacyclohexanol framework), and rac-3 are compared (comparison of the pharmacological selectivity profiles with respect to serotonin, noradrenaline, and dopamine reuptake inhibition).

Full text of DOI:10.1016/j.jorganchem.2006.04.015

Journal of Organometallic Chemistry 691 (2006) 3589–3595
www.elsevier.com/locate/jorganchem
Synthesis, crystal structure analysis, and pharmacological
characterization of desmethoxy-sila-venlafaxine, a derivative
of the serotonin/noradrenaline reuptake inhibitor sila-venlafaxine
a
a
b
b
Jurgen O. Daiß , Christian Burschka , John S. Mills , John G. Montana ,
¨
b
b
a,
*
Graham A. Showell , Julie B.H. Warneck , Reinhold Tacke
a
Institut fu¨r Anorganische Chemie, Universita¨t Wu¨rzburg, Am Hubland, D-97074 Wu¨rzburg, Germany
Paradigm Therapeutics Ltd., 162 Cambridge Science Park, Milton Road, Cambridge CB4 0GP, UK
b
Received 14 December 2005; received in revised form 7 April 2006; accepted 18 April 2006
Available online 26 April 2006
Dedicated to Professor Siegfried Hunig on the occasion of his 85th birthday.
¨
Abstract
rac-Desmethoxy-sila-venlafaxine (rac-3) is a derivative of the noradrenaline-selective serotonin/noradrenaline reuptake inhibitor rac-
sila-venlafaxine (rac-1b), a silicon analogue of the serotonin-selective serotonin/noradrenaline reuptake inhibitor rac-venlafaxine (rac-1a)
(rac-1a, rac-1-[2-(dimethylamino)-1-(4-methoxyphenyl)ethyl]cyclohexan-1-ol; rac-1b, rac-1-[2-(dimethylamino)-1-(4-methoxyphenyl)ethyl]-
1-silacyclohexan-1-ol). The synthesis and crystal structure analyses of rac-3 and rac-3 Æ HCl are reported, and the pharmacological prop-
erties of rac-1a, rac-1b, rac-2 (a sila-venlafaxine derivative with a silacyclopentanol skeleton instead of a silacyclohexanol framework),
and rac-3 are compared (comparison of the pharmacological selectivity profiles with respect to serotonin, noradrenaline, and dopamine
reuptake inhibition).
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Sila-venlafaxine; Silicon; Silicon-based drugs; Serotonin/noradrenaline reuptake inhibitor; Structure–activity relationships; Venlafaxine
1. Introduction
of rac-3 and its hydrochloride rac-3 Æ HCl, and we compare
the pharmacological properties of rac-1a, rac-1b, rac-2, and
rac-3. Preliminary results of these studies have been
reported elsewhere [6]. For a recent review on silicon-based
drugs, see Ref. [7].
In a series of recent publications, we have reported on
the synthesis and pharmacological characterization of
rac-sila-venlafaxine (rac-1b) [1–3], a sila-analogue of the
serotonin/noradrenaline reuptake inhibitor rac-venlafaxine
(rac-1a) [4,5]. In context with structure–activity relation-
ship (SAR) studies, the enantiomers (R)-1b and (S)-1b
and the derivative rac-2 have also been investigated [3].
In continuation of these SAR studies, we have now suc-
ceeded in synthesizing and characterizing the sila-venlafax-
ine derivative rac-desmethoxy-sila-venlafaxine (rac-3). We
report here on the syntheses and crystal structure analyses
NMe2
NMe2
HO
HO
El
Si
OMe
OMe
El = C: 1a
El = Si: 1b
2
NMe2
HO
Si
*
Corresponding author. Tel.: +49 931 888 5250; fax: +49 931 888 4609.
E-mail address: r.tacke@mail.uni-wuerzburg.de (R. Tacke).
3
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.04.015

3590
J.O. Daiß et al. / Journal of Organometallic Chemistry 691 (2006) 3589–3595
Racemic venlafaxine hydrochloride (Effexore, Wyeth-
yield), which upon treatment with lithium aluminum
hydride afforded 1-(1-phenylvinyl)-1-silacyclohexane (6)
(83% yield). Compound 6 was then reacted with dimethyl-
amine, in the presence of lithium dimethylamide, to
give rac-1-(dimethylamino)-1-(2-(dimethylamino)-1-phen-
ylethyl)-1-silacyclohexane (rac-7) (40% yield). Hydrolysis
of rac-7 yielded rac-3 (91% yield), which upon treatment
with an ethereal hydrogen chloride solution gave the corre-
sponding hydrochloride rac-3 Æ HCl (93% yield).
In the course of the synthesis of rac-7, the formation of
1,1-bis(dimethylamino)-1-silacyclohexane (8) and dimethyl-
(2-phenylethyl)amine (9) was observed (comparison with
authentic samples, GC–MS analysis of the reaction mix-
ture), resulting from an Si–C bond cleavage induced by a
nucleophilic attack of LiNMe2 at the silicon atom. This
cleavage reaction is mainly responsible for the poor yield
of rac-7.
Ayerst; Efexore, Wyeth, Wyeth-Lederle; Trevilore, Wyeth)
is in clinical use as an antidepressant [8]. Venlafaxine is a
monoamine reuptake inhibitor, with an activity profile dif-
ferent from traditional tricyclic antidepressants. It potently
inhibits serotonin and noradrenaline synaptic transport,
but exhibits weak potency at dopamine transporters as well
as a range of other receptors (e.g. histaminic, muscarinic,
adrenergic) responsible for the adverse effects of traditional
antidepressants [9]. Venlafaxine is rapidly absorbed
but undergoes extensive metabolism to active metabolites
[10].
