J. Plant Physiol. 160.

607 614 (2003)

Urban & Fischer Verlag
http://www.urbanfischer.de/journals/jpp

Regulation of anthraquinone biosynthesis in cell cultures of Morinda

citrifolia
Marc Stalman, Anne-Marie Koskamp, Rianne Luderer, Juanita H. J. Vernooy, Jobien C. Wind, George J. Wullems, Anton F.
Croes*
Department of Experimental Botany, University of Nijmegen, Toernooiveld, 6525 ED Nijmegen, The Netherlands
Received March 1, 2002 Accepted August 16, 2002

Summary
Cell cultures of Morinda citrifolia L. are capable of accumulating substantial amounts of anthraquinones. Chorismate formed by the shikimate pathway is an important precursor of these secondary metabolites. Isochorismate synthase (EC 5.4.99.6), the enzyme that channels chorismate into the direction of the anthraquinones, is involved in the regulation of anthraquinone biosynthesis. Other
enzymes of the shikimate pathway such as deoxy-D-arabino-heptulosonate 7-phosphate synthase
(EC 4.1.2.15) and chorismate mutase (EC 5.4.99.5) do not play a regulatory role in the process. The
accumulation of anthraquinones is correlated with isochorismate synthase activity under a variety of
conditions, which indicates that under most circumstances the concentration of the branchpoint metabolite chorismate is not a rate-limiting factor. Anthraquinone biosynthesis in Morinda is strongly
inhibited by 2,4-D, but much less by NAA. Both auxins inhibit the activity of isochorismate synthase
proportionally to the concomitant reduction in the amount of anthraquinone accumulated. However,
the correlation between enzyme activity and rate of biosynthesis is less clear when the activity of the
enzyme is very high. In this case, a limiting concentration of precursor may determine the extent of
anthraquinone accumulation. Partial inhibition of chorismate biosynthesis by glyphosate leads to less
anthraquinone accumulation, but also to a reduction in ICS activity. The complexity of the interference
of glyphosate with anthraquinone biosynthesis is illustrated by the effect of the inhibitor in cell cultures of the related species Rubia tinctorum L. In these cells, glyphosate leads to an increase in
anthraquinone content and a concomitant rise in ICS activity. All data indicate that the main point of
regulation in anthraquinone biosynthesis is located at the entrance of the specific secondary route.
Key words: Anthraquinones auxins glyphosate isochorismate synthase Morinda citrifolia
Rubia tinctorum
Abbreviations: 2,4-D = 2,4-dichlorophenoxyacetic acid. CM = chorismate mutase. DAHP = 3-deoxy-D-arabino-heptulosonate 7-phosphate. E4P = erythrose-4-phosphate. ICS = isochorismate
synthase. NAA = 1-naphthalene acetic acid. PEP = phospoenolpyruvate

* E-mail corresponding author: croes@sci.kun.nl

0176-1617/03/160/06-607 $ 15.00/0

608

Marc Stalman et al.

Introduction
Plant cells have to partition available resources between primary and secondary metabolic pathways. Since secondary
pathways derive their precursors from primary metabolism,
efficient regulatory mechanisms are required to prevent any
wasteful loss of precursors if both primary and secondary
metabolism are active at the same time. The most efficient
place to control the partitioning of precursors is at the branching point between primary and secondary metabolic routes.
Two factors that play an important regulatory role at such a
branching point are substrate concentration and enzyme
activity. In a model where only substrate concentration determines the partitioning of precursors, activities of the branchpoint enzymes remain constant. Any change in distribution of
precursors over different metabolic pathways is the result of a
rise or a decline in the amount of substrate available at the
branchpoint. Such a change in substrate concentration leads
to a shift in the partitioning because of the different catalytic
properties of the enzymes competing for the substrate. The
overflow model of regulation (Jensen 1986, Poulsen and Verpoorte 1991) describes the extreme situation that precursors
are channeled into a secondary route only under saturating
concentrations for the primary pathway.
Adjusting the activity of a key enzyme at the start of the
secondary metabolic route is a more direct form of regulation
at the branchpoint that leads to the control of the use of precursors for secondary metabolism. In this model, changes in
the activities of other enzymes that are involved in the regulation of production or consumption of the branchpoint metabolite will have little or no effect on secondary metabolite synthesis. In in vivo situations, the biosynthesis of secondary metabolites is most likely to be determined by a combination of
both substrate concentration and enzyme activity. Modulation
of a key enzyme is used to switch secondary metabolism on
and off, whereas fine-tuning occurs by reacting to the substrate levels. For example, substrate levels may become limiting when precursors are in low supply.
An important branchpoint, where precursors are distributed across primary and secondary metabolic pathways, is
located in the shikimate pathway at chorismate. Chorismate is
the last compound of the common shikimate pathway and
marks the starting point for several metabolic routes, including those involved in the biosynthesis of primary metabolites
such as the aromatic amino acids, and the pathway leading
to the secondary metabolites known as anthraquinones (Haslam 1996, Jensen 1986, Zenk et al. 1975).
If the partitioning of chorismate between these various routes is achieved via substrate regulation, the concentration of
chorismate and, presumably, the activity of the enzyme DAHP
synthase (EC 4.1.2.15) will play an important role. DAHP synthase is located at the beginning of the shikimate pathway
and uses PEP and E4P to form DAHP. By controlling this first
reaction, DAHP synthase is believed to regulate the intake of
carbon into the shikimate pathway (Herrmann 1995, Herr-

mann and Weaver 1999) and, as a result, to affect chorismate

production in the cell. As in this model, the amount of anthraquinones formed is controlled via the chorismate concentration, a large production of anthraquinones might require an
increase in DAHP synthase, whereas low activity of the enzyme may result in a situation where less or even no chorismate is channeled towards secondary metabolic pathways.
The effect of low DAHP synthase activity may be mimicked by
exposing the cells to glyphosate, a herbicide that specifically
blocks the shikimate pathway and prevents the formation of
chorismate (Amrhein et al. 1980). If anthraquinone production
is regulated by the chorismate concentration, a low DAHP
synthase activity as well as sublethal doses of glyphosate
would be expected to cause a shift in the balance between
secondary and primary metabolism in favor of the latter.
If the flow of chorismate towards secondary metabolism is
regulated by a key enzyme at the start of the secondary pathway, isochorismate synthase (EC 5.4.99.6) is likely to be the
regulating factor. This enzyme catalyses the first committed
reaction of the route involved in anthraquinone production by
converting chorismate into isochorismate. An increase in ICS
activity would channel more chorismate towards the anthraquinones (Van Tegelen 1999). Regulation of anthraquinone
synthesis via ICS would result in a direct correlation between
anthraquinone accumulation and ICS activity, whereas
changes in activity of other chorismate-utilising enzymes
such as chorismate mutase (EC 5.4.99.5), which converts
chorismate into prephenate for phenylalanine and tyrosine
synthesis, should have only limited effects on the biosynthesis of anthraquinones. Unlike in the substrate regulation
model, low DAHP synthase activity or sublethal doses of
glyphosate would not immediately lead to a reduced anthraquinone formation because anthraquinone production is limited by ICS activity. It is easy to imagine that the actual situation is intermediate and that plant cells regulate the production of anthraquinones via modulation of ICS activity, but that
the availability of chorismate may become a limiting factor at
low production rates of chorismate or at (extremely) high ICS
activities.
For a study on the role of the enzymes DAHP synthase, CM
and ICS in chorismate partitioning, cell cultures of Morinda
citrifolia L. are a very suitable model system. First, they are
capable of producing very large amounts of anthraquinones,
up to 40 mol g 1 fresh weight in Morinda. Such an extreme
anthraquinone production might cause a serious drain on the
chorismate available for primary metabolism, especially in
Morinda cultures where anthraquinone biosynthesis may
coincide with active growth (Hagendoorn et al. 1994). Another
advantage of Morinda cultures is the fact that anthraquinone
production can be switched on and off by adjusting the auxin
content of the growth medium (Zenk et al. 1984). The auxin
2,4-D prevents anthraquinone formation, whereas NAA or absence of any auxin leads to anthraquinone production. The
latter situation is known to be unstable, which raises the question of whether or not anthraquinone production and normal

Anthraquinone biosynthesis in Morinda

growth are mutually exclusive (Van der Plas et al. 1995). If
anthraquinone synthesis is regulated via ICS activity, 2,4-D
should be a potent inhibitor of this activity.
In this study, the activities of enzymes potentially involved
in chorismate partitioning under conditions permitting active
anthraquinone production and conditions repressing their
biosynthesis are compared. These comparisons, together
with experiments on the effects of glyphosate on anthraquinone biosynthesis, will give insight in the problem under what
conditions anthraquinone production responds to modulation
of the chorismate level and when it is enzymatically regulated
by ICS.

Materials and Methods

609

1 mmol/L dithiothreitol and 0.2 mmol/L PMSF. The homogenate was

centrifuged at 12,000 g for 10 min and 1mL of the supernatant was applied to a Sephadex G25 PD-10 column (Pharmacia) pre-washed with
extraction buffer. The protein fraction containing ICS and CM was collected from the column, frozen in liquid nitrogen and stored at 80 C
until further use.
Enzyme extracts for the determination of DAHP synthase activity
were made by a method based on the procedure described by Pinto
et al. (1986). Frozen cells (2 g) were ground in liquid nitrogen with
80 mg polyvinylpolypyrrolidone and 1.5 mL extraction buffer (N-2hydroxyethyl-piperazine-N-3-propanesulfonic acid, 50 mmol/L, pH 8)
containing 1.5 mmol/L tryptophan, 0.1% -mercapto-ethanol, 2 mmol/L
MgCl2, 2 mmol/L MnCl2, 5 mmol/L Na2S2O5 and 2 mmol/L PEP. The
mixture was centrifuged for 30 min at 12,000 g after which the supernatant was collected, centrifuged again for 15 min and passed over a
Sephadex G25 PD-10 column (Pharmacia).

Chemicals

Protein content

Chemicals were of analytical grade and purchased from Merck

(Darmstadt, Germany) or ICN (Costa Mesa, U.S.A.). Glyphosate was
obtained from Duchefa (Haarlem, Netherlands). Barium chorismate
(65 % purity) was purchased from Sigma (St Louis, USA).

The protein content of enzyme extracts was determined according to

Bradford (1976). Bovine serum albumin was used as a standard.

ICS activity
Cell cultures
Cell cultures of Morinda citrifolia L. (Rubiaceae) were grown in 300 mL
Erlenmeyer flasks containing 50 mL of Gamborgs B5 medium (Gamborg et al. 1968) supplemented with 0.12 mol/L sucrose and 0.6 mol/
L kinetin. Auxin was supplied as 2,4-D at 0.1 mol/L unless stated
otherwise, or as NAA at 4.5 mol/L. Glyphosate used in some experiments was added to the growth medium prior to autoclaving as it is
stable and does not lose any biological activity under these conditions (Haderlie et al. 1977, Cooley and Foy 1992). Cultures were maintained in the dark at 25 C on a gyratory shaker (125 rpm). Cells were
subcultured every 14 days by diluting 10 mL culture with 50 mL of
fresh medium. Anthraquinone synthesis was initiated by transferring
cells to medium without 2,4-D. Cell cultures of Rubia tinctorum L. (Rubiaceae) were grown under similar conditions, but the sucrose content of the growth medium was 0.06 mol/L and the concentration of
2,4-D was 25 mol/L. The cells were harvested by filtration, and the
fresh weight was determined. The cell mass was frozen in liquid nitrogen and stored at 20 C until further use.

Anthraquinone determination
Anthraquinones were extracted from the cells by heating 0.3 g of cells
in 2 mL of 80 % ethanol at 80 C for 1h after which the ethanol was collected. This procedure was repeated twice after which the ethanol
fractions were pooled and centrifuged (5 min, 5000 g). The anthraquinone content was determined spectrophotometrically at 434 nm using
alizarin as a standard.

ICS activity was measured according to Poulsen et al. (1991) with

some modifications. The reaction mixture (50 L) consisted of Tris-HCl
(0.1mol/L, pH 7.5), 10 mmol/L MgCl2, 2 mmol/L barium chorismate and
25 L of crude enzyme extract. After incubation at 30 C for 60 min, the
reaction was stopped by the addition of 62.5 L methanol-butanol
(1: 1). The mixture was centrifuged at 4 C for 10 min at 12,000 g and a
10 L sample was injected into a SMART-system (Pharmacia)
equipped with a Sephasil C8 column. The mobile phase was 35 %
methanol in 50 mmol/L H3PO4 (pH 2.5) and the flow rate was 200 L/
min. Enzyme activity was calculated from the area of the isochorismate peak measured at 280 nm according to Poulsen et al. (1991).

CM activity
Activity of CM was measured according to Grisch (1978) with modifications as described in Poulsen and Verpoorte (1992). Assays were
performed using microtiter plates. Each well contained 100 L reaction mixture consisting of Tris-HCl (0.1mol/L, pH 7.5), 1mmol/L barium
chorismate and 50 L crude enzyme extract. The plates were incubated for 30 min at 30 C after which the reaction was stopped by
addition of 25 L 4 mol/L HCl. Plates were kept at room temperature
for 15 min and 30 L 2,4-dinitrophenylhydrazine (1 mg/mL) in 2 mol/L
HCl was added. After incubation for 15 min at room temperature, 50 L
6 mol/L NaOH was pipetted into each well, followed by a second addition of 50 L 6 mol/L NaOH 2 min later. The absorption was measured
at 450 nm in a microtiter plate reader and CM activity was calculated
according to Poulsen and Verpoorte (1992).

DAHP synthase activity

Enzyme extraction
Crude extracts of ICS and CM were obtained by homogenising 1 g of
frozen cells in 2 mL Tris-HCl extraction buffer (0.1 mol/L, pH 7.5) containing 10 % glycerol, 1 mmol/L ethylenediaminetetraacetic acid,

The activity of of DAHP synthase was measured using the method of

Ganson et al. (1986) with some modifications. The assay mixture
(150 L) contained N-2-hydroxyethylpiperazine-N-3-propane-sulfonic
acid (0.7 mmol/L, pH 8.0), 5.7 mmol/L PEP, 2 mmol/L E4P, 0.7 mmol/L

610

Marc Stalman et al.

MnCl2, 0.5 mmol/L tryptophan, 1.7 mmol/L Na2S2O5, 0.033 % -mercapto-ethanol and 50 L enzyme extract. After an incubation for
45 min at 37 C, the reaction was stopped by addition of 40 L 20 %
TCA. The amount of DAHP formed was measured using the thiobarbituric acid assay as described by Jensen and Nester (1966).

Shikimate detection
Accumulation of shikimate was analysed by homogenising 0.5 g of
cells grown in the presence of 0.5 mmol/L glyphosate in 1 mL of 1 %
phosphoric acid containing 80 % acetonitrile. The homogenate was
centrifuged for 5 min at 6000 g and the supernatant was diluted 4-fold
in acetonitrile. The sample was filtered (0.45 m) and 100 L was injected into a SMART system (Pharmacia) equipped with a Spherisorb
S3 Amino column (Waters). The flow rate of the mobile phase was
0.6 mL/min and peaks were detected at 215 nm.

Results
Auxins had a profound influence on anthraquinone production in Morinda cell cultures (Fig. 1). In a growth medium with
5 mol/L 2,4-D, a stable cell culture was established in which
virtually no anthraquinone was synthesised. When transferred
to medium without 2,4-D, the cells started to produce large
amounts of anthraquinones after a lag-phase of 3 4 weeks.
Cell cultures without auxin were unstable. When the anthraquinone concentration reached about 40 mol/g fw, growth
was arrested and the cells died, probably due to absence of
growth hormone or toxic effects of the anthraquinones.
To investigate whether anthraