Preparation of secolycorines against acetylcholinesterase

Article abstract of DOI:10.1016/j.bmc.2006.10.026

5,6-Secolycorines possessing a 5,6-dihydrophenanthridine skeleton were facilely prepared from lycorine through chemical transformations, mainly including N-alkylation, Hofmann degradation type reaction, reductive cleavage of trichloroethylcarbonyl moiety, and hydrogenation. Several secolycorine derivatives showed potent inhibitory activity against acetylcholinesterase with the IC₅₀ value at micromolar range and are more potent than galanthamine.

Full text of DOI:10.1016/j.bmc.2006.10.026

Bioorganic & Medicinal Chemistry 15 (2007) 1034–1043
Preparation of secolycorines against acetylcholinesterase
Shoei-Sheng Lee,a,* Uppala Venkatesham,a Chitneni Prasad Rao,a
Sio-Hong Lama and Jung-Hsin Lina,b,c
aSchool of Pharmacy, College of Medicine, National Taiwan University, 1 Jen-Ai Rd., Sec. 1, Taipei 100, Taiwan, ROC
bDivision of Mechanics, Research Center for Applied Sciences, Academia Sinica, Taipei 115, Taiwan, ROC
cInstitute for Biomedical Sciences, Academia Sinica, Taipei 115, Taiwan, ROC
Received 1 July 2006; revised 12 October 2006; accepted 14 October 2006
Available online 18 October 2006
Abstract—5,6-Secolycorines possessing a 5,6-dihydrophenanthridine skeleton were facilely prepared from lycorine through chemical
transformations, mainly including N-alkylation, Hofmann degradation type reaction, reductive cleavage of trichloroethylcarbonyl
moiety, and hydrogenation. Several secolycorine derivatives showed potent inhibitory activity against acetylcholinesterase with the
IC50 value at micromolar range and are more potent than galanthamine.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
450 lM) against AChE.10 Assoanine was found to be
the most active AChE inhibitor (IC50 3.87 lM) among
more than 20 lycorine related alkaloids.11,12 Its higher
activity was explained by the presence of an aromatic
ring C, providing a planar-like skeleton.11,12 Whether
the intact D-ring is essential for anti-AChE activity,
however, had not been reported yet. Based upon these,
we planned to prepare a series of N-alkyl-5,6-secolyco-
rine derivatives, possessing a planar-like dihydrophe-
nanthridine moiety devoid of D-ring, and evaluate
their anti-AChE activity. Since lycorine (1) possesses a
skeleton similar to the target designed and was found
to be a major alkaloid in Crinum asiaticum var sinicum
in our preliminary study, it was chosen as starting mate-
rial in this study. The following describes the outcome of
this effort.
Alzheimer’s disease (AD) is a progressive, degenerative,
and irreversible disorder that causes intellectual impair-
ment and cognitive dysfunction.1 In the last decade,
treatment for AD has been based on the ‘cholinergic
hypothesis’.2 This hypothesis suggested that patients
with AD suffer from a deficit of cholinergic function in
the brain such as decrease in hippocampal and cortical
levels of the acetylcholine (ACh) and associated enzyme
choline transferase. Inhibition of acetylcholinesterase
(AChE), an enzyme responsible for the metabolic break-
down of ACh, can restore the level of ACh in the brain.3
Nowadays, four AChE inhibitors, including tacrine,
donepezil, rivastigmine, and galantamine, are used clin-
ically for the treatment of mild to moderate disease
stage, which is a short interval in the course of AD.4
Hence the development of drugs for more serious AD
is essential. Lycorine (1) is the first Amaryllidaceous
alkaloid isolated in 1877 from Narcissus pseudonarcis-
sus.5 It has a tetracyclic pyrrole[d,e]phenanthridine (gal-
anthan) skeleton and has been demonstrated to possess
several biological activities, such as inhibition of protein
and DNA synthesis,6 cell growth and division,7 and
antiviral activity.8,9 It also displays weak activity (IC50
2. Results and discussion
Lycorine (1) was facilely isolated from the MeOH ex-
tract of C. asiaticum var sinicum, simply by filtering
the precipitate obtained during alkalinization of the acid
soluble fraction as described in Section 5. Peracetylation
of 1 by Ac2O/py yielded diacetyllycorine (2),13 dAc–Me
2.00 and 2.20. N-Alkylation of 2 with various alkyl ha-
lides (CnH2n+1X, n = 1–6, X = I or Br) under refluxing
CH3CN gave the corresponding N-alkyl quaternary
ammonium salts 3a–f as diastereomers in excellent yields
(Table 1). While reacting with iodomethane under
refluxing CHCl3, 3a was obtained in a poor yield. Sever-
Keywords: Lycorine; Chemical modification; Secolycorines; Anti-
acetylcholinesterase.
*
Corresponding author. Tel./fax: +886 2 23916127; e-mail: shoeilee@
ha.mc.ntu.edu.tw
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.10.026

S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
1035
7.22 (1H, dd, J = 17.8, 11.0 Hz). The 1H NMR data
incorporating with the high resolution EI-MS, affording
a molecular formula C17H15NO2, established 4a as 5-
methyl-8,9-methylenedioxy-4-vinyl-5,6-dihydrophenan-
thridine. Hence, under such reaction conditions, not
only the Hofmann degradation but also elimination of
two acetoxyl groups was undertaken. Under similar
conditions, compounds 4b–f were obtained from 3b–f,
respectively, in good yields (72–88%) (Scheme 1). The
reason why another series of the Hofmann degradation
products, that is, breakdown between C-13 and N,6 were
not obtained could be attributable to the reaction order.
That is the elimination step took place first to form the
dihydrophenanthridine (40), followed by the Hofmann
degradation to yield 4a–f (Scheme 2). In addition, the
C-13, N6-seco products 400 could not be isolated because
their skeleton was highly strained, making them decom-
posed easily. Respective catalytic hydrogenation of 4a–f
(H2/Pd–C) yielded the corresponding 5-alkyl-4-ethyl-
Table 1. Synthesis of diastereomeric mixtures of N-alkyl quaternary
ammonium salts 3a–f, and yields (%) of secolycorines 4a–f and 5a–f
Compounda
Reaction conditions for
preparing 3a–f
Yield
Temp. (ꢁC) RXb
Sub 4c,d 5 (h)e
1
2
3
4
5
6
40
70
70
82
82
82
CH3I
C2H5I
a
b
c
d
e
f
50
76
75
72
88
84
66 (3.5)
54 (7)
55 (4)
n-C3H7Br
n-C4H9Br
n-C5H11Br
n-C6H13Br
61 (16)
52 (17)
60 (24)
a The reactions for preparing 3a–b, d–f were carried out in MeCN for
24 h while for 3c 36 h.
b The low boiling reagents (methyl, ethyl, and propyl halides) were
added in excess with frequent intervals.
c Two-step yields from 2 to 4a–f.
d Reactions were carried out for 3.5–4 h both for the conversion of 2–3
and 3–4.
e Compounds 4b–c were reacted in cyclohexane and the rest were run
in acetone.