In context with our systematic studies on sila-substituted
drugs [7], we were interested in the pharmacological proper-
ties of racemic sila-venlafaxine (rac-1b) and its derivatives
rac-2 and rac-3. Sila-substitution of rac-1a (!rac-1b) was
expected to affect the chemical and physicochemical proper-
ties and the structure of venlafaxine and therefore to alter
its biological properties, potentially resulting in improve-
ments on the venlafaxine safety profile. Structural varia-
tions of rac-1b (!rac-2, rac-3) could lead to further
alterations of the biological properties.
NMe2
Me2N
NMe2
Si
8
9
2. Results and discussion
Compounds rac-3 and rac-3 Æ HCl were isolated as col-
orless crystalline solids, whereas 5, 6, rac-7, and 8 were
obtained as colorless liquids. The identities of all these
compounds were established by elemental analyses
(C,H,N) and NMR studies (1H,13C,29Si), and rac-3 and
rac-3 Æ HCl were additionally characterized by crystal struc-
ture analyses.
2.1. Syntheses
The title compound rac-3 and its hydrochloride rac-
3 Æ HCl were prepared in multistep syntheses, starting from
1,1-dimethoxy-1-silacyclohexane (4) (Scheme 1). Thus,
reaction of 4 with (1-phenylvinyl)magnesium bromide gave
1-methoxy-1-(1-phenylvinyl)-1-silacyclohexane (5) (59%
2.2. Crystal structure analyses
CH2
Compounds rac-3 and rac-3 Æ HCl were structurally
characterized by single-crystal X-ray diffraction. The crys-
tal data and experimental parameters used for these studies
are given in Table 1. The molecular structure of rac-3 and
the structure of the cation of rac-3 Æ HCl are depicted in
Figs. 1 and 2; selected bond distances and angles are given
in the respective figure legends.
Compounds rac-3 and rac-3 Æ HCl crystallize in the
space group P21/n. The asymmetric unit of rac-3 contains
two molecules (A and B), with very similar structures.
The crystal structures of rac-3 and rac-3 Æ HCl are governed
by hydrogen bonds [11]. Compound rac-3 forms intermo-
lecular O–Hꢀ ꢀ ꢀN hydrogen bonds that lead to infinite
chains of the molecules along [100]. These chains are built
up by molecules A and B in an alternating manner
(ꢀ ꢀ ꢀAꢀ ꢀ ꢀBꢀ ꢀ ꢀAꢀ ꢀ ꢀBꢀ ꢀ ꢀ), the absolute configurations of A
and B in a given chain being opposite. Compound rac-
3 Æ HCl forms O–Hꢀ ꢀ ꢀCl and N–Hꢀ ꢀ ꢀCl hydrogen bonds,
leading to the formation of infinite chains along [100].
These chains are built up by the ammonium cations and
chloride anions, the absolute configurations of all the cat-
ions in a given chain being identical.
BrMg
CH2
MeO
OMe
MeO
Si
Si
4
5
LiAlH4
CH2
HNMe2
[LiNMe2]
NMe2
Me2N
H
Si
Si
rac-7
6
+
H2O [H3O ]
NMe2
NMe2 HCl
HCl
HO
HO
Si
Si
rac-3
rac-3 HCl
Scheme 1.

J.O. Daiß et al. / Journal of Organometallic Chemistry 691 (2006) 3589–3595
3591
Table 1
Crystal data and experimental parameters for the crystal structure
analyses of rac-3 and rac-3 Æ HCl
rac-3
rac-3 Æ HCl
Empirical formula
Formula mass (g molꢁ1
Collection T (K)
C15H25NOSi
263.45
173(2)
C15H26ClNOSi
299.91
173(2)
)
˚
k (Mo Ka) (A)
0.71073
0.71073
Crystal system
Space group (no.)
Monoclinic
P21/n (14)
11.9830(10)
11.1445(10)
23.160(2)
99.162(11)
3053.5(5)
8
Monoclinic
P21/n (14)
6.0012(5)
26.724(3)
10.7674(9)
103.940(9)
1676.0(3)
4
˚
a (A)
˚
b (A)
˚
c (A)
b (ꢁ)
3
˚
V (A )
Z
Dcalc (g cmꢁ3
)
1.146
0.144
1152
1.189
0.293
648
l (mmꢁ1
F(000)
)
Crystal dimensions (mm)
2h Range (ꢁ)
0.5 · 0.5 · 0.3
5.02–54.16
ꢁ15 6 h 6 15,
ꢁ14 6 k 6 14,
ꢁ29 6 l 6 19
17801
0.5 · 0.4 · 0.2
4.94–53.92
ꢁ7 6 h 6 7,
ꢁ34 6 k 6 34,
ꢁ13 6 l 6 13
14764
Fig. 1. Structure of rac-3 in the crystal. The unit cell contains two
molecules (A and B), with very similar structures (only one enantiomer of
Index ranges
˚
molecule A depicted). Selected bond distances (A) and angles (ꢁ) for
molecule A: Si1–O1 1.6200(11), Si1–C1 1.8655(17), Si1–C5 1.8608(16),
Si1–C6 1.8931(15), C1–C2 1.529(2), C2–C3 1.524(2), C3–C4 1.519(2), C4–
C5 1.533(2), O1–Si1–C1 113.78(7), O1–Si1–C5 115.47(7), O1–Si1–C6
105.45(6), C1–Si1–C5 103.76(7), C1–Si1–C6 109.26(7), C5–Si1–C6
109.03(7), Si1–C1–C2 110.99(11), C1–C2–C3 113.03(15), C2–C3–C4
114.65(15), C3–C4–C5 113.42(15), C4–C5–Si1 110.50(11). Data for
molecule B (the atoms are labelled by adding ‘‘20’’ to the label number
of the corresponding atoms in molecule A): Si21–O21 1.6259(12), Si21–
C21 1.8609(16), Si21–C25 1.8560(18), Si21–C26 1.8988(15), C21–C22
1.528(2), C22–C23 1.518(3), C23–C24 1.515(3), C24–C25 1.536(3), O21–
Si21–C21 111.95(7), O21–Si21–C25 113.66(8), O21–Si21–C26 108.25(7),
C21–Si21–C25 103.91(8), C21–Si21–C26 109.09(7), C25–Si21–C26
109.87(7), Si21–C21–C22 112.40(12), C21–C22–C23 113.44(15), C22–
C23–C24 114.22(16), C23–C24–C25 112.50(15), C24–C25–Si21
111.13(13).
Number of collected reflections
Number of independent
reflections
6592
3483
Rint
0.0441
6592
331
1.016
0.0572/0.4079
0.0391
0.1063
+0.272/ꢁ0.315
0.0515
3483
178
1.040
0.0551/0.4269
0.0355
0.0972
+0.307/ꢁ0.361
Number of reflections used
Number of parameters
Sa
Weight parameters a/bb
c
R1 [I > 2r(I)]
d
wR2 (all data)
Maximum/minimum residual
ꢁ3
˚
electron density (e A
)
P
2
S ¼ f ½wðF 2o ꢁ F c2Þ ꢂ=ðn ꢁ pÞg0:5; n is the number of reflections; p is
a
the number of parameters.
2
b
wꢁ1 ¼ r2ðF 2oÞ þ ðaPÞ þ bP; with P ¼ ½maxðF o2; 0Þ þ 2F 2c ꢂ=3.
P
P
c
R1 = iFo|ꢁ|Fci/ |Fo|.
P
P
2
2
0:5
d
The silicon compound rac-sila-venlafaxine (rac-1b)
exhibits a substantially altered monoamine reuptake inhibi-
tion profile when compared with its carbon analogue rac-
venlafaxine (rac-1a). Activity at the noradrenaline and
dopamine transporters is essentially unaffected by sila-
substitution (within experimental biological variation),
whereas the potency at serotonin transporters is reduced
by two orders of magnitude. In contrast, whilst the potency
of the sila-venlafaxine derivative rac-2 (a compound with a
silacyclopentanol skeleton instead of a silacyclohexanol
framework) at serotonin receptors has also been reduced
by two orders of magnitude when compared to the parent
drug rac-venlafaxine (rac-1a), the dopamine transporter
selectivity is increased, leading to a compound essentially
equipotent at all three monoamine transporters. The title
compound rac-desmethoxy-sila-venlafaxine (rac-3) exhibits
a significant reduction in activity at both serotonin and
noradrenaline reuptake transporters, resulting in a non-
selective compound with a weak monoamine reuptake
inhibitor activity profile.
wR2 ¼ f ½wðF 2o ꢁ F c2Þ ꢂ= ½wðF 2oÞ ꢂg
.
The silacyclohexane skeletons of rac-3 and rac-3 Æ HCl
adopt a chair conformation, with the OH groups in an
equatorial position. In the crystal structures of the closely
related compounds rac-1b Æ HCl, (R)-1b Æ HBr, and rac-2,
the OH groups occupy axial sites. The structures of
rac-3 and rac-3 Æ HCl are characterized by relatively
long Si1–C6 (rac-3, molecule A), Si21–C26 (rac-3, mole-
cule B), and Si–C6 (rac-3 Æ HCl) distances (1.8931(15)–
˚
1.9041(14) A), the reason for this elongation being unclear.
Similar Si–C bond elongations were also observed for rac-
1b Æ HCl, (R)-1b Æ HBr, and rac-2 [2]. All the other bond
distances and angles of rac-3 and rac-3 Æ HCl are in the
expected range and do not need further discussion.
2.3. Pharmacological studies
Compounds rac-1a, rac-1b, rac-2, and rac-3 were tested
as their hydrochlorides for their efficacy in serotonin, nor-
adrenaline, and dopamine reuptake inhibition assays
(Table 2). The data are shown in Table 3 and Fig. 3.
These data demonstrate the importance of the p-meth-
oxy moiety in maintaining serotonin and noradrenaline
reuptake inhibition, whereas sila-substitution effects

3592
J.O. Daiß et al. / Journal of Organometallic Chemistry 691 (2006) 3589–3595
Fig. 3. In vitro efficacy of compounds rac-1a, rac-1b, rac-2, and rac-3
regarding serotonin, noradrenaline, and dopamine reuptake inhibition.
pIC50 denotes the negative decadic logarithm of the half-maximum effect
concentration (M). The monoamine reuptake inhibition profiles of rac-1a,
rac-1b, rac-2, and rac-3 were generated via radioligand transporter assays
using recombinant human monoamine transporter proteins. The data
represent the mean of at least 2 determinations.
Fig. 2. Structure of the cation of rac-3 Æ HCl in the crystal (only one
˚
enantiomer depicted). Selected bond distances (A) and angles (ꢁ): Si–O
1.6348(12), Si–C1 1.8602(15), Si–C5 1.8481(15), Si–C6 1.9041(14), C1–C2
1.537(2), C2–C3 1.524(2), C3–C4 1.531(2), C4–C5 1.526(2), O–Si–C1
113.03(7), O–Si–C5 110.04(7), O–Si–C6 108.25(6), C1–Si–C5 105.73(7),
C1–Si–C6 108.34(7), C5–Si–C6 111.47(7), Si–C1–C2 111.57(10), C1–C2–
C3 113.39(15), C2–C3–C4 114.23(13), C3–C4–C5 113.42(15), C4–C5–Si
110.93(11).
