HPLC-based bioactivity profiling of plant extracts: A kinetic assay for the identification of monoamine oxidase-A inhibitors using human recombinant monoamine oxidase-A

Article abstract of DOI:10.1016/j.phytochem.2004.07.024

An assay for the HPLC-based search for monoamine oxidase-A (MAO-A) inhibitors in plant extracts was established. It combines human recombinant MAO-A, expressed as GST-fusion protein in yeast, with a kinetic measurement of the conversion of kynuramine to 4-hydroxyquinoline. Substrate selectivity and kinetic parameters of the GST-fusion protein were comparable to the wild-type enzyme. The applicability of the assay to HPLC-based activity profiling was tested with plant extracts spiked with small amounts of known MAO inhibitors.

Full text of DOI:10.1016/j.phytochem.2004.07.024

PHYTOCHEMISTRY
Phytochemistry 65 (2004) 2885–2891
www.elsevier.com/locate/phytochem
HPLC-based bioactivity profiling of plant extracts: a kinetic assay
for the identification of monoamine oxidase-A inhibitors
using human recombinant monoamine oxidase-A
*
Kathrin Dittmann, Ulrike Riese, Matthias Hamburger
Institute of Pharmacy, Friedrich-Schiller-University Jena, Semmelweisstrasse 10, D-07743 Jena, Germany
Received 5 April 2004; accepted 13 July 2004
Available online 22 September 2004
Dedicated to Prof. Dr. Kurt Hostettmann on the occasion of his 60th birthday
Abstract
An assay for the HPLC-based search for monoamine oxidase-A (MAO-A) inhibitors in plant extracts was established. It com-
bines human recombinant MAO-A, expressed as GST-fusion protein in yeast, with a kinetic measurement of the conversion of
kynuramine to 4-hydroxyquinoline. Substrate selectivity and kinetic parameters of the GST-fusion protein were comparable to
the wild-type enzyme. The applicability of the assay to HPLC-based activity profiling was tested with plant extracts spiked with
small amounts of known MAO inhibitors.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Keywords: Monoamine oxidase A; Human recombinant; GST-fusion protein; HPLC; Activity profiling; Plant extract; Lead discovery
1. Introduction
libraries in a highly automated high-throughput process,
the search for natural product leads has come under
increasing pressure. The need for rapid and miniaturized
approaches enabling assessment of individual com-
pounds in an extract without macroscopic isolation
has led to various concepts for the interfacing of biolog-
ical and chemical analysis of complex matrices (van Els-
wijk et al., 2002; Hamburger, 2003). Among these,
HPLC-based activity profiling has been successfully ap-
plied to the isolation of novel lead structures (Ham-
burger et al., 2002; Danz et al., 2001; Fitch et al., 2003).
In the context of our search for new natural products
with action on various pharmacological targets of the
central nervous system (CNS) (Schmidt et al., 2002,
2003a,b), we needed a robust assay for profiling of
extracts with monoamine oxidase (MAO) inhibitory
activity. Monoamine oxidase (E.C. 1.4.3.4) is an FAD-
containing enzyme of the outer mitochondrial mem-
brane which catalyzes the oxidative deamination of a
Compounds of natural origin still play a major role
as drugs and as lead structures for the development of
synthetic drug substances. About 40% of the worldÕs
best selling drugs derive directly or indirectly from com-
pounds of biogenic origin (Newman et al., 2000). In the
past, bioactive natural products were typically discov-
ered and purified in a preparative, bioactivity-directed
approach involving substantial quantities of biological
material, manpower and time (Hamburger and Hostett-
mann, 1991). As a consequence of the pradigm shift in
drug discovery towards screening of large compound
*
Corresponding author. Present address: Department of Phar-
macy, Institute of Pharmaceutical Biology, University of Basel,
Klingelbergstrasse 50, CH-4056 Basel, Switzerland. Tel.: +41 61 267
14 25; fax: +41 61 267 14 74.
E-mail address: matthias.hamburger@unibas.ch (M. Hamburger).
