Tissue distribution and biosynthesis of 1,2-saturated pyrrolizidine alkaloids in Phalaenopsis hybrids (Orchidaceae)

Article abstract of DOI:10.1016/j.phytochem.2006.05.031

Phalaenopsis hybrids contain two 1,2-saturated pyrrolizidine monoesters, T-phalaenopsine (necine base trachelanthamidine) and its stereoisomer Is-phalaenopsine (necine base isoretronecanol). T-Phalaenopsine is the major alkaloid accounting for more than 90% of total alkaloid. About equal amounts of alkaloid were genuinely present as free base and its N-oxide. The structures were confirmed by GC-MS. The quantitative distribution of phalaenopsine in various organs and tissues of vegetative rosette plants and flowering plants revealed alkaloid in all tissues. The highest concentrations were found in young and developing tissues (e.g., root tips and young leaves), peripheral tissues (e.g., of flower stalks) and reproductive organs (flower buds and flowers). Within flowers, parts that usually attract insect visitors (e.g., labellum with colorful crests as well as column and pollinia) show the highest alkaloid levels. Tracer feeding experiments with ¹⁴C-labeled putrecine revealed that in rosette plants the aerial roots were the sites of phalaenopsine biosynthesis. However active biosynthesis was only observed in roots still attached to the plant but not in excised roots. There is a slow but substantial translocation of newly synthesized alkaloid from the roots to other plant organs. A long-term tracer experiment revealed that phalaenopsine shows neither turnover nor degradation. The results are discussed in the context of a polyphyletic molecular origin of the biosynthetic pathways of pyrrolizidine alkaloids in various scattered angiosperm taxa. The ecological role of the so called non-toxic 1,2-saturated pyrrolizidine alkaloids is discussed in comparison to the pro-toxic 1,2-unsaturated pyrrolizidine alkaloids. Evidence from the plant-insect interphase is presented indicating a substantial role of the 1,2-saturated alkaloids in plant and insect defense.

Full text of DOI:10.1016/j.phytochem.2006.05.031

PHYTOCHEMISTRY
Phytochemistry 67 (2006) 1493–1502
www.elsevier.com/locate/phytochem
Tissue distribution and biosynthesis of 1,2-saturated
pyrrolizidine alkaloids in Phalaenopsis hybrids (Orchidaceae)
,1
*
Cordula Fro¨lich, Thomas Hartmann, Dietrich Ober
Institut fu¨r Pharmazeutische Biologie, Technische Universita¨t Braunschweig, Mendelssohnstrasse 1, D-38106 Braunschweig, Germany
Received 15 March 2006; received in revised form 18 May 2006
Available online 3 July 2006
Abstract
Phalaenopsis hybrids contain two 1,2-saturated pyrrolizidine monoesters, T-phalaenopsine (necine base trachelanthamidine) and its
stereoisomer Is-phalaenopsine (necine base isoretronecanol). T-Phalaenopsine is the major alkaloid accounting for more than 90% of
total alkaloid. About equal amounts of alkaloid were genuinely present as free base and its N-oxide. The structures were confirmed
by GC–MS. The quantitative distribution of phalaenopsine in various organs and tissues of vegetative rosette plants and flowering plants
revealed alkaloid in all tissues. The highest concentrations were found in young and developing tissues (e.g., root tips and young leaves),
peripheral tissues (e.g., of flower stalks) and reproductive organs (flower buds and flowers). Within flowers, parts that usually attract
insect visitors (e.g., labellum with colorful crests as well as column and pollinia) show the highest alkaloid levels. Tracer feeding exper-
iments with 14C-labeled putrecine revealed that in rosette plants the aerial roots were the sites of phalaenopsine biosynthesis. However
active biosynthesis was only observed in roots still attached to the plant but not in excised roots. There is a slow but substantial trans-
location of newly synthesized alkaloid from the roots to other plant organs. A long-term tracer experiment revealed that phalaenopsine
shows neither turnover nor degradation. The results are discussed in the context of a polyphyletic molecular origin of the biosynthetic
pathways of pyrrolizidine alkaloids in various scattered angiosperm taxa. The ecological role of the so called non-toxic 1,2-saturated
pyrrolizidine alkaloids is discussed in comparison to the pro-toxic 1,2-unsaturated pyrrolizidine alkaloids. Evidence from the plant-insect
interphase is presented indicating a substantial role of the 1,2-saturated alkaloids in plant and insect defense.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Keywords: Phalaenopsis hybrids; Orchidaceae; 1,2-Saturated pyrrolizidine alkaloid; Pyrrolizidine alkaloid biosynthesis; Alkaloid tissue distribution;
Phalaenopsine; Molecular evolution
1. Introduction
pathway was recruited during evolution and integrated into
the plant metabolism.
Pyrrolizidine alkaloids represent an excellent system to
study not only phytochemical and biochemical aspects of
plant secondary metabolism (Hartmann and Witte, 1995;
Hartmann and Ober, 2000) but also its molecular evolution
(Hartmann and Ober, 2000; Ober et al., 2003; Reimann
et al., 2004). Using these alkaloids as a model system we
are trying to understand in more detail, how a biogenetic
Pyrrolizidine alkaloids are esters of a necine base with
one or more necic acids. They are produced constitutively
by the plant most probably as defense against herbivores.
Pyrrolizidine alkaloids are strong feeding deterrents for
most herbivores and are part of many fascinating interac-
tions between the producing plant and adapted herbivores
(Hartmann, 1999; Hartmann and Ober, 2000). The occur-
rence of more than 400 structures is restricted to certain
unrelated angiosperm taxa. About 95% of the species con-
taining pyrrolizidine alkaloids belong to only five families,
viz. the Asteraceae (tribes Senecioneae and Eupatorieae),
the Boraginaceae, the Apocynaceae, the genus Crotalaria
within the Fabaceae and in certain genera of the Orchida-
*
Corresponding author. Tel.: +49 431 880 4299; fax: +49 431 880 1527.
E-mail address: dober@bot.uni-kiel.de (D. Ober).
Present address: Botanisches Institut, Universita¨t Kiel, Olshausen-
1
strasse 40, D-24098 Kiel, Germany.
0031-9422/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2006.05.031

1494
C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
ceae. Further occurrences are reported from single species
of other families like the Celastraceae, Santalaceae, Sapot-
aceae, Ranunculaceae and Convolvulaceae (Hartmann and
Witte, 1995; Jenett-Siems et al., 2005). In plants the pyrro-
lizidine alkaloids are mostly found in the form of their
polar, salt-like N-oxides (Hartmann and Ober, 2000).
The occurrence of pyrrolizidine alkaloids in the Orchida-
ceae have been first described in the early 1960s and the 1970s
by Branda¨nge and coworkers (Luning et al., 1966; Bran-
da¨nge and Luning, 1969, 1970; Branda¨nge et al., 1970,
1971, 1972; Branda¨nge and Granelli, 1973). To date, in the
Orchidaceae pyrrolizidine alkaloids have been found in eight
genera, viz. Liparis, Malaxis, Cysis, Phalaenopsis, Vanda,
Vandopsis (Hartmann and Witte, 1995), Pleurothallis (Borba
et al., 2001), and Cremastra (Ikeda et al., 2005) belonging to
five tribes of the subfamily Epidendroideae (Cameron,
2004). More than ten unique orchid alkaloids are known,
representing the phalaenopsine type of pyrrolizidine alka-
loids (Hartmann and Witte, 1995). Alkaloids of the phalae-
nopsine type are monoesters of 1,2-saturated simple necines
bases like trachelanthamidine and its stereoisomers (Fig. 1).
The typical necic acids are mostly aralkyl acids (Fig. 1).
The genus Phalaenopsis with about 70 species is native in
the humid forests from South Asia to North Australia. Pha-
laenopsis species grow as epiphytes mainly on trees without
parasitizing them. Phalaenopsis species are monopodial
plants as they grow as a single stem without pseudobulbs.
The inflorescence emerges from the leaf rosette. Aerial roots
allow the plants to attach to their support and ensure water
and nutrient supply. Phalaenopsis species are popular as
ornamental plants and are used for many decades for breed-
ing, resulting in several thousands of intra and interspecific
hybrids. Since Branda¨nge’s pioneering work no other stud-
ies dealt with the orchid alkaloids. The occurrence of pyrr-
olizidine alkaloids in monocots deserves closer attention
since we showed that the first specific enzyme of pyrrolizi-
dine alkaloid biosynthesis, i.e. homospermidine synthase
(EC 2.5.1.45), was recruited at least four times indepen-
dently during angiosperm evolution. The independent ori-
gin of the pathways leading to pyrrolizidine alkaloids in
these lineages is likely and would account for the scattered
occurrence of pyrrolizidine alkaloids in isolated, unrelated
families of the angiosperms (Reimann et al., 2004). As a
basis for more detailed molecular studies the present inves-
tigation was carried out. Using Phalaenopsis hybrids as
experimental plant we established the alkaloid profiles
and distribution pattern within the various plant organs
and performed tracer feeding studies to outline the biosyn-
thetic pathway and identify its tissue localization. The
results are discussed in an evolutionary context, particularly
addressing the ecological role of the so-called non-toxic 1,2-
saturated pyrrolizidine alkaloids.
2. Results
2.1. Alkaloid profiles
Alkaloid extracts of Phalaenopsis hybrids revealed the
presence of two pyrrolizidine alkaloids which could be iden-
tified by GC–MS as T-phalaenopsine and Is-phalaenopsine
(Branda¨nge et al., 1972; Hartmann et al., 2005a,b) (Fig. 1).
These two alkaloids are characterized by a unique necic
acid, 2-benzylmalate-4-methyl ester, esterified to the epi-
meric necine bases trachelanthamidine and isoretronecanol,
respectively. In all studied Phalaenopsis hybrid plants
T-phalaenopsine was found to be the dominating alkaloid,
generally accounting for more than 90% of total alkaloids.
The same ratio was found in different organs and tissues of
individual plants analyzed. Since a separation of the two
isomeric phalaenopsines was only achieved by GC but not
HPLC or TLC, the term ‘phalaenopsine’ used in the follow-
ing text refers to the mixture of both isomers.
In the various plant tissues phalaenopsine was always
present as a mixture of the free base and its N-oxide
(Fig. 1). The proportion of N-oxide accounted for 40–60%
of total alkaloid. Since the N-oxides are easily reduced in
the presence of weak reducing agents (Hartmann and Top-
pel, 1987), a control experiment was performed in which the
standard extraction procedure was carried out in the pres-
ence of added 14C-labeled phalaenopsine N-oxide. Analysis
of the final extract by TLC confirmed that no artificial
reduction had occurred during the extraction procedure.
Furthermore, no phalaenopsine could be detected by GC
if its N-oxide was injected. Direct HPLC analysis of crude
plant extracts confirmed the presence of almost equal
Fig. 1. Structures of the 1,2-saturated pyrrolizidine alkaloids found in
Phalaenopsis hybrids. Both alkaloids are found as mixture of the free base
and its N-oxide.
Fig. 2. HPLC separation of phalaenopsine (1) and its N-oxide (2) in crude
methanolic extracts of Phalaenopsis roots. Separation was achieved using
the gradient system; detection by UV at 280 nm.

C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
1495
amounts of the alkaloid free base and its N-oxide as illus-
trated in Fig. 2.
A dramatic decrease in the phalaenopsine content was
observed in withering flowers (Table 1D). Whether this loss
is due to alkaloid export or degradation remains ambiguous.
2.2. Tissue-specific alkaloid distribution
2.3. Biogenetic tracer feeding experiments
The quantitative distribution of phalaenopsine between
various organs and tissues as well as its tissue concentra-
tions were analyzed for a four-month-old Phalaenopsis
plant in the rosette stage (Table 1A), a flowering plant
(Table 1B) and an inflorescence with flower buds (Table
1C). In addition various organs and tissues of a single
flower were analyzed (Table 1D). The major results are
summarized as follows. In the rosette plant (Table 1A)
about equal portions of phalaenopsine are found in roots
(34%), leaves (34%), and the shoot apex (32%), respec-
tively. The highest phalaenopsine concentration, i.e. 6-fold
higher than the average concentration of the whole plant,
was found in the shoot apex. About 75% of total leaf-alka-
loid was found in the youngest leaf compared to only 6% in
the oldest one. The phalaenopsine content and concentra-
tion in root tips is twice as high as the value measured
for the basal part of the roots. Thus, in the rosette plant,
the young still developing and growing organs or tissues,
i.e. apical root parts, young leaves, shoot meristems show
the highest alkaloid concentrations.
The analysis of a plant in bloom (Table 1B) revealed
that the five flowers contained more than 50% of total plant
phalaenopsine, the stalk and leaves about 20% and total
roots a bit less than 30%. The highest alkaloid concentra-
tions (i.e., 1.2 mg/g) were found in the flowers; they were
about two to six times higher than those found in the vege-
tative organs. However, the total alkaloid concentration
measured in the flowering plant was more than twice the
concentration found in the rosette plant (Table 1A and
B). By far the highest phalaenopsine levels are found in
inflorescences with young flower buds (Table 1C). The
alkaloid concentration of the whole inflorescence was
about 10-fold higher than in rosette plants and 4–5-fold
higher than the mean value of a plant in bloom. Young
flower buds (6.3 mg alkaloid per g) contain about 2–3-fold
higher alkaloid concentrations than old buds (i.e., 2.8 mg
per g). Within the inflorescence stalk the alkaloid concen-
tration decreased from the top (2 mg/g) to less than 0.3
in the basal part. The peripheral stem tissue showed a 6-
fold higher alkaloid concentrations than the central tissue.