8,9-methylenedioxy-5,6-dihydrophenanthridines
5a–f
1
(Scheme 1). The H NMR spectrum of 5a showed sig-
nals for 4-ethyl group at d 1.28 (3H, t) and 2.77 (2H,
q), and N-Me at d 2.45 (s), and the 13C NMR showed
signals of 4-ethyl group at d 14.8 (q, C-12) and 23.2 (t,
C-11), while those of C-4a and C-6 at d 145.5 (s) and
55.3 (t).
al attempts to run the Hofmann degradation of 3a–f
were made using various conditions. It was found that
the optimal reaction conditions were using potassi-
um-tbutoxide as base and under refluxing tbutanol
(Table 1). Thus 3a afforded 4a (50%). Relative to that
1
of 2, the H NMR spectrum of 4a exhibited additional
For comparison of anti-AChE activity with compounds
4a–f and 5a–f, the preparation of secolycorine deriva-
tives with a partially saturated (8-10) and fully saturated
ring C (14–16) was carried out (Schemes 3 and 4). Treat-
two AMX systems, one in the aromatic region appear-
ing at d 7.55 (1H, dd, J = 7.7, 1.3 Hz), 7.44 (1H, dd,
J = 7.7, 1.3 Hz), 7.13 (1H, t, J = 7.7 Hz), and the other
in the olefinic region appearing at d 5.29 (1H, dd,
J = 11.0, 1.5 Hz), 5.72 (1H, dd, J = 17.8, 1.5 Hz), and
ment of
(TCECF) in the presence of potassium carbonate under
2
with 2,2,2-trichloroethylchloroformate
OH
C
OAc
OAc
3
12
1
HO
AcO
AcO
11
C
MeO
D
O
O
O
O
O
O
a
b
A
B
13
B
D
N
A
N
N
N+
R
MeO
X-
5
7
8
Assoanine
1
2
3a-f
R
X
CH3
I
I
a
b
c
d
e
f
3
1
10
7
n-C2H5
n-C3H7
n-C4H9
n-C5H11
n-C6H13
c
d
O
O
O
11
12
10b
4a
Br
Br
Br
Br
N
N
O
R
R
6
5a-f
4a-f
Scheme 1. Reagents and conditions: (a) Ac2O, py, 50 ꢁC, 13 h, 96%; (b) RX, CH3CN, 24 h, reflux, 99%; (c) tBuOK, tBuOH, reflux, 4 h, 50–88%; (d)
H2, 10% Pd–C, acetone or cyclohexane, rt, 16 h, 52–66%.
OAc
OAc
AcO
AcO
c
O
O
O
O
O
O
c
O
O
N+
R
N
N+
N
R
X-
X-
R
R
3a-f
4a-f
4"
4'
t
Scheme 2. Reagents and conditions: (c) BuOK, BuOH, reflux, 4 h.
t

1036
S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
OR
OAc
N
OAc
OAc
NH
2
RO
1
AcO
AcO
AcO
3
10
10b
4
a, b
O
O
O
O
O
O
11
12
OR
O
O
c
d, e
4a
N
N
O
6 R= Cl
7 R= OAc
R
CCl3
OAc
CH3
6
7
O
2
9 R= Ac
10 R= H
8
Scheme 3. Reagents and conditions: (a) ClCO2CH2CCl3, K2CO3, dry toluene, reflux, 14 h; (b) NaOAc, DMSO, 105 ꢁC, 58%; (c) Zn/90% AcOH,
50 ꢁC, 6 h, 58%; (d) HCHO, NaBH3CN, HOAc–MeOH, rt, 2 h, 90%; (e) K2CO3, MeOH–H2O 9:1, 2 h, 72%.
OAc
OAc
N
OAc
3
AcO
AcO
AcO
O
O
13
b, c
O
O
O
O
d
a
12
7
N
N
R
CCl3
6
5
O
O
2
12 R= Cl
13 R= OAc
11
OH
OAc
OAc
NH
2
1
HO
AcO
AcO
10
10b
11
12
OH
O
O
e
O
O
O
O
f
N4a
N
OAc
OAc
CH3
CH3
6
7
15
16
14
Scheme 4. Reagents and conditions: (a) H2 (300 psi)/10% Pd–C, AcOH, rt, 18 h, 60%; (b) ClCO2CH2CCl3, K2CO3, dry toluene, reflux, 14 h, 57%; (c)
NaOAc, DMSO, 105 ꢁC, 6.5 h, 94%; (d) Zn/90% AcOH, 50ꢁC, 4 h, 43%; (e) HCHO/NaBH3CN, AcOH–MeOH, rt, 2 h, 100%; (f) K2CO3, MeOH–
H2O 9:1, 2 h, 90%.
refluxing toluene yielded a major secolycorine carba-
mate 6, obtained by cleaving C5–N6 bond in the strained
ring D. That a reductive cleavage of the N-protecting
group in 6 with zinc dust in 90% HOAc regenerated 2
suggested the spacious vicinity between N6 and the C-4
chloroethyl group as depicted in 6. Nucleophilic dis-
placement of the halide by an acetoxyl group in 6 with
NaOAc in DMSO, affording 7, followed by a reductive
cleavage of the N-protecting group with freshly pre-
pared zinc dust in 90% AcOH, yielded 5-acetoxy-diace-
tyl-5,6-secolycorine (8) (58%), whose 1H NMR spectrum
showed signals of three Ac–Me (d 1.89, 1.98, and 2.04),
H-4a (d 3.43, d, J = 10.0 Hz), and H-6s (d 3.98 and 4.07,
each d, Jgem = 16.3 Hz). Reductive N-methylation of 8
synthesis of 10 (Scheme 4). Treatment of 11 with
TCECF, followed by NaOAc-DMSO and reductive
cleavage of the N-protecting group (Zn–90% HOAc),
1
yielded 14 (43%), whose H NMR spectrum (CDCl3)
showed the signals of H-4a, H-6, and H-10b at d 3.26
(dd), 3.95 (d) and 4.08 (d) (Jgem 15.6 Hz) and 3.19 (br
d), respectively. Reductive N-methylation of 14 yielded
15 (ꢀ100%), whose 1H NMR spectrum (CDCl3) dis-
played the signals of N-Me, H-4a, H-6, and H-10b at
d 2.29 (s), 2.47 (dd), 3.35 and 3.73 (Jgem 14.4 Hz) and
3.25 (br d), respectively. O-Deacetylation under alkaline
conditions gave 16 (90%), whose 1H NMR spectrum
(CD3OD) exhibited the signals of N-Me, H-4a, H-6,
and H-10b at d 2.35 (s), 2.66 (dd), 3.41 (d) and 3.72
(d) (Jgem 14.7 Hz), and 3.27 (br d), respectively. The high
resolution EI-MS of 16, affording a molecular formula
C17H23NO5, was also supportive for this structure
assignment.
1
(HCHO/NaBH3CN) yielded 9 (90%), whose H NMR
spectrum (CDCl3) showed signals of N-Me (d 2.16),
H-4a (d 3.62, d, J = 10.7 Hz) and H-6s (d 3.57 and
4.28, both doublet, J = 16.0 Hz). Subsequent O-deacety-
lation of 9 (K2CO3/MeOHaq) gave the desired 5-hy-
droxy-N-methyl-5,6-secolycorine (10) (72%). The 1H
NMR spectrum of 10 (CD3OD) showed the signals of
H-4a (d 3.69, d, J = 11.0 Hz), H-6s (d 3.62/4.21, both
d, J = 16.6 Hz), and H-12s (d 3.66, 2H, m). Its high res-
3. Bioactivity
The anti-AChE activity of these prepared 5,6-secolyco-
rine derivatives was evaluated by in vitro AChE inhibi-
tion assay, modified from Ellman’s method.14,15 The
results were expressed as IC50 values and are summa-
rized in Table 2. From these data, four conclusions are
drawn. First, the cleavage of ring D does not decrease
the anti-AChE activity. Instead, some advantages of
such modification, including enhancing both potency
olution EI-MS, affording
a
molecular formula
C17H21NO5, was also supportive for this structure
assignment for 10.
Starting from diacetyldihydrolycorine (11), obtained by
catalytic hydrogenation of 2, the 5,6-seco-dihydrolyco-
rine 16 was prepared by a similar procedure as for the

S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
1037
Table 2. Inhibitory effect of secolycorine derivatives against EelAChE (IC50, lM) and the calculated binding free energies (DGcalc, kcal/mol) of
compounds 4 and 5 series to EelAChE
a
a
a
Compound
IC50
Compound
IC50
DGcalc
Compound
IC50
DGcalc
1
3a
ndb
nd
4a
4b
4c
4d
4e
4f
2.89
8.35
10.59
2.66
<5
ꢁ9.36
ꢁ8.91
5a
5b
5c
5d
5e
5f
3.00
8.82
9.98
5.44
ꢁ9.28
ꢁ9.28
9
10
nd
nd
ꢁ10.67
ꢁ10.30
ꢁ11.80
ꢁ12.51
ꢁ11.58
ꢁ11.54
ꢁ11.30
ꢁ10.10
15
16
nd
nd
<5
<5
<5
Galanthamine
3.18
a The IC50 values were calculated from the dose–response curve of six concentrations of each test compound in triplicate.
b No significant anti-AChE activity was observed at 10 lM.
and solubility relative to those of assoanine, are ob-
served. For example, 4-ethyl-N-methyl-dihydrophenan-
thridine (4a) (IC50 2.89 lM; for galanthamine IC50
3.18 lM) is more potent than the reported assoanine
(IC50 3.87 lM; for galanthamine IC50 1.07 lM).11 Sec-
ond, the length of N-substitution in dihydrophenanthri-
dines of series 4 and 5 is critical to the activity. N-Methyl
or N-butyl substitution seems to be optimal for anti-
AChE activity, which decreases two to four times for
N-ethyl (4b, 5b) or N-propyl derivatives (4c and 5c).
