Synthesis and in vitro cytotoxicity of 1,2,3,4-tetrahydroisoquinoline derivatives

Article abstract of DOI:10.1016/j.ejmech.2005.11.003

Several 1-alkyl-1,2,3,4-tetrahydroisoquinoline (TIQ) derivatives, which may play a role in Parkinson's disease, have been synthesized via Pummerer-type cyclization of the sulfonium ion formed in situ from N-formyl sulfoxide. Using an in vitro trypan blue

Full text of DOI:10.1016/j.ejmech.2005.11.003

European Journal of Medicinal Chemistry 41 (2006) 241–252
http://france.elsevier.com/direct/ejmech
Original article
Synthesis and in vitro cytotoxicity
of 1,2,3,4-tetrahydroisoquinoline derivatives
Toshiaki Saitoh a, Kenji Abe b, Masami Ishikawa a, Masanao Nakatani c, Seiichiro Shimazu c,
Noriyuki Satoh c, Fumio Yoneda c, Kyoji Taguchi d, Yoshie Horiguchi d,*
a
Department of Organic Reaction Chemistry, Showa Pharmaceutical University, 3-3165 Higashitamagawagakuen, Machida, Tokyo 194-0042, Japan
b
School of Pharmaceutical Sciences, Ohu University, 31-1 Tomitamachi, Koriyama, Fukushima 963-8611, Japan
c
Research Institute, Fujimoto Pharmaceutical Corp., Matsubara, Osaka 580-8503, Japan
d
Department of Medicinal Chemistry, Showa Pharmaceutical University, 3-3165 Higashitamagawagakuen, Machida, Tokyo 194-0042, Japan
Received 1 September 2005; received in revised form 22 November 2005; accepted 24 November 2005
Available online 10 January 2006
Abstract
Several 1-alkyl-1,2,3,4-tetrahydroisoquinoline (TIQ) derivatives, which may play a role in Parkinson’s disease, have been synthesized via
Pummerer-type cyclization of the sulfonium ion formed in situ from N-formyl sulfoxide. Using an in vitro trypan blue exclusion assay, high
concentrations of TIQ derivatives possessing bulky alkyl group substituents such as 1-cyclobutyl-, 1-cyclohexyl-, 1-phenyl- or 1-benzyl- at the C-
1 position were found to significantly affect the viability of PC12 cells. Moreover, TIQ derivatives that moderately or strongly induced apoptosis
(e.g., 1-phenyl-TIQ and 1-cyclohexyl-TIQ, respectively) paralleled the results obtained using the trypan blue exclusion assay. These results
suggest that the size and electron-donating properties of functional groups may affect the cytotoxicity of TIQ derivatives.
© 2006 Elsevier SAS. All rights reserved.
Keywords: Pummerer-type cyclization; 1,2,3,4-Tetrahydroisoquinoline; Structure–activity relationships; Cytotoxicity; Parkinson’s disease
1. Introduction
and 1-benzyl-TIQ-induced behavioral abnormalities in rats [2,
4]. Furthermore, the concentration of 1-methyl-TIQ is mark-
edly lower in brain tissue from Parkinson’s patients compared
to healthy individuals [5]. We have previously shown that 1-
methyl-TIQ inhibits TIQ-induced bradykinesia, and have also
shown that (R)-1-methyl-TIQ is a more potent inhibitor than
(S)-1-methyl-TIQ in mice [6]. Recently, (R)-1-methyl-TIQ
was found to exert a dose-dependent neuroprotective effect,
while (S)-1-methyl-TIQ had little effect [7]. However, addi-
tional studies using in vivo binding of radioligands to presy-
naptic dopamine transporters in the mouse striatum showed
that (S)-1-methyl-TIQ had an enhanced protective effect on
TIQ-induced loss of ligand-transporter binding compared to
(R)-1-methyl-TIQ [8]. These findings suggest that both (R)-
and (S)-enantiomers may be endogenous agents that prevent
PD by different mechanisms. Thus, 1-methyl-TIQ may play
an important role in the prevention of the onset of neurotox-
in-induced parkinsonism, and suggests that the substituent
group at the C-1-position of the TIQ ring system may have
biologically-important implications.
Parkinson’s disease (PD) is a neurodegenerative disorder
characterized clinically by bradykinesia, rigidity and tremor,
and pathologically by the death of dopaminergic neurons in
the substantia nigra pars compacta; additionally, Lewy bodies
may be present in surviving neurons. It has been reported that
1,2,3,4-tetrahydroisoquinoline (TIQ) derivatives may act as en-
dogenous or environmental neurotoxins causing PD [1,2]. One
TIQ derivative, 1-benzyl-1,2,3,4-tetrahydroisoquinoline (1-
benzyl-TIQ), occurs naturally in animal brain tissue and was
found to induce parkinsonism in monkeys and mice [2,3]. 1-
Benzyl-TIQ has been implicated as a neurotoxin causing PD in
humans because its concentration in the cerebro-spinal fluid of
Parkinsonian patients is approximately three times that found
in healthy individuals [2]. In contrast, 1-methyl-TIQ is present
in normal mouse brain, and has been shown to prevent TIQ-
* Corresponding author. Tel.: +81 42 721 1570; fax: +81 42 721 1570.
E-mail address: horiguti@ac.shoyaku.ac.jp (Y. Horiguchi).
0223-5234/$ - see front matter © 2006 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2005.11.003

242
T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
The efficient synthesis of 1-methyl-TIQ, 1-phenyl-TIQ and
amine 3 by treatment with acetic-formic anhydride, prepared
from acetic anhydride and formic acid, followed by oxidation
of the resulting N-formyl sulfide with NaIO4, produced the cor-
responding sulfoxide 5e–j in high overall yields (see experi-
mental).
1-benzyl-TIQ using a Pummerer-type cyclization reaction as
the key step has been previously reported [9]. In this cycliza-
tion reaction, the sulfonium ion formed in situ from a sulfoxide
is a powerful electrophile that reacts with nucleophilic carbon
species. In a series of papers we reported the syntheses of TIQs
[10–13], 1,2,3,4-tetrahydroquinolines [14], 2,3,4,5-tetrahydro-
1H-3-benzazepines [15], 1,2,4,5-tetrahydro-3H-3-benzazepines
[16], 2,3,4,5-tetrahydro-1H-2-benzazepines [17], 2-quinolones
[18], erythrinan [19], homoerythrinan [20], isopavine and pa-
vine alkaloids [21], and heteroaromatic analogues of TIQ [22].
All compounds were synthesized using a Pummerer-type cycli-
zation reaction, thus demonstrating the utility of this reaction
for the synthesis of various heterocyclic compounds.
As previously described [10,11], sulfoxides lacking a meth-
oxy group as an electron- donating group on the benzene ring
do not cause cyclization under normal Pummerer-type cycliza-
tion reaction conditions (i.e., treating the sulfoxide with tri-
fluoroacetic anhydride (TFAA) in benzene). However, cycliza-
tion of these sulfoxides can be achieved by using boron
trifluoride diethyl etherate (BF3·Et2O) as an additional acid.
Sulfoxide 5e in benzene was treated with TFAA (5 mol eq.)
for 0.5 h, then BF3·Et2O (3 mol eq.) was added; the expected
1-ethyl-2-formyl-4-phenylsulfanyl-TIQ 6e was obtained in
99% yield after 1.0 h. Sequential treatment of sulfoxides 5f–h
with TFAA and BF3·Et2O gave the corresponding 1-alkyl-2-
formyl-4-phenylsulfanyl-TIQs 6f–h in excellent yield. How-
ever, under these reaction conditions 5i, steric hindrance
caused by the bulkiness of cyclohexane ring resulted in only
14% yield of the expected product, 1-cyclohexyl-2-formyl-4-
phenylsulfanyl-TIQ 6i, and 44% yield of aldehyde 7i. The cy-
clization of 5i was improved by adding trifluoromethane sulfo-
nic acid (TFSA), an exceptionally strong acid, rather than
BF3·Et2O; under these conditions, 6i was obtained in 63%
yield. Pummerer-type cyclization of 5j also inefficient due to
decreased nucleophilicity of the aromatic ring caused by the
introduction of a highly electron-withdrawing fluorine in the
aromatic ring. Therefore, cyclization of 5j was achieved using
In the present study, we describe the synthesis of several 1-
alkyl-TIQs and investigate their neurotoxicity in PC12 pheo-
chromocytoma cells.
2. Chemistry
Alkyl-TIQs (1-ethyl-, 1-propyl-, 1-cyclopropyl-, 1-cyclobu-
tyl-, 1-cyclohexyl-TIQ 10e–i and 6-fluoro-1-methyl-TIQ 10j),
were synthesized from N-formyl sulfoxides 5e–j using a Pum-
merer-type cyclization reaction as a key step (Fig. 1). Sub-
strates for the Pummerer-type cyclization reaction, compounds
5e–j, were prepared from alkyl phenyl ketones 1e–j in three
steps. Condensation of 1 and 2-phenylsulfanylethylamine 2 in
titanium(IV) isopropoxide, followed by NaBH4 reduction of
the resulting imines, afforded amines 3e–j. Formylation of
Fig. 1. Synthesis of 1-alkyl-1,2,3,4-tetrahydroisoquinolines.