To get a picture of phalaenopsine distribution within
flowers, a single flower was dissected and analyzed as
shown in Table 1D. Alkaloid was present in all flower
organs. Almost equal alkaloid concentrations were mea-
sured in sepals and petals, including the labellum. The
whole perianth contained 58% of total alkaloid at tissue
concentrations of 0.5–0.7 mg/g, whereas the remaining
42% were concentrated in the crests on the upper surface
of the labellum and in the reproductive organs, i.e. the col-
umn and anther cup with the pollinia. These tissues reach
2–3-fold higher phalaenopsine concentration than the aver-
age of the whole flower.
To identify the site of phalaenopsine synthesis in Pha-
laenopsis plants, detached organs were incubated with
14C-labeled putrescine as it was described earlier for Sene-
cio vulgaris (Hartmann et al., 1989) and Cynoglossum
officinale (Van Dam et al., 1995). Although the applied tra-
cer was completely absorbed by detached roots, leaves,
flowers, shoot apices and inflorescences within five days,
no labeled phalaenopsine could be isolated from the tis-
sues. The same negative results were obtained with
[14C]arginine as an alternative alkaloid precursor. In Sene-
cio roots the diamine oxidase inhibitor, b-hydrox-
yethylhydrazine (HEH), causes an accumulation of
homospermidine, the first pathway-specific intermediate
of pyrrolizidine alkaloid biosynthesis, by preventing its
incorporation into the alkaloids (Bo¨ttcher et al., 1993).
In none of the orchid tissues did tracer feeding in the pres-
ence of 2 mM HEH cause accumulation of labeled homo-
spermidine. Substantial tracer incorporation was only
obtained when the tracer was fed via the aerial roots of
undisturbed intact plants (Table 2). In this experiment
the tracer was completely absorbed within two days and
after another five days 38% of applied radioactivity could
be recovered from the roots; 26% of the recovered radioac-
tivity accounted for phalaenopsine. This corresponds to an
incorporation rate of ca. 10%. Labeled putrescine was no
longer detectable in the alkaloid extracts. The labeled pha-
laenopsine was present in the state of the free base; phalae-
nopsine N-oxide could not be detected. The identity of the
labeled alkaloid was confirmed by radio TLC and radio
HPLC in comparison to reference alkaloid. Since small
amounts of labeled phalaenopsine (less than 3% of the
recovered radioactivity) were also detected in young leaves,
the ability of leaves to synthesize alkaloids could not be
excluded. In a comparative experiment the potential bio-
synthetic capacity of roots and shoots were compared.
The results are summarized in Table 3. Plant A received
the tracer via an aerial root, plant B via a leaf. In both
plants the tracer was completely absorbed within four
days. Small amounts of tracer were only detectable in
those organs, to which the tracer was applied and in which
most of the recovered radioactivity was detected. This sug-
gests a slow transport of the labeled compounds within the
plant. In plant A, as expected, newly biosynthesized pha-
laenopsine was found in root extracts, whereas in plant B
no alkaloid could be detected in the leaves. However the
roots of plant B contained small but substantial amounts
of labeled alkaloids. This indicates that the leaf is not capa-
ble of synthesizing phalaenopsine, but the tracer taken up
by the leaf must have been translocated to the roots where
it was incorporated into phalaenopsine. As traces of
labeled phalaenopsine were also detected in the basal part

1496
C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
Table 1
Total content, concentration and relative abundance (alkaloid distribution) of phalaenopsine in various organs and tissues of Phalaenopsis hybrids
Plant organ
Fresh weight (g)
Alkaloid content (mg)
Alkaloid concentration (mg/g FW)
Alkaloid distribution (%)
(A) Rosette plant (4-months)
Root tips (1)
Root basal parts (2)
Shoot apex (3)
Young leaf (4)
Middle leaf (5)
0.57
0.53
0.14
0.84
0.46
0.22
0.11
0.06
0.16
0.13
0.03
0.01
0.20
0.12
1.15
0.15
0.07
0.03
22
12
32
26
6
Old leaf (6)
2
Whole plant
2.76
0.50
0.18
100
(B) Flowering plant
5 Flowers
Stalk
3 Leaves
Roots
28.9
9.0
54.9
60.5
33.8
5.3
8.5
1.2
0.6
0.2
0.3
52
8
13
28
18.8
Whole plant
153.3
66.4
0.4
100
(C) Inflorescence with buds
Young flower buds (1)
Middle flower bud (2)
Old flower bud (3)
Stalk tip (4)
Stalk middle part (5)
Stalk basal part (6)
Total
0.16
0.24
0.51
0.32
0.50
1.01
0.91
1.43
0.65
0.64
6.31
3.80
2.80
2.03
1.28
20.7
18.6
29.3
13.3
13.1
1.03
0.41
0.62
0.24
0.20
0.04
0.23
0.49
0.06
4.9
Peripheral tissues
Central tissues
Whole inflorescence
2.76
4.88
1.77
100
(D) Single flower
3 Sepals (1)
2 Petals (2)
Labellum (3)
Crests (4)
1.00
1.13
0.55
0.19
0.39
0.006
0.69
0.63
0.27
0.31
0.89
0.01
0.69
0.56
0.49
1.63
2.28
2.00
24.6
22.5
9.6
11.1
31.8
0.4
Column (5)
Pollinia (6)
Whole flower
3.27
2.58
0.86
100
Flower slightly withered
Flower completely withered
2.41
2.03
0.15
tr
0.07
tr
(A) Rosette plant (vegetative state); the term ‘shoot apex’ refers to the apical shoot part engulfed by the leave sheaths; numbers in parenthesis refer to
illustration A above the table. (B) Flowering plant. (C) Abscised inflorescence with flower buds; numbers in parenthesis refer to illustration C. (D) Single
flower, numbers in parenthesis refer to the illustration D. Scale bars in illustrations: 1 cm.
tr, traces.

C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
1497
Table 2
weeks, respectively. No tracer was detected in any of the
alkaloid extracts. The results are summarized in Table 4.