With longer side chain (N-pentyl 4e/5e and N-hexyl 4f/
5f), the bioassay, however, was hampered by the poor
solubility in the test solution, even as the tartaric acid
salts. From the saturated solution, the IC50 values of
4e and 4f were estimated to be less than 5 lM. Third,
although the structures of 4-vinyl-5,6-dihydrophenan-
thridines (4a–f) are more planar than those of the corre-
sponding 4-ethyl derivatives (5a–f), their anti-AChE
activities seem to show no significant difference. Fourth,
the increase of solubility with nonplanar ring C causes
the dramatic loss of the anti-AChE activity as that ob-
served for 5-hydroxy-5,6-secolycorine (10) and 5-hy-
tion can be found for the effect of different lengths for
the N-alkyl chain of the compounds 4a–f and 5a–f. Ini-
tially, the binding affinity decreases by increasing the N-
alkyl chain length, but then increases again with even
longer alkyl chain, although the predicted alkyl chain
length for the minimum of binding affinity has been off-
set from that of experimental results. This trend is iden-
tical for both series of compounds, indicating that there
is a second binding pocket just adjacent to that of the
active site pocket. The second binding pocket should
be mainly hydrophobic, while the first binding pocket
should be more hydrophilic. On the other hand, based
on the analysis using LIGPLOT,18 it was also suggested
that number of residues having hydrophobic contacts
with the ligand can be considered as a good measure
for prediction of binding affinities for compound 4 ser-
ies. For example, the number of residues of AChE in-
volved in the hydrophobic contacts with compound 4a
and compound 4d are 12 and 11, respectively, while
those for compound is 4b and 4c it is only 8 and 6
(Fig. 1), respectively, correlating very well with the
reduction of their binding affinities. Although the IC50
values of compounds 4e, 4f, 5e, and 5f cannot be deter-
mined due to poor solubility, the incomplete inhibitory
assays did indicate that their potencies should be at least
as good as the best compound in each series. This sug-
gests that hydrophobic interaction plays a key role for
binding these compounds to AChE, and therefore the
inhibitory effect of the these compounds. It should be
noted that the N-alkyl chain needs be tailored to an opti-
mal length in order to make a balance between the bind-
ing affinity and solubility.
droxy-5,6-seco-dihydrolycorine (16). Compounds
9
and 15, the peracetyl derivatives of 10 and 16, are much
less polar but more bulky relative to the parent com-
pounds. They did not show significant anti-AChE activ-
ity at 10 lM level similar to that observed for 10 and 16.
Accordingly, steric effect might be another factor to be
clarified concerning anti-AChE activity.
4. Docking calculations
In order to elucidate the relationships between activities
and the structures of these compounds, we performed
docking calculations using AutoDock 3.0516 and ME-
Dock.17 The IC50 values from the enzyme assay and
the calculated binding free energies, obtained from
docking of active compounds to an analogy to the eel
AChE (EelAChE) (for details, see Section 5), are listed
in Table 1. The IC50 values of some compounds cannot
be determined due to poor solubility, which is a natural
result of long hydrophobic moieties of these com-
pounds. Due to the sequence mismatch between Eel-
AChE and mAChE (mouse AChE), as well as the
accuracy of the AutoDock scoring function, the predict-
ed binding free energy does not match the experimental
values on a number-by-number basis. However, an
interesting correlation between experiments and simula-
5. Experimental
5.1. General
All reagents were used as purchased from the commercial
suppliers without further purification. NMR spectra
were recorded on a Bruker Avance 400 NMR spectrom-
eter. Chemical shifts are reported in d ppm referenced to
CDCl3 for 1H NMR (d 7.26) and for 13C NMR (d 77.0).
Infrared spectra were recorded on Jasco FT/IR-410 spec-
trometer and only the characteristic peaks were quoted.
EI mass spectra were recorded on a Finnigan TSQ-700
mass spectrometer (70 eV) and HRMS spectra were
recorded on a JEOL SX-102A mass spectrometer. Silica
gel TLC was performed on aluminum sheet-backed TLC

1038
S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
Figure 1. The diagrammatic representation of the interactions between 4 and EelAChE by LIGPLOT.18
plates (Merck Kieselgel 60F254), visualized under UV
254 nm and Dragendorf’s spray agent. Flash chromatog-
raphy was performed on silica gel (230–400 mesh).
dride (9.5 mL) in a sealed flask was stirred at 50 ꢁC for
18 h, then abs. ethanol (25 mL) was added and was stir-
red for 3 h at room temperature. Condensation with the
aid of toluene (4· 25 mL) afforded diacetyllycorine (2)
(3.89 g, yield 96%),13 Rf = 0.72 [CHCl3–MeOH (94:6)].
5.2. Isolation of lycorine (1)
The dry powdered leaves (2.84 kg) of C. asiaticum var. sin-
icum, collected in May 2004 in Pin-Lin area, Taipei Coun-
ty, Taiwan, were percolated with 95% EtOH (10 L · 5) at
room temperature. Evaporation under reduced pressure
gave an EtOH extract (1.02 kg) which was triturated with
2% HOAc(aq) (500 mL · 3). The combined acidic solu-
tions were extracted with EtOAc (1.2 L · 3) to give an
EtOAc soluble fraction (13.63 g). The aqueous layer
was adjusted to pH 9 with 25% ammonia water, and the
crystalline lycorine (1, 13.70 g), Rf = 0.42 [CHCl3–
MeOH–NH4OH (85:15:0.1)], was collected by filtration.
5.3.2. General procedure for preparing diacetyllycorine
N-alkyl salt (3a–f). To the solution of diacetyllycorine
(2, 25 mg, 0.067 mmol) in acetonitrile (2 mL, LC grade)
was added alkyl halide (0.5 mL, excess) via cannula. The
reaction mixture was stirred under N2 and heating
conditions (see Table 1). After completion of the
reaction, the reaction mixture was concentrated and
the residue (3a–f) without further purification was
subjected to Hofmann degradation reaction.
5.3.3. General procedure for preparing 5-alkyl-8,9-methy-
lenedioxy-4-vinyl-5,6-dihydrophenanthridine (4a–f). To
one of compounds 3a–f (1.5 mmol) dissolved in tert-
BuOH (5 mL) was added potassium tert-butoxide
(10.5 mmol, 7 equiv). The reaction mixture was refluxed
under nitrogen. After the reaction was complete, water
(50 mL) was added and the pH value was adjusted to 9
5.3. Preparation of 5-alkyl-8,9-methylenedioxy-4-vinyl-
5,6-dihydrophenanthridine (4a–f)
5.3.1. Preparation of diacetyllycorine (2). The mixture of
1 (3.17 g, 11 mmol), pyridine (8.0 mL), and acetic anhy-

S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
1039
with saturated NH4Cl solution. The resultant solution
was partitioned with chloroform (3· 50 mL). The organ-
ic layer was dried over anhyd Na2SO4, concentrated to
give a residue, which was subjected to silica gel flash col-
umn chromatography (FCC), eluted with hexane–CHCl3
(25:75) to afford the corresponding secolycorine 4.
H-30), 1.49 (2H, m, H-20), 2.65 (2H, m, H-10); other 13C
NMR (CDCl3) d 11.4 (q, C-30), 21.3 (t C-20), 55.1 (t, C-
10); EIMS m/z [M]+ 293 (90), [MꢁH]+ 292 (73), 264 (M+-
29) (88), 250 (M+-43) (100); HREIMS m/z [M]+ 293.1409
(calcd for C19H19NO2, 293.1410).