3. Experimental
3.1. Syntheses
3.1.1. General procedures
Except for the hydrolysis rac-7 ! rac-3, all syntheses
were carried out under dry nitrogen. The organic solvents
used were dried and purified according to standard proce-
dures and stored under dry nitrogen. A Buchi GKR 50
¨
apparatus was used for the bulb-to-bulb distillations. Melt-
changes in serotonin and dopamine reuptake activities
therefore identifying rac-sila-venlafaxine (rac-1b) as a drug
with a refined selectivity profile consistent with selective
noradrenaline reuptake inhibition. Compounds with this
pharmacological profile may provide therapeutic benefit
in the treatment of various CNS disorders.
ing points were determined with a Buchi Melting Point B-
¨
1
540 apparatus using open glass capillaries. The H, 13C,
Table 2
Experimental conditions for pharmacological assays
Assaya
SERT
NET
DAT
Cell line
Radioligand
Human HEK-293
3H-serotonin
Human MDCK
3H-noradrenaline
Human CHO-K1
3H-dopamine
Incubation time (min)
Incubation temperature (ꢁC)
Incubation bufferb
10
25
10
25
10
25
5 mM Tris–HCl, 7.5 mM HEPES, 120 mM NaCl, 5.4 mM KCl, 1.2 mM CaCl2, 1.2 mM MgSO4,
5 mM glucose, 1 mM ascorbic acid, pH 7.1
References
[18]
[19]
[20,21]
a
SERT, serotonin reuptake transporter; NET, noradrenaline (norephinephrine) reuptake transporter; DAT, dopamine reuptake transporter.
The incubation buffer was identical for all three assays.
b
Table 3
Monoamine reuptake transporter inhibition profilesa,b
Compound
SERT
NET
DAT
rac-1a
rac-1b
rac-2
rac-3
a
0.020 0.007
1.063 0.538
1.155 0.166
5.698 1.701
0.184 0.056
0.138 0.041
0.305 0.057
2.310 0.452
4.425 0.005
2.739 1.531
0.941 0.537
31.44 1.83
SERT, serotonin reuptake transporter; NET, noradrenaline (norephinephrine) reuptake transporter; DAT, dopamine reuptake transporter.
b
Data expressed as IC50 values (lM, mean SD) and represent the mean of at least 2 determinations.

J.O. Daiß et al. / Journal of Organometallic Chemistry 691 (2006) 3589–3595
3593
and 29Si NMR spectra were recorded on a Bruker DRX-
300 NMR spectrometer (1H, 300.1 MHz; 13C, 75.5 MHz;
29Si, 59.6 MHz). CD2Cl2, [D6]DMSO, or [D8]THF were
used as the solvent. All spectra were recorded at 22 ꢁC.
Chemical shifts were determined relative to internal
CHDCl2 (1H, d 5.32; CD2Cl2), CD2Cl2 (13C, d 53.8;
CD2Cl2), [D5]DMSO (1H, d 2.49; [D6]DMSO), [D6]DMSO
(13C, d 39.5; [D6]DMSO), [D7]THF (1H, d 1.73; [D8]THF),
[D8]THF (13C, d 25.3; [D8]THF), or external TMS (29Si, d
0; CD2Cl2, [D6]DMSO, [D8]THF). Analysis and assign-
7.03–7.09 (m, 2H, H-2/H-6, Aryl), 7.09–7.14 (m, 1H,
H-4, Aryl), 7.20–7.29 (m, 2H, H-3/H-5, Aryl). 13C NMR
(CD2Cl2):
d
12.5 (SiCH2C), 14.5 (SiCH2C), 24.5
(SiCH2CH2C), 24.6 (SiCH2CH2C), 29.9 (Si(CH2)2CH2C),
34.3 (SiCHC2), 45.5 (NCH3), 61.6 (NCH2C), 125.2 (C-4,
Aryl), 127.9 (C-2/C-6, Aryl), 128.6 (C-3/C-5, Aryl), 141.9
(C-1, Aryl). 29Si NMR (CD2Cl2): d 9.5. Anal. Calc. for
C15H25NOSi: C, 68.39; H, 9.56; N, 5.32. Found: C, 68.0;
H, 9.7; N, 5.5%.
ment of the H NMR data was supported by H,1H and
13C,1H correlation experiments and partially by simula-
tions using the WIN-DAISY software package (version 4.05,
Bruker) [12]. Assignment of the 13C NMR data was sup-
ported by DEPT 135 and 13C,1H correlation experiments.
3.1.3. rac-[2-(1-Hydroxy-1-sila-1-cyclohexyl)-2-
phenylethyl]dimethylammonium chloride (rac-desmethoxy-
sila-venlafaxine hydrochloride, rac-3 Æ HCl)
1
1
A 2 M ethereal hydrogen chloride solution (10.5 ml,
21.0 mmol of HCl) was added in one single portion at
20 ꢁC to a stirred solution of rac-3 (5.27 g, 20.0 mmol) in
dichloromethane (50 ml). The resulting solid was dissolved
in dichloromethane (150 ml) at reflux temperature, and the
solution was then kept undisturbed at 4 ꢁC for 1 day and at
ꢁ20 ꢁC for a further 3 days. The precipitate was isolated by
filtration at ꢁ20 ꢁC, washed with diethyl ether (20 ml), and
dried in vacuo (0.001 mbar, 20 ꢁC, 6 h) to give rac-3 Æ HCl
in 93% yield (including workup of the mother liquor) as a
colorless crystalline solid (5.60 g, 18.7 mmol); m.p. 186–
187 ꢁC (dec.). 1H NMR ([D6]DMSO) [13]: d 0.24–0.41,
0.51–0.71, and 1.12–1.59 (m, 10H, Si(CH2)5), 2.56 (dM),
2.61 (dN), 2.83 (dC), 3.42 (dA), 3.81 (dB), and 9.8 (br, dG)
2
The JHH coupling constants reported for the C@CH2
groups represent absolute values.
3.1.2. rac-1-(2-(Dimethylamino)-1-phenylethyl)-1-
silacyclohexan-1-ol (rac-desmethoxy-sila-venlafaxine,
rac-3)
A solution of rac-7 (7.88 g, 27.1 mmol) in diethyl ether
(50 ml) was added in one single portion at 0 ꢁC to a stirred
two-phase mixture of diethyl ether (50 ml) and 2 M potas-
sium acetate/acetic acid buffer (pH 4.5, 150 ml). The pH of
the aqueous phase changed to pH ca. 7 within 10 min and
was readjusted to pH 5.0 by addition of small portions of
glacial acetic acid. The mixture was stirred at 0 ꢁC for a fur-
ther 1 h, with the pH of the aqueous phase remaining con-
stantly at pH 5.0 during this time. The aqueous layer was
separated, the organic phase was extracted with 1 M potas-
sium acetate/acetic acid buffer (pH 5.0) (3 · 100 ml), and
the aqueous solutions were combined. Diethyl ether
(50 ml) was added, and the pH of the aqueous phase was
adjusted to pH 10.5 by addition of small portions of a sat-
urated aqueous potassium carbonate solution. The organic
layer was separated, the aqueous phase was extracted with
diethyl ether (5 · 100 ml), and the organic extracts were
combined, followed by addition of n-hexane (200 ml).
The solvent was removed under reduced pressure in a water
bath (5–15 ꢁC) until a residual volume of 150 ml was
obtained, whereupon residual water separated from the
organic phase (formation of a two-phase system). The
organic layer was separated, the aqueous phase was
extracted with n-hexane (2 · 100 ml), and the organic solu-
tions were combined. The solvent was removed completely
under reduced pressure in a water bath (5–15 ꢁC) to give a
colorless oil. Crystallization of this oil from n-pentane
(120 ml) at 4 ꢁC over a period of 1 day and then at
ꢁ20 ꢁC for a further 6 days afforded rac-3 in 91% yield
as a colorless crystalline solid (6.50 g, 24.7 mmol) (isolated
by quick decantation of the cold solvent, followed by dry-
ing in vacuo (0.001 mbar, 20 ꢁC, 6 h)); m.p. 63 ꢁC. 1H
NMR (CD2Cl2): d 0.45–0.78 and 1.00–1.69 (m, 10H,
Si(CH2)5), 2.30 (s, 6H, NCH3), 2.54 (dC), 2.58 (dA),
2
3
3
(10H, JAB = ꢁ13.1 Hz, JAC = 2.5 Hz, JBC = 12.5 Hz,
3JGM and JGN not resolved, SiCHCCHAHBNHG
3
(C(HM)3)(C(HN)3)), 6.06 (s, 1H, SiOH), 7.11–7.19 (m,
1H, H-4, Aryl), 7.23–7.34 (m, 4H, H-2/H-6, H-3/H-5,
Aryl). 13C NMR ([D6]DMSO): d 12.3 (SiCH2C), 13.2
(SiCH2C), 23.6 (SiCH2CH2C), 23.7 (SiCH2CH2C), 29.1
(Si(CH2)2CH2C), 32.1 (SiCHC2), 41.6 (NCH3), 42.9
(NCH3), 57.3 (CCH2N), 125.4 (C-4, Aryl), 127.7 (C-2/C-6,
Aryl), 128.5 (C-3/C-5, Aryl), 139.1 (C-1, Aryl). 29Si
NMR ([D6]DMSO): d 2.8. Anal. Calc. for C15H26ClNOSi:
C, 60.07; H, 8.74; N, 4.67. Found: C, 60.0; H, 8.9; N, 4.8%.
3.1.4. 1,1-Dimethoxy-1-silacyclohexane (4)
This compound was synthesized according to Ref. [3].