Three major points need to be stressed: (1) There is a slow
but substantial translocation of the newly synthesized alka-
loid from the ‘tracer-root’ to the other roots, the shoot
apex and young leaves, whereas older leaves did not receive
newly synthesized alkaloids. (2) Phalaenopsine (free base)
is accompanied by a small proportion of its N-oxide. How-
ever, in the various tissues, with one exception, the frac-
tions of N-oxide are between 8% and 14% of total
alkaloid and thus considerably lower than the 40–60% nor-
mally found in Phalaenopsis (see Fig. 2). It might be possi-
ble that N-oxidation is a slow continuous process affecting
total plant phalaenopsine and is not yet recognizable
within the population of newly synthesized labeled mole-
cules even after four weeks. (3) The total incorporation
of phalaenopsine is almost 3% after 2 weeks. This value
does not change after another two weeks indicating that
there is no substantial degradation or turnover of the alka-
loids during this time interval.
Incorporation of [14C]putrescine (6 lCi) into phalaenopsine by a young
Phalaenopsis plant
Plant organ
analyzed
Recovery of
radioactivity
applied (%)
14C-labeled phalaenopsine
% of recovered Total incorporation
radioactivity
(% of radioactivity
applied)
Roots and
shoot apex
2 Young leaves
3 Old leaves
38.2
25.6
9.8
0.7
0.1
2.7
n.d.
tr
n.d.
The tracer was applied via aerial root (2 days). Analysis after five days
further incubation in water.
tr, traces; n.d., not detectable.
of the tracer root of plant A, this tissue was tested sepa-
rately for its capacity to synthesize phalaenopsine. The api-
cal part of an aerial root was removed. The remaining
basal part of the root was fed with tracer for five days as
well as the detached root tip. Another undamaged aerial
root of the plant was fed as control (Table 3, plant C).
As observed earlier the detached root tip was in contrast
to the control unable to synthesize alkaloids. In compari-
son to the control root the decapitated root showed a
7-fold lower incorporation of tracer into phalaenopsine,
indicating a reduced but significant ability of the basal root
part to synthesize alkaloid.
To get more information on the transport, turnover and
N-oxidation of phalaenopsine a biogenetic long-term
experiment was performed. 14C-labeled putrescine was
applied to aerial roots of two rosette plants. In both plants
the tracer was completely absorbed within 24 h. The two
plants were dissected and analyzed after two and four
The incorporation of various 14C-labeled potential alka-
loid precursors into phalaenopsine was tested to establish
the polyamine-dependent alkaloid biosynthesis in Phalaen-
opsis (Table 5). Besides putrescine homospermidine and less
pronounced spermidine are alkaloid precursors, whereas
the amino acids arginine and ornithine, the direct precur-
sors of putrecine and spermidine, are not incorporated at
all. Labeled trachelanthamidine and isoretronecanol are
incorporated into phalaenopsine but with low efficiency,
i.e. 5%, respectively, 1% total incorporation. The higher rate
with trachelanthamidine in comparison to isoretronecanol
may reflect the dominating role of T-phalaenopsine in the
mixture of the two stereoisomeric orchid alkaloids.
Table 3
Biosynthesis of phalaenopsine by Phalaenopsis rosette plants
Plant organ analyzed
Recovery of radioactivity applied (%)
% of recovered radioactivity
Total incorporation into alkaloid
Putrescine
Phalaenopsine
Plant A
Tracer root (apical)
Tracer root (basal)
Remaining roots
Shoot apex
27.1
1.3
0.2
0.2
0.5
5.8
tr
n.d.
n.d.
n.d.
28.7
1.8
n.d.
n.d.
n.d.
7.8
0.02
Leaves
Plant B
Tracer leaf
Remaining leaf
Shoot apex
Roots
23.5
n.d.
3.0
7.7
n.d.
tr
n.d.
n.d.
tr
2.6
n.d.
5.6
0.2
Plant C
Undamaged root
Decapitated root
Detached root tip
29.3
29.7
18.9
6.1
7.0
10.0
28.2
3.9
n.d.
Plant A received [14C] putrescine (3 lCi) via an aerial root; plant B received the same amount via a leaf. In plant C one root received the tracer in the same
manner as plant A (undamaged root, control), a second root was decapitated and both, the detached root tip and the basal root part still attached to the
plant, were fed with tracer. Analysis was carried out after a feeding time of four days.
tr, traces; n.d., not detectable.

1498
C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
3. Discussion
3.1. Tissue distribution of phalaenopsine suggests ecological
relevance
T-phalaenopsine accompanied by a small portion of its
stereoisomer Is-phalaenopsine were the only pyrrolizidine
alkaloids detected in Phalaenopsis hybrids. Phalaenopsine
was found in all plant organs and tissues. The allover tissue
distribution indicates always highest concentrations in
young and developing tissues (i.e., root tips, shoot apex
and young leaves), peripheral tissues (e.g., of the flower
stalk) and reproductive organs. The highest alkaloid con-
centrations with 0.63% (fresh weight basis) were found in
the youngest flower buds. This is more than 7-fold of the
concentrations found in fully extended flowers (0.09%).
The apparent decrease in the alkaloid concentration during
flower development is not due to alkaloid loss or export, it
is a ‘dilution’ by increase of fresh weight. In fact, during
flower development a ‘middle old bud’ (Table 1C) increases
its fresh weight almost 14-fold but its alkaloid content only
3-fold (see Table 1C and D). Within the open flower high-
est alkaloid concentrations were found in the column, the
pollinia and the crests of the labellum. The latter exhibits
a conspicuous bait for insect visitors. These are precisely
the tissues that are visited by desired and undesired orchid
pollinators and that are most important for the reproduc-
tive success of the plant. In general, the quantitative alka-
loid distribution within an orchid plant resembles the
distribution reported for pro-toxic 1,2-unsaturated pyrro-
lizidine alkaloids in S. vulgaris (Asteraceae), i.e. highest
concentrations in flowers and peripheral stem tissues
(Hartmann and Zimmer, 1986; Hartmann et al., 1989). It
is well in accordance with the ‘‘optimal defense theory’’
that explains the allocation of defense metabolites prefer-
entially in those tissues that have a high probability of her-
bivore attack and in those of higher value for reproductive
fitness, e.g., meristems, other young organs, flower parts
near ovaries and pollen (McKey, 1974, 1979).
3.2. Are 1,2-saturated pyrrolizidine alkaloids defense
compounds? Chemoecological evidence
An assumed role of phalaenopsine in plant defense
against herbivores raises the question of its biological activ-
ity. Pyrrolizidine alkaloids are feared for their toxicity.
Pyrrolizidine alkaloid-containing plants are probably the
most common poisonous plants affecting livestock, wild-
life, and humans (Mattocks, 1986; Stegelmeier et al.,
1999). The molecular basis of pyrrolizidine alkaloid toxic-
ity is well understood. There are at least two essential struc-
tural features for toxicity: (1) a 1,2-double bond and (2)
associated with it an allylic hydroxyl group at C-9 that
needs to be esterified. A second OH-group at C-7, free or
esterified, intensifies the toxicity. Alkaloids of this structure
are not toxic per se but they require metabolic activation to
exert toxicity. In vertebrates this bioactivation is mediated

C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
1499
Table 5
Incorporation of potential 14C-labeled precursors into phalaenopsine
Tracer applied
Recovery of radioactivity
applied (%)
Percent of radioactivity
recovered
Total incorporation
into alkaloid (%)a
Tracer
Phalaenopsine
L-[U–14C]Arginine
L-[U–14C]Ornithine
13.2
14.3
25.6
23.0
27.4
55.9
26.7
25.3
30.2
6.2
tr
16.4
18.8
13.6
68.6
63.6
n.d.
n.d.
40.4
1.8
n.d.
13.4
18.8
4.1
[
[
[
[
[
14C]Putrescine
10.3
0.3
14C]Spermidine
14C]Spermine
14C]Homospermidine
14C]Trachelanthamidine
[3H]Isoretronecanol
7.5
5.0
1.0
The tracer was applied to the root-tip of an aerial root of an intact plant. After 4 days the tracer root was analyzed.
tr, traces; n.d., not detectable.
a
Radioactivity applied, 100%.
by liver cytochrome P-450 enzymes that convert the pro-
toxic alkaloids into pyrrolic intermediates which rapidly
react with biological nucleophiles, i.e. proteins and nucleic
acids, causing severe cell toxicity and even liver cancer (Fu
et al., 2004). Since insects possess a similar cytochrome P-
450 system like vertebrates (Brattsten, 1992) pyrrolizidine
alkaloids are also cytotoxic and genotoxic for insects (Frei
et al., 1992; Narberhaus et al., 2005).
The 1,2-saturated pyrrolizidine alkaloids like phalae-
nopsine lack the 1,2-double bond and thus an allylic OH-
group and therefore are regarded as non-toxic, at least as
far as P-450-dependent bioactivation is concerned. Gener-
ally they are neglected when toxicity of pyrrolizidine alka-
loids is discussed. Many species of the poisonous genera
Senecio, Eupatorium (Asteraceae) and Heliotropium (Bora-
ginaceae) contain mixtures of pro-toxic 1,2-unsaturated
and non-toxic 1,2-saturated pyrrolizidine alkaloids (Hart-
mann and Witte, 1995). Among them there are even single
species like Heliotropium floridum (Reina et al., 1997) that
contain only 1,2-saturated alkaloids. From two plant fam-
ilies only 1,2-saturated pyrrolizidine alkaloids are known.
These are the Orchidaceae discussed here, and the Convol-
vulaceae, where a few species belonging to the subgenus
Quamoclit of the large genus Ipomoea contain the ipangu-
lines, i.e. monoesters and diesters of 1,2-saturated platyne-
cine (Jenett-Siems et al., 1998, 2005).
The biological activity of 1,2-saturated pyrrolizidine
alkaloids is largely unknown. We are aware only two
reports: insect antifeedant activity has been demonstrated
for some trachelanthamidine esters (Reina et al., 1997)
and most interestingly cremastrine, an isoretronecanol
ester with isocaproic acid, isolated from the orchid Crema-
stra appendiculata inhibits the binding of anticholinergic
drugs to the muscarinic M3 receptor (Ikeda et al., 2005).
The tissue-specific alkaloid distribution as outlined above
for Phalaenopsis, represents a strategy one would expect
for a plant chemical defense system that evolved under
the pressure of herbivory. Further strong but indirect evi-
dence suggesting a defense role of 1,2-saturated pyrrolizi-
dine alkaloids comes from the fascinating field of
chemical ecology of plant-insect interactions. A number
of specialized insects, e.g. leaf beetles, butterflies, moths,
developed highly specific adaptations that allow them to
sequester pro-toxic pyrrolizidine alkaloids from their
host-plants and utilize them for their own benefit particu-
larly as defense against their predators (Hartmann, 1999,
2004; Hartmann and Ober, 2000; Eisner et al., 2002).
Among tiger moths (Arctiidae) there are species that as lar-
vae not only sequester pro-toxic 1,2-unsaturated pyrrolizi-
dines from their food plants but also 1,2-saturated
alkaloids. The highly polyphagous larvae of the arctiids
Estigmene acrea and Grammia geneura locate pyrrolizidine
alkaloid-containing plants incidentally but immediately
recognize them by taste receptors in their mouthparts that
are phagostimulatory (Bernays et al., 2002a,b). These
receptors specifically recognize all kinds of pro-toxic pyrr-
olizidine alkaloids but interestingly also non-toxic 1,2-satu-
rated alkaloids (Hartmann et al., 2005b). Both kinds of
pyrrolizidine alkaloids are sequestered by larvae and are
transmitted to the adult stages during metamorphosis.
However pyrrolizidine alkaloids that do not fulfill certain
structural requirements cannot be directly transmitted to
the pupal stage, before transmission; they need to be con-
verted into transmittable ‘insect alkaloids’ (i.e., creatono-
tines and callimorphines) by trans-esterification.
Surprisingly, not only the 1,2-unsaturated pyrrolizidines
are recovered in this way but also 1,2-saturated structures,
including phalaenopsine (Hartmann et al., 2005a,b). This
specific biochemical behavior suggests strongly that not
only the 1,2-unsaturated alkaloids are involved in the well
documented defense of herbivorous insects against their
predators and parasitoids (Eisner et al., 2002) but that also
the largely neglected class of 1,2-saturated alkaloids must
be of ecological importance.
3.3. Aspects of phalaenopsine biosynthesis
When discussing phalaenopsine biosynthesis it should
be recalled that pyrrolizidine alkaloid biosynthesis in
angiosperms is of polyphyletic origin. Evidence has been
presented that the gene encoding homospermidine syn-
thase, the first pathway-specific enzyme of necine base

1500
C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
formation, has been recruited several times independently
from gene duplicates of ubiquitous deoxyhypusine syn-
thase (EC 2.5.1.46) in dicots and at least once within
monocots (Ober and Hartmann, 1999; Ober et al., 2003;
Reimann et al., 2004; Ober, 2005). Putrescine, spermidine
and homospermidine are precursors of phalaenopsine indi-
cating that the basic route of necine biosynthesis should
be the same as in Senecio that has been studied in detail
(for review see Hartmann and Ober, 2000). However, there
are some differences. In contrast to S. vulgaris roots
(Hartmann et al., 1988) spermidine is a much less efficient
precursor in Phalaenopsis and the polyamine precursors
arginine and ornithine are not incorporated at all. This
may reflect differences of polyamine metabolism at the
interphase between primary and secondary metabolism
that need to be studied in more detail. In the rosette plant
the root is the only site of alkaloid biosynthesis and active
synthesis requires the intact root-shoot connection and
comes to rest if the root is detached. This behavior shows
some parallels to alkaloid biosynthesis in Senecio and
Eupatorium roots that only synthesize alkaloids when they
are actively growing (Hartmann et al., 1988; Sander and
Hartmann, 1989; Anke et al., 2004). Since in both plants
the alkaloids synthesized in roots and are translocated into
the shoot and do not underlie any turnover it is important
to adjust the rate of alkaloid synthesis to plant growth. In
Phalaenopsis alkaloid biosynthesis requires apparently an
endogenous signal transmitted from the root base to the
apex.
