Main glucosidase conversion products of the gluco-alkaloids dolichantoside and palicoside

Article abstract of DOI:10.1016/S0031-9422(01)00085-1

The enzymatic glucose cleavage of palicoside revealed the biosynthetic pathway to akagerine, whereas the conversion of dolichantoside led to a new quaternary heteroyohimbine alkaloid named N_b-methyl-21-β-hydroxy-mayumbine. The hypothetical models of reactions occurring after the conversion of both substrates are proposed. Dolichantoside and palicoside, as well as Strychnos mellodora stem bark crude ethanol extract, exhibit significant antimycotic activity against human pathogens in presence of specific glucosidase.

Full text of DOI:10.1016/S0031-9422(01)00085-1

Phytochemistry 57 (2001) 653–659
www.elsevier.com/locate/phytochem
Main glucosidase conversion products of the gluco-alkaloids
dolichantoside and palicoside
Viviane Brandt*, Monique Tits, Jacques Penelle, Michel Frederich, Luc Angenot
´ `
Service de Pharmacognosie, Institut de Pharmacie, Universite de Liege, C.H.U., Tour 4, Avenue de l’Hopital 1, 4000 Liege 1, Belgium
ˆ
Received 9 October 2000; received in revised form 16 February 2001
Abstract
The enzymatic glucose cleavage of palicoside revealed the biosynthetic pathway to akagerine, whereas the conversion of doli-
chantoside led to a new quaternary heteroyohimbine alkaloid named Nb-methyl-21-b-hydroxy-mayumbine. The hypothetical
models of reactions occurring after the conversion of both substrates are proposed. Dolichantoside and palicoside, as well as
Strychnos mellodora stem bark crude ethanol extract, exhibit significant antimycotic activity against human pathogens in presence
of specific glucosidase. # 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Strychnos mellodora; Loganiaceae; Biosynthesis; Gluco-alkaloids; Dolichantoside; Palicoside; Akagerine; Nb-Methyl-21-b-hydroxy-
mayumbine
1. Introduction
glucosidase from Catharanthus roseus (SGD) and by a
specific glucosidase isolated from Strychnos mellodora
(SMGD) (Brandt et al., 2000). Here we present the iso-
lation and identification of the main conversion pro-
ducts. An evaluation of the antimycotic activity before
and after glucosidase conversion was carried out on the
crude ethanol extract of Strychnos mellodora stem bark,
as well as on dolichantoside (1) and palicoside (2).
In our search for alkaloids in Strychnos mellodora S.
Moore, an endemic species of East-Africa (Leeuwen-
berg, 1969) we isolated a wide range of gluco-alkaloids
(Tits et al., 1996; Brandt et al., 1999) and particularly
interesting amounts of the Nb-methylated alkaloids
dolichantoside (1) and palicoside (2). Their structures
are closely related to and present the same 3a confi-
guration as strictosidine (3), the central intermediate in
the biosynthesis of monoterpenoid indole and quinoline
alkaloids (Rueffer et al., 1978; Bisset, 1980). The enzy-
matic cleavage of glucose in strictosidine (3) gives rise to
a large variety of structure groups (Kisakurek et al.,
1983). The intermediate isolated by Stevens (1994) after
strictosidine conversion was 3a, 15a-ꢀ20(21)-didehy-
drocathenamine, a ꢀ20(21)-didehydro-heteroyohimbine.
The amounts of dolichantoside (1) and palicoside (2)
obtained from Strychnos mellodora gave us the oppor-
tunity to start phylogenetical studies in order to evalu-
ate the potential biosynthetic pathway from those
gluco-alkaloids. In our previous work, we mentioned
the conversion of these two substrates by strictosidine
2. Results and discussion
2.1. Identification and determination of conversion
products
To check the invariability of the pattern of conversion
products of dolichantoside (1) and palicoside (2), incu-
bations were first compared in the presence of three
different glucosidases: SGD, SMGD and a less specific
one, the glucosidase from sweet almonds. Independent
of the glucosidase used, the pattern of conversion pro-
ducts was the same. The kinetic parameters Km and
Vmax were determined for the enzymes SGD and SMGD,
in the presence of the substrates 1 and 2 (Brandt et al.,
2000). This study clearly revealed that the conversion of
substrates 1 and 2 was quicker and more complete with
the enzyme SMGD (Table 1). The Vmax or maximal
* Corresponding author at Laboratoires Ortis, Hinter der Heck, 46,
4750 Elsenborn, Belgium. Tel.: +32-80-440055; fax: +32-80-444280.
E-mail address: viviane.brandt@ortis.com (V. Brandt).
0031-9422/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0031-9422(01)00085-1

654
V. Brandt et al. / Phytochemistry 57 (2001) 653–659
Scheme 2. Hypothetical model for the conversion of palicoside into akagerine.
conversion velocities were higher in the presence of
SMGD and reached with lower substrate concentra-
tions, as Km represents half of the substrate concentra-
tion necessary to reach the maximal conversion velocity.
We thus incubated a larger amount of dolichantoside
(1) in the presence of SMGD. After 72 h of incubation,
dolichantoside was nearly completely converted into a
main product (4) that was more polar. This product was
compared in TLC and HPLC and stated to be identical
to the one obtained after the 15 min incubation of a
1 mM dolichantoside solution. It was separated (see
Experimental) from the incubation mixture and its
structure determined by spectroscopic methods as Nb-
methyl-21-b-hydroxy-mayumbine (4), a new hetero-
yohimbine quaternary alkaloid. Its UV spectrum revealed
maxima of tetrahydro-ꢀ-carboline alkaloids. The high
resolution mass spectrum exhibited a molecular peak
corresponding to the molecular formula C22H27N2O+4 .

V. Brandt et al. / Phytochemistry 57 (2001) 653–659
655
Table 1
with a C/D cis ring junction. Indeed, this stereo-
chemistry is supported by the relatively low field
position (1H: 3.27; 13C: 47.0 ppm) of the quaternary
Nb-methyl, suggesting a cis-relationship between the
Nb-methyl and the H-3 (Arata et al., 1975; Sugiura et
al., 1977). The resonances of H-14a, H-14b and H-3
appear respectively at 2.45, 1.90 and 5.0 ppm. This latter
deshielded value is likewise consistent with the presence
of the adjacent quaternary nitrogen. All aromatic pro-
tons, including H-17 at 7.49 ppm, are detected. Carbons
chemical shifts were determined on the base of HMQC
spectra.
Kinetic parameters of SGD and SMGD towards dolichantoside (1)
and palicoside (2). Conditions used to determine these parameters are
found in literature (Brandt et al., 2000)
Substrate
SGD
SMGD
Km (mM)
Vmax
(nkat mg À1
protein)
Km (mM)
Vmax
(nkat mg À1
protein)
(1)
(2)
1130
7015
21.2
11.1
564
335
224
39
The molecular peak also present in the ES–MS–MS
spectrum at m/z 383 (M)+ exactly corresponded to the
loss of a glucose moiety from dolichantoside (1); an ion
at m/z 365 suggested the easy loss of water and thus the
The second substrate, palicoside (2), was incubated
with SMGD for 72 h. Its unfinished conversion gave
rise to two main products. The first one, 5, was appar-
ently less polar than palicoside on TLC, in contrast to
the second one, 6, more polar. After a rapid chloroform
extraction of 5 from the incubation mixture, its UV and
MS spectra were measured. The UV spectrum exhibited
maxima of an indole chromophore. The mass spectrum
showed a molecular peak at m/z 325, corresponding to
the molecular formula C20H24N2O2. We confronted this
product 5 with available reference substances of iden-
tical molecular mass, and in particular to akagerine.
They showed to possess the same behaviour when com-
pared by HPLC and TLC in different systems. More-
over, the revelation of TLC plates by iodoplatinate
reagent (Svendsen and Verpoorte, 1983) produced in
each case a pink colour turning into red after a few
minutes. We completed this identification by comparison
1
presence of a hydroxyl function. In the H-1H NMR
auto-correlation spectra, the most characteristic point
for the formation of a heteroyohimbine structure was
the presence of a methyl group at 1.55 ppm (d, J18À19
=6.7 Hz) (CH3-18) correlated to H-19 at 3.75 ppm (q,
J19-20=8.6 Hz). Chemical shifts of those protons, in
particular shielding of H-19, as well as high coupling
constants, by comparison with the literature data con-
cerning the eight possible diastereoisomers of hetero-
yohimbine type (Phillipson and Supavita, 1983),
suggested a-configuration of H-19 and consequently
b-configuration of the methyl group C-18 (19 R series).
This configuration is related to 19-epiajmalicine or
mayumbine (3a, 20b, normal diastereoisomeric type)
(Uusvuori and Lounasmaa, 1981). The formation of a
normal type alkaloid is predictable considering the
initial H3-a, H20-a configuration of dolichantoside. On
the other hand, the configuration a of H19, charac-
teristic of 19-epiajmalicine, is less evident but plausible
if one considers the biomimetic scheme named ‘‘La
Ronde’’ established by Lounasmaa and Hanhinen (1998).
This scheme shows interconversions existing between the
four ꢀ20(21)-didehydroheteroyohimbines (cathenamines),
which, when combined with reduction steps, allow
access to the eight basic heteroyohimbines. The option
of H-19 a or b configuration takes place during the Z–E
epimerisation of the ethylidene side chain (Lounasmaa
et al., 1992; Uskokovic et al., 1979). Closing of E ring is
then achieved by a Michael addition of the enolate ste-
reochemically exclusively possible on the b-face on the
ethylidene chain (Brown, 1980). This ethylidene chain is
moreover directly connected to H-20 (3.15 ppm), itself
coupling with H-21 appearing in the form of a deshiel-
ded doublet at 4.45 ppm (J20À21=7.5 Hz), compatible
with the presence of an hydroxyl function on C-21 and
the presence of an adjacent quaternary nitrogen. The
coupling constant J20À21 suggested the antiperiplanar
position of H-20 and H-21, and thus the a position of
H-21. H-20 is also correlated to H-15 at 2.85 ppm. The
proton chemical shift of the N-methyl group is in accord
1
of their 1D H NMR spectra. The recorded resonances
were identical, which confirmed that akagerine was the
non-polar conversion product of palicoside (2). Aka-
gerine (5) is a perhydroazepine alkaloid that accumu-
lates in numerous species of Strychnos as Strychnos
usambarensis (Angenot et al., 1975). It has showed con-
vulsivant (Rolfsen et al., 1980) and antiprotozoal
(Wright et al., 1991) properties on mice. Its enantio-
selective synthesis was proposed by Danieli et al. (1995).
We completed the spectral data of this reference alka-
loid by further 1H and 13C NMR contribution. The
polar conversion product 6 could not be separated in
sufficient amount to be identified and is, thus, still under
studies.
2.2. Hypothetical models for the conversion reactions
For both conversions, we present hypothetical models
of reactions occurring after enzymatic hydrolysis and
removal of the glucose moiety. It should be noted that
none of the intermediates depicted in Schemes 1 and 2
has ever been isolated as a product from any incubation
mixture, or has ever been demonstrated in any other
direct way. Their occurrence is described by mere theo-
retical organic chemistry. It has been reported that the
deglucosylation of strictosidine leads also to reactive

656
V. Brandt et al. / Phytochemistry 57 (2001) 653–659
and unstable intermediates, which may readily and
quickly undergo different intra-molecular rearrangements
(Hemscheidt and Zenk, 1980; Stockigt, 1980; Stevens,
1994). The same situation is observed in our hands
using systems with dolichantoside or palicoside as ana-
logous substrates.
Table 2
Minimal inhibitory concentration (MIC) of dolichantoside (1), pali-
coside (2) and the crude EtOH extract of Strychnos mellodora stem
bark in presence of the glucosidase from Strychnos mellodora (SMGD)
against three ATCC reference strains
Minimum inhibitory concentration (mg/ml)
Reactions leading from dolichantoside (1) to
Nb-methyl-21-b-hydroxy-mayumbine (4) are similar to
those proposed in the conversion model of strictosidine
(Stevens, 1994). They can be schematised as follows
(scheme 1): removal of the glucose moiety of dolichan-
toside leads to an unstable hemiacetal (1a), which is
rapidly converted into a high reactive aldehyde by
opening of E ring. This aldehyde (1b) is in equilibrium
with its enol tautomer (1c). This intermediate is then
stabilised by a nucleophilic attack of the nitrogen Nb on
C-21 to close D ring, with concomitant transformation
of the vinyl chain into a Z-ethylidene chain (1d). Proto-
nation of C-18 is ensured by the C-17 hydroxyl func-
tion, acting as proton donor. Closing of E ring is then
achieved by a Michael addition of the enolate on the
b face of the ethylidene side chain to form the quaternary
alkaloid Nb-methyl-21-b-hydroxy-mayumbine (4).
Besides the cleavage of the glucose moiety, palicoside
(2), during its conversion, further loses a group of 44
mass units equivalent to the carbonic gas CO2. The con-
version is thought to start by the two first steps occurring
with dolichantoside (scheme 2). The loss of CO2, hap-
pening in totally soft conditions (30ꢀC, neutral pH), can
take place at the stage of the tautomeric equilibrium
between aldehyde and enol functions. This step is just
characterised by the presence of a carboxylic function in b
position from an aldehydic function where the start of
CO2 is favoured by a cyclic elimination process at six
centres, identical to what occurs in the decarboxylation of
malonic acid or b-cetonic acids (Carey and Sundberg,
1997). The double bond shift and the appearance of an
ethylidene chain take place at the same time. Closing of
perhydroazepine ring is then realised by the nucleophilic
attack of the indole nitrogen Na on C-17, to form a
carbinolamine, akagerine (5). The loss of CO2 would not
be allowed in the case of dolichantoside, because of the
presence of the methyl group on the carboxylic function,
which prevents the six centres elimination process to
take place.
Candida albicans Candida glabrata Aspergillus niger
10231
90030
16404
Dolichantoside 0.38
0.76
0.38
(1)/SMGD
Palicoside
(2)/SMGD
EtOH extract
/SMGD
0.14
2.28
1.48
0.38
0.74
0.76
alone showed no activity at the higher tested con-
centrations (1.9 mg/ml of 1; 1.8 mg/ml of 2; 3.8 mg/ml
of EtOH extract). This is in contrast with the significant
antimycotic activity found against those same three
pathogens when the substrates were tested in the pre-
sence of SMGD. The minimum inhibitory concentra-
tion (MIC) measured for each substrate in presence of
SMGD against each strain is presented in Table 2.
These products are, of course, far from competing with
other antifungal agents such as miconazole that inhibits
the growth of yeast strains C. albicans (1.6 mg/l), C.
glabrata (0.2 mg/l) and Aspergillus niger (0.8 mg/l) at
much lower concentrations in the same conditions (Piel
et al., 1999).
3. Conclusion
The hypothesis formulated by Massiot and Delaude
(1988) putting forward that Nb-methylated precursors
lead to the formation of akagerine-type bases, mono-
terpenoid alkaloids with Na–C17 bond, has been verified
by the isolation and identification of a stable alkaloid,
akagerine (5), after the conversion of palicoside (2). We
have shown, moreover, that even in the presence of a
methyl group, addition on Nb is possible and produces
polar alkaloid, as in the case of the conversion of
dolichantoside (1) into a new quaternary alkaloid,
Nb-methyl-21-b-hydroxy-mayumbine (4).
2.3. Antimycotic activity of the substrates in presence of
specific glucosidase
We detected no presence of those conversion products
in Strychnos mellodora organs, which may indicate that
at the plant level substrates and glucosidase are con-
tained in different cell compartments. Although the
results of antimycotic activity do not concern vegetal
pathogens, they may nevertheless suggest a constitutive
‘‘trigger mechanism’’ in the plant, in which an infection,
by damage of the plant cell, results in substrate deglu-
cosidation, thus allowing the active aglycone(s) to act
against the intruder.
The antimycotic activity of dolichantoside (1) and
palicoside (2), as well as of the crude ethanol extract of
Strychnos mellodora stem bark (EtOH extract), was tes-
ted before and after enzymatic hydrolysis by SMGD
against three standard strains of human pathogens, the
yeasts Candida albicans, Candida glabrata and a fila-
mentous fungus, Aspergillus niger. The substrates applied

V. Brandt et al. / Phytochemistry 57 (2001) 653–659
657
4. Experimental
The dolichantoside (1) incubation mixture was then
frozen and lyophilised. The dry residue was suspended
in 2 ml MeOH, stirred and centrifuged 10 min at 3500 g.
The supernatant was separated and evaporated to give a
residue which was taken in 2 ml H2O. This aq. soln was
then extracted twice by n-BuOH and evaporated under
reduced pressure. The residue dissolved in a few drops
of MeOH gave by addition of Et2O a voluminous white
ppt. This latter was separated by centrifugation 10 min
at 2500 g, dried and washed with Et2O to obtain
Nb-methyl-21-b-hydroxy-mayumbine (4) (27 mg) as an
amorphous white powder; UV (MeOH) lmax nm (log "):
4.1. Materials and methods
Dolichantoside (1) and palicoside (2) were isolated
from Strychnos mellodora S. Moore stem bark accord-
ing to Tits et al. (1996). Strictosidine b-glucosidase
(SGD) was obtained by purification from suspension
cultured cells of Catharanthus roseus L. G. Don accord-
ing to Geerlings (1999). Strictosidine b-glucosidase from
Strychnos mellodora (SMGD) was obtained by purifica-
tion from dry leaves of Strychnos mellodora (Brandt et
al., 2000). The b-glucosidase from sweet almonds was
purchased from Merck (Germany). The reference aka-
gerine has been isolated from Strychnos usambarensis
and is kept in our laboratory.
The comparison of conversion products pattern was
based on a HPLC method described by Pennings et al.
(1989), but with a 4 Â 250 mm Lichrospher 60 RP-Select
B column (particle size 5 mm, cat. No. 00961; Merck).
With this system the retention time of the substrates and
conversion products were: 1 (18.5 min), 2 (5.4 min), 4
(24.2 min), 5 (4.3 min).