3.1.5. 1-Methoxy-1-(1-phenylvinyl)-1-silacyclohexane (5)
A solution of 1-bromo-1-phenylethene (28.0 g, 153
mmol) in diethyl ether (140 ml) was added dropwise within
15 min to a suspension of magnesium turnings (4.10 g,
169 mmol) in diethyl ether (10 ml), followed by heating
under reflux for an additional 1 h. (The Grignard reaction
proceeded smoothly, but required gentle heating to get
started.) The resulting dark brown Grignard reagent [14]
was cooled to 20 ꢁC, separated from the excess magnesium
turnings by decantation, and then added dropwise at 20 ꢁC
within 10 min to a stirred solution of 4 (24.6 g, 153 mmol)
in diethyl ether (50 ml). The resulting mixture was heated
under reflux for 3 days (precipitation of magnesium salts)
and was then cooled to 20 ꢁC, followed by filtration. The
filter cake was washed with n-hexane (300 ml), the filtrate
2
3
and 3.16 (dB) (3H, JAB = ꢁ12.0 Hz, JAC = 5.3 Hz,
3JBC = 12.1 Hz, SiCHCCHAHBN), 5.1 (br s, 1H, SiOH),

3594
J.O. Daiß et al. / Journal of Organometallic Chemistry 691 (2006) 3589–3595
and the wash solution were combined, and the solution was
concentrated under reduced pressure at 5–15 ꢁC to a vol-
ume of 200 ml and then kept undisturbed at 20 ꢁC for
1 day (postprecipitation of magnesium salts). The precipi-
tate was separated by filtration, the filter cake was washed
with n-hexane (50 ml), the filtrate and the wash solution
were combined, the solvent was removed completely under
reduced pressure at 5–15 ꢁC, and the residue was distilled
in vacuo to give 5 in 59% yield as a colorless liquid
within 15 min to a stirred solution of dimethylamine
(17.3 g, 384 mmol) in tetrahydrofuran (100 ml). The result-
ing mixture was warmed to ꢁ12 ꢁC within 2 h and then
cooled to ꢁ40 ꢁC, followed by dropwise addition of a solu-
tion of 6 (14.9 g, 73.6 mmol) in tetrahydrofuran (20 ml)
within a period of 20 min (evolution of hydrogen; rise in
temperature from ꢁ40 ꢁC to ꢁ33 ꢁC; change of color from
colorless to scarlet red). The resulting solution was stirred
at ꢁ30 ꢁC for 3 h and then kept undisturbed at 4 ꢁC for
16 h. Subsequently, the solution was placed in an ice bath
and stirred again, followed by addition of chlorotrimeth-
ylsilane (16.0 g, 147 mmol) in one single portion (change
of color from scarlet red to colorless). The mixture was stir-
red at 0 ꢁC for 5 min, warmed to 20 ꢁC within 30 min, and
then stirred at 20 ꢁC for a further 30 min. The solvent was
removed completely under reduced pressure at 5–15 ꢁC,
followed by addition of n-hexane (70 ml). The mixture
was stirred at 20 ꢁC for 15 min, the resulting precipitate
was separated by filtration, and the filter cake was washed
with n-hexane (20 ml). The filtrate and the wash solution
were combined, the solvent was removed completely under
reduced pressure at 5–15 ꢁC, and the residue was distilled
1
(21.2 g, 91.2 mmol), b.p. 80–81 ꢁC/0.001 mbar. H NMR
(CD2Cl2): d 0.71–0.98 (m, 4H, SiCH2C), 1.31–1.57 (m,
2H, Si(CH2)2CH2C), 1.60–1.81 (m, 4H, SiCH2CH2C),
3.45 (s, 3H, OCH3), 5.73 (dA) and 6.05 (dB) (2H,
2JAB = 2.8 Hz, C@CHAHB), 7.21–7.28 (m, 1H, H-4, Aryl),
7.28–7.34 (m, 4H, H-2/H-6, H-3/H-5, Aryl). 13C NMR
(CD2Cl2): d 12.5 (SiCH2C), 24.6 (SiCH2CH2C), 30.2
(Si(CH2)2CH2C), 50.6 (OCH3), 126.99 (C-2/C-6, Aryl),
127.00 (C-4, Aryl), 128.6 (C-3/C-5, Aryl), 129.5 (C@CH2),
143.7 (C-1, Aryl), 148.9 (C@CH2). 29Si NMR (CD2Cl2): d
3.4. Anal. Calc. for C14H20OSi: C, 72.36; H, 8.67. Found:
C, 72.4; H, 8.8%.
3.1.6. 1-(1-Phenylvinyl)-1-silacyclohexane (6)
in vacuo in a Kugelrohr apparatus (first fraction,
A solution of 5 (20.8 g, 89.5 mmol) in diethyl ether
(40 ml) was added at 20 ꢁC within 10 min to a stirred sus-
pension of lithium aluminum hydride (1.70 g, 44.8 mmol)
in diethyl ether (145 ml). The mixture was heated under
reflux for 6.5 h and then added carefully at 0 ꢁC to a stirred
mixture of 4 M hydrochloric acid (165 ml) and diethyl
ether (80 ml). The organic phase was separated, the aque-
ous layer was extracted with diethyl ether (3 · 100 ml),
and the organic solutions were combined and dried over
anhydrous magnesium sulfate in an ice bath, followed by
an additional thorough dynamic drying using a chromato-
graphic column densely packed with anhydrous magne-
sium sulfate (column diameter, 3.5 cm; column length,
15 cm). The magnesium sulfate was finally washed with
diethyl ether (2 · 200 ml), the organic solutions were com-
bined, the solvent was removed under reduced pressure,
and the residue was distilled in vacuo (Vigreux column,
15 cm) to give 6 in 83% yield as a colorless liquid (15.1 g,
74.6 mmol); b.p. 60–61 ꢁC/0.001 mbar. 1H NMR
(CD2Cl2): d 0.69–0.85, 0.92–1.05, 1.25–1.41, 1.51–1.70,
and 1.79–1.94 (m, 10H, Si(CH2)5), 4.26–4.33 (dX), 5.69
690 ꢁC/0.002 mbar, 8.3 g (mainly consisting of 8 and 9
[15]); second fraction, 90–125 ꢁC/0.0005 mbar, 12.5 g
(crude product)). The crude product was redistilled in
vacuo (Vigreux column, 5 cm) to give rac-7 in 40% yield
as a colorless oily liquid (8.63 g, 29.7 mmol); b.p. 112–
114 ꢁC/0.02 mbar. 1H NMR ([D8]THF): d 0.36–0.71,
0.85–0.98, 1.12–1.30, and 1.35–1.79 (m, 10H, Si(CH2)5),
2.12 (s, 6H, CNCH3), 2.36–2.46 (m, 1H, SiCHC2), 2.42
(s, 6H, SiNCH3), 2.59–2.76 (m, 2H, CCH2N), 6.96–7.05
(m, 3H, H-2/H-6, H-4, Aryl), 7.11–7.20 (m, 2H, H-3/H-5,
Aryl). 13C NMR ([D8]THF): d 11.9 (SiCH2C), 12.7
(SiCH2C), 24.90 (SiCH2CH2C), 24.93 (SiCH2CH2C), 31.1
(Si(CH2)2CH2C), 37.4 (SiCHC2), 38.6 (SiNCH3), 45.7
(CNCH3), 61.2 (CCH2N), 124.9 (C-4, Aryl), 128.5 (C-3/
C-5, Aryl) 128.7 (C-2/C-6, Aryl), 143.9 (C-1, Aryl). 29Si
NMR ([D8]THF): d 0.8. Anal. Calc. for C17H30N2Si: C,
70.28; H, 10.41; N, 9.64. Found: C, 70.0; H, 10.3; N, 9.5%.