5.3.4.5. 5-Butyl-8,9-methylenedioxy-4-vinyl-5,6-dihydrophe-
nanthridine (4d). Yield 72%; Rf = 0.54, hexane–CHCl3 (4:6);
5.3.4. General procedure for preparing 5-alkyl-4-ethyl-
8,9-methylenedioxy-5,6-dihydro-phenanthridine
1
(5a–f).
other H NMR (CDCl3) d 0.85 (3H, t, J = 7.3 Hz, H-40),
To compound 4 (0.127 mmol) dissolved in a suitable sol-
vent was added 10% Pd/C (20% w/w). The suspension
was degassed under vacuum and then flushed with H2
gas. This procedure was repeated three times and then
the reaction was undertaken under H2 (1 atm) at room
temperature overnight. The reaction mixture was then
filtered through a Celite pad and concentrated to give
a residue, which was subjected to silica gel FCC, eluted
with hexane–EtOAc (1:1), to give 5.
1.25 (2H, m, H-30), 1.46 (2H, m, H-20), 2.65 (2H, m, H-10);
13
other C NMR (CDCl3) d 14.0 (q, C-40), 20.2 (t, C-30),
30.4 (t, C-20), 52.8 (t, C-10); EIMS m/z [M]+ 307 (72),
[MꢁH]+ 306 (53), 264 (M+–C3H7) (91), 250 (M+–C4H9)
(100); HREIMS m/z [M]+ 307.1570 (calcd for C20H21NO2,
307.1567).
5.3.4.6. 8,9-Methylenedioxy-5-penty-4-vinyl-5,6-dihydro-
phenanthridine (4e). Yield 88%; Rf = 0.53, hexane–CHCl3
1
(4:6); other H NMR (400 MHz; CDCl3) d 0.84 (3H, t,
5.3.4.1. 5-Methyl-8,9-methylenedioxy-4-vinyl-5,6-
dihydrophenanthridine (4a). Yield 50%; Rf = 0.38,
hexane–CHCl3 (4:6); H NMR (CDCl3) d 2.49 (3H, s,
J = 7.2 Hz, H-50), 1.19 (4H, m, H-30 and 40), 1.49 (2H, m,
H-20), 2.63 (2H, m, H-10); other 13C NMR (CDCl3) d 14.0
(q, C-50), 22.5 (t, C-40), 27.9 (t, C-30), 29.1 (t, C-20), 53.0 (t,
C-10); EIMS m/z [M]+ 321 (52), [MꢁH]+ 320 (41), 264
(M+-57) (79), 250 (M+-71) (100); HREIMS m/z [M]+
321.1714 (calcd for C21H23NO2, 321.1723), [MꢁH]+
320.1648 (calcd for C21H22NO2, 320.1645).
1
N-Me), 4.01 (2H, s, H-6), 5.29 (1H, dd, J = 11.0,
1.5 Hz, H-12E), 5.73 (1H, dd, J = 17.8, 1.5 Hz, H-12Z),
5.96 (2H, s, –OCH2O–), 6.69 (1H, s, H-7), 7.13 (1H, t-
like, J = 7.7 Hz, H-2), 7.21 (1H, s, H-10), 7.22 (1H,
dd, J = 17.8, 11.0 Hz, H-11), 7.44 (1H, dd, J = 7.7,
1.3 Hz, H-3), 7.55 (1H, dd, J = 7.7, 1.3 Hz, H-1); 13C
NMR (CDCl3) d 41.6 (q, N-Me), 54.9 (t, C-6), 101.0
(t, –OCH2O–), 103.7 (d), 107.1 (d), 114.3 (t, C-12),
122.7 (d), 124.3 (d), 124.9 (d), 125.9 (s), 126.5 (s),
129.3 (s), 133.2 (s), 133.5 (d, C-11), 145.2 (s, C-4a),
147.3 (s) and 147.4 (s) (C-8 and C-9); EIMS m/z [M]+
265 (89), [MꢁH]+ 264 (100), 250 (M+-15) (14); HREIMS
m/z [M]+ 265.1083 (calcd for C17H15NO2, 265.1097).
5.3.4.7. 5-Hexyl-8,9-methylenedioxy-4-vinyl-5,6-dihydro-
phenanthridine (4f). Yield 84%; Rf = 0.53, hexane–CHCl3
1
(4:6); other H NMR (CDCl3) d 0.84 (3H, t, J = 6.8 Hz,
H-60), 1.24 (6H, m, H-30, 40 and 50), 1.47 (2H, m, H-20),
2.63 (2H, t, J = 7.5 Hz, H-10); other 13C NMR (CDCl3) d
14.0 (q, C-60), 22.6 (t, C-50), 26.6 (t, C-40), 28.2 (t, C-30),
31.7 (t, C-20), 53.1 (t, C-10); EIMS m/z [M]+ 335 (60),
[MꢁH]+ 334 (48), 264 (M+-71) (87), 250 (M+-85) (100);
HREIMS: m/z [M]+ 335.1874 (calcd for C22H25NO2,
335.1880).
5.3.4.2. Common 1H and 13C NMR data for 5-alkyl-
8,9-methylenedioxy-4-vinyl-5,6-dihydrophenanthridine
(4b–f). 1H NMR (CDCl3) d 4.02–4.04 (2H, s, H-6), 5.26–
5.27 (1H, dd, J = 11.0, 1.5 Hz, H-12E), 5.69–5.71 (1H,
dd, J = 17.8, 1.5 Hz, H-12Z), 5.96 (2H, s, –OCH2O–),
6.68–6.69 (1H, s, H-7), 7.11–7.13 (1H, t-like,
J = 7.7 Hz, H-2), 7.20–7.21 (1H, s, H-10), 7.17–7.23
(1H, dd, J = 17.8, 11.0 Hz, H-11), 7.44–7.46 (1H, dd,
J = 7.7, 1.3 Hz, H-3), 7.54–7.55 (1H, dd, J = 7.7,
1.3 Hz, H-1); 13C NMR (CDCl3) d 49.8–50.2 (t, C-6),
100.9 (t, –OCH2O–), 103.7 (d), 106.8–106.9 (d), 113.8–
113.9 (t, C-12), 122.6 (d), 123.9 (d), 125.0–125.1 (d),
126.5 (s), 127.2–127.3 (s), 129.6 (s), 133.2 (s), 133.7–
133.9 (d, C-11), 145.3–145.5 (s, C-4a), 147.2 (s) and
147.3 (s) (C-8 and C-9).
5.3.4.8. 4-Ethyl-5-methyl-8,9-methylenedioxy-5,6-dihydro-
phenanthridine (5a). Yield 66%; Rf = 0.28, hexane–CHCl3
1
(4:6); H NMR (CDCl3) d 1.28 (3H, t, J = 7.6 Hz, H-12),
2.45 (3H, s, N-Me), 2.78 (2H, q, J = 7.6 Hz, H-11), 3.96
(2H, s, H-6), 5.96 (2H, s, –OCH2O–), 6.69 (1H, s, H-7),
7.13 (1H, t-like, J = 7.6 Hz, H-2), 7.16 (1H, dd, J = 7.6,
2.1 Hz, H-3), 7.22 (1H, s, H-10), 7.47 (1H, dd, J = 7.6,
2.1 Hz, H-1); 13C NMR (CDCl3) d 14.8 (q, C-12), 23.2 (t,
C-11), 41.0 (q, N-Me), 55.3 (t, C-6), 100.9 (t, –OCH2O–),
103.8 (d), 107.2 (d), 121.1 (d), 124.6 (d), 126.4 (s), 126.7 (d),
127.7 (s), 129.4 (s), 139.5 (s), 145.5 (s, C-4a), 147.2 (s) and
147.3 (s) (C-8 and C-9); EIMS: m/z 267 [M]+ (50), 266
[MꢁH]+ (100), 251 (13); HREIMS m/z[M]+ 267.1243 (calcd
for C17H17NO2, 267.1254).