Protein concentrations were determined using the
method described by Bradford (1976). UV spectra were
recorded on a spectrophotometer Kontron Uvikon 922
with Hellma type 100-QS 10.00 mm thickness precision
cuvettes.
1H NMR and 13C NMR 1D and 2D spectra of 4 were
determined at 400 and 100 MHz respectively on a Bru-
ker 400 MHz spectrometer. H NMR and H–1H auto-
correlation spectra of 5 and reference akagerine were
determined at 600 MHz; 13C NMR and HMQC spectra
of reference akagerine were recorded at 150 MHz on a
Bruker DMX 600 MHz spectrometer. Samples were run
in deuterated solvents with TMS as internal standard.
HRMS was performed on a Micromass Ultima spec-
trometer (5.0 mDa resolution) in electrospray (MS–
MS), positive ion; ES–MS–MS spectra was taken on a
Micromass VG-Quattro II in electrospray, positive ion,
ionisation energy 20 eV.
1
222 (4.19), 267 (3.54), 278 (3.55), 287 (3.43); H NMR
spectral data (400 MHz, CD3OD): ꢁ 1.55 (3H, d, J=6.7
Hz, H-18), ꢁ 1.90 (1H, m, H-14b), ꢁ 2.45 (1H, m, H-14a),
ꢁ 2.85 (1H, m, H-15), ꢁ 3.15 (1H, m, J=8.6, 7.5 Hz, H-
20), ꢁ 3.20 (1H, H-6b), ꢁ 3.27 (3H, s, N+CH3), ꢁ 3.50
(1H, d, H-5a), ꢁ 3.71 (3H, s, OCH3), ꢁ 3.75 (1H, q,
J=6.7, 8.6 Hz, H-19), ꢁ 4.30 (1H, H-5b), ꢁ 4.45 (1H, d,
J=7.5 Hz, H-21), ꢁ 5.00 (1H, d, H-3), ꢁ 7.07 (1H, t,
J=7.8 Hz, H-0), ꢁ 7.17 (1H, t, J=8.2 Hz, H-11), ꢁ 7.37
(1H, d, J=8.2 Hz, H-12), ꢁ 7.49 (1H, s, H-17), ꢁ 7.50
(1H, d, J=7.8 Hz, H-9); 13C NMR spectral data (100
MHz, CD3OD): ꢁ 10.8 (C-18), ꢁ 17.5 (C-6), ꢁ 25.6 (C-
14), ꢁ 32.7 (C-15), ꢁ 47.0 (N+CH3), ꢁ 51.3 (OCH3), ꢁ
59.5 (C-5), ꢁ 61.7 (C-19), ꢁ 69.0 (C-3), ꢁ 75.0 (C-20), ꢁ
97.0 (C-21), ꢁ 114.4 (C-12), ꢁ 118.2 (C-9), ꢁ 119.6 (C-10),
ꢁ 122.4 (C-11), ꢁ 144.0 (C-17); HRMS-MS–(EI) m/z (rel.
int.): 383.1978 [C22H27N2O4]+ (100), 365.2310 [M-
H2O]+ (16), 352.2046 (20), 340.1962 (69), 322.1725 (41),
308.1646 (27), 144.0979 (53); Positive ES-MS-MS m/z
(rel. int.): 383 [M]+ (100), 365 [M–H2O]+ (13), 340 (28),
222 (30), 194 (64), 170 (50), 144 (97), 130 (42);
1
1
4.3. Conversion of palicoside (2) into akagerine (5)
Twenty milligrams of palicoside (2) dissolved in 200 ml
of Na–Pi buffer pH 6.3 and 0.5 ml of enzymatic pre-
paration (36.2 mg proteinmlÀ1) obtained from the leaves
of Strychnos mellodora were mixed, homogenised and
incubated in a bain-marie at 30ꢀC. Lower specificity of
the glucosidase in the presence of palicoside prevented to
obtain a complete conversion of this substrate. Incuba-
tion was stopped after 72 h. In the TLC monitoring sys-
tem described above palicoside (2) reaches a Rf of 0.09.
Its conversion products, 5 and 6, are characterised by Rf
of 0.05 and 0.78 respectively. The palicoside (2) incuba-
tion mixture was then frozen and lyophilised. The resi-
due was taken in 2 ml H2O and then extracted by 2 ml
CH2Cl2 until complete disappearance of 5 from the aq.
layer. Each organic layer was separated by centrifuga-
tion 10 min at 2500 g. The joined organic layers were
then conc. under reduced pressure and submitted to a
prep. TLC in the system composed of silica gel and the
mobile phase EtOAc–iso–prop-NH4OH 4.25% (60:25:
15). The elution was performed by CH2Cl2 and allowed
4.2. Conversion of dolichantoside (1) into Nb-methyl-
21-ꢀ-hydroxy-mayumbine (4)
Half a milliliter of Na-Pi buffer pH 6.3 containing 50
mg dolichantoside (1) and 1 ml of enzymatic prepara-
tion (36.2 mg protein mlÀ1) obtained from the leaves of
Strychnos mellodora were mixed, homogenised and
incubated in a bain-marie at 30ꢀC until complete con-
version of dolichantoside, after 72 h. The conversion
was monitored by TLC analysis on silica gel with the
mobile phase EtOAc–iso–prOH–NH4OH 4.25%
(60:25:15). In this system, dolichantoside (1) appears at
Rf 0.43 and Nb-methyl-21-b-hydroxy-mayumbine (4) at
Rf 0.13.

658
V. Brandt et al. / Phytochemistry 57 (2001) 653–659
to obtain 7.38 mg of 5 as an amorphous white powder.
UV (MeCN) lmax nm (log "): 229 (4.51), 276 (3.78), 283
(3.80), 293 (3.68); H NMR spectral data (600 MHz,
mg/ml hydrolysed EtOH extract, 3.04 mg/ml hydrolysed
1 and 2.88 mg/ml hydrolysed 2. Strains were pre-culti-
vated following the NCCLS guidelines. Cultures were
realised in 96 wells microplates with lid. Wells 2–11
contained twofold decreasing dilutions of the different
substrates. Well 1 served as sterility control with 0.1 ml
RPMI 1640 medium and 0.1 ml sterile physiologic liquid
while well 12 served as growing control with 0.1 ml strain
suspension and 0.1 ml sterile physiologic liquid. Each
well was inoculated with 0.1 ml of one of the three
inoculum suspensions. The plates were then incubated
at 35ꢀC and examined on a spectrophotometer at 625
nm every 24 h until positive control reaches a optical
density of 0.2. CMI is determined as the substrate con-
centration that allows a reduction of the strain growth
to 20% of the positive control.
1
CDCl3): ꢁ 1.99 (1H, d, J=12.8 Hz, H-14b), ꢁ 2.07 (3H,
d, J=7.2 Hz, H-18), ꢁ 2.09 (1H, m, H-16b), ꢁ 2.19 (1H,
q, J=10.9, 12.8 Hz, H-14a), ꢁ 2.33 (1H, t, H-16a), ꢁ 2.53
(3H, s, NCH3), ꢁ 2.72–2.78 (3H, m, H-5b, H-6a, H-6b),
ꢁ 3.07 (1H, m, H-5a), ꢁ 3.66 (1H, t, J=11.7, 1.4 Hz, H-
15), ꢁ 3.82 (1H, d, J=10.9 Hz, H-3), ꢁ6.24 (1H, d,
J=3.7 Hz, H-17), ꢁ 6.55 (1H, q, J=7.2 Hz, H-19), ꢁ
7.11 (1H, t, J=7.7 Hz, H-10), ꢁ7.18 (1H, t, J=8.2 Hz,
H-11), ꢁ 7.29 (1H, d, J=8.2 Hz, H-12), ꢁ 7.48 (1H, d,
J=7.7 Hz, H-9), ꢁ 9.31 (1H, d, J=1.4 Hz, H-21); Posi-
tive ES–MS–MS m/z (rel. int.): 325 [M+H]+ (93), 307
[M+H-H2O]+ (5), 282 (100), 264 (73), 144 (5), 130 (3);
5 has the same behaviour as reference akagerine in the
TLC system described above (Rf=0.78), and with the
mobile phases CHCl3–CH3OH (80:20) (Rf=0.5) and
EtOAc–iso–prOH–NH4OH 4.25% (45:35:5) (Rf=0.73);
iodoplatinate reagent produces a pink color turning into
red in a few minutes.
Acknowledgements
This research was supported by the ‘‘Fondation Leon
Fredericq’’. We wish to thank the NMR laboratory
of the Gorleaus Laboratories (University of Leiden-
Netherlands) where the spectra of akagerine were
measured; Mrs C. Prosperi (Physical Institute,
University of Liege) for measuring the spectra of
Nb-methyl-21-b-hydroxy-mayumbine; Dr. M.-P. Hayette
(Department of Medical Microbiology, University of
Liege) and Dr. G. Piel (Department of Pharmaceutical
Technology, Institute of Pharmacy, University of Liege)
who realised the antimycotic tests. The manuscript has
benefited from the competence of Dr. P. DeTullio
(Department of Pharmaceutical Organic Chemistry,
Institute of Pharmacy, University of Liege).
4.4. Reference akagerine
1H NMR spectral data (600 MHz, CDCl3) exactly
match with 5 ones; 13C NMR spectral data (150 MHz,
CDCl3): ꢁ 15.2 (C-8), ꢁ 19.8 (C-6), ꢁ 29.1 (C-15), ꢁ 36.3
(C-14), ꢁ 37.4 (C-16), ꢁ 42.5 (NCH3), ꢁ 50.4 (C-5), ꢁ 60.8
(C-3), ꢁ 75.7 (C-17), ꢁ 108.3 (C-12), ꢁ 108.8 (C-7), ꢁ
118.3 (C-9), ꢁ 119.4 (C-10), ꢁ 121.3 (C-11), ꢁ 126.7 (C-8),
ꢁ 136.0 (C-13), ꢁ 137.0 (C-2), ꢁ 147.9 (C-20), ꢁ 150.5 (C-
19), ꢁ 194.9 (C-21).
4.5. Measure of antimycotic activity
The minimal inhibitory concentration (MIC) of 1, 2
and the crude Strychnos mellodora EtOH extract against
three ATCC1 (American Type Culture Collection)
reference strains were determined by using a broth
microdilution method, based upon the National Com-
mittee for Clinical Laboratory Standards (NCCLS-
USA) guidelines for broth microdilution technique
(M-27Aand M-38P) (NCCLS, 1997 and 1998). The
three references strains, Candida albicans (ATCC 10231),
Candida glabrata (ATCC 90030) and Aspergillus niger
(ATCC 16404), were tested in duplicate and in parallel
with the substrates alone and in presence of SMGD.
Preparation of tested antifungal stock solns was not
possible because of the low availability of the substrates.
1, 2 and EtOH extract were dissolved in H2O as acetates
(by addition of the HOAc molar equivalent), filtered,
frozen and then lyophilised. The residues were then
weighed and dissolved in RPMI-1640 standard medium
to give 7.6 mg/ml EtOH extract, 3.8 mg/ml 1 and 3.6
mg/ml 2. These solns were incubated for 24 h in the
presence of SMGD, then filtered to give solns of 6.08
References
Angenot, L., Dideberg, O., Dupont, L., 1975. Isolation and structure
of akagerine: a new type of indole alkaloid. Tetrahedron Letters 16,
1357–1358.
Arata, Y., Aoki, T., Hanaoka, M., Kamei, M., 1975. Studies on 1-
azabicyclo compounds. XXII. Stereochemistry of 9a-substituted
quinolizidine methiodides. Chemical and Pharmaceutical Bulletin
23, 333–339.
Bisset, N.G., 1980. Alkaloids of the loganiaceae. In: Phillipson, J.D.,
Zenk, M.H. (Eds.), Indole and Biogenetically Related Alkaloids.
Academic Press, London, pp. 34–35.
Bradford, M.M., 1976. Arapid and sensitive method for the quanti-
tation of microgram quantities of protein utilizing the principle of
protein dye binding. Analytical Biochemistry 72, 248–254.
Brandt, V., Geerlings, A., Tits, M., Delaude, C., van der Heijden, R.,
Verpoorte, R., Angenot, L., 2000. New strictosidine b-glucosidase
from Strychnos mellodora. Plant Physiology and Biochemistry 38,
187–192.
Brandt, V., Tits, M., Geerlings, A., Frederich, M., Penelle, J.,
Delaude, C., Verpoorte, R., Angenot, L., 1999. b-Carboline glu-
coalkaloids from Strychnos mellodora. Phytochemistry 51, 1171–
1176.
Brown, R.T., 1980. Biomimetic conversion of secologanin into alkaloids.

V. Brandt et al. / Phytochemistry 57 (2001) 653–659
659
In: Phillipson, J.D., Zenk, M.H. (Eds.), Indole and Biogenetically
Related Alkaloids. Academic Press, London, pp. 171–184.
Carey, F.A., Sundberg, R.J., 1997. Chimie organique avancee. In: De
Boeck (Ed.), Tome 2. Reactions et Syntheses. Universite Paris,
Bruxelles, pp. 14–15.
Phillipson, J.D., Supavita, N., 1983. Alkaloids of Uncaria elliptica.
Phytochemistry 22, 1809–1813.
Piel, G., Hayette, M.-P., Evrard, B., Van Hees, T., De Mol, P.,
Delattre, L., 1999. Acomparative study of the antimycotic activity
of a miconazole HP-b-cyclodextrin solution and a surfactant solu-
tion. Journal de Pharmacie de Belgique 54, 87–88.