T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
243
Table 1
3. Biological results and discussion
Yields of products for compound 6–10
3.1. Effect of cell culture conditions on the viability
and differentiation of PC12 cells
Using pheochromocytoma cells (PC12 cells), the 1-alkyl-
TIQs described above were tested in vitro for their cytotoxic
activity and their effect on cell differentiation. We first exam-
ined the relationship between cell culture conditions and PC12
cell viability. In differentiation medium (DM; Dulbecco’s mod-
ified Eagle’s medium (DMEM) containing serum and nerve
growth factor (NGF)), over 95% of PC12 cells survived (data
not shown). Viability decreased when serum was removed
from the medium (NGF + DMEM), and when the cells were
grown in trophic withdrawal medium (DMEM without serum
and NGF). The reduction in cell viability was statistically sig-
nificant (Table 2; the viability of the control (DM-cultured
cells) is taken as 100%). Outgrowth of neurites was also sig-
nificantly decreased if serum and/or NGF were removed from
the medium. In trophic withdrawal medium, the percentage of
cytoplasmic neurite-bearing cells was significantly reduced.
These results suggest that the ratio of viable and neurite-bear-
ing cells depends on the presence of trophic factors in the
growth medium (Table 2).
TFSA, affording 6-fluoro-2-formyl-1-methyl-TIQ 6j in 63%
yield. Reductive removal of the phenylthio group of TIQ 6
readily proceeded in the presence of NiCl2/NaBH4 in
MeOH/H2O to give 1-alkyl-2-formyl-TIQs 8 in good yield,
although in some cases 1,2-dihydroisoquinolines 9 were pro-
duced as by-products (Fig. 1). Deprotection of the N-formyl
group was achieved in the conventional manner, as described
below. Alkaline hydrolysis of 1-alkyl-2-formyl-TIQs 8 gave 1-
alkyl-TIQs 10 in excellent yield (Fig. 1). The yields of all pro-
ducts are summarized in Table 1.
3.2. Effects of TIQ derivatives on PC12 cell viability
The effect of each TIQ derivative on trophic withdrawal-
cultured PC12 cells was estimated as the percentage of trophic
withdrawal-cultured cell viability in the absence of any TIQ.
Cell viability was unaffected by low (10–5M) and medium
(10–4M) concentrations of TIQ derivatives (10a–10j). How-
We have previously reported the synthesis of chiral TIQs,
(R)- and (S)-1-methyl TIQs 16a, b, from (R)- and (S)-1-pheny-
lethylamines 11a, b via Pummerer-type cyclization in 6 steps
as shown in Fig. 2 [12].
Table 2
Effects of serum and NGF on the growth of PC12 cells
Treatment
% Of trophic withdrawal
(trypan blue exclusion)
100 0.5
% Of trophic withdrawal
(bearing neurite)
100 4.0
Serum + NGF (as differential medium, DM)
Without serum (added NGF only)
92.7 0.4 **
83.2 3.2 **
Without serum and NGF (as trophic withdrawal)
76.9 0.9 **
57.6 2.2 **
**P < 0.01, vs. Serum + NGF. Data are shown as mean S.E.
Fig. 2. Synthesis of (R)-(S)-1-methyl-1,2,3,4-tetrahydroisoquinoline (16).

244
T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
ever, high concentrations (10–3M) of TIQ derivatives with re-
latively large substituents (e.g., 1-cyclobutyl- 10h, 1-cyclohex-
yl- 10i, 1-phenyl- 10c or 1-benzyl-TIQ 10d) at the C-1 position
significantly decreased cell viability. Small substituents at the
C-1 position such as TIQ 10a, 1-methyl- 16a and 16b (10b), 1-
ethyl- 10e, 1-propyl- 10f and 1-cyclopropyl-TIQ 10g had no
effect on cell viability even at high concentrations; the addition
of fluorine to small substituent groups attached to the TIQ (e.g.
, 6-fluoro-1-methyl-TIQ 10j) also had no effect. One TIQ de-
rivative, 1-methyl-6,7-dihydroxy-TIQ (salsolinol), induced cell
death at lower concentrations than did TIQ derivatives with
large substituents. Of the 1-alkyl-TIQs synthesized and tested,
only TIQ 10a protected cells from the effects of trophic with-
drawal. These results suggest that the size of the substituent at
the C-1 position of TIQs affects cell viability (Table 3).
derivatives that reduced cell viability at high concentration,
such as 1-cyclohexyl- 10i and 1-benzyl-TIQ 10d, significantly
decreased neurite-bearing cells at lower concentrations. Other
TIQs that reduced cell viability (e.g., 1-phenyl- 10c, 1-cyclo-
butyl-TIQ 10h and salsolinol) also inhibited neuronal differen-
tiation, as did TIQ derivatives that did not influence cell viabi-
lity (e.g., (R)- and (S)-1-methyl- 16a and 16b, 1-ethyl- 10e, 1-
propyl- 10f, 1-cyclopropyl- 10g and 6-fluoro-1-methyl-TIQ
10j). In contrast, low concentrations of 1-cyclobutyl-TIQ 10h
showed increasing effects on neurite-bearing cells (Table 3).
3.4. Evaluation of the cytotoxic effects of TIQ derivatives using
the comet assay
At high concentrations (10–3 M), 1-phenyl- 10c and cyclo-
hexyl-TIQ 10i, were clearly cytotoxic (cell viability was 41.4
10.3% and 0.2 0.2%, respectively; Table 3). The ability of
these compounds to induce DNA damage in individual cells
was evaluated using the comet assay. Fig. 3A depicts a typical
migrated cell nucleus with fragmented DNA caused by treating
with 1-cyclohexyl-TIQ 10i. Small portions of DNA remain at
3.3. Effects of TIQ derivatives on the differentiation of PC12
cells
The effect of TIQ derivatives on the outgrowth of neurites
was more pronounced than their effect on cell viability. TIQ
Table 3
Effects of TIQ derivatives on trophic withdrawal-induced apoptosis and differentiation in PC12 cells
Drugs
Concentration
(M)
n
% Of trophic withdrawal
(trypan blue exclusion)
95.2 3.1
0.0 0.0**
0.3 0.2**
106.5 2.2
106.6 1.8
114.6 2.5**
94.8 5.5
99.9 3.2
98.2 2.9
95.5 4.5
101.1 2.3
97.3 4.7
105.2 3.1
114.2 2.7**
41.4 10.3**
95.0 3.0
107.8 2.1
0.0 0.0**
100.2 5.0
104.9 4.1
88.8 6.9
104.7 4.5
104.3 2.8
88.8 4.8
98.7 4.0
97.1 3.4
% Of trophic withdrawal
(bearing neurite)
75.4 14.2*
0.0 0.0**
0.0 0.0**
N.D.
N.D.
N.D.
89.7 5.0
83.8 12.3
45.1 5.4**
67.8 10.5
89.6 11.5
38.4 6.3**
81.5 5.6
41.0 3.3**
3.6 1.5**
79.9 4.4
55.8 4.6**
0.0 0.0**
91.9 5.5
80.3 6.7
31.7 4.1**
77.7 8.3
65.2 8.3*
27.8 4.5**
85.2 13.4
76.2 18.4
34.8 11.8**
134.2 28.0*
75.4 9.8
12.0 8.3**
72.7 13.3
35.3 5.9**
0.0 0.0**
82.8 13.1
63.2 11.0*
35.8 4.1**
Name
Compound
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
10–5
10–4
10–3
8
8
8
16
12
12
4
4
4
4
4
4
12
12
12
12
12
12
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
Salsolinol
–
TIQ
10a
(R)-1-methyl-TIQ
(S)-1-methyl-TIQ
1-Phenyl-TIQ
16a (10b)
16b (10b)
10c
1-Benzyl-TIQ
10d
1-Ethyl-TIQ
10e
1-Propyl-TIQ
10f
1-Cyclopropyl-TIQ
1-Cyclobutyl-TIQ
1-Cyclohexyl-TIQ
6-Fluoro-1-methyl-TIQ
10g
89.1 7.2
104.7 2.7
104.9 2.6
34.9 7.4**
107.9 1.6
110.2 4.8
0.2 0.2**
100.6 2.5
95.8 4.8
10h
10i
10j
4
87.9 8.0
N.D.: not done, *P < 0.05 and **P < 0.01, vs. wo/Serum and NGF. Data are shown as mean S.E.

T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
245
Fig. 3. Fluorescence photomicrograph of a comet in PC12 cells after a 3-h exposure to 1-cyclohexyl- or 1-phenyl-TIQ. Panel A shows a comet typically observed in
cells treated with high concentrations (10–3 M) of 1-cyclohexyl-TIQ. A small portion of the DNA remains at the origin, but most of the DNA has fragmented and
migrated to form the comet tail. In the present study, damaged cells such as this are termed apoptotic damaged cells. Panels B and C shows representative
photomicrographs of PC12 cells treated with either 1-cyclohexyl-TIQ (B) or 1-phenyl-TIQ (C). Images were obtained using a fluorescence microscope at × 200
magnification and at a wavelength of 568 nm.

246
T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
Table 4
Evaluation of 1-phenyl- and 1-cyclohexyl-TIQ-induced PC12 cell damage assessed using the comet assay
Drugs
Concentration
n
Length and width ratio
of DNA mass
1.10 0.04
1.37 0.15
3.12 0.30**
The frequency of tailed cells
(%)
8.3 5.4
23.2 2.2
69.6 5.1**
Name
Non-treat
1-Phenyl-TIQ
1-cyclohexyl-TIQ
Compound
–
10c
10i
(M)
–
8
12
4
10–3
10–3
**P < 0.05 and **P < 0.01, vs. non-treated cells. Data are shown as mean S.E.
the origin, but most of the DNA has fragmented and migrated
to form a ‘comet tail’. Diffuse, migrated DNA is indicative of
apoptosis in a cell. In the present study, a cell exhibiting such a
‘comet tail’ is considered apoptotic. The ratio of the length of
the mass of DNA in the tail to the width of the DNA mass in
the tail, together with the frequency of cells exhibiting tails,
was used to estimate the cytotoxicity of the TIQ derivatives
described above. 1-Cyclohexyl-TIQ 10i significantly increased
both the length: width ratio and the frequency of cells with tails
(Fig. 3A, B and Table 4), whereas 1-phenyl-TIQ 10c was sig-
nificantly less cytotoxic (Fig. 3C and Table 4). These observa-
tions are in agreement with results obtained using the trypan
blue exclusion assay (Table 3) and suggest that both the size
of the functional groups, and the presence of electron-donating
group(s) attached to the isoquinoline structure, may affect the
toxicity of TIQs towards PC12 cells. This tentative conclusion
is further supported by the fact that although high concentra-
tions of 1-methyl-TIQ did not induce cell death, dihydroxide-1-
methyl-TIQ (salsolinol) is a potent cytotoxin (Table 3).
mass spectrometer at 70 eV using a direct inlet probe. High-
resolution EIMS (HR-EIMS) and low-resolution chemical ioni-
zation mass spectra (LR-CIMS) were taken on a JEOL JMS-
D300 mass spectrometer at 30 eV (EIMS) or at 270 eV
[(CIMS), reactant gas: iso-butane] using a direct inlet probe.
High-resolution fast atom bomb ionization mass spectra (HR-
FABMS) were measured with a JEOL HX-110A mass spectro-
meter (reactants gas: xenon, matrix: glycerol). TLC was per-
formed on Merck pre-coated Silica gel 60 F254 plates. Column
chromatography was carried out with silica gel (Wakogel C-
200). Flash column chromatograph was performed with silica
gel (KANTOH Silica gel 60N (40–100 μm)). The organic ex-
tract from each reaction mixture was washed with brine, dried
over anhydrous Na2SO4, and concentrated in vacuo to dryness.
5.2. Compounds
With regard to compounds 3b–10d, see references [9], and
11–16, see references [12].
4. Conclusions
5.3. Preparation of 3e. Typical procedure
A number of 1-alkyl-TIQ derivatives that may be implicated
in PD were synthesized using a Pummerer-type cyclization re-
action. Cyclizations involving sterically-hindered 1-alkyl sub-
stituents were accelerated with TFSA. High concentrations of
TIQ derivatives with large substituents at the C-1 position sig-
nificantly affected PC12 cell viability; in addition, the electron-
donating properties of functional groups may also contribute
the cytotoxicity of TIQ derivatives. Further investigations into
the synthesis and biological activity of TIQ derivatives with
different C-1 substituents may be useful for developing new
therapeutic agents for treating PD.
A mixture of 1e (13.4 g, 0.10 mol), 2-phenylsulfanylethyla-
mine (15.3 g, 0.10 mol), and titanium(IV) isopropoxide (43.0 g,
0.15 mol) was heated at 80°C for 4.0 h under an argon atmo-
sphere. After cooling, the reaction mixture was diluted with
MeOH (500 ml). To this solution, NaBH4 (7.50 g, 0.20 mol)
was added in small portions while cooling the reaction mixture
on ice. The reaction mixture was stirred at room temperature for
1.0 h and concentrated in vacuo. Water (ca. 40 ml) was added to
the residue, and the mixture was diluted with MeOH (ca. 500
ml). After removal of precipitated inorganic materials by filtra-
tion, the filtrate was concentrated in vacuo. The residue was
dissolved in water extracted with CHCl3. The residue was dis-
tilled under reduced pressure (180–185°C/5 mmHg) to give
5. Experimental
1-ethyl-1-phenylmethyl-(2-phenylsulfanyl)ethylamine
3e
(25.2 g, 93%) as a pale yellow oil. 1H-NMR: 0.80 (3H, t,
J = 7 Hz, CH2CH3), 1.50–1.85 (2H, m, CH2CH3), 2.58–0.73
(2H, m, SCH2CH2), 2.90–3.09 (2H, m, SCH2CH2), 3.46 (1H,
dd, J = 8, 6 Hz, CHPh), 7.17–7.28 (10H, m, PhHx2). 13C-NMR:
10.8 (CH2CH3), 31.0 (CH2CH3), 34.3 (SCH2CH2), 45.8
(SCH2CH2), 64.5 (CHPh), 126.0 (PhCH), 126.9 (PhCH),
127.3 (PhCHx2), 128.3 (PhCHx2), 128.8 (PhCHx2), 129.5
(PhCHx2), 135.8 (PhC), 143.8 (PhC). LR-CIMS m/z: 272
(MH+).
5.1. Chemistry
Unless otherwise noted, the following procedures were
adopted. Melting points were taken on a Yanagimoto SP-M1
hot-stage melting point apparatus and are uncorrected. IR spec-
tra were measured as films for oils and gums, and KBr disks for
solids, with a HORIBA FT-710 Fourier transform infrared spec-
trophotometer, and the values are given in cm–1. NMR spectra
were measured on a JEOL JNM-AL300 (1H-NMR: 300 MHz,
13C-NMR: 75 MHz) NMR spectrometer in CDCl3 with tetra-
methylsilane as an internal standard, and the chemical shifts
are given in δ values. Low-resolution electron impact ionization
mass spectra (LR-EIMS) were taken on a JEOL JMS-AM20
5.3.1. 1-Phenylmethyl-1-propyl-(2-phenylsulfanyl)ethylamine 3f
1
Yield, 94%. A pale yellow oil (180–185°C/3 mmHg). H-
NMR: 0.86 (3H, t, J = 7 Hz, CH2CH2CH3), 1.07–1.36 (2H, m,

T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
247
CH2CH2CH3), 1.52–1.69 (2H, m, CH2CH2CH3), 2.61–2.72
(2H, m, SCH2CH2), 2.93–3.01 (2H, m, SCH2CH2), 3.55 (1H,
dd, J = 7, 6 Hz, CHPh), 7.13–7.55 (10H, m, PhHx2). 13C-
NMR: 14.0 (CH2CH2CH3), 19.5 (CH2CH2CH3), 34.3 (S
CH2CH2), 40.5 (CH2CH2CH3), 45.7 (SCH2CH2), 62.7
(CHPh), 126.1 (PhCH), 126.9 (PhCH), 127.2 (PhCHx2),
128.3 (PhCHx2), 128.8 (PhCHx2), 129.5 (PhCHx2), 135.8
(PhC), 144.1 (PhC). LR-CIMS m/z: 286 (MH+).
57.1 (CHPh), 115.1 (ArCHx2), 126.1 (PhCH), 128.0
(ArCHx2), 128.8 (PhCHx2), 129.5 (PhCHx2), 135.6 (PhC),
141.0 (ArC). 161.7 (ArC). HR-EIMS m/z (M+): Calcd for
C16H18FNS: 275.1144, found: 275.1178.
5.4. Formylation of 3e. Typical procedure
A solution of formic-acetic anhydride prepared from formic
acid (84.0 ml, 2.22 mol) and acetic anhydride (50.6 ml, 0.74
mol), was added to 3e (20.0 g, 74.0 mmol) at 0°C in one por-
tion and then the mixture was heated at 60°C for 1.0 h. After
removal the formic-acetic anhydride in vacuo, 10% NaOH so-
lution was added to the residue and extracted with CHCl3. The
residue was chromatographed (AcOEt/hexane = 1:2) over silica
gel to give N-(1-ethyl-1-phenylmethyl)-N-(2-phenylsulfany-
lethyl)formamide 4e (22.0 g, 99%) as a pale yellow oil. IR:
5.3.2. 1-Cyclopropyl-1-phenylmethyl-(2-phenylsulfanyl)
ethylamine 3g
Yield, 90%. A colorless oil (190–195°C/4 mmHg). 1H-
NMR: 0.18–0.25 (1H, m, cyclopropyl), 0.27–0.32 (1H, m, cy-
clopropyl), 0.36–0.43 (1H, m, cyclopropyl), 0.53–0.62 (1H, m,
cyclopropyl), 1.02–1.13 (1H, m, cyclopropyl), 2.64–2.75 (2H,
m, SCH2CH2), 2.79 (1H, d, J = 9 Hz, CHPh), 2.95–3.06 (2H,
m, SCH2CH2), 7.14–7.32 (10H, m, PhHx2). 13C-NMR: 2.56
(CH2), 4.81 (CH2), 18.9 (CH2), 34.2 (SCH2CH2), 45.7
(SCH2CH2), 68.0 (CHPh), 126.0 (PhCH), 127.0 (PhCH),
127.1 (PhCHx2), 128.2 (PhCHx2), 128.8 (PhCHx2), 129.5
(PhCHx2), 135.7 (PhC), 143.6 (PhC). HR-EIMS m/z (M+):
Calcd for C18H21NS: 283.1395, found: 283.1397.
1
1664. H-NMR: 0.95 (3H, t, J = 7 Hz, CH2CH3), 1.84–2.08
(2H, m, CH2CH3), 2.48–2.58 (1H, m, SCH2CH2), 2.92–3.22
(2H, m, SCH2CH2 and SCH2CH2), 3.38–3.53 (1H, m,
SCH2CH2), 4.36, 5.48 (total 1H, each dd, J = 9, 6 Hz, CHPh),
7.15–7.37 (10H, m, PhHx2), 8.18, 8.36 (total 1H, each s,
CHO). HR-EIMS m/z (M+): Calcd for C18H21NOS: 299.1344,
found: 299.1314.
5.3.3. 1-Cyclobutyl-1-phenylmethyl-(2-phenylsulfanyl)
ethylamine 3h
Yield, 84%. A colorless oil (210–212°C/8 mmHg). 1H-
NMR: 1.53–1.81 (7H, m, cyclobutyl), 2.55–2.70 (2H, m,
SCH2CH2), 2.92–3.07 (2H, m, SCH2CH2), 3.46 (1H, d, J = 9
Hz, CHPh), 7.11–7.31 (10H, m, PhHx2). 13C-NMR: 17.6
(CH2), 25.2 (CH2), 26.5 (CH2), 34.2 (SCH2CH2), 42.2 (CH),
45.6 (SCH2CH2), 68.8 (CHPh), 126.0 (PhCH), 126.9 (PhCH),
127.4 (PhCHx2), 128.1 (PhCHx2), 128.7 (PhCHx2), 129.4
(PhCHx2), 135.7 (PhC), 142.2 (PhC). LR-CIMS m/z: 298
(MH+).
5.4.1. N-(1-phenylmethyl-1-propyl)-N-(2-phenylsulfanylethyl)
formamide 4f
1
Yield, 99%. A pale yellow oil. IR: 1666. H-NMR: 0.93–
0.99 (3H, m, CH2CH2CH3), 1.20–1.45 (2H, m, CH2CH2CH3),
1.85–1.95 (2H, m, CH2CH2CH3), 2.49–2.60, 2.91–3.00 (total
2H, each m, SCH2CH2), 3.06–3.22, 3.37–3.46 (total 2H, each
m, SCH2CH2), 4.48, 5.58 (total 1H, each t, J = 7 Hz, CHPh),
7.13–7.35 (10H, m, PhHx2), 8.16, 8.36 (total 1H, each s,
CHO). HR-EIMS m/z (M+): Calcd for C19H23NOS: 313.1501,
found: 313.1514.
5.3.4. 1-Cyclohexyl-1-phenylmethyl-(2-phenylsulfanyl)
ethylamine 3i
5.4.2. N-(1-cyclopropyl-1-phenylmethyl)-N-(2-
phenylsulfanylethyl)formamide 4g
Yield, 20%. Colorless needles, m.p. 58–60°C, recrystallized
1
from Et2O. H-NMR: 0.76–1.76 (10H, m, cyclohexyl), 1.93
1
Yield, 99%. A pale yellow oil. IR: 1657. H-NMR: 0.35–
(1H, m, cyclohexyl), 2.56 (1H, dd, J = 12, 6 Hz, SCH2CH2),
2.64 (1H, dd, J = 12, 6 Hz, SCH2CH2), 2.95 (1H, dd, J = 13, 6
Hz, SCH2CH2), 3.01 (1H, dd, J = 13, 6 Hz, SCH2CH2), 3.28
(1H, d, J = 9 Hz, CHPh), 7.11–7.29 (10H, m, PhHx2). 13C-
NMR: 26.2 (CH2), 26.3 (CH2), 26.5 (CH2), 29.8 (CH2), 60.2
(CH2), 34.3 (SCH2CH2), 44.2 (CH), 45.9 (SCH2CH2), 68.4
(CHPh), 126.0 (PhCH), 126.7 (PhCH), 127.9 (PhCHx2),
127.94 (PhCHx2), 128.8 (PhCHx2), 129.5 (PhCHx2), 135.9
(PhC), 142.7 (PhC). HR-EIMS m/z (M+): Calcd for
C21H27NS: 325.1864, found: 325.1897.
0.90 (4H, m, cyclopropyl), 1.27−1.36 (1H, m, cyclopropyl),
2.79–2.85, 2.94–3.08 (total 2H, each m, SCH2CH2), 3.25–
3.43 (2H, m, SCH2CH2), 3.64, 4.70 (total 1H, each d, J = 10
Hz, CHPh), 7.12–7.39 (10H, m, PhHx2), 8.23, 8.42 (total 1H,
each s, CHO). HR-EIMS m/z (M+): Calcd for C19H21NOS:
311.1341, found: 311.1341.
5.4.3. N-(1-cyclobutyl-1-phenylmethyl)-N-(2-
phenylsulfanylethyl)formamide 4h
1
Yield, 99%. A pale yellow oil. IR: 1668. H-NMR: 1.66–
2.13 (6H, m, cyclobutyl), 2.53–2.70 (1H, m, cyclobutyl), 2.84–
3.17 (3H, m, SCH2CH2 and SCH2CH2), 3.29–3.39 (1H, m,
SCH2CH2), 4.37, 5.66 (total 1H, each d, J = 11 Hz, CHPh),
7.12–7.37 (10H, m, PhHx2), 8.18, 8.37 (total 1H, each s,
CHO). HR-EIMS m/z (M+): Calcd for C20H23NOS: 325.1500,
found: 325.1508.
5.3.5. 1-(4-Fluorophenyl)ethyl-(2-phenylsulfanyl)ethylamine 3j
1
Yield, 83%. A pale yellow oil (185°C/4 mmHg). H-NMR:
1.30 (3H, d, J = 7 Hz, CH3), 1.96 (1H, br-s, NH), 2.62–2.74
(2H, m, SCH2CH2), 3.03 (2H, t, J = 6 Hz, SCH2CH2), 3.76
(1H, q, J = 6 Hz, CHAr), 6.95–7.30 (9H, m, PhH and ArH).
13C-NMR: 24.4 (CH3), 34.3 (SCH2CH2), 45.7 (SCH2CH2),

248
T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
5.4.4. N-(1-cyclohexyl-1-phenylmethyl)-N-(2-
5.5.3. N-(1-cyclobutyl-1-phenylmethyl)-N-(2-
phenylsulfanylethyl)formamide 4i
phenylsulfinylethyl)formamide 5h
1
1
Yield, 99%. A pale yellow oil. IR: 1668. H-NMR: 0.75–
1.29 (5H, m, cyclohexyl), 1.53–1.82 (5H, m, cyclohexyl),
2.06–2.17 (2H, m, SCH2CH2), 2.33–2.63, 2.92–3.09 (total
2H, each m, SCH2CH2), 3.17–3.42, 3.64–3.66 (total 1H, each
m, cyclohexyl), 4.02, 5.12 (total 1H, each d, J = 12 Hz,
CHPh), 7.10–7.37 (10H, m, PhHx2), 8.13, 8.30 (total 1H, each
s, CHO). LR-CIMS m/z: 354 (MH+).
Yield, 99%. A pale yellow gum. IR: 1668. H-NMR: 1.70–
2.10 (7H, m, cyclobutyl), 2.50–2.62, 2.68–2.75 (total 1H, each
m, SCH2CH2), 2.92–3.12 (2H, m, SCH2CH2 and SCH2CH2),
3.32–3.59 (1H, m, SCH2CH2), 4.39, 4.44 (total 1H, each d,
J = 11 Hz, CHPh), 7.18–7.55 (10H, m, PhHx2), 8.18, 8.33,
8.36, 8.39 (total 1H, each s, CHO). LR-CIMS m/z: 342
(MH+).
5.4.5. N-[1-(4-fluorophenyl)ethyl]-N-(2-Phenylsulfanylethyl)
5.5.4. N-(1-cyclohexyl-1-phenylmethyl)-N-(2-
formamide 4j
phenylsulfinylethyl)formamide 5i
1
1
Yield, 99%. A yellow oil. IR: 1668. H-NMR: 1.50, 1.59
Yield, 99%. A pale yellow gum. IR: 1668. H-NMR: 0.66–
(total 3H, each d, J = 7 Hz, CH3), 2.60–2.66, 2.75–2.85,
2.94–3.03 (total 2H, each m, SCH2CH2), 3.16–3.37 (2H, m,
SCH2CH2), 4.74, 5.71 (total 1H, each q, J = 7 Hz, CHAr),
7.00–7.06 (2H, m, ArH), 7.14–7.28 (7H, m, PhH and ArH),
8.13, 8.36 (total 1H, each s, CHO). HR-EIMS m/z (M+): Calcd
for C17H18FNOS: 303.1093, found: 303.1112.
4.10 (16H, m, cyclohexyl, CHPh, SCH2CH2 and SCH2CH2),
7.20–7.60 (10H, m, PhHx2), 8.06, 8.11, 8.28, 8.30 (total 1H,
each s, CHO). LR-CIMS m/z: 370 (MH+).
5.5.5. N-[1-(4-Fluorophenyl)ethyl]-N-(2-Phenylsulfinylethyl)
formamide 5j
1
Yield, 86%. A pale yellow gum. IR: 1664, 1602. H-NMR:
5.5. Oxidation of 4e with NaIO4. Typical procedure
1.50, 1.59, 1.65, 1.70 (total 3H, each d, J = 7 Hz, CH3), 2.70–
3.54 (4H, m, SCH2CH2 and SCH2CH2), 4.76–4.82 (1H, m,
CHAr), 7.01–7.10, 7.21–7.35, 7.48–7.59 (total 9H, each m,
ArH and PhH), 8.14, 8.25, 8.34, 8.40 (total 1H, each s,
CHO). LR-CIMS m/z: 320 (MH+).
A solution of 4e (8.97 g, 30.0 mmol) and NaIO4 (9.63 g, 45.0
mmol) in MeOH (200 ml) and H2O (100 ml) was stirred at room
temperature for 13 h. After removal of inorganic precipitates by
filtration, the filtrate was concentrated in vacuo. The residue was
extracted with CHCl3. The product was chromatographed with
AcOEt and AcOEt/MeOH = 9:1 to give N-(1-ethyl-1-phenyl-
methyl)-N-(2-phenylsulfinylethyl)formamide 5e (9.10 g,
96%) as a pale yellow gum. IR: 1666. 1H-NMR: 0.91–1.00
(3H, m, CH2CH3), 1.99–2.07, 2.42–2.51 (total 2H, each m,
CH2CH3), 2.79 (1H, t, J = 7 Hz, SCH2CH2), 3.03–3.70 (3H,
m, SCH2CH2 and SCH2CH2), 4.41 (1H, d, J = 8 Hz, CHPh),
7.22–7.55 (10H, m, PhHx2), 8.34, 8.37 (total 1H, each s,
CHO). LR-CIMS m/z: 316 (MH+).
5.6. Pummerer reaction of sulfoxide 5e. Typical procedure
TFAA (2.21 ml, 15.9 mmol) was added into a solution of 5e
(1.00 g, 3.17 mmol) in benzene (50.0 ml) at room temperature.
After the mixture was stirred for 0.5 h, BF3·Et2O (1.20 ml,
9.51 mmol) was added, and the reaction mixture was further
stirred at the same temperature for 1.0 h. The reaction mixture
was diluted with H2O and extracted with CHCl3. Purification
by column chromatography (AcOEt/hexane = 1:2) of the pro-
duct gave N-formyl-1-ethyl-4-phenylsulfanyl-1,2,3,4-tetra-
hydro-isoquinoline 6e (932 mg, 99%) as a pale yellow gum.
5.5.1. N-(1-phenylmethyl-1-propyl)-N-(2-phenylsulfinylethyl)
1
formamide 5f
IR: 1673. H-NMR: 0.98, 0.99 (total 3H, each t, J = 7 Hz,
1
Yield, 99%. A pale yellow oil. IR: 1664. H-NMR: 0.96,
CH2CH3), 1.76, 1.91 (total 2H, each dq, J = 7, 4 Hz, C
H2CH3), 3.15, 3.72 (total 1H, each dd, J = 14, 3 Hz, C3-H),
3.55, 4.62 (total 1H, each d, J = 14 Hz, C3-H), 4.35, 4.51 (total
1H, each d, J = 3 Hz, C4-H), 4.45, 5.41 (total 1H, each dd,
J = 10, 4 Hz, C1-H), 7.13–7.49, 7.68–7.69 (total 9H, each m,
ArH and PhH), 8.14, 8.37 (total 1H, each s, CHO). LR-CIMS
m/z: 298 (MH+).
0.98 (total 3H, each t, J = 7 Hz, CH2CH2CH3), 1.25–1.40
(2H, m, CH2CH2CH3), 1.93–2.03 (2H, m, CH2CH2CH3),
2.41–2.51, 2.76–2.80, 3.00–3.07 (total 2H, each m, SC
H2CH2) ,3.18–3.45, 3.60–3.70 (total 2H, each m, SCH2CH2),
4.99, 4.52 (total 1H, each t, J = 7 Hz, CHPh), 7.21–7.61 (10H,
m, PhHx2), 8.33, 8.37 (total 1H, each s, CHO). HR-EIMS m/z
(M+): Calcd for C19H23NO2S: 329.1450, found: 329.1450.
5.6.1. N-formyl-4-phenylsulfanyl-1-propyl-1,2,3,4-
5.5.2. N-(1-cyclopropyl-1-phenylmethyl)-N-(2-
phenylsulfinylethyl)formamide 5g
tetrahydroisoquinoline 6f
1
Yield, 99%. Colorless prisms, m.p. 62–64°C. IR: 1674. H-
1
Yield, 99%. A pale yellow gum. IR: 1668. H-NMR: 0.33–
NMR: 0.95, 0.97 (total 3H, each t, J = 7 Hz, CH2CH2CH3),
1.26–1.54 (2H, m, CH2CH2CH3), 1.69–1.83 (2H, m, C
H2CH2CH3), 3.17, 3.76 (1H, each dd, J = 14, 3 Hz, C3-H),
3.52, 4.61 (1H, each d, J = 14 Hz, C3-H), 4.35, 4.52 (1H, each
d, J = 3 Hz, C4-H), 5.48 (1H, dd, J = 9, 5 Hz, C1-H), 7.10–
7.60, 7.65–7.68 (total 9H, each m, ArH and PhH), 8.13, 8.37
(total 1H, each s, CHO). LR-CIMS m/z: 312 (MH+).
0.96 (4H, m, cyclopropyl), 1.32–1.56 (1H, m, cyclopropyl),
2.82–2.93 (1H, m, SCH2CH2), 3.13–3.39, 3.46–3.62 (total
3H, each m, SCH2CH2 and SCH2CH2), 3.66, 3.73, 4.81, 4.86
(total 1H, each d, J = 10 Hz, CHPh), 7.30–7.58 (10H, m,
PhHx2), 8.27, 8.33, 8.35, 8.49 (total 1H, each s, CHO). LR-
CIMS m/z: 328 (MH+).

T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
249
5.6.2. N-formyl-1-cyclopropyl-4-phenylsulfanyl-1,2,3,4-
5.7.2. N-(1-cyclohexyl)ethyl-N-(2-oxothyl)formamide 7i
1
tetrahydroisoquinoline 6g
A yellow gum. IR: 1734, 1674, 1653. H-NMR: 0.81–1.33
(5H, m, cyclohexyl), 1.60–1.87 (5H, m, cyclohexyl), 2.00–2.12
(1H, m, cyclohexyl), 3.55 (1H, dd, J = 18, 2 Hz, CH2CHO),
4.10 (1H, dd, J = 18, 2 Hz, CH2CHO), 4.22 (1H, J = 11 Hz,
CHPh), 7.21–7.40 (5H, m , PhH), 8.79 (1H, s, NCHO), 8.89
(1H, t, J = 2 Hz, CH2CHO). LR-CIMS m/z: 260 (MH+).
1
Yield, 99%. A pale yellow gum. IR: 1674. H-NMR: 0.36–
0.48 (1H, m, cyclopropyl), 0.57–0.83 (3H, m, cyclopropyl),
1.11–1.25 (1H, m, cyclopropyl), 3.38, 3.93 (total 1H, each
dd, J = 14, 3 Hz, C3-H), 3.59, 4.64 (total 1H, each d, J = 14
Hz, C3-H), 4.38, 4.54 (total 1H, each br-s, C4-H), 4.70, 4.76,
4.85 (total 1H, each d, J = 9 Hz, C1-H), 7.24–7.48 (9H, m,
ArH and PhH), 8.07, 8.13, 8.33 (total 1H, each s, CHO).
HR-EIMS m/z (M+): Calcd for C19H19NOS: 309.1184, found:
309.1184.
5.8. Pummerer reaction of sulfoxide 5j
Method A: 5j (500 mg, 1.57 mmol) was treated with TFAA
(1.10 ml, 7.85 mmol) and BF3·Et2O (0.67 ml, 4.71 mmol) in
benzene (25.0 ml) at room temperature to give 6j (62.0 mg,
13%), 7j (148 mg, 45%).
Method B: 5j (500 mg, 1.57 mmol) was treated with TFAA
(1.10 ml, 7.85 mmol) and TFSA (0.14 ml, 4.71 mmol) in ben-
zene (25.0 ml) at room temperature to give 6j (202 mg, 63%).
5.6.3. N-formyl-1-cyclobutyl-4-phenylsulfanyl-1,2,3,4-
tetrahydroisoquinoline 6h
1
Yield, 99%. A pale yellow gum. IR: 1674. H-NMR: 1.73–
1.84 (3H, m, cyclobutyl), 1.98–2.23 (3H, m, cyclobutyl), 2.58–
2.64 (1H, m, cyclobutyl), 3.18, 3.71 (total 1H, each dd, J = 14,
3 Hz, C3-H), 3.51, 4.58 (total 1H, each d, J = 14 Hz, C3-H),
4.36, 4.51 (total 1H, each d, J = 3Hz, C4-H), 4.38, 5.41 (total
1H, each d, J = 10 Hz, C1-H), 7.10–7.48 (8H, m, ArH and
PhH), 7.64–7.67 (1H, m, ArH), 8.10, 8.41 (total 1H, each s,
CHO). LR-CIMS m/z: 324 (MH+).
5.8.1. N-formyl-6-fluoro-1-methyl-4-phenylsulfanyl-1,2,3,4-
tetrahydroisoquinoline 6j
1
A pale yellow gum. IR: 1670. H-NMR: 1.44, 1.50 (total
3H, each d, J = 7 Hz, CH3), 3.22, 3.71 (total 1H, each dd,
J = 14, 3 Hz, C3-H), 3.54, 4.58 (total 1H, each d, J = 14 Hz,
C3-H), 4.28, 4.43 (total 1H, each s, C4-H), 4.81, 5.49 (total
1H, each q, J = 7 Hz, C1-H), 6.87–7.65 (8H, m, ArH and
PhH), 8.00, 8.42 (total 1H, each s, CHO). HR-EIMS m/z
(M+): Calcd for C17H16FNOS: 301.0937, found: 301.0932.
5.7. Pummerer reaction of sulfoxide 5i
Method A: TFAA (1.88 ml, 13.6 mmol) was added into a
solution of 5i (1.00 g, 2.71 mmol) in benzene (50.0 ml) at
room temperature. After the mixture was stirred for 0.5 h,
BF3·Et2O (1.00 ml, 8.13 mmol) was added then the reaction
mixture was further stirred at the same temperature for 1.0 h.
The reaction mixture was diluted with H2O and extracted with
CHCl3. The residue was purified by chromatography (AcOEt/
hexane = 1:2) to give 6i (133 mg, 14%), 7i (310 mg, 44%).
5.8.2. N-[1-(4-fluorophenyl)ethyl]-N-(2-oxothyl)formamide 7j
1
A yellow gum. IR: 1735, 1662, 1604. H-NMR: 1.48, 1.64
(total 3H, each d, J = 7 Hz, CH3), 3.73, 3.76 (total 1H, each d,
J = 18 Hz, CH2CHO), 3.83, 3.89 (total 1H, each d, J = 18 Hz,
CH2CHO), 4.89 (1H, q, J = 7 Hz, CHPh), 7.00–7.10 (2H, m ,
PhH), 7.23–7.28 (2H, m, PhH), 8.12, 8.50 (total 1H, each s,
NCHO), 9.30, 9.35 (total 1H, each t, J = 1 Hz, CH2CHO).
HR-EIMS m/z (M+): Calcd for C11H12FNO2: 209.0852, found:
209.0890.
Method B: TFAA (1.88 ml, 13.6 mmol) was added to a
solution of 5i (1.00 g, 2.71 mmol) in benzene (50.0 ml) at
room temperature. After the mixture was stirred for 0.5 h,
TFSA (0.70 ml, 8.13 mmol) was added into the mixture, then
the reaction mixture was further stirred at the same conditions
for 1.0 h. The reaction mixture was diluted with H2O and ex-
tracted with CHCl3. The residue was purified by chromatogra-
phy (AcOEt/hexane = 1:2) to give 6i (599 mg, 63%).
5.9. Reductive desulfurization of 7e. Typical procedure
NaBH4 (1.87 g, 49.0 mmol) was added in small portions to
a stirred solution of 7e (700 mg, 2.35 mmol) and NiCl2·6H2O
(3.92 g, 16.5 mmol) in MeOH-THF (3:1) (50.0 ml) under ice-
cooling. The mixture was stirred at room temperature for
further 0.5 h. To the reaction mixture, H2O (10.0 ml) was
added and filtered. The filtrate was diluted with H2O and acid-
ified by 5% HCl solution, and the mixture was extracted with
CHCl3. The residue was purified by flash column chromatogra-
phy (AcOEt/hexane = 2:3) to give 8e (369 mg, 83%) and 9e
(39 mg, 9%).
5.7.1. N-formyl-1-cyclohexyl-4-phenylsulfanyl-1,2,3,4-
tetrahydroisoquinoline 6i
Colorless needles, m.p. 123–125°C, recrystallized from hex-
1
ane–AcOEt. IR: 1674, 1660. H-NMR: 0.90–1.23 (5H, m, cy-
clohexyl), 1.51–1.91 (6H, m, cyclohexyl), 3.29, 3.83 (total 1H,
each dd, J = 14, 3 Hz, C3-H), 3.58, 4.65 (total 1H, each d,
J = 14 Hz, C3-H), 4.33, 5.28 (total 1H, each d, J = 6 Hz,
C1-H), 4.39, 4.55 (total 1H, each d, J = 3 Hz, C4-H), 7.28–
7.47, 7.61–7.64 (total 9H, each m, ArH and PhH), 8.12, 8.32
(total 1H, each s, CHO). HR-EIMS m/z (M+): Calcd for
C22H25NOS: 351.1667, found: 351.1703.
5.9.1. N-formyl-1-ethyl-1,2,3,4-tetrahydroisoquinoline 8e
Colorless prisms, m.p. 63–65°C, recrystallized from hex-
ane-AcOEt. IR: 1674. 1H-NMR: 0.98 (3H, t, J = 7 Hz,
CH2CH3), 1.75–1.96 (2H, m, CH2CH3), 2.73–3.13, 3.51–
3.73, 4.44–4.51 (total 4H, each m, C4-H and C3-H), 4.35–

250
T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
4.40, 5.32–5.37 (total 1H, each m, C1-H), 7.08–7.22 (4H, m,
ArH), 8.21, 8.25 (total 1H, each s, CHO). HR-EIMS m/z (M+):
Calcd for C12H15NO: 189.1151, found: 189.1134.
5.9.7. N-formyl-1-cyclobutyl-1,2,3,4-tetrahydroisoquinoline 8h
A colorless gum. IR: 1672, 1653. 1H-NMR: 1.76–2.20 (6H,
m, cyclobutyl), 2.64–3.00, 3.13–3.22 (total 4H, each m, cyclo-
butyl and C4-H, C3-H), 3.49–3.68 (1H, m, C3-H), 4.33, 5.35
(total 1H, each d, J = 10 Hz, C1-H), 7.08–7.22 (4H, m, ArH),
8.21, 8.26 (total 1H, each s, CHO). HR-EIMS m/z (M+): Calcd
for C14H17NO: 215.1307, found: 215.1291.
5.9.2. N-formyl-1-ethyl-1,2-dihydroisoquinoline 9e
1
A pale yellow gum. IR: 1674. H-NMR: 0.84, 0.90, (total
3H, each t, J = 7 Hz, CH2CH3), 1.54–1.85 (2H, m, CH2CH3),
4.60 (dd, J = 8, 6 Hz) and 5.45 (t, J = 7 Hz): total 1H, C1-H,
5.91, 6.06 (total 1H, each d, J = 8 Hz, C4-H), 6.56, 7.06 (total
1H, d, J = 8 Hz, C3-H), 7.05–7.61 (4H, m, ArH), 8.16, 8.35
(total 1H, each s, CHO). LR-CIMS m/z: 188 (MH+).
8f (486 mg, 83%) and 9f (14 mg, 2%) were obtained from
6f (900 mg).
8i (254 mg, 73%) was obtained from 6i (500 mg).
5.9.8. N-formyl-1-cyclohexyl-1,2,3,4-tetrahydroisoquinoline 8i
1
A colorless gum. IR: 1670. H-NMR: 0.93–1.26 (5H, m,
cyclohexyl), 1.63–1.76 (6H, m, cyclohexyl), 2.94 (2H, t,
J = 7 Hz, C4-H), 3.32–3.41, 4.14–4.22 (total 1H, each m,
C3-H), 3.66–3.70, 4.13–4.20 (total 1H, each m, C3-H), 4.15,
5.14 (total 1H, each d, J = 8 Hz, C1-H), 7.08–7.36 (4H, m,
ArH), 8.18, 8.23 (total 1H, each s, CHO). HR-EIMS m/z
(M+): Calcd for C16H21NO: 243.1623, found: 243.1641.
8j (279 mg, 87%) was obtained from 6j (500 mg)
5.9.3. N-formyl-1-propyl-1,2,3,4-tetrahydroisoquinoline 8f
1
A pale yellow gum. IR: 1674. H-NMR: 0.96, 0.98 (total
3H, each t, J = 7 Hz, CH2CH2CH3), 1.30–1.51 (2H, m,
CH2CH2CH3), 1.66–1.92 (2H, m, CH2CH2CH3), 2.75–3.16,
3.52–3.73, 4.43–4.51 (total 4H, each m, C3-H and C4-H),
4.49, 5.42 (total 1H, each t, J = 7 Hz, C1-H), 7.08–7.21 (4H,
m, ArH), 8.20, 8.22 (total 1H, each s, CHO). HR-EIMS m/z
(M+): Calcd for C13H17NO: 203.1311, found: 203.1349.
5.9.9. N-formyl-6-fluoro-1-methyl-1,2,3,4-
tetrahydroisoquinoline 8j
A pale yellow oil. IR: 1616. 1H-NMR: 1.46, 1.54 (total 3H,
each d, J = 7 Hz, CH3), 2.70–3.00 (2H, m, C3-H and C4-H),
3.15, 3.53 (total 1H, each ddd, J = 13, 5, 2 Hz, C3-H), 3.70,
4.45 (total 1H, each ddd, J = 13, 6, 2 Hz, C3-H), 4.77, 5.44
(total 1H, each q, J = 7 Hz, C1-H), 6.79–7.13 (3H, m, ArH),
8.14, 8.29 (total 1H, each s, CHO).
5.9.4. N-formyl-1-propyl-1,2-dihydroisoquinoline 9f
1
A pale yellow gum. IR: 1687, 1631. H-NMR: 0.84–0.99
(3H, m, CH2CH2CH3), 1.22–1.29 (2H, m, CH2CH2CH3),
1.32–1.79 (2H, m, CH2CH2CH3), 4.69 (dd, J = 7, 6 Hz) and
5.51 (t, J = 7 Hz): total 1H, C1-H, 5.22, 6.06 (total 1H, each d,
J = 7 Hz, C4-H), 6.54, 7.06 (total 1H, d, J = 8 Hz, C3-H),
7.04–7.35 (4H, m, ArH), 8.14, 8.33 (total 1H, each s, CHO).
LR-CIMS m/z: 202 (MH+).
5.10. Hydrolysis of 8e. Typical procedure
A solution of 8e (360 mg, 1.90 mmol) in EtOH (15.0 ml)/
20% NaOH (15.0 ml) was refluxed for 3.0 h. The reaction
mixture was diluted with H2O and the mixture was extracted
with CHCl3. The product was purified by column chromato-
graphy (AcOEt) to give 1-ethyl-1,2,3,4-tetrahydroisoquino-
line 10e (274 mg, 59%) as a colorless oil. 1H-NMR: 1.01
(3H, t, J = 7 Hz, CH2CH3), 1.66–1.81 (1H, m, CH2CH3),
1.86–1.99 (1H, m, CH2CH3), 2.70–2.89 (2H , m, C4-H), 2.97
(1H, ddd, J = 12, 8, 5 Hz, C3-H), 3.24 (1H, dt, J = 12, 5 Hz,
C3-H), 3.90 (1H, dd, J = 9, 4 Hz, C1-H), 7.06–7.18 (4H, m,
ArH). 13C-NMR: 10.5 (CH2CH3), 29.0 (CH2CH3), 30.0 (C4),
41.1 (C3), 57.0 (C1), 125.7 (C7), 125.8 (C6), 126.1 (C5),
129.2 (C8), 135.2 (C8a), 139.5 (C4a). HR-FABMS m/z
(MH+): Calcd for C11H16N: 162.1282, found: 162.1295.
8g (415 mg, 71%) and 9g (32 mg, 6%) were obtained from
6g (700 mg).
5.9.5. N-formyl-1-cyclopropyl-1,2,3,4-tetrahydroisoquinoline
8g
Colorless prisms, m.p. 82–84°C, recrystallized from hex-
1
ane-AcOEt. IR: 1664. H-NMR: 0.33–0.48 (1H, m, cyclopro-
pyl), 0.60–0.80 (3H, m, cyclopropyl), 1.22–1.31 (1H, m, cy-
clopropyl), 2.86–2.97 (2H, m, C4-H), 3.31–3.40, 4.38–4.45
(total 1H, each m , C3-H), 3.72–3.76 (1H, m, C3-H), 3.79,
4.83 (total 1H, each d, J = 9 Hz, C1-H), 7.08–7.32 (4H, m,
ArH), 8.19, 8.38 (total 1H, each s, CHO). HR-EIMS m/z
(M+): Calcd for C13H15NO: 201.1151, found: 201.1119.
5.10.1. 1-Propyl-1,2,3,4-tetrahydroisoquinoline 10f
1
Yield, 99%. A pale yellow oil. H-NMR: 0.98 (3H, t, J = 7
5.9.6. N-formyl-1-cyclopropyl-1,2-dihydroisoquinoline 9g
Hz, CH2CH2CH3), 1.37–1.58 (2H, m, CH2CH2CH3), 1.65–
1.88 (2H, m, CH2CH2CH3), 2.69–2.88 (2H, m, C4-H), 2.98
(1H, ddd, J = 12, 7, 5 Hz, C3-H), 3.24 (1H, dt, J = 12, 5 Hz,
C3-H), 3.97 (1H, dd, J = 9, 4 Hz, C1-H), 7.07–7.17 (4H, m,
ArH). 13C-NMR: 14.9 (CH2CH2CH3), 19.4 (CH2CH2CH3),
30.0 (C4), 38.7 (CH2CH2CH3), 41.0 (C3), 55.5 (C1), 125.7
(C7), 125.7 (C6), 126.1 (C5), 129.2 (C8), 135.1 (C8a), 139.8
(C4a). HR-FABMS m/z (MH+): Calcd for C12H18N: 176.1439,
found: 176.1441.
1
A pale yellow gum. IR: 1685, 1631. H-NMR: 0.36–0.65
(4H, m, cyclopropyl), 1.20–1.31 (1H, m, cyclopropyl), 3.97,
5.04 (total 1H, each d, J = 8 Hz, C1-H), 5.97, 6.09 (total 1H,
each d, J = 7 Hz, C4-H), 6.62, 7.08 (total 1H, each d, J = 7 Hz,
C3-H), 7.10–7.30 (4H, m, ArH), 8.16, 8.38 (total 1H, each s,
CHO). HR-FABMS m/z (MH+): Calcd for C13H14NO:
200.1076, found: 200.1077.
8h (466 mg, 78%) was obtained from 6h (900 mg).

T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
251
5.10.2. 1-Cyclopropyl-1,2,3,4-tetrahydroisoquinoline 10g
peatedly with phosphate-buffered saline (PBS, pH 7.4) to re-
move serum and NGF, and placed in DM, DMEM without
serum (as NGF + DMEM) or DMEM without either serum or
NGF (as trophic withdrawal medium). TIQs were added at fi-
nal concentrations ranging from 10–5 to 10–3 M to trophic
withdrawal medium, and cells were cultured for 24 h.
1
Yield, 87%. A yellow oil. H-NMR: 0.27–1.25 (5H, m, cy-
clopropyl), 2.62–3.37 (5H, m, C1-H, C3-H and C4-H), 7.01–
7.51 (4H, m, ArH). 13C-NMR: 1.7 (CH2), 5.4 (CH2), 17.8
(CH), 29.7 (C4), 42.5 (C3), 61.8 (C1), 125.5 (C7), 125.9
(C6), 126.1 (C5), 128.9 (C8), 134.7 (C8a), 139.1 (C4a). LR-
CIMS m/z: 174 (MH+).
5.11.2. Determination of cell viability
5.10.3. 1-Cyclobutyl-1,2,3,4-tetrahydroisoquinoline 10h
Viable cell ratios were determined using the dye exclusion
assay. This assay is based on the exclusion of trypan blue dye
from viable cells. After 24 h treatment with TIQs, PC12 cells
were dispersed with trypsin/EDTA, then incubated with 0.1%
trypan blue for 5 min. At least 1000 cells per sample were
observed under a microscope and counted as either stained or
unstained. Viable cell ratios of NGF + DMEM and trophic
withdrawal medium were expressed as a percentage of the
DM-cultured cells’ viability. The effects of TIQs on trophic
withdrawal-cultured PC12 cells were estimated as the percen-
tage of trophic withdrawal-cultured cell viability in the absence
of any TIQ.
1
Yield, 71%. A yellow oil. H-NMR: 1.81–2.16 (6H, m, cy-
clobutyl), 2.62–2.83 (3H, m, cyclobutyl and C4-H), 2.96 (1H,
ddd, J = 12, 7, 6 Hz, C3-H), 3.23 (1H, dt, J = 12, 6 Hz, C3-H),
3.83 (1H, d, J = 9 Hz, C1-H), 7.06–7.15 (4H, m, ArH). 13C-
NMR: 18.1 (CH2), 25.8 (CH2), 27.9 (CH2), 29.9 (C4), 41.1
(C3), 41.3 (CH), 60.7 (C1), 125.3 (C7), 125.8 (C6), 125.9
(C5), 129.1 (C8), 134.7 (C8a), 138.8 (C4a). LR-CIMS m/z:
188 (MH+).
5.10.4. 1-Cyclohexyl-1,2,3,4-tetrahydroisoquinoline 10i
1
Yield, 86%. A yellow oil. H-NMR: 1.05–1.44 (6H, m, cy-
clohexyl), 1.70–1.97 (5H, m, cyclohexyl), 2.63–2.70 (1H, m,
C4-H), 2.81–2.97 (2H, m, C3-H and C4-H), 3.24–3.31 (1H, m,
C3-H), 3.90 (1H, d, J = 4 Hz, C1-H), 7.05–7.16 (4H, m, ArH).
13C-NMR: 26.4 (CH2), 26.57 (CH2), 26.62 (CH2), 27.0 (CH2),
30.3 (CH2), 30.9 (C4), 42.3 (C3), 42.9 (CH), 60.7 (C1), 125.5
(C7), 125.6 (C6), 126.0 (C5), 129.1 (C8), 136.2 (C8a) 138.5
(C4a). HR-EIMS m/z (M+): Calcd for C15H21N: 215.1709,
found: 215.1692.
5.11.3. Measurement of neurite outgrowth
Cells bearing neurite-like processes longer than the cell
body diameter (measured at the widest point), or bearing a
neurite with a length at least double that of the cell diameter,
were counted as differentiated. Data were expressed as the ratio
of neurite-bearing cells to total cells counted. Cell aggregates
were not scored, and cells with more than one neurite were
only counted once. Estimates of the effects of growth medium
and TIQs on neurite bearing were expressed in the same man-
ner as “Determination of cell viability”, described above.
5.10.5. 6-Fluoro-1-methyl-1,2,3,4-tetrahydroisoquinoline 10j
1
Yield, 91%. A colorless oil. H-NMR: 1.44 (3H, d, J = 7
Hz, CH3), 2.72 (1H, dt, J = 16, 5 Hz, C4-H), 2.86 (1H, m, C4-
H), 3.00 (1H, ddd, J = 13, 9, 5 Hz, C3-H), 3.50 (1H, dt, J = 13,
5 Hz, C3-H), 4.06 (1H, q, J = 7 Hz, C1-H), 6.77 (1H, dd,
J = 10, 3 Hz, C8-H), 6.84 (1H, td, J = 9, 3 Hz, C7-H), 7.09
(1H, dd, J = 9, 6Hz, C6-H). 13C-NMR: 22.6 (CH3), 30.0 (C4),
41.4 (C3), 51.1 (C1), 115.0 (C5 or C7), 127.7 (C5 or C7),
127.2 (C8), 136.0 (C8a), 136.8 (C4a), 160.8 (C6).
5.11.4. Comet assay
The reagents and equipment for the comet assay were pur-
chased from TREVIGEN (Gaithersburg, MD, USA). PC12
cells were incubated with the test substances at 37°C for 3 h.
After exposure, cells were washed with PBS(-), then suspended
(1×105 cells/100 μl PBS(–)); 50 μl of this suspension was
mixed with 500 μl of 1% low-melting point agarose. The mix-
ture (75 μl aliquots) was applied to the comet slides, and the
slides were kept horizontal for 10 min at 4°C. The slides were
then treated with Lysis solution (2.5 M NaCl, 100 mM EDTA,
10 mM Tris base, 1% sodium laurylsarcosinate, and 1% Triton
X-100) for 1 h at 4°C. The slides were immersed in alkaline
solution (300 mM NaOH and 1 mM EDTA) at room tempera-
ture for 30 min, followed by two immersions in TBE buffer for
5 min. Electrophoresis was conducted at 15 V at room tem-
perature for 30 min, then the slides were dipped in 70% EtOH
for 5 min. After drying, DNA on the slides was stained with
SYBR Green (0.1 μl/ml) solution.
5.11. Biological experiments
5.11.1. Cell culture
PC12 cells were obtained from the American Tissue Type
Culture Collection (reference no. CRL 1721). Dulbecco’s mod-
ified Eagle’s medium (DMEM), horse serum and fetal bovine
serum (FBS) were purchased from GIBCO (Grand Island, NY,
USA). Mouse nerve growth factor (2.5S NGF) was from Ta-
kara Shuzo Co. Ltd. (Shiga, Japan). PC12 cells were routinely
maintained in DMEM containing 15% heat-inactivated serum
(10% FBS, 5% horse serum) under an atmosphere containing
5% CO2 at 37°C. The medium was changed every 2–3 days.
The cells were transferred to 24-well plates and cultured for 2
days. These cells were neuronally differentiated for 5 days in
DMEM containing 7% heat-inactivated serum (2% FBS and
1% horse serum) supplemented with 50 ng/ml 2.5S NGF (dif-
ferentiation medium, DM). Thereafter, cells were washed re-
The comet image was observed through a fluorescence mi-
croscope at × 200 magnification and at 568 nm. Routinely, 250
cells (50 cells/slide) were screened per sample. In selecting
cells for measurement, straight line scanning of a slide was be-
gun at an arbitrary point and cells were measured as they came
into the field, provided there was no overlap with patterns from

252
T. Saitoh et al. / European Journal of Medicinal Chemistry 41 (2006) 241–252
[5] S. Ohta, M. Kohno, Y. Makino, O. Tachikawa, M. Hirobe, Biomed. Res.
8 (1987) 453–456.
[6] K. Abe, K. Taguchi, T. Wasai, J. Ren, I. Utsunomiya, T. Shinohara, T.
Miyatake, T. Sano, Brain Res. Bull. 56 (2001) 55–60.
[7] Y. Kotake, R. Taguchi, K. Okuda, K. Sekiya, Y. Tasaki, M. Hirobe, S.
Ohta, Brain Res. 1033 (2005) 143–150.
[8] K. Ishiwata, Y. Koyanagi, K. Abe, K. Kawamura, K. Taguchi, T. Saitoh,
J. Toda, M. Senda, T. Sano, J. Neurochem. 79 (2001) 868–876.
[9] T. Shinohara, A. Takeda, J. Toda, T. Sano, Heterocycles 46 (1997) 555–
565.
other cells. The length and width of the DNA mass was mea-
sured using an ocular micrometer disk. In the present study, the
length: width ratios of the DNA mass and the frequency of
cells with prominent tails (i.e., a small portion of the DNA
remains at the origin, but most of the DNA has fragmented
and migrated to form a ‘tail’) were used for estimating the cy-
totoxicity of the TIQ derivatives.
5.11.5. Data analysis
[10] T. Shinohara, J. Toda, T. Sano, Chem. Pharm. Bull. (Tokyo) 45 (1997)
813–819.
[11] T. Shinohara, A. Takeda, J. Toda, Y. Ueda, M. Kohno, T. Sano, Chem.
Pharm. Bull. (Tokyo) 46 (1998) 917–918.
[12] T. Shinohara, A. Takeda, J. Toda, T. Sano, Chem. Pharm. Bull. (Tokyo)
46 (1998) 430–433.
[13] J. Toda, S. Matsumoto, S. Saitoh, T. Sano, Chem. Pharm. Bull. (Tokyo)
48 (2000) 91–98.
Data were expressed as mean standard error (S.E.) for
each group. The significance of difference was analyzed by
Dunnett’s multiple comparison test or Tukey–Kramer’s multi-
ple comparison test. A value of P < 0.05 was regarded as sta-
tistically significant.
[14] J. Toda, M. Sakagami, T. Sano, Chem. Pharm. Bull. (Tokyo) 47 (1999)
1269–1275.
Acknowledgements
[15] J. Toda, T. Ichikawa, T. Sithoh, Y. Horiguchi, T. Sano, Heterocycles 53
(2000) 2009–2018.
[16] T. Saitoh, T. Ichikawa, Y. Horiguchi, J. Toda, T. Sano, Chem. Pharm.
Bull. (Tokyo) 49 (2001) 979–984.
[17] Y. Horiguchi, T. Saitoh, S. Terakado, K. Honda, T. Kimura, J. Toda, T.
Sano, Heterocycles 54 (2001) 967–987.
[18] J. Toda, M. Sakagami, Y. Goan, M. Shimakata, T. Saitoh, Y. Horiguchi,
T. Sano, Chem. Pharm. Bull. (Tokyo) 48 (2000) 1854–1861.
[19] J. Toda, Y. Niimura, K. Takeda, T. Sano, Y. Tsuda, Chem. Pharm. Bull.
(Tokyo) 46 (1998) 906–912.
The work was supported in part by the Kimi Imai Memorial
Foundation for Research of Incurable Neuromuscular Diseases
(T.S.).
References
[1] K. Abe, T. Saitoh, Y. Horiguchi, I. Utsunomiya, K. Taguchi, Biol.
Pharm. Bull. 28 (2005) 1355–1362.
[20] J. Toda, Y. Niimura, T. Sano, Y. Tsuda, Heterocycles 48 (1998) 1599–
1607.
[2] Y. Kotake, Y. Tasaki, Y. Makino, S. Ohta, M. Hirobe, J. Neurochem. 65
(1995) 2633–2638.
[3] Y. Kotake, M. Yoshida, M. Ogawa, Y. Tasaki, M. Hirobe, S. Ohta, Neu-
rosci. Lett. 217 (1996) 69–71.
[21] T. Shinohara, A. Takeda, J. Toda, T. Sano, Heterocycles 48 (1997) 981–
992.
[4] Y. Tasaki, Y. Makino, S. Ohta, M. Hirobe, J. Neurochem. 57 (1991)
1940–1943.
[22] Y. Horiguchi, K. Ogawa, T. Saitoh, T. Sano, Chem. Pharm. Bull. (To-
kyo) 52 (2004) 214–220.