5.3.4.3. 5-Ethyl-8,9-methylenedioxy-4-vinyl-5,6-dihydro-
phenanthridine (4b). Yield 76%; Rf = 0.46, hexane–CHCl3
1
5.3.4.9. Common H and 13C NMR data for 5-alkyl-4-
1
(4:6); other H NMR (CDCl3) d 1.03 (3H, t, J = 7.1 Hz,
ethyl-8,9-methylenedioxy-5,6-dihydrophenanthridine (5b–f). 1H
NMR (CDCl3) d 1.26–1.29 (3H, t, J = 7.6 Hz, H-12), 2.75–
2.77 (2H, q, J = 7.6 Hz, H-11), 3.97–3.99 (2H, s, H-6),
5.95–5.96 (2H, s, –OCH2O–), 6.68–6.69 (1H, s, H-7), 7.11–
7.13 (1H, t-like, J = 7.6 Hz, H-2), 7.16–7.17 (1H, dd,
J = 7.6, 2.1 Hz, H-3), 7.21–7.22 (1H, s, H-10), 7.45–7.47
(1H, dd, J = 7.6, 2.1 Hz, H-1); 13C NMR (CDCl3) d 14.6–
14.7 (q, C-12), 23.3–23.4 (t, C-11), 49.4–50.2 (t, C-6),
100.8–100.9 (t, –OCH2O–), 103.8 (d), 107.0–107.1 (d),
H-20), 2.73 (2H, q, J = 7.1 Hz, H-10); other 13C NMR
(CDCl3) d 13.5 (q, C-20), 47.4 (t, C-10); EIMS m/z [M]+
279 (91), [MꢁH]+ 278 (100), 250 (M+-29)+ (88); HREIMS
m/z [M]+ 279.1245 (calcd for C18H17NO2, 279.1254).
5.3.4.4. 5-Propyl-8,9-methylenedioxy-4-vinyl-5,6-dihydro-
phenanthridine (4c). Yield 75%; Rf = 0.54, hexane–CHCl3
1
(4:6); other H NMR (CDCl3) d 0.82 (3H, t, J = 7.4 Hz,

1040
S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
120.9–121.0 (d), 124.3–124.4 (d), 126.9–127.1 (s), 127.2–127.4
(s), 127.8–127.9 (d), 129.6–129.7 (s), 139.4 (s), 145.9–146.1 (s,
C-4a), 147.0–147.1 (s) and 147.2–127.3 (s) (C-8 and C-9).
5.4.2. Preparation of the triacetyl-5,6-secolycorine 8. The
solution of compound 6 (50 mg, 86 lmol) and anhy-
drous sodium acetate (50 mg) dissolved in dry DMSO
(1.5 mL) was stirred at 105 ꢁC for 6 h. After cooling,
the reaction mixture was partitioned between water
(35 mL) and toluene (3· 20 mL). The organic layer
was dried over anhyd Na2SO4 and concentrated to give
a residue, which was purified on a silica gel FCC, eluted
with hexane–CHCl3 (35:65), to afford 7 (30 mg, 58%).
The suspension of secolycorine 7 (30 mg, 50 lmol)
and activated Zn (105 mg, 1.60 mmol) in 90% acetic
acid was stirred at room temperature for 4 h. Then
the reaction mixture was diluted with chloroform
(20 mL) and filtered through a Celite cake. The filtrate
was washed with water (2· 5 mL), dried over anhyd
Na2SO4, and concentrated to give a residue, which
was flash chromatographed over a silica gel column,
eluted with CHCl3–MeOH (96:4), to afford 8 (8 mg,
yield 37%).
5.3.4.10. 4,5-Diethyl-8,9-methylenedioxy-5,6-dihydrophe-
nanthridine (5b). Yield 54%; Rf = 0.40, hexane–CHCl3 (4:6);
1
other H NMR (CDCl3) d 1.06 (3H, t, J = 7.1 Hz, H-20),
2.65 (2H, q, J = 7.1 Hz, H-10); other 13C NMR (CDCl3) d
13.6 (q, C-20), 46.4 (t, C-10); EIMS m/z 281 [M]+ (76), 280
[MꢁH]+ (100), 266 [M-15]+ (18), 252 [M-29]+ (18); HREIMS
m/z [M]+ 281.1403 (calcd for C18H19NO2, 281.1410).
5.3.4.11. 4-Ethyl-8,9-methylenedioxy-5-propyl-5,6-dihydro-
phenanthridine (5c). Yield 55%; Rf = 0.48, hexane–CHCl3
1
(4:6); other H NMR (CDCl3) d 0.85 (3H, t, J = 7.4 Hz,
H-30), 1.53 (2H, m, H-20), 2.56 (2H, m, H-10); other 13C
NMR (CDCl3) d 11.4 (q, C-30), 21.3 (t, C-20), 54.0 (t, C-
10); EIMS m/z [M]+ 295 (77), [MꢁH]+ 294 (44), 266 (M+-
29) (100), 252 (M+-43) (21); HREIMS m/z [M]+ 295.1567
(calcd for C19H21NO2, 295.1567).
5.4.3. Preparation of 5-hydroxy-N-methyl-5,6-secolyco-
rine (10). To 8 (6 mg, 14 lmol) dissolved in MeOH
(1 mL) were added formalin solution (55 rmuL, 37%),
sodium cyanoborohydride (21 mg, 334 lmol), and acetic
acid (0.2 mL) in sequence at room temperature. The
reaction mixture was stirred for 2 h and then concentrat-
ed under reduced pressure. The residue obtained was
partitioned between chloroform (25 mL) and satd NaH-
CO3 (2· 5 mL). The CHCl3 layer was washed with brine
solution (2· 5 mL), dried over anhyd Na2SO4 and con-
centrated to give a residue, which was purified by a flash
silica gel column, eluted with chloroform, to give 9
(6.1 mg, 100%). The mixture of 9 (67 mg, 150 lmol),
potassium carbonate (125 mg, 0.9 mmol), and MeOH–
H2O (9:1) (2.0 mL) was stirred at room temperature
for 2 h. The reaction mixture was concentrated under re-
duced pressure and the dry residue was subjected to alu-
mina FCC (neutral), eluted with CHCl3–MeOH (94:6),
to give 10 (34.5 mg, yield 72%).
5.3.4.12. 5-Butyl-4-ethyl-8,9-methylenedioxy-5,6-dihydro-
phenanthridine (5d). Yield 61%; Rf = 0.50, hexane–CHCl3
1
(4:6); other H NMR (CDCl3) d 0.91 (3H, t, J = 7.3 Hz,
H-40), 1.31 (2H, m, H-30), 1.53 (2H, m, H-20), 2.61 (2H, m,
13
H-10); other C NMR (CDCl3) d 14.0 (q, C-40), 20.2 (t,
C-30), 30.5 (t, C-20), 51.9 (t, C-10); EIMS m/z [M]+ 309 (76),
[MꢁH]+ 308 (39), 266 (M+-43) (100), 252 (M+-57) (21); HRE-
IMS m/z [M]+ 309.1724 (calcd for C20H23NO2, 309.1723).
5.3.4.13. 4-Ethyl-8,9-methylenedioxy-5-pentyl-5,6-dihydro-
phenanthridine (5e). Yield 52%; Rf = 0.48, hexane–CHCl3
1
(7:13); other H NMR (CDCl3) d 0.86 (3H, t, J = 6.8 Hz,
H-50), 1.27 (4H, m, H-30 and 40), 1.52 (2H, m, H-20), 2.57
(2H, t, J = 7.0 Hz, H-10); other 13C NMR (CDCl3) d 14.0
(q, C-50), 22.6 (t, C-40), 27.9 (t, C-30), 29.2 (t, C-20), 52.1 (t,
C-10); ESIMS: [M+H]+m/z 324.
5.3.4.14. 4-Ethyl-5-hexyl-8,9-methylenedioxy-5,6-dihydro-
phe-nanthridine (5f). Yield 60%; Rf = 0.49, hexane–
1
CHCl3 (7:13); other H NMR (CDCl3) d 0.85 (3H, t,
5.4.3.1. 4aa,10bb-1a,2b-Diacetoxy-4-(2-chloroethyl)-
8,9-methylenedioxy-1,2,4a,5,6,10b-hexahydrophenanthri-
dine-5-carboxylic acid 2,2,2-trichloroethyl ester (6).
Rf = 0.63, hexane–CHCl3 (2:8), developed three times;
1H NMR (CDCl3) d 1.90 (3H, s), 2.07 (3H, s), 2.71
(2H, m), 3.58 (1H, m), 3.66 (2H, t-like, J = 6.2 Hz),
4.09 (1H, m), 4.37 (1H, m), 4.66 (1H, m), 4.76 (1H,d,
J = 10.8 Hz), 5.13 (2H, br s), 5.70 (2H, m, H-1 and H-
3), 5.92 (2H, s, –OCH2O–), 6.56 (1H, s, H-7), 6.74
(1H, s, H-10); ESIMS m/z [M+Na]+ 604: [M+Na+2]+
606: [M+Na+4]+ 608: [M+Na+6]+ 610: [M+Na+8]+
612 (80:100:50:8:1).
J = 6.8 Hz, H-60), 1.24 (6H, m, H-30, 40 and 50), 1.51
(2H, m, H-20), 2.56 (2H, t, J = 7.4 Hz, H-10); other 13C
NMR (CDCl3) d 14.0 (q, C-60), 22.6 (t, C-50), 26.7 (t,
C-40), 28.2 (t, C-30), 31.7 (t, C-20), 52.2 (t, C-10); ESIMS:
[M+H]+m/z 338.
5.4. Synthesis of 4aa,10bb-1a,2b-dihydroxy-4-(2-hydroxyethyl)-5-
methyl-8,9-methylenedioxy-1,2,4a,5,6,10b-hexahydrophenanthri-
dine (10)
5.4.1. Preparation of the 5,6-secolycorine carbamate 6.
To the suspension of 2 (4.83 g, 13.02 mmol), dry toluene
(120 mL), and potassium carbonate (5.40 g, 39.1 mmol)
was added 2,2,2-trichloroethylchloroformate (TCECF,
3.5 mL, 26.0 mmol) under nitrogen. The reaction mix-
ture was heated for 20 h at 120 ꢁC. After cooling to room
temperature, the reaction mixture was washed with water
(2· 100 mL) and brine (2· 50 mL) dried over anhyd
Na2SO4, and concentrated under reduced pressure to
give a liquid residue (12.1 g), which was subjected to sil-
ica gel FCC, eluted with hexane–CHCl3 (1:3), to give a
diacetyl-5,6-secolycorine carbamate (6) (4.31 g, 57%).
5.4.3.2. 4aa,10bb-1a,2b-Diacetoxyl-4-(2-acetoxyleth-
yl)-8,9-methylenedioxy-1,2,4a,5,6,11b-hexahydrophenan-
thridine (8). 1H NMR (CDCl3, 400 MHz): d 1.89 (3H, s),
1.98 (3H, s), 2.04 (3H, s) (3· Ac–Me), 2.47 (1H, m) and
2.70 (1H, m) (H-11), 2.89 (1H, br d, J = 10 Hz, H-10b),
3.43 (1H, br d, J = 10 Hz, H-4a), 3.98 (1H, d,
J = 16.8 Hz) and 4.07 (1H, br d, J = 16.8 Hz) (H-6),
4.25 (2H, m, H-12), 5.06 (1H, s, H-2), 5.61 (2H, s, H-1
and 3), 5.85 (2H, s, –OCH2O–), 6.45 (1H, s, H-7), 6.71
(1H, s, H-10).

S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
1041
5.4.3.3. 4aa,10bb-1a,2b-Diacetoxy-4-(2-acetoxyleth-
yl)-5-methyl-8,9-methylenedioxy-1,2,4a,5,6,11b-hexahydro-
EtOAc (6:4), to yield 13 (49 mg, yield 94%). The suspen-
sion of 13 (153 mg, 252 lmol) and activated Zn (535 mg,
8.19 mmol) in 90% acetic acid was stirred at 50 ꢁC for
4 h. After a similar workup as for the preparation of
8, the residue obtained was subjected to silica gel
FCC, eluted with CHCl3–MeOH (95:5), to afford the
product 14 (47 mg, yield 43%).
1
phenanthridine (9). Rf = 0.59, CHCl3–MeOH (98:2); H
NMR (CDCl3) d 1.90 (3H, s), 2.00 (3H, s), 2.07 (3H,
s) (3· Ac–Me), 2.16 (3H, s, N-Me), 2.38 (1H, m) and
2.68 (1H, m) (H-11), 3.15 (1H, d, J = 10.7 Hz, H-10b),
3.62 (1H, d, J = 10.7 Hz, H-4a), 4.17 (1H, m) and 4.29
(1H, m) (H-12), 3.57 (1H, d) and 4.28 (1H, d) (H-6,
Jgem = 16.0 Hz), 5.06 (1H, br s, H-2), 5.68 (1H, br s,
H-1), 5.72 (1H, br s, H-3), 5.88 (2H, s, –OCH2O–),
6.49 (1H, s, H-7), 6.78 (1H, s, H-10); 13C NMR (CDCl3)
d 20.9 (q, Ac–Me), 21.0 (q, Ac–Me), 21.2 (q, Ac–Me),
30.2 (d, C-10b), 32.4 (t, C-11), 34.9 (q, N-Me), 56.9 (d,
C-4a), 57.2 (t, C-6), 62.8 (t, C-12), 67.7 (d) and 69.7
(d) (C-1 and C-2), 100.9 (t, –O–CH2O–), 104.9 (d),
107.2 (d), 121.6 (d, C-3), 126.2 (s), 128.1 (s), 143.4 (s,
C-4), 146.5 (s), 146.6 (s), 169.8 (s), 170.1 (s), 170.9 (s)
(3· Ac–CO); EIMS: m/z [M]+ 445 (80), 430 (M+-15)
(11), 386 (M+-59) (75), 385 (M+-60) (63), 372 (M+-73)
(100).
5.5.3. Preparation of 5-hydroxy-N-methyl-5,6-seco-dihy-
drolycorine (16). To 14 (23 mg, 0.053 mmol) dissolved in
MeOH (1 mL) were added formalin solution (0.2 mL,
37%), NaCNBH3 (81 mg, 1.27 mmol) and acetic acid
(0.7 mL) in sequence at room temperature. The reaction
mixture was stirred for 2 h. After similar workup as for
the preparation of 9, the residue obtained was subjected
to silica gel FCC, eluted with CHCl3–MeOH (97:3), to
afford the product 15 (23.5 mg, yield 100%). The mix-
ture of 15 (18 mg, 40.3 lmol), potassium carbonate
(17 mg, 121 lmol), and MeOH–H2O (9:1) (2.0 mL)
was stirred at room temperature for 2 h. The reaction
mixture was concentrated under reduced pressure and
the dry residue was flash chromatographed over an alu-
mina column (neutral), eluted with CHCl3–MeOH
(95:5), to give 16 (11.6 mg, yield 90%).
5.4.3.4. 4aa,10bb-1a,2b-Dihydroxy-4-(2-hydroxylethyl)-
5-methyl-8,9-methylenedioxy-1,2,4a,5,6,11b-hexahydrophe-
nanthridine (10). Rf = 0.44, CHCl3–MeOH–NH4OH
1
(85:15:0.1); H NMR (CD3OD) d 2.21 (3H, s, N-Me),
2.32 (1H, m) and 2.63 (1H, m) (H-11), 3.07 (1H, br d,
J = 11.0 Hz, H-10b), 3.68 (2H, m, H-12), 3.69 (1H, d,
J = 11.0 Hz, H-4a), 4.00 (1H, br s) and 4.50 (1H, br s)
(H-1 and H-2), 3.62 (1H, d) and 4.21 (1H, d) (H-6,
Jgem = 16.6 Hz), 5.68 (1H, br s, H-3), 5.88 (2H, s, –
OCH2O–), 6.56 (1H, s, H-7), 6.91 (1H, s, H-10); 13C
NMR (CD3OD) d 32.4 (d, C-10b), 35.1 (q, N-Me),
38.4 (t, C-11), 57.88 (t, C-6), 57.91 (d, C-4a), 62.2 (t,
C-12), 70.4 (d) and 72.7 (d) (C-1 and C-2), 102.3 (t, –
OCH2O–), 106.1 (d) and 108.3 (d) (C-7 and C-10),
126.8 (d, C-3), 128.8 (s) and 129.6 (s) (C-6a and C-
10a), 142.9 (s, C-4), 147.9 (s) and 148.4 (s) (C-8 and
C-9); EIMS m/z [M]+ 319 (79), 304 (M+-15) (18), 288
(M+-31) (100%). HREIMS m/z [M]+ 319.1414 (calcd
for C17H21NO5, 319.1408).
5.5.3.1. 4aa,10bb-1a,2b-Diacetoxy-4-(2-chloroethyl)-
8,9-methylenedioxy-1,2,3,4,4a,5,6,10b-octahydrophenan-
thridine-5-carboxylic acid 2,2,2-trichloroethyl ester (12).
Rf = 0.54 hexane–EtOAc (1:1); 1H NMR (CDCl3,
400 MHz): d 2.07 (6H, s, 2· Ac–Me), 2.18 (2H, m),
2.83 (1H, m), 3.28 (1H, d, J = 12.5 Hz), 3.46 (1H, m),
3.52 (1H, m), 3.91 (1H, m), 4.08 (1H, m), 4.73 (1H, br
s, H-2), 5.02 (2H, m), 5.65 (1H, br s, H-1), 5.90 (2H,
br s, –OCH2O–), 6.52 (1H, s, H-7), 6.68 (1H, s, H-10);
EI-MS: m/z [M]+ 583: [M+2]+ 585: [M+4]+ 587:
[M+6]+ 589: [M+8]+ 591 [21:28:14:3:4:0.28, theoretically
21:28:14:3.1:0.26]. 463 (M+–2HOAc): 465 (M+–
2HOAc + 2): 467 (M+–2HOAc + 4): 469 (M+–
2HOAc + 6) (68.5: 91.1: 46.3: 11.5: 1.2, theoretically
68.5:91.3:10.1:0.85).
5.5. Synthesis of 4a,4aa,10bb-1a,2b-dihydroxy-4-(2-hydroxy-
ethyl)-5-methyl-8,9-methylenedioxy-1,2,3,4,4a,5,6,10b-octa-
hydrophenanthridine (16, 5-hydroxy-N-methyl-5,6-seco-
dihydrolycorine)
5.5.3.2. 4aa,10bb-1a,2b-Diacetoxy-4-(2-acetoxyethyl)-
8,9-methylenedioxy-1,2,3,4,4a,5,6,10b-octahydrophenanthri-
dine-5-carboxylic acid 2,2,2-trichloroethyl ester (13).
Rf = 0.56 hexane–EtOAc (1:1); 1H NMR (CDCl3) d
1.86 (2H, m), 2.03 (3H, s) and 2.07 (6H, s) (3· Ac–
Me), 2.08 (1H, m), 2.72 (1H, m), 3.29 (1H, d,
J = 12.3 Hz), 4.03 (1H, m), 4.11 (3H, m), 4.50 (1H, br
s), 4.74 (1H, s), 4.96 (1H, m), 5.02 (1H, s), 5.66 (1H,
br s), 5.91 (2H, s, –OCH2O–), 6.53 (1H, s, H-8), 6.70
(1H, s, H-11); EIMS m/z [M]+ 607: [M+2]+ 609:
[M+4]+ 611: [M+6]+ 613 (53.9:55.2:19.4:3.0), 487 (M+–
2HOAc): 489 (M+–2HOAc + 2): 491 (M+–2HOAc + 4):
493 (M+–2HOAc + 6) (81.8:100:46.1:10.3).
5.5.1. Preparation of 5,6-seco-dihydrolycorine carbamate
12. To the suspension of diacetyldihdrolycorine (11,
225 mg, 603 lmol), dry toluene (50 mL), and potassium
carbonate (3 equiv, 250 mg, 1.81 mmol) was added
TCECF (0.16 mL, 1.21 mmol) under nitrogen. The reac-
tion mixture was heated at 120 ꢁC for 14 h. After similar
workup for the preparation of 6, the residue obtained
was subjected to silica gel FCC, eluted with hexane–
CHCl3 (3:7), to give the 5,6-secodihydrolycorine carba-
mate 12 (200 mg, yield 57%).
5.5.3.3. 4aa,10bb-1a,2b-Diacetoxy-4-(2-acetoxyethyl)-
8,9-methylenedioxy-1,2,3,4,4a,5,6,10b-octahydrophenanthri-
1
5.5.2. Preparation of the triacetyl-5,6-seco-dihydrolyco-
rine 14. The solution of compound 12 (50 mg, 86 lmol)
and sodium acetate trihydrate (24 mg, ca. 2 equiv) in
DMSO was stirred at 105 ꢁC for 6.5 h. After similar
workup for the preparation of 7, the residue obtained
was subjected to silica gel FCC, eluted with hexane–
dine (14). H NMR (CDCl3) d 1.87 (3H, s), 2.02 (3H, s)
and 2.09 (3H, s) (3· Ac–Me), 1.91 (2H, m), 2.20 (1H,
m), 3.19 (1H, br d, J = 11.6 Hz, H-10b), 3.26 (1H, dd,
J = 11.6, 4.3 Hz, H-4a), 3.95 (1H, d) and 4.08 (1H, d)
(H-6s, Jgem = 15.6 Hz), 4.07 (1H, dt, J = 11.0, 4.3 Hz)
and 4.19 (1H, dt, J = 11.0, 6.4 Hz) (H-5s), 4.34 (1H,

1042
S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
NH, D2O exchangeable), 5.00 (1H, m, H-2), 5.54 (1H, br
s, H-1), 5.88 (2H, s, –OCH2O–), 6.46 (1H, s, H-7), 6.69
(1H, s, H-10); EI–MS m/z [M]+ 433 (29), 432 (16), 373
(32), 372 (73), 314 (38), 313 (14), 254 (68), 204 (100);
HREIMS m/z [M]+ 433.1721 (calcd for C22H27NO8,
433.1731).
sample (10 mM in MeOH diluted with water to 1–
10 lM) or blank (1% MeOH in water) were added. The
reaction was started after 13 lL acetylcholinesterase
(0.22 U/ml in 50 mM Tris–HCl containing 0.1% bovine
serum albumin) was added to the 96-well plate. Absor-
bance of the yellow anion product due to the spontaneous
hydrolysis of substrate was measured at 405 nm for
12 min on a Microtiter plate reader (uQuant, Bio-TEK).
Galanthamine used as a positive control. The inhibition
percentage (%) was calculated by the following equation:
5.5.3.4. 4aa,10bb-1a,2b-Diacetoxy-4-(2-acetoxyethyl)-5-
methyl-8,9-methylenedioxy-1,2,3,4,4a,5,6,10b-octahydrophe-
nanthridine (15). Rf = 0.66, CHCl3–MeOH (23:2); 1H
NMR (CDCl3) d 1.81 (1H, dt, J = 15.8, 4.4 Hz, H-3ax),
1.86 (3H, s), 2.01 (3H, s) and 2.09 (3H, s) (3· Ac–Me),
2.00 (2H, m, H-4s), 2.13 (1H, br d, J = 15.8 Hz, H-
3eq), 2.21 (1H, m, H-4), 2.29 (3H, s, N-Me), 2.47 (1H,
dd, J = 11.0, 4.0 Hz, H-4a), 3.25 (1H, br d,
J = 11.0 Hz, H-10b), 3.35 (1H, d) and 3.73 (1H, d) (H-
6s Jgem = 14.0 Hz), 4.13 (2H, m, H-12), 4.94 (1H, m,
H-2), 5.51 (1H, br s, H-1), 5.85 (1H, br s) and 5.86
(1H, br s) (–OCH2O–), 6.48 (1H, s, H-7), 6.68 (1H, s,
H-10); 13C NMR (CDCl3) d 20.9 (q), 21.0 (q), and
21.4 (q) (3· Ac–Me), 24.5 (t, C-11), 27.0 (t, C-3), 31.6
(d, C-4), 36.3 (d, C-10a), 41.6 (q, N-Me), 59.6 (t, C-6),
62.1 (d, C-4a), 63.7 (t, C-12), 69.3 (d) and 70.7 (d) (C-
1 and C-2), 100.8 (t, –OCH2O–), 105.4 (d) and 106.2
(d) (C-7 and C-10), 125.9 (s) and 128.5 (s) (C-6a and
C-10a), 146.0 (s) and 146.7 (s) (C-8 and C-9), 169.5
(s), 169.6 (s) and 171.1 (s) (3· Ac–CO); EIMS m/z
[M]+ 447 (41), 388 (7), 328 (7), 260 (40), 218 (100);
HRFABMS m/z [M+H]+ 448.1952 (calcd for
C23H29NO8+H, 448.1966).