Danieli, B., Lesma, G., Mauro, M., Palmisano, G., Passarella, D., 1995.
First enantioselective synthesis of (À)-akagerine by a chemoenzy-
matic approach. Journal of Organic Chemistry 60, 2506–2513.
Geerlings, A., 1999. Strictosidine b-d-glucosidase. PhD thesis, Uni-
versity of Leiden, The Netherlands.
Rolfsen, W., Olaniyi, A., Hylands, P.J., 1980. New tertiary alkaloids
of Strychnos decussata. Journal of Natural Products 43, 97–102.
Rueffer, M., Nagakura, N., Zenk, M.H., 1978. Strictosidine, the
common precursor for monoterpenoid indole alkaloids with 3a and
3b configuration. Tetrahedron Letters 18, 1593–1596.
Stevens, L.H., 1994. Formation and conversion of strictosidine in the
biosynthesis of monoterpenoid indole and quinoline alkaloids. PhD
thesis, University of Leiden, The Netherlands.
Hemscheidt, T., Zenk, M.H., 1980. Glucosidases involved in indole
alkaloid biosynthesis of Catharanthus cell cultures. FEBS Letters
110, 187–191.
Kisakurek, M.V., Leeuwenberg, A.J.M., Hesse, M., 1983. A chemo-
taxonomic investigation of the plant families of Apocynaceae,
Loganiaceae, and Rubiaceae by their indole alkaloid content. In:
Pelletier, S.W. (Ed.), Alkaloids — Chemical and Biological Per-
spectives, Vol. 1. John Wiley, New York, pp. 211–376.
Stockigt, J., 1980. The biosynthesis of heteroyohimbine alkaloids. In:
Phillipson, J.D., Zenk, M.H. (Eds.), Indole and Biogenetically
Related Alkaloids. Academic Press, London, pp. 113–141.
Sugiura, M., Takao, N., Sasaki, Y., 1977. Stereochemistry of quinoli-
zidines. II. Carbon-13 magnetic resonance spectra of N-methylqui-
nolizidinium ions. Chemical and Pharmaceutical Bulletin 25, 960–
963.
Leeuwenberg, A.J.M., 1969. The Loganiaceae of Africa VIII–Strych-
nos III. University of Wageningen, pp. 180–181.
Lounasmaa, M., Hanhinen, P., 1998. Biomimetic formation and
interconversion in the heteroyohimbine series. Heterocycles 48,
1483–1492.
Svendsen, A.B., Verpoorte, R., 1983. Chromatography of alkaloids,
part A: thin-layer chromatography. In: Journal of Chromatography
Library, Vol. 23A. Elsevier Scientific Publishing Company,
Amsterdam.
Lounasmaa, M., Jokela, R., Miettinen, J., Halonen, M., 1992. New,
short synthesis of (Æ)-geissoschizine. Heterocycles 34, 1497–1502.
Massiot, G., Delaude, C., 1988. African strychnos alkaloids. In: Brossi,
A. (Ed.), The Alkaloids. Academic Press, San Diego, pp. 211–239.
National Committee for Clinical Laboratory Standard, 1997. Refer-
ence method for broth dilution antifungal susceptibility testing of
yeasts; approved standard, M27-A.
Tits, M., Brandt, V., Wauters, J.-N., Delaude, C., Llabres, G., Ange-
not, L., 1996. Glucoindole alkaloids from stem bark of Strychnos
mellodora. Planta Medica 62, 73–74.
Uskokovic, M.R., Lewis, R.L., Partridge, J.J., Despreaux, C.W.,
Pruess, D.L., 1979. Asymmetric synthesis of allo-heteroyohimbine
alkaloids. Journal of American Chemical Society 101, 6742–6744.
Uusvuori, R., Lounasmaa, M., 1981. 13C NMR data of 3-iso-
ajmalicine and 19-epiajmalicine. Planta Medica 41, 406–407.
Wright, C.W., Bray, D.H., O’Neill, M.J., Warhurst, D.C., Phillipson,
J.-D., Quetin-Leclercq, J., Angenot, L., 1991. Antiamoebic and
antiplasmodial activities of alkaloids isolated from Strychnos usam-
barensis. Planta Medica 57, 337–340.
National Committee for Clinical Laboratory Standard, 1998. Refer-
ence method for broth dilution antifungal susceptibility testing of
Conidium-forming filamentous fungi; proposed standard, M38-P.
Pennings, E.J.M., van den Bosch, R.A., van der Heijden, R., Stevens,
L.H., Duine, J.A., Verpoorte, R., 1989. Assay of strictosidine syn-
thase from plant cell cultures by high performance liquid chroma-
tography. Analytical Biochemistry 176, 412–415.