Zn dust and stirred for 3 h at room temperature for com-
plete reduction of alkaloid N-oxides. Then the mixture
was made basic and further processed as given above. This
fraction contains total pyrrolizidine alkaloids, i.e. free base
plus N-oxide.
4.3. Identification of phalaenopsines by GC–MS
The GC–MS data were obtained with a Hewlett Pack-
ard 5890A gas chromatograph equipped with a 2 m fused
silica guard column (deactivated, ID 0.32 mm) and a
30 m · 0.32 mm analytical column (DB-1 J&W Scientific
or ZB-1 Phenomenex). The capillary column was directly
coupled to a triple quadrupole mass spectrometer (TSQ
700, Finnigan). The conditions applied were: Injector and
transfer line were set at 280 ꢁC; the temperature program
used was: 100 ꢁC (3 min)–300 ꢁC at 6 ꢁC minÀ1 for the sep-
aration of pyrrolizidine alkaloids and 70 ꢁC (6 min)–300 ꢁC
at 10 ꢁC minÀ1 for the separation of necic acid methyl
esters, respectively. The injection volume was 1 ll. Depend-
ing on the concentration the split ratio was 1:20 or splitless,
the carrier gas flow was 1.6 ml minÀ1 He, and the mass
spectra were recorded at 70 eV.
The alkaloids were identified by their RI, molecular ions
and mass fragmentation patterns in comparison to
reference data. In addition the purified alkaloids were
hydrolyzed and the respective necine bases identified by
GC–MS. Hydrolysis was performed in 8% NaOH at
60 ꢁC for 90 min. GC–MS data of the phalaenopsines:
T-phalaenopsine: RI (ZB-1), 2522; [M+], 361; necine
base, trachelanthamidine, RI (DB-1), 1236; [M+], 141.
Is-phalaenopsine: RI (ZB-1), 2560; [M+], 361; necine
base, isoretronecanol, RI (DB-1), 1257; [M+], 141.
4. Experimental
4.1. Plant material
Young up to 4-month-old rosette plants of Phalaenopsis
hybrids were obtained from local orchid breeding stations
(Hennis Orchideen, Hildesheim and Wichmann Orchideen,
Celle, Germany). The plants were kept in a greenhouse till
use. Flowering Phalaenopsis hybrids were obtained from
the same breeders or as ornamental plants from the local
market.
4.4. Alkaloid separation by GC and HPLC
Routine GC was performed using a capillary column
(15 m · 0.25 mm fused-silica; DB-1, J&W Scientific) (Witte
et al., 1993). All other GC conditions were the same as
given for GC–MS. Detectors, FID and PND. Quantitative
analyses were performed via the FID signals using helio-
trine as internal standard.
The separation of phalaenopsine and phalaenopsine N-
oxides was achieved by HPLC using the method according
to Hartmann et al. (1988) that had to be adapted due to the
low polarity of the phalaenopsines. For a gradient elution
system two mobile phases were used. Mobile phase A: mix-
ture of 150 ml acetonitrile and 850 ml H3PO4 (1.5% v/v).
Mobile phase B: mixture of 500 ml acetonitrile and 500
ml H3PO4 (1.5% v/v). A linear gradient was formed start-
ing with 100% eluent A leading in 30 min to 100% eluent
B. For an isocratic system a mobile phase of 250 ml aceto-
nitrile and 750 ml TFA in water (pH adjusted to 2.0) was
applied. The flow-rate was 1.0 ml per min. The alkaloids
were detected at 280 nm after a retention time for phalae-
4.2. Alkaloid extraction and purification for GC and HPLC
analysis (Witte et al., 1993)
Fresh plant material was weighed and ground with
liquid nitrogen and sea sand (Merck) in a mortar. Samples
were extracted twice for 30 min each with 0.1 M H2SO4
and centrifuged. The supernatants were combined and half
of the solution was made basic with 25% ammonia and
applied to an Extrelut (Merck) column (1.4 ml solution/g
Extrelut). The alkaloid free bases were eluted with CH2Cl2
(6 ml/g Extrelut). The solvent was evaporated and the res-
idue dissolved in 10–100 ll MeOH prior to GC, GC–MS or
HPLC analysis. The remaining half of the acid supernatant
was adjusted to 0.25 M H2SO4 and mixed with an excess of

C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
1501
nopsine and phalaenopsine N-oxide of 13.1 min and
13.4 min, respectively, with the gradient system, and
7.8 min and 8.5 min, respectively, with the isocratic system.
Detection of 14C-labeled phalaenopsine was achieved with
Radiochemical purity of the phalaenopsine N-oxide was
confirmed by radio TLC and radio HPLC.
a
HPLC radioactivity monitor LB-506D (Berthold)
Acknowledgements
equipped with a 2 ml flow cell and the splitt-mixer LB-
5034. Rialuma (BAKER) was used as liquid scintillator.
This work was supported by grants of the Deutsche
Forschungsgemeinschaft and Fonds der Chemischen
Industrie. We thank Claudine Theuring for excellent tech-
nical assistance and Ludger Witte and Till Beuerle for the
GC–MS measurements.
4.5. Tracer-feeding experiments and product analysis
Most experiments were performed with [14C]putrescine
as tracer. Tracer was applied via aerial roots or leaves:
the tip of an aerial root or a carved leaf tip of an intact
plant was dipped into a reaction vial containing the tracer
(3–6 lCi) solved in tap water. The absorption of the tracer-
fluid was usually completed after 24 h. Afterwards the lost
fluid volume was replaced by pure tap water till the end of
the experiment. In experiments with b-hydroxyethylhydr-
azine the inhibitor was added 2 h prior to tracer applica-
tion; the final inhibitor concentration was 2 mM.
For extraction, the plant organs were washed with tap
water, dabbed dry, weighed, and ground in a mortar with
liquid nitrogen and sea sand (Merck) before they were
extracted twice for 30 min each with methanol containing
1% HCl (25%) and centrifuged. The supernatant of the
combined methanol extracts was evaporated, the residue
was dissolved in a defined volume of methanol and kept
at À20 ꢁC till analysis. Total radioactivity was determined
by scintillation counting. Aliquots were subjected to radio
TLC or radio HPLC for separation and quantification of
the labeled products according to Hartmann and Dierich
(1998). TLC separation was achieved on silica gel 60 F254
with the solvent system CH2Cl2–MeOH–NH3 (25%)
(82:15:3). Rf values: phalaenopsine N-oxide, 0.56; phalae-
nopsine (free base), 0.65.
References
Anke, S., Niemuller, D., Moll, S., Ha¨nsch, R., Ober, D., 2004.
¨
Polyphyletic origin of pyrrolizidine alkaloids within the Asteraceae.
Evidence from differential tissue expression of homospermidine
synthase. Plant Physiol. 136, 4037–4047.
Bernays, E.A., Chapman, R.F., Hartmann, T., 2002a. A highly sensitive
taste receptor cell for pyrrolizidine alkaloids in the lateral galeal
sensillum of a polyphagous caterpillar, Estigmene acraea. J. Comp.
Physiol. A 188, 715–723.
Bernays, E.A., Chapman, R.F., Hartmann, T., 2002b. A taste receptor
neurone dedicated to the perception of pyrrolizidine alkaloids in the
medial galeal sensillum of two polyphagous arctiid caterpillars.
Physiol. Entomol. 27, 312–321.
Borba, E.L., Trigo, J.R., Semir, J., 2001. Variation of diastereoisomeric
pyrrolizidine alkaloids in Pleurothallis (Orchidaceae). Biochem. Syst.
Ecol. 29, 45–52.
Bo¨ttcher, F., Adolph, R.D., Hartmann, T., 1993. Homospermidine
synthase, the first pathway-specific enzyme in pyrrolizidine alkaloid
biosynthesis. Phytochemistry 32, 679–689.
Branda¨nge, S., Granelli, I., 1973. Orchidaceae alkaloids. XXXVI. Alka-
loids from some Vanda and Vandopsis species. Acta Chem. Scand. 27,
1096–1097.
Branda¨nge, S., Granelli, I., Luning, B., 1970. Orchidaceae alkaloids.
XVIII. Isolation of phalaenopsin La from Kingiella taenialis (Lindl.)
Rolfe. Acta Chem. Scand. 24, 354.
Other radiochemicals were obtained or prepared as fol-
lows: L-[U–14C]arginine (320 mCi/mmol), L-[U–14C] orni-
thine (282 mCi/mmol), [14C]spermidine, and [14C]
spermine were purchase by Amersham Biosciences.
[14C]Homospermidine was prepared from [14C]putrescine
enzymatically using bacterial homospermidine synthase
(EC 2.5.1.44) (Graser et al., 1998). [14C]Trachelanthami-
dine was prepared by hydrolysis of [14C]trachelanthamine
obtained from Eupatorium cannabinum root cultures that
had been fed with labeled putrescine (Hartmann, 1994)
and [3H]isoretronecanol prepared synthetically by Dr. Xia.
Branda¨nge, S., Luning, B., 1969. Studies on orchidaceae alkaloids. XII.
Pyrrolizidine alkaloids from Phalaenopsis amabilis B1. and Ph. mannii
Rchb.f. Acta Chem. Scand. 23, 1151–1154.
Branda¨nge, S., Luning, B., 1970. Studies on Orchidaceae alkaloids. XVII.
Alkaloids from Vandopsis longicaulis Schltr. Acta Chem. Scand. 24,
353–354.
Branda¨nge, S., Luning, B., Moberg, C., Sjostrand, E., 1971. Studies on
orchidaceae alkaloids. XXIV.A pyrrolizidine alkaloid from Phalaen-
opsis cornu-cervi Rchb. Acta Chem. Scand. 25, 349–350.
Branda¨nge, S., Luning, B., Moberg, C., Sjostrand, E., 1972. Studies on
¨
Orchidaceae alkaloids. XXX. Investigations of fourteen Phalaenopsis
species. A new pyrrolizidine alkaloid from Phalaenopsis equestris. Acta
Chem. Scand. 26, 2558–2560.
Brattsten, L.B., 1992. Metabolic defenses against plant allelochemicals. In:
Rosenthal, G.A., Berenbaum, M.R. (Eds.), Herbivores
Interactions with Secondary Plant Metabolites, vol. 2. Academic
Press, San Diego, pp. 175–242.
– Their
4.6. Preparation of phalaenopsine N-oxides
14C-Labeled phalaenopsine was isolated, reduced and
purified from a tracer experiment with [14C]putrescine as
described. The oxidation was performed according to
Cymerman Craig and Purushothaman (1970). The purified
[14C]phalaenopsine was incubated in 1 ml 1 M m-chlorop-
erbenzoic acid in chloroform for 30 min on an ice bath.
The resulting N-oxide was purified via an alkaline alumin-
ium oxide column using CH1Cl3–MeOH (3:1) as solvent.
Cameron, K.M., 2004. Utility of plastid psaB gene sequences for
investigating intrafamilial relationships within Orchidaceae. Mol.
Phylogenet. Evol. 31, 1157–1180.
Cymerman Craig, J., Purushothaman, K.K., 1970. An improved prepa-
ration of tertiary amine N-oxides. J. Org. Chem. 35, 1721–1722.
Eisner, T., Rossini, C., Gonzales, A., Iyengar, V.K., Siegler, M.V.S.,
Smedley, S.R., 2002. Paternal investment in egg defence. In: Hilker,
M., Meiners, T. (Eds.), Chemoecology of Insect Eggs and Egg
Deposition. Blackwell, Oxford, pp. 91–116.

1502
C. Fro¨lich et al. / Phytochemistry 67 (2006) 1493–1502
Frei, H., Luthy, J., Brauchli, J., Zweifel, U., Wurgler, F.E., Schlatter, C.,
1992. Structure/activity relationships of the genotoxic potencies of
sixteen pyrrolizidine alkaloids assayed for the induction of somatic
mutation and recombination in wing cells of Drosophila melanogaster.
Chem. Biol. Interact. 83, 1–22.
Fu, P.P., Xia, Q., Lin, G., Chou, M.W., 2004. Pyrrolizidine alkaloids –
genotoxicity, metabolism enzymes, metabolic activation, and mecha-
nisms. Drug Metab. Rev. 36, 1–55.
Ikeda, Y., Nonaka, H., Furumai, T., Igarashi, Y., 2005. Cremastrine, a
pyrrolizidine alkaloid from Cremastra appendiculata. J. Nat. Prod. 68,
572–573.
Jenett-Siems, K., Ott, S.C., Schimming, T., Siems, K., Mu¨ller, F., Hilker,
M., Witte, L., Hartmann, T., Austin, D.F., Eich, E., 2005. Ipangulines
and minalobines, chemotaxonomic markers of the infrageneric
Ipomoea taxon subgenus Quamoclit, section Mina. Phytochemistry
66, 223–231.