3.1.8. 1,1-Bis(dimethylamino)-1-silacyclohexane (8)
A 2.5 M solution of n-butyllithium in n-hexane (18.0 ml,
45.0 mmol of n-BuLi) was added dropwise at ꢁ50 ꢁC within
15 min to a stirred solution of dimethylamine (7.12 g,
158 mmol) in tetrahydrofuran (70 ml). The resulting mix-
ture was warmed to ꢁ15 ꢁC within 2 h, followed by drop-
wise addition of 1,1-dichloro-1-silacyclohexane (3.73 g,
22.1 mmol; synthesized according to Ref. [3]) at ꢁ15 ꢁC
within a period of 25 min. The resulting solution was
warmed to 20 ꢁC within 4 h and then stirred at 20 ꢁC for
a further 12 h. The solvent was removed completely under
reduced pressure at 5–15 ꢁC, followed by addition of n-
hexane (100 ml) (formation of a precipitate). The mixture
was stirred at 20 ꢁC for 1 day, the resulting precipitate
was separated by filtration, and the filter cake was washed
with n-hexane (20 ml). The filtrate and the wash solution
2
4
(dA), and 6.03 (dB) (3H, JAB = 2.6 Hz, JBX = 0.4 Hz,
HXSiC@CHAHB), 7.21–7.28 (m, 1H, H-4, Aryl), 7.28–
7.36 (m, 4H, H-2/H-6, H-3/H-5, Aryl). 13C NMR
(CD2Cl2): d 10.8 (SiCH2C), 25.2 (SiCH2CH2C), 30.1
(Si(CH2)2CH2C), 126.8 (C-2/C-6, Aryl), 127.1 (C-4, Aryl),
128.5 (C@CH2), 128.7 (C-3/C-5, Aryl), 144.0 (C-1, Aryl),
148.7 (C@CH2). 29Si NMR (CD2Cl2): d ꢁ19.8. Anal. Calc.
for C13H18Si: C, 77.16; H, 8.97. Found: C, 77.0; H, 9.2%.
3.1.7. rac-1-(Dimethylamino)-1-(2-(dimethylamino)-1-
phenylethyl)-1-silacyclohexane (rac-7)
A 2.5 M solution of n-butyllithium in n-hexane (32.4 ml,
81.0 mmol of n-BuLi) was added dropwise at ꢁ50 ꢁC

J.O. Daiß et al. / Journal of Organometallic Chemistry 691 (2006) 3589–3595
3595
[4] G.E.M. Husbands, J.P. Yardley, E.A. Muth (Inventors; American
Home Products Corp., USA). Eur. Pat. Appl. EP 0112669 A2
(04.07.1984);
were combined, the solvent was removed completely under
reduced pressure at 5–15 ꢁC, and the residue was distilled in
vacuo (Vigreux column, 5 cm) to give 8 in 75% yield as a
colorless liquid (3.07 g, 16.5 mmol); b.p. 69–70 ꢁC/3 mbar.
1H NMR (CD2Cl2): d 0.61–0.70 (m, 4H, SiCH2C), 1.33–
1.44 (m, 2H, Si(CH2)2CH2C), 1.57–1.67 (m, 4H,
SiCH2CH2C), 2.46 (s, 12H, NCH3). 13C NMR (CD2Cl2):
d 11.7 (SiCH2C), 25.1 (SiCH2CH2C), 30.7 (Si(CH2)2CH2C),
37.8 (NCH3). 29Si NMR (CD2Cl2): d ꢁ6.0. Anal. Calc. for
C9H22N2Si: C, 58.00; H, 11.90; N, 15.03. Found: C, 57.9; H,
12.0; N, 15.1%.
Chem. Abstr. 102 (1985) 5895e.
[5] J.P. Yardley, G.E.M. Husbands, G. Stack, J. Butch, J. Bicksler, J.A.
Moyer, E.A. Muth, T. Andree, H. Fletcher III, M.N.G. James, A.R.
Sielecki, J. Med. Chem. 33 (1990) 2899.
[6] J.O. Daiß, B. Muller, C. Burschka, R. Tacke, W. Bains, J. Warneck,
¨
in: N. Auner, J. Weis (Eds.), Organosilicon Chemistry VI, Wiley–
VCH, Weinheim, Germany, 2005, pp. 575–581.
[7] W. Bains, R. Tacke, Curr. Opin. Drug Discovery Dev. 6 (2003) 526.