0031-9422/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2004.07.024

2886
K. Dittmann et al. / Phytochemistry 65 (2004) 2885–2891
number of biogenic amines in brain and peripheral tis-
sues, and of dietary amines and xenobiotics. MAO oc-
curs in two isoforms, MAO-A and MAO-B, which
differ in substrate selectivity, inhibitor sensitivity and tis-
sue distribution (Shih et al., 1999; Abell and Kwan,
2001). Inhibitors of MAO are of considerable pharma-
cological and therapeutic interest. MAO-A inhibitors
are useful in the therapy of mental disorders, mainly
as antidepressants (Wouters, 1998), whereas MAO-B
inhibitors can be used in the treatment of neurodegener-
ative disorders such as ParkinsonÕs and AlzheimerÕs dis-
ease (Foley et al., 2000). Various compounds have been
described, including reversible and irreversible inhibi-
tors, with selectivity for MAO-A or MAO-B. Clinically
used drug substances include, among others, moclobe-
et al., 1997). It was, however, prone to false positive re-
sults. Extract constituents with a redox potential lower
than that of the reagent 4-aminoantipyrin acted as pro-
ton donors in the peroxidase reaction, and widely occur-
ring phenolic substances could condense with the
oxidized form of 4-aminoantipyrin to coloured products
(Schmidt et al., 2003a). An assay with kynuramine as
substrate (Weissbach et al., 1960) proved to be more ro-
bust in that respect, but the significant turbidity and low
specific activity of the microsomal preparation rendered
the assay unsuitable for HPLC-based activity profiling.
Here, we describe the development of a kinetic assay
with human recombinant MAO-A and kynuramine as
substrate, and we validate its suitability for profiling
purposes with the example of plant extracts spiked with
small concentrations of known MAO inhibitors.
mide, a selective reversible MAO-A inhibitor, and
L-
deprenyl, an irreversible selective inhibitor of MAO-B.
A number of plant extracts reportedly inhibit MAO
(McKenna et al., 1984; Schurr and Rigor, 1984; Dar
and Khatoon, 2000; Khalil et al., 2000), and xanthones
(Suzuki et al., 1981; Gnerre et al., 2001a), stilbenoids
(Han et al., 1990), b-carbolines (McKenna et al.,
19984), naphthoquinones (Khalil et al., 2000) and ben-
zopyranes (Gnerre et al., 2001b), have been identified
as plant-derived MAO inhibitors.
2. Results and discussion
Saccharomyces cerevisiae was chosen for the expres-
sion of recombinant MAO-A. Full enzymatic function-
ality of the enzyme could be expected in this common
eukaryotic system, because of recent reports of MAO
expression in yeast cells (Li et al., 2002). The MAO-A
cDNA was cloned into the high-copy yeast expression
vector pEG-KT, to generate a GST MAO-A fusion pro-
tein. Orientation and correctness of the cloned cDNA
construct was confirmed by sequencing. The expression
of the protein occurred under the control of a galactose-
inducible promoter. The fusion protein was expressed as
a mitochondrial membrane protein. The mitochondrial
fraction was, therefore, separated from cell homogen-
ates via differential centrifugation steps. Extraction with
Triton X 100, precipitation with ammonium sulfate and
dialysis afforded a MAO preparation with high specific
activity. A high content of fusion protein in the dialysate
Various protocols for assaying MAO activity have
been reported. Enzyme preparations of MAO-A and B
have been obtained mostly from various animal tissues,
such as brain, kidney, and beef, pig or rat liver (Salach,
1979; Lu et al., 1996), whereas human placenta has
served as a source for MAO-A (Weyler and Salach,
1985). A common drawback of these preparations is
the moderate specific activity and the turbidity of the
microsomal pellets which negatively affects the accuracy
of photometric measurements. Recently, recombinant
MAO-A and B (Urban et al., 1991; Newton-Vinson
et al., 2000; Li et al., 2002) from yeast expression sys-
tems, and MAO-B from baculovirus-infected insect cells
(Rebrin et al., 2001) were reported but have not found
so far application for screening purposes. The expres-
sion of functional MAO-A in Escherichia coli has not
been successful (Bach et al., 1988). Activity measure-
ments use photometric, fluorimetric and radiometric
readouts with typically single point, rarely continuous
measurement (Zhou et al., 1996) of conversion products.
These assay protocols were orginally developed for
the testing of discrete drug substances. Screening of ex-
tracts and semi-purified fractions, however, poses a
number of characteristic problems related to the nature
of these materials. They are usually highly coloured and
contain complex mixtures of unknown compounds,
some of which may interfere with the detection. In the
context of a screening programme for new natural prod-
ucts with action on CNS-related targets (Schmidt et al.,
2003a), we initially used a MAO assay with a rat mit-
ochondrial preparation and a peroxidase readout (Holt
Fig. 1. SDS–PAGE of GST MAO-A fusion protein preparations.The
dialysate after precipitation with ammonium sulfate (a) and the eluate
of fusion protein after Glutathion–Sepharose affinity chromatography
(b) are shown.