Inhibition ð%Þ ¼ ½1 ꢁ ðAsample=AcontrolÞꢂ ꢃ 100
5.7. Docking calculations of active compounds to
EelAChE
The docking calculations, using AutoDock 3.0516 and
MEDock,17 of the active compounds to the eel AChE
(EelAChE) were performed. The structure of EelAChE
was taken from the Protein Data Bank19 (accession
code: 1C2B), which was obtained by fitting the structure
of the mouse AChE (mAChE) structure (accession code:
1MAA) into a low resolution X-ray electron density
map. The ligand files were prepared on the PRODRG
2 server (http://davapc1.bioch.dundee.ac.uk/programs/
prodrg/).20 For the docking protocol, the number of
generation in both evolutionary algorithms was set as
large as 5000, so that the evolution of each run can reach
a converged state. The iteration number for the local
search algorithm, that is, the Solis-Wet searcher, is set
to 600. The grid center is set at the amine nitrogen posi-
tion of the Serine 203, which forms the catalytic triad
with GLU 334 and HIS 447. The docked conformations
were analyzed with LIGPLOT 4.4.2.18
5.5.3.5. 4aa,10bb-1a,2b-Dihydroxy-4-(2-hydroxyethyl)-
5-methyl-8,9-methylenedioxy-1,2,3,4,4a,5,6,10b-octahydr-
ophenanthridine (16). Rf = 0.34, CHCl3–MeOH–NH4
OHaq (85:15:0.1); 1H NMR (CD3OD) d 1.76 (1H, m, H-
11), 1.87 (1H, dt, J = 15.1, 4.0 Hz, H-3ax), 1.99 (1H, br
d, J = 15.0 Hz, H-3eq), 2.15 (1H, ddt, J = 13.9, 11.7,
4.6 Hz, H-11), 2.29 (1H, m, H-4), 2.35 (3H, s, N-Me),
2.66 (1H, dd, J = 11.3, 4.1 Hz, H-4a), 3.27 (1H, br d,
J = 11.4 Hz, H-10b), 3.41 (1H, d) and 3.72 (1H, d) (H-
6s, Jgem = 14.7 Hz), 3.69 (1H, ddd, J = 10.6, 9.0, 4.8 Hz)
and 3.76 (1H, ddd, J = 10.6, 5.7, 4.9 Hz) (H-12s), 3.88
(1H, q-like, J = 2.8 Hz, H-2), 4.31 (1H, br s, H-1), 5.87
(2H, s, –OCH2O–), 6.53 (1H, s, H-7), 6.85 (1H, s, H-10);
13C NMR (CD3OD) d 29.5 (t, C-11), 29.8 (t, C-3), 32.5
(d, C-4), 37.8 (d, C-10b), 41.7 (q, N-Me), 60.8 (t, C-6),
61.4 (t, C-12), 63.5 (d, C-4a), 71.8 (d) and 74.0 (d) (C-1
and C-2), 102.1 (t, –OCH2O–), 106.7 (d), 107.1 (d) (C-7
and C-10), 129.6 (s) and 129.7 (s) (C-6a and C-10a),
147.2 (s) and 148.4 (s) (C-8 and C-9); EIMS m/z [M]+
321 (14), 218 (100), 188 (28); HREIMS m/z [M]+
321.1581 (calcd for C17H23NO5, 321.1571).
Acknowledgments
We gratefully acknowledge the financial support from
the National Science Council (NSC), Taiwan, ROC un-
der Grant NSC92-2320-B-002-079, and the Ministry of
Economics, Taiwan, ROC under Grant 93-EC-17-A-
20-S1-0010.
References and notes
1. Scapini, E.; Scheltens, P.; Feldman, H. Lancet Neurol.
2003, 2, 539.
2. Perry, E. K. Br. Med. Bull. 1986, 42, 63.
3. Frisoni, G. B. J. Neurol. 2001, 248, 551.
4. Birks, J. Cochrane Database Syst. Rev. 2006, 25,
CD005593.
5. Cook, J. W.; Loudon, J. D.. In The alkaloids; Manske, R.
H. F., Holmes, H. L., Eds.; Academic Press: New York,
1952; Vol. II, p 331.
5.6. Biological assay
6. (a) Mineshita, T.; Yayaguchi, K.; Takeda, K.; Kotera, K.
Ann. Rep. Shinogi Res. Lab. 1956, 6, 119; (b) Chattopad-
hyay, U.; Chaudhuri, L.; Das, S.; Kumar, Y.; Ghosal, S.
Pharmazie 1984, 39, 855.
7. De Leo, P.; Dalessandro, G.; De Santis, A.; Arigoni, O.
Plant Cell Physiol. 1973, 14, 487.
8. For a review of the chemistry and biological effects of
lycorine and related Amaryllidaceae alkaloids, see Ghosal,
S.; Saini, K. S.; Razdan, S. Phytochemistry 1985, 24, 2141.
AChE inhibitory activity was measured by modification
of Ellman’s method.14,15 This colorimetric method is
based on the amount of thiocholine produced when the
substract ACh is hydrolyzed by the enzyme. In the 96-well
plates, 13 lL acetylthiocholine iodide (ATCI, 15 mM in
water), 64 lL of 5,50-dithiobis[2-nitrobenzoic acid]
(DTNB, 3 mM in PBS containing 50 mM Tris–HCl,
0.1 M NaCl, and 0.02 M MgCl2Æ6H2O, pH 8), 10 lL of

S.-S. Lee et al. / Bioorg. Med. Chem. 15 (2007) 1034–1043
1043
9. Li, S.-Y.; Chen, C.; Zhang, H.-Q.; Guo, H.-Y.; Wang, H.;
Wang, L.; Zhang, X.; Hua, S.-N.; Yu, J.; Xiao, P.-G.; Li,
R.-S.; Tan, X. Antiviral Res. 2005, 67, 18.
10. Houghton, P. J.; Agbedahunsi, J. M.; Adegbulugbe, A.
Phytochemistry 2004, 65, 2893.
15. Rhee, I. K.; van de Meent, M.; Ingkaninan, K.; Ver-
poorte, R. J. Chromatogr. A 2001, 915, 217.
16. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.;
Hart, W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem.
1998, 19, 1639.
11. Lopez, S.; Bastida, J.; Viladomat, F. Life Sci. 2002, 71,
2521.
17. Chang, D. T. H.; Oyang, Y. J.; Lin, J. H. Nucleic Acids
Res. 2005, 33, W233.
12. Houghton, P. J.; Ren, Y.; Howes, M. J. Nat. Prod. Rep.
2006, 23, 181.
18. Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. Protein
Eng. 1995, 8, 127.
13. Tsuda, Y.; Kashiwaba, N.; Kumar, V. Chem. Pharm. Bull.
1984, 32, 3023.
14. Ellman, G. L.; Courtney, K. D.; Andres, V., Jr.;
Featherstone, R. M. Biochem. Pharmacol. 1961, 7,
88.
19. Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.;
Bhat, T. N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E.
Nucleic Acids Res. 2000, 28, 235.
20. Schuttelkopf, A. W.; van Aalten, D. M. F. Acta Crystal-
logr. D 2004, 60, 1355.