[8] (a) V.L. Ellingrod, P.J. Perry, Am. J. Hosp. Pharm. 51 (1994) 3033;
(b) R.J. Goldberg, Drugs & Aging 11 (1997) 119;
(c) E. Schweizer, R.J. Thielen, A. Frazer, Exp. Opin. Invest. Drugs 6
(1997) 65;
3.1.9. Dimethyl(2-phenylethyl)amine (9)
This compound was commercially available.
(d) M. Briley, Hum. Psychopharmacol. Clin. Exp. 13 (1998) 99;
(e) K. Wellington, C.M. Perry, CNS Drugs 15 (2001) 643;
(f) R.L. Rudolph, Acta Psychiatr. Scand. 106 (Suppl. 415) (2002) 24;
(g) M. Dierick, A. De Nayer, M. Ansseau, H. D’Haenen, P. Cosyns,
W. Verbruggen, A. Seghers, I. Pelc, P. Fossion, G. Stefos, J.
Peuskens, M. Malfroid, S. Leyman, A. Mignon, Curr. Ther. Res.
63 (2002) 475;
(h) H. Sauer, S. Huppertz-Helmhold, W. Dierkes, Pharmacopsychi-
atry 36 (2003) 169;
(i) M.A. Gutierrez, G.L. Stimmel, J.Y. Aiso, Clin. Ther. 25 (2003)
2138.
3.2. Crystal structure analyses
Suitable single crystals of rac-3 and rac-3 Æ HCl were
obtained directly from the preparation of these com-
pounds. The crystals were mounted in inert oil (perflu-
oroalkyl ether, ABCR) on a glass fiber and then
transferred to the cold nitrogen gas stream of the diffrac-
tometer (Stoe IPDS; graphite-monochromated Mo Ka
[9] S.M. Holliday, P. Benfield, Drugs 49 (1995) 280.
[10] K.J. Klamerus, V.D. Parker, R.L. Rudolph, A.T. Derivan, S.T.
Chiang, Pharmacotherapy 16 (1996) 915.
˚
radiation (k = 0.71073 A)). The structures were solved by
[11] The hydrogen-bonding systems were analyzed by using the program
system PLATON. A.L. Spek, PLATON, University of Utrecht,
Utrecht, The Netherlands, 1998;
direct methods [16]. All non-hydrogen atoms were refined
anisotropically [17]. The NH hydrogen atoms were local-
ized in difference Fourier syntheses and refined freely. A
riding model was employed in the refinement of the CH
and OH hydrogen atoms.
In this context, see also: G.A. Jeffrey, W. Saenger, Hydrogen Bonding
in Biological Structures, Springer-Verlag, Berlin, Germany, 1991, pp.
15–24.
[12] (a) Program WIN-DAISY 4.05, Bruker-Franzen GmbH, Bremen, Ger-
many, 1998;
3.3. Pharmacological studies
(b) U. Weber, A. Germanus, H. Thiele, Fresenius J. Anal. Chem. 359
(1997) 46.
[13] The 1H NMR data of rac-3 Æ HCl depend significantly on the
concentration of this compound, especially for the SiCHC-
CHAHBNHG(C(HM)3)(C(HN)3) moiety. The data given were
Receptor binding activities were determined using radi-
oligand cellular uptake inhibition assays via contract
research services (MDS Pharma Services, Taipei, Taiwan).
Radioactivity levels were detected by scintillation counting.
The experimental conditions for each assay are given in
Table 2.
3
3
obtained at a concentration of 80 mM. The JAG and JBG couplings
are not resolved, but recognizable by line broadening of the signals
for the CHAHBN protons.
[14] Surprisingly, this Grignard reagent did not react with 1,1-dichloro-1-
silacyclohexane in refluxing diethyl ether (analogous reaction condi-
tions, GC control).
Appendix A. Supplementary data
[15] Attempts to separate 8 and 9 by distillation were unsuccessful.
However, the identities of 8 and 9 could be established unequivocally
by comparing the GC–MS data of the mixture with those of authentic
samples.
[16] (a) G.M. Sheldrick, SHELXS-97, University of Go¨ttingen, Go¨ttingen,
Germany, 1997;
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.jorgan
chem.2006.04.015.
(b) G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[17] G.M. Sheldrick, SHELXL-97, University of Go¨ttingen, Go¨ttingen,
Germany, 1997.
References
[1] R. Tacke, J. Daiss (Inventors; Amedis Pharmaceuticals Ltd., UK).
PCT Int. Pat. Appl. WO 03/037905 A1 (08.05.2003);
Chem. Abstr. 138 (2003) 354097d.
[18] H. Gu, S.C. Wall, G. Rudnick, J. Biol. Chem. 269 (1994) 7124.
[19] A. Galli, L.J. DeFelice, B.-J. Duke, K.R. Moore, R.D. Blakely, J.
Exp. Biol. 198 (1995) 2197.
[2] J.O. Daiss, M. Penka, C. Burschka, R. Tacke, Organometallics 23
(2004) 4987.
[3] J.O. Daiss, C. Burschka, J.S. Mills, J.G. Montana, G.A. Showell,
J.B.H. Warneck, R. Tacke, Organometallics 25 (2006) 1188.
[20] B. Giros, S. El Mestikawy, N. Godinot, K. Zheng, H. Han, T. Yang-
Feng, M.G. Caron, Mol. Pharmacol. 42 (1992) 383.
[21] Z.B. Pristupa, J.M. Wilson, B.J. Hoffman, S.J. Kish, H.B. Niznik,
Mol. Pharmacol. 45 (1994) 125.