K. Dittmann et al. / Phytochemistry 65 (2004) 2885–2891
2887
was visualized by SDS–PAGE (Fig. 1(a)). Further puri-
fication of the GST fusion protein on Glutathione–
Sepharose yielded a preparation which showed four
distinct bands on SDS–PAGE with the correct molecu-
lar mass for GST MAO-A (94 kDa), MAO-A (65 kDa),
Schistosoma japonicum GST from pEG-KT vector (29
kDa) and yeast GST (26 kDa) (Fig. 1(b)). This purifica-
tion step, however, led to an almost complete loss of
enzymatic activity of the membrane-bound enzyme. It
appears that the fusion protein undergoes a partial deg-
radation during elution from the affinity column, result-
ing in truncated protein fragments. Therefore, the
dialysate (Fig. 1(a)) was used for subsequent analyses.
Kinetic data were measured with kynuramine as sub-
strate. From the Lineweaver–Burk plot of the recombi-
nant enzyme, a Km of 150 lM was calculated. This value
is in good agreement with data reported earlier for
MAO-A (Weyler and Salach, 1985; Tan et al., 1991).
Next, we assessed the stability of the enzyme prepara-
tion. Enzyme solutions (0.05 mg/ml protein concentra-
tion) were kept at 22 and 4 ꢁC for 72 h, and the
activity determined at defined intervals (Fig. 2). At room
temperature, the enzyme showed a gradual loss of activ-
ity to approx. 50% of its initial value. The activity re-
mained at >80% of the initial value when the enzyme
was kept at 4 ꢁC. Selectivity of the enzyme preparation
was determined with clorgyline and deprenyl, which
are selective inhibitors of MAO-A and B, respectively
(Fig. 3). The IC50 determined for clorgyline was 0.014
lM, and 20.0 lM in case of deprenyl. The IC50 values
for the two inhibitors and their ratio were comparable
to data reported with the wild type enzyme (Shih
et al., 1999).
Fig. 3. Functional characterization of recombinant MAO-A. The dose
inhibitory curves for the selective MAO-A inhibitor clorgyline (1.95–
62.5 nM) and the selective MAO-B inhibitor deprenyl (0.5–50 lM) are
shown. Each data point was determined as triplicate.
absorption and/or fluorescence in order to test the
robustness of the assay and its applicability towards
the profiling of complex matrices. We chose chamomile
flowers and angelica root as two medicinal plants which
are not known to possess any CNS-related activity of
significance, and we spiked them with defined amounts
of known MAO-A inhibitors. Chamomile (Matricaria
recutita L., Asteraceae) is widely used as a antiphlogis-
tic, spasmolytic and antimicrobial in topical and oral
application. The flower heads contain lipophilic sequi-
terpenoids, polyacetylenic compounds and numerous
flavonoids (Wichtl, 2002). Angelica root (Angelica arch-
angelica L, Apiaceae) is used as a stimulant of appetite
and digestion. It is rich in simple coumarins, linear
and angular furanocoumarins, and in phenylpropanoids
such as caffeic acid derivatives (Wichtl, 2002). Dichloro-
methane extracts were prepared by accelerated solvent
extraction (ASE). The chamomile extract was spiked
with 0.5% of harmane, a plant derived b-carbolin alka-
loid and known reversible MAO-A inhibitor. The angel-
ica extract was spiked with 0.5% of the synthetic,
irreversible MAO-A inhibitor clorgyline.
The applicability of the assay was tested with crude
plant extracts. The extracts were selected to contain
structurally diverse compounds with strong UV/Vis
Gradient HPLC profiles of the extracts with and
without spiked MAO inhibitor are shown in Fig. 4.
The MAO inhibition of corresponding HPLC-micro-
fractions are shown above the respective chromato-
grams. The UV trace of chamomile extract recorded at
230 nm (Fig. 4(a)) shows a few distinct peaks of flavo-
noids, whereas the terpenoids are not detected due to
lack of a chromophore. The peak of harmane is easily
detected at Rt = 3.5 min in the chromatogram of the
spiked extract due to the strong UV absorption of the
b-carboline chromophore. A total of seven microfrac-
tions were collected from the effluent on the basis of
the HPLC profile, and inhibition of MAO was deter-
mined. In the spiked extract (black bars), a significant
Fig. 2. Stability of the enzyme activity after time-dependent pre-
incubation of the crude enzyme at 4 and 22 ꢁC. Each datapoint was
determined as triplicate.

2888
K. Dittmann et al. / Phytochemistry 65 (2004) 2885–2891
The HPLC profile of the angelica extract (Fig. 4(b))
showed numerous compounds with strong absorption
at 230 nm. DAD spectra suggested that they were
mainly coumarins. The clorgyline peak (Rt = 9.5 min)
in the spiked extract, in contrast, was hardly detectable.