Graser, G., Witte, L., Robins, D.J., Hartmann, T., 1998. Incorporation of
chirally deuterated putrescines into pyrrolizidine alkaloids: a reinves-
tigation. Phytochemistry 47, 1017–1024.
Hartmann, T., 1994. Senecio spp.: Biochemistry of the formation of
pyrrolizidine alkaloids in root cultures. In: Bajaj, Y.P.S. (Ed.),
Medicinal and Aromatic Plants, vol. VI. Springer, Berlin, Heidelberg,
pp. 339–355.
Hartmann, T., 1999. Chemical ecology of pyrrolizidine alkaloids. Planta
207, 483–495.
Hartmann, T., 2004. Plant-derived secondary metabolites as defensive
chemicals in herbivorous insects: a case study in chemical ecology.
Planta 219, 1–4.
Jenett-Siems, K., Schimming, T., Kaloga, M., Eich, E., Siems, K., Gupta,
M.P., Witte, L., Hartmann, T., 1998. Pyrrolizidine alkaloids of
Ipomoea hederifolia and related species. Phytochemistry 47, 1551–1560.
Luning, B., Trankner, H., Branda¨nge, S., 1966. Studies on orchidaceae
alkaloids V. A new alkaloid from Phalaenopsis amabilis Bl. Acta
Chem. Scand. 20, 2011.
Mattocks, A.R., 1986. Chemistry and Toxicology of Pyrrolizidine
Alkaloids. Academic Press, London.
McKey, D., 1974. Adaptive patterns in alkaloid physiology. Am. Nat. 108,
305–320.
McKey, D., 1979. The distribution of secondary compounds within plants.
In: Rosenthal, G.A., Janzen, D.H. (Eds.), Herbivores: Their Interac-
tion with Secondary Plant Metabolites. Academic Press, New York,
pp. 55–133.
Hartmann, T., Dierich, B., 1998. Chemical diversity and variation of
pyrrolizidine alkaloids of the senecionine type: biological need or
coincidence? Planta 206, 443–451.
Narberhaus, I., Zintgraf, V., Dobler, S., 2005. Pyrrolizidine alkaloids on
three trophic levels – evidence for toxic and deterrent effects on
phytophages and predators. Chemoecology 15, 121–125.
Ober, D., 2005. Seeing double: gene duplication and diversification in
plant secondary metabolism. Trends Plant Sci. 10, 444–449.
Ober, D., Harms, R., Witte, L., Hartmann, T., 2003. Molecular evolution
by change of function: alkaloid-specific homospermidine synthase
retained all properties of deoxyhypusine synthase except binding the
eIF5A precursor protein. J. Biol. Chem. 278, 12805–12812.
Ober, D., Hartmann, T., 1999. Homospermidine synthase, the first
pathway-specific enzyme of pyrrolizidine alkaloid biosynthesis,
evolved from deoxyhypusine synthase. Proc. Natl. Acad. Sci. USA
96, 14777–14782.
Hartmann, T., Ehmke, A., Eilert, U., von Borstel, K., Theuring, C., 1989.
Sites
of
synthesis,
translocation
and
accumulation
of
pyrrolizidine alkaloid N-oxides in Senecio vulgaris L. Planta 177, 98–
107.
Hartmann, T., Ober, D., 2000. Biosynthesis and metabolism of pyrrolizi-
dine alkaloids in plants and specialized insect herbivores. In: Leeper,
F.J., Vederas, J.C. (Eds.), Topics in Current Chemistry, vol. 209.
Springer, Berlin, pp. 207–244.
Hartmann, T., Sander, H., Adolph, R.D., Toppel, G., 1988. Metabolic
links between the biosynthesis of pyrrolizidine alkaloids and polyam-
ines in root cultures of Senecio vulgaris. Planta 175, 82–90.
Hartmann, T., Theuring, C., Beuerle, T., Bernays, E.A., Singer, M.S.,
2005a. Acquisition, transformation and maintenance of plant pyrro-
lizidine alkaloids by the polyphagous arctiid Grammia geneura. Insect
Biochem. Mol. Biol. 35, 1083–1099.
Reimann, A., Nurhayati, N., Backenko¨hler, A., Ober, D., 2004. Repeated
evolution of the pyrrolizidine alkaloid-mediated defense system in
separate angiosperm lineages. Plant Cell 16, 2772–2784.
Hartmann, T., Theuring, C., Beuerle, T., Klewer, N., Schulz, S.,
Singer, M.S., Bernays, E.A., 2005b. Specific recognition, detoxifi-
cation and metabolism of pyrrolizidine alkaloids by the polyph-
agous arctiid Estigmene acrea. Insect Biochem. Mol. Biol. 35,
391–411.
Hartmann, T., Toppel, G., 1987. Senecionine N-oxide, the primary
product of pyrrolizidine alkaloid biosynthesis in root cultures of
Senecio vulgaris. Phytochemistry 26, 1639–1644.
Hartmann, T., Witte, L., 1995. Chemistry, biology and chemoecology of
the pyrrolizidine alkaloids. In: Pelletier, S.W. (Ed.), Alkaloids:
Chemical and Biological Perspectives, vol. 9. Pergamon Press, Oxford,
pp. 155–233.
Reina, M., Gonzalez-Coloma, A., Gutierrez, C., Cabrera, R., Henriquez,
J., Villarroel, L., 1997. Bioactive saturated pyrrolizidine alkaloids from
Heliotropium floridum. Phytochemistry 46, 845–853.
Sander, H., Hartmann, T., 1989. Site of synthesis, metabolism and
translocation of senecionine N-oxide in cultured roots of Senecio
erucifolius. Plant Cell Tissue Organ Cult. 18, 19–32.
Stegelmeier, B.L., Edgar, J.A., Colegate, S.M., Gardner, D.R., Schoch,
T.K., Coulombe, R.A., Molyneux, R.J., 1999. Pyrrolizidine alkaloid
plants, metabolism and toxicity. J. Nat. Toxins 8, 95–116.
Van Dam, N.M., Witte, L., Theuring, C., Hartmann, T., 1995. Distribu-
tion, biosynthesis and turnover of pyrrolizidine alkaloids in Cynoglos-
sum officinale. Phytochemistry 39, 287–292.
Hartmann, T., Zimmer, M., 1986. Organ-specific distribution and
accumulation of pyrrolizidine alkaloids during the life history of two
annual Senecio species. J. Plant Physiol. 122, 67–80.
Witte, L., Rubiolo, P., Bicchi, C., Hartmann, T., 1993. Comparative
analysis of pyrrolizidine alkaloids from natural sources by gas
chromatography–mass spectrometry. Phytochemistry 32, 187–196.