Interestingly, the coumarins themselves possessed some
MAO-A inhibitory properties, as the bar graph of the
untreated extract shows. Given that the degree of inhibi-
tion was not proportional to the peak areas in the chro-
matogram indicated that this was true inhibition of the
enzyme and not a detection artifact due to absorption
of the coumarins. In the activity profile of the spiked ex-
tract (black bars), the clorgyline peak in fraction V was
clearly discernible due to a virtually 100% inhibition of
MAO. Again, activity of other microfractions was not
significantly altered.
3. Conclusion
A kinetic assay for the detection of MAO-A inhibi-
tors in plant extracts was established, using recombinant
human MAO-A expressed as GST-fusion protein in
yeast. The enzyme showed functional characteristics
and kinetic parameters comparable to those of wild-type
MAO-A. As a membrane-associated enzyme, highly
purified MAO-A exhibited low activity. The purified
yeast mitochondrial preparation, however, showed high
MAO-A content, satisfactory stability and, in contrast
to liver microsomal preparations, no turbidity. Repro-
ducibility of data and robustness of the assay were sig-
nificantly improved. The suitability of the assay for
HPLC-based profiling was validated with two spiked
plant extracts. In both cases, low concentrations of
MAO-inhibitor were reliably detected. The example of
angelica extract showed that the activity of minor peaks
could be located in complex chromatograms containing
numerous major extract constituents with strong
chromophores.
4. Experimental
Fig. 4. HPLC-based activity profiling with chamomile (a) and angelica
(b) extracts for MAO-A activity. The HPLC chromatograms show an
overlay of the original extracts (grey), and extracts spiked with 0.5%
harmane or clorgyline, respectively (black lines). Detection was in both
extracts at 230 nm. The amount of extract injected was 500 lg.
Chromatographic conditions are given in Section 4. The superposed
bar graphs above the chromatograms show the inhibitory activity of
individual HPLC fractions collected from a single separation of the
original extract (grey bars) and the spiked extract (black bars). Each
fraction was tested in triplicate.
4.1. Chemicals and reagents
Kynuramine dihydrobromide, aprotinin, leupeptin,
R-(ꢀ)-deprenyl hydrochloride and clorgyline hydrochlo-
ride were purchased from Sigma Chemical (Taufkirchen,
Germany). EDTA, tris(hydroxymethyl)aminomethane
and Tritonꢂ X 100 were from Merck (Darmstadt, Ger-
many). Restriction endonucleases were obtained from
Fermentas (St. Leon-Rot, Germany), and primers for
PCR were synthesized by MWG Biotech (Ebersberg,
Germany).
activity was observed in fraction II which corresponded
to the harmane peak. The corresponding fraction of the
original extract, in contrast, showed only marginal inhi-
bition (grey bars). For all other fractions, no significant
activity differences were found between the spiked and
non-spiked extracts.
Media constituents were purchased from Roth (Kar-
lsruhe, Germany), with the exception of yeast nitrogen
base (Sigma Chemical, Taufkirchen, Germany). HPLC

K. Dittmann et al. / Phytochemistry 65 (2004) 2885–2891
2889
grade solvents were from Roth (Karlsruhe, Germany).
Pharmacopoeia-grade chamomile flowers and angelica
root were purchased from Caesar & Loretz GmbH (Hil-
den, Germany).
mg/l,
Isoleu 50 mg/l,
20 mg/l, -Phe 50 mg/l,
l, -Tyr 50 mg/l, and valine 140 mg/ml) was grown
L
-Arg 50 mg/l,
-Leu 100 mg/l,
-Thr 100 mg/l,
L
-Asp 80 mg/l,
-Lys 50 mg/l,
-Trp 50 mg/
L
-His 20 mg/l,
L-
L
L
L
-Met
L
L
L
L
up to a OD600 of 1.52. The starter culture was trans-
ferred into the fermenter containing 5.5 l of fresh glu-
cose (1%) ‘‘fermentation-medium’’. In the batch phase
of 10 h, a OD600 of 4.84 was reached. In the following
feed batch phase of 9 h, the OD600 increased to 8.84.
Induction was achieved with galactose (3%) ‘‘fermenta-
tion-medium’’. The final volume was 9 l. At the end of
a 48 h induction phase, a OD600 of 21 was measured.
Cells (wet weight 210 g) were separated by centrifuga-
tion and stored at ꢀ70 ꢁC until further processing.
4.2. Culturing of microbial strains
Plasmid bearing E. coli strains were grown at 37 ꢁC in
LB medium (trypton 10 g/l, yeast extract 5 g/l, NaCl 10
g/l) containing 100 lg/ml ampicillin. Transfected clones
of S. cerevisiae were cultured at 30 ꢁC in glucose mini-
mal medium (glucose 2 g/l, yeast nitrogen base 6,7 g/l,
L
-Trp 20 mg/l, L-His 20 mg/l, L-Leu 30 mg/l). For cultur-
ing in Petri dishes, the medium contained 1.5% (w/v)
agar.
4.5. Preparation of recombinant MAO-A
4.3. Cloning, sequencing and expression
All purification steps were carried out at 4 ꢁC. Por-
tions (70 g) of yeast cells were resuspended in 1 l of
TRIS/EDTA buffer (containing 0.15 lM aprotinin and
10 lM leupeptin) and treated with a EmulsiFlex C5
homogenizer. The homogenization was repeated two
times to achieve quantitative cell disruption. Cell debris
were removed by centrifugation (1590g, 15 min). Mit-
ochondrial fraction was obtained from the supernatant
by ultracentrifugation (70,000g, 20 min). The pellet
was stored at ꢀ70 ꢁC.
The cDNA sequence encoding human liver and pla-
centa MAO-A contained in pUC plasmid (clone
HM11) was purchased from ATCC (Manassas, USA).
The MAO-A coding region contained in HM11 was
amplified by PCR using the following primers flanked
by XbaI- and SalI-sites: 50-GCT CTA GAT AGG
CAC AGG AGT GG-30 and 30-ACG CGT CGA
CGG CCC AAT GAC ACA G-50. The PCR-fragment
was ligated into the XbaI and SalI-sites of pEG-KT
(Mitchell et al., 1993) and transformed into E. coli strain
DH5a (Clontech, Heidelberg, Germany) by calcium-de-
pendent transformation (Mandel and Higa, 1970).
Insertion and correct orientation were confirmed by
restriction analysis and partial sequencing performed
by commercial services (MWG, Ebersberg, Germany).
The pEG-KT/MAO-A construct was introduced into
S. cerevisiae strain CEN-PK2 (received from K.D. Enti-
an, Frankfurt) via polyethyleneglycol-induced genetic
transformation (Klebe et al., 1983). The strain is defi-
cient for the his3, trp1, leu2 and ura3 genes. Because of
the URA3 marker on the plasmid, transformants were
screened on plates lacking uracil.
Portions of the pellet (3 g) were resuspended in 15 ml
Tris/EDTA buffer (containing 0.15 lM aprotinin and 10
lM leupeptin) with addition of Triton X 100 (0.5% v/v)
and homogenized in a Wheaton homogenizer. The sup-
ernatant obtained by ultracentrifugation (160,000g, 40
min) was stored on ice, the pellet resuspended and trea-
ted as above. The supernatants were combined and pre-
cipitated with addition of portions of aqueous
ammonium sulfate (65%, m/v). The pellet obtained by
centrifugation (4000g, 30 min) was resuspended in
Tris/EDTA buffer (20 ml) and dialyzed against Tris/
EDTA buffer. The dialysate was stored at ꢀ70 ꢁC or
used for purification by affinity chromatography.
An aliquot of the dialysate was purified over Glutath-
ion–Sepharosee 4B (Amersham Biosciences, Freiburg,
Germany) following the manufacturerÕs protocol. The
MAO-A GST fusion protein was eluted with glutath-
ion/Tris HCl/NaCl buffer but was devoid of activity.
4.4. Fermentation
The fermentation of transformed S. cerevisiae was
carried out in a Biostat ED 10 fermentor equipped with
a digital control unit for standard process control and a
multi fermentor control system for data management.
The reactor was operated at 30 ꢁC, 200 mbar and at
a pH 5, an aeration rate of 10 l/min and a stirring
speed of 310 rpm. On-line analysis and adjustment of
carbon dioxide and oxygen, of glucose and pH was car-
ried out. The fermentation process was divided into a
batch phase (phase I), a feed batch phase (phase II)
and a induction phase (phase III). A starter culture
(1 l) in glucose (2%) ‘‘fermentation-medium’’ (yeast
nitrogen base without amino acids 6.7 g/l, adenine 10
4.6. Characterization of enzyme
Purity and molecular mass of the recombinant pro-
tein was checked by SDS–PAGE on 4–12% Bis–Tris-
Gels (Invitrogen, Karlsruhe, Germany) in comparison
with a prestained protein ladder 10–160 kDa (Fermen-
tas, St. Leon-Rot, Germany) as marker. The gels were
stained with Coomassie Brilliant Blue R-250.
Kinetic data were determined at 30 ꢁC with the dial-
ysate (protein concentration 0.05 mg/ml) and varying

2890
K. Dittmann et al. / Phytochemistry 65 (2004) 2885–2891
concentrations of kynuramin dihydrobromide (31.25–
250 lM) in Soerensen-phosphate buffer. Absorption at
360 nm was recorded every 60 s over 10 min and meas-
ured against a blank (dialysate in buffer).
Specific inhibition of the enzyme was determined
with clorgyline (1.95–62.5 nM), and with deprenyl
(0.5–50 lM) according to the assay protocol below.
For testing of the temperature stability of the enzyme
preparation, the dialysate was kept at room tempera-
ture and at 4 ꢁC. Activity measurements were carried
out at 0, 4, 8, 12, 24, 32, 48, 56 and 72 h, whereby
the activity of the freshly thawed preparation was con-
sidered as 100%.
station. Separations were performed on a LiChrospher
100 RP-18e cartridge (5 lm particle size, 125 mm · 4
mm i.d.; Merck, Darmstadt). Gradient elution was car-
ried out with acetonitrile (solvent A) and 10 mM aque-
ous ammonium acetate (solvent B1, angelica extract), or
25 mM aqueous ammonium acetate (solvent B2, cham-
omile extract). The separation of chamomile extract was
achived with the following gradient profile: 40% A (0
min) to 70% A (20 min), 100% A (22 min), hold for 1
min. The gradient profile for angelica root extract was
as follows: 50% A (0 min) to 85% (20 min), 100% (22
min), hold for 2 min. The flow rate was 1 ml/min, and
detection was carried out at 230 nm. The injection vol-
ume was 10 or 20 ll, depending on the concentration
of the extract.
4.7. Assay for monoamine oxidase-A inhibitory activity
The dialysate was diluted to 0.05 mg/ml protein with
Soerensen-phosphate buffer. 495 ll of this dilution was
incubated under shaking for 5 min at 30 ꢁC with
DMSO solutions (5 ll) of test samples. About 140 ll
of the reaction mixture were transferred to a 96-well
microtiter plate, and 10 ll of 7.5 mM kynuramine
dihydrobromide in Soerensen-phosphate buffer were
added (final concentration in the assay 500 lM). After
brief shaking, absorption measurement at 360 nm was
carried out at 30 ꢁC in a thermostatted plate reader
(FLUOStar Galaxy, BMG Labtechnologies, Offenburg)
every 60 s over a period of 10 min. Inhibition (%) was
calculated from the slope DE/Dt, with control experi-
ments without enzyme as 0% and incubations without
sample as 100% activity. All assays were carried out
in triplicate.
4.10. Activity profiling
For profiling studies, dried dichloromethane extracts
were dissolved in acetonitrile. The concentrations were
10 mg/ml for angelica extract, and 5 mg/ml for the
chamomile extract. Aliquots of 10 ll (angelica) and
20 ll (chamomile) were injected, corresponding to
500 lg of dry extract. Spiked extracts contained in
addition 0.5% harman (chamomile extract) or 0.5%
clorgyline (angelica extract). The column effluent was
fractionated into vials to obtain seven (chamomile) or
eight (angelica) fractions, respectively. The solvent
was removed with a vacuum concentrator (Evaporator
Centrifuge RC 10.22, Jouan, Saint Herblain, France).
The dry film was redissolved in 50 ll DMSO for
bioassay.
4.8. Extraction
Acknowledgements
Extracts were prepared by accelerated solvent extrac-
tion (ASE) using a Dionex ASE 200 extractor with sol-
vent module (Dionex, Sunnyvale, USA). The plant
material was ground in a SMZ ultracentrifugal mill
(Retsch, Haan, Germany), and 1 g of powdered drug
were filled into extraction cartridges (volume 22 ml).
The herbs were extracted at 80 ꢁC and 120 bar with
dichloromethane, followed by methanol. For each sol-
vent, two extraction cycles of 5 min were carried out.
These conditions have been previously shown to provide
exhaustive extraction of medicinal plants (Benthin et al.,
1999). The extracts were evaporated to dryness with a
rotary evaporator and stored at 4 ꢁC until use.
The author are grateful to Prof. G. Diekert, Jena, for
access to lab. The CEN-PK2 strain was a gift from Dr.
K.D. Entian, Frankfurt. We thank Dr. M. Ramm and
U. Knupfer, Hans-Kno¨ll-Institute for Natural Product
¨
Research, Jena, for the high density fermentation.
Financial Support by the Fonds der Chemischen Indust-
rie (to M.H.) is gratefully acknowledged.
References
Abell, C.W., Kwan, S.W., 2001. Molecular characterization of
monoamine oxidases A and B. Prog. Nucl. Acids Res. Mol. Biol.
65, 129–156.
4.9. HPLC
Bach, A.W.J., Lan, N.C., Johnson, D.L., Abell, C.W., Bembenek,
M.E., Kwan, S.-W., Seeburg, P.H., Shih, J.C., 1988. cDNA cloning
of human liver monoamine oxidase A and B: molecular basis of
differences in enzymatic properties. Proc. Natl. Acad. Sci. USA 85,
4934–4938.
The HPLC equipment consisted of a HP 1100 system
(Hewlett–Packard, Waldbronn) with high pressure
mixing pump, solvent degasser, column thermostat,
photodiode array detector, autosampler and HP Chem-
Benthin, B., Danz, H., Hamburger, M., 1999. Pressurized liquid
extraction of medicinal plants. J. Chromatogr. A 837, 211–219.

K. Dittmann et al. / Phytochemistry 65 (2004) 2885–2891
2891
Danz, H., Stoyanova, S., Wippich, P., Brattstro¨m, A., Hamburger, M.,
2001. Identification and isolation of the cyclooxygenase-2 inhibi-
tory principle in Isatis tinctoria. Planta Med. 67, 411–416.
Dar, A., Khatoon, S., 2000. Behavioral and biochemical studies of
dichloromethane fraction from the Areca catechu nut. Pharmacol.
Biochem. Behav. 65, 1–6.
Mitchell, D.A., Marshall, T.K., Deschenes, R.J., 1993. Vectors for the
inducible overexpression of glutathione S-transferase fusion pro-
teins in yeast. Yeast 9, 715–723.
Newman, D.J., Cragg, G.M., Snader, K.M., 2000. The influence of
natural products upon drug discovery. Nat. Prod. Rep. 17, 215–
234.
Fitch, R.W., Garraffo, H.M., Spande, T.S., Yeh, H.J.C., Daly, J.W.,
2003. Bioassay-guided isolation of epiquinamide, a novel quino-
lizidine alkaloid and nicotinic agonist from an Equadoran poison
frog Epidobates tricolor. J. Nat. Prod. 66, 1345–1350.
Newton-Vinson, P., Hubalek, F., Edmondson, D.E., 2000. High level
expression of human liver monoamine oxidase-B in Pichia pastoris.
Protein Expr. Purif. 20, 334–345.
Rebrin, I., Geha, R.M., Chen, K., Shih, J.C., 2001. Effects of carboxyl-
terminated truncations on the activity and solubility of human
monoamine oxidase-B. J. Biol. Chem. 276, 29499–29506.
Salach, J.I., 1979. Monoamine oxidase from beef liver mitochondria:
simplified isolation procedure, properties, and determination of its
cysteinyl flavin content. Arch. Biochem. Biophys. 192, 128–137.
Foley, P., Gerlach, M., Youdim, M.B.H., Riederer, P., 2000. MAO-B
inhibitors: multiple roles in the therapy of neurodegenerative
disorders?. Parkinson Rel. Disorders 6, 25–47.
Gnerre, C., Thull, U., Gaillard, P., Carrupt, P.A., Testa, B.,
Fernandes, E., Silva, F., Pinto, M., Pinto, M.M.M., Wolfender,
J.L., Hostettmann, K., Cruciani, G., 2001a. Natural and synthetic
xanthones as monoamine oxidase inhibitors: biological assay and
3D-QSAR. Helv. Chim. Acta 84, 552–570.
Schmidt, K., Gunther, W., Stoyanova, S., Schubert, B., Li, Z.,
¨
Hamburger, M., 2002. Militarinone A, a neurotrophic pyridone
alkaloid from Paecylomyces militaris. Org. Lett. 4, 197–199.
Schmidt, K., Li, Z., Schubert, B., Huang, B., Stoyanova, S.,
Hamburger, M., 2003a. Screening of entomopathogenic Deutero-
mycetes for activities on targets involved in degenerative diseases of
the central nervous system. J. Ethnopharmacol. 89, 251–260.
Schmidt, K., Riese, U., Li, Z., Hamburger, M., 2003b. Novel tetramic
acids and pyridone alkaloids, militarinones B, C and D, from the
insect pathogenic fungus Paecylomyces militaris. J. Nat. Prod. 66,
378–383.
Gnerre, C., von Poser, G.L., Ferraz, A., Viana, A., Testa, B., Rates,
S.M.K., 2001b. Monoamine oxidase inhibitory activity of some
Hypericum species native to South Brazil. J. Pharm. Pharmacol. 53,
1273–1279.
Hamburger, M., Hostettmann, K., 1991. Bioactivity in plants: the link
between phytochemistry and medicine. Phytochemistry 30, 3864–
3874.
Hamburger, M., Danz, H., Dittmann, K., 2002. HPLC-based activity
profiling of plant extracts. Screening Trends Drug Discov. 3, 32–35.
Hamburger, M., 2003. Tracking bioactivity in plant extracts new
concepts and approaches. In: Sprecher, E., Caesar, W. (Eds.),
Society for Medicinal Plant Research – 50 Years 1953–2003.
Wissenschaftliche Verlagsgesellschaft, Stuttgart, pp. 109–122.
Han, Y.N., Ryu, S.Y., Han, B.H., 1990. Antioxidant activity of
resveratrol closely correlates with its monoamine oxidase-A inhib-
itory activity. Arch. Pharm. Res. 13, 132–135.
Schurr, A., Rigor, B.M., 1984. Cannabis extract, but not D1-
tetrahydrocannabinol, inhibits human brain and liver monoamine
oxidase. Gen. Pharmacol. 15, 171–174.
Shih, J.C., Chen, K., Ridd, M.J., 1999. Monoamine oxidase: from gene
to behaviour. Annu. Rev. Neurosci. 22, 197–217.
Suzuki, O., Katsumata, Y., Oya, M., Chari, Vermes, V.M., Wagner,
B., Hostettmann, H.K., 1981. Inhibition of type A and type B
monoamine oxidases by naturally occurring xanthones. Planta
Med. 42, 17–21.
Holt, A., Sharman, D.S., Baker, G.B., Palcic, M.M., 1997.
A
continuous spectrophotometric assay for monoamine oxidase and
related enzymes in tissue homogenates. Anal. Biochem. 244, 384–
392.
Tan, A.K., Weyler, W., Salach, J.I., Singer, T.P., 1991. Differences in
substrate specificities of monoamine oxidase-A from human liver
and placenta. Biochem. Biophys. Res. Commun. 181, 1084–1088.
Urban, P., Andersen, J.K., Hsu, H.P.P., Pompon, D., 1991. Compar-
ative membrane locations and activities of human monoamine
oxidases expressed in yeast. FEBS Lett. 286, 142–146.
van Elswijk, D.A., Irth, H., 2002. Analytical tools for the detection
and characterization of biologically active compounds from nature.
Phytochem. Rev. 1, 427–439.
Khalil, A.A., Steyn, S., Castagnoli, N., 2000. Isolation and character-
ization of a monoamine oxidase inhibitor from tobacco leaves.
Chem. Res. Toxicol. 13, 31–35.
Klebe, J.K., Harris, J.V., Sharp, D., Douglas, M.G., 1983. A general
method for polyethyleneglycol-induced genetic transformation of
bacteria and yeasts. Gene 25, 333–341.
Li, M., Hubalek, F., Newton-Vinson, P., Edmondson, D.E., 2002.
High level expression of human liver monoamine oxidase-A in
Pichia pastoris: comparison with the enzyme expressed in Saccha-
romyces cerevisiae. Protein Expr. Purif. 24, 152–162.
Lu, G., Unge, T., Owera-Atepo, J.B., Shih, J.C., Ekblom, J., Oreland,
L., 1996. Characterization and partial purification of human
monoamine oxidase-B expressed in Escherichia coli. Protein Expr.
Purif. 7, 315–322.
Weissbach, H., Smith, T.E., Daly, J.W., Witkop, B., Udenfried, S.,
1960. A rapid spectrophotometric assay of monoamine oxidase
based on the rate of dissapearance of kynuramine. J. Biol. Chem.
235, 1160–1163.
Weyler, W., Salach, J.I., 1985. Purification and properties of mitoch-
ondrial monoamine oxidase type A from human placenta. J. Biol.
Chem. 260, 13199–13207.
Wichtl, M. (Ed.), 2002. Teedrogen und Phytopharmaka, fourth ed.
Wissenschaftliche Verlagsgesellschaft, Stuttgart.
Mandel, M., Higa, A., 1970. Calcium-dependent bacteriophage DNA
infection. J. Mol. Biol. 53, 159–162.
Wouters, J., 1998. Structural aspects of monoamine oxidase and its
reversible inhibition. Curr. Med. Chem. 5, 137–162.
McKenna, D.J., Towers, G., Abbott, H.N.F., 1984. Monoamine
oxidase inhibitors in South American hallucinogenic plants:
tryptamine and b-carboline constituents of Ayahuasca. J. Ethno-
pharmacol. 10, 195–223.
Zhou, J.J.P., Zhong, B., Silverman, R.B., 1996. Direct continuous
flurometric assay for monoamine oxidase-B. Anal. Biochem. 234,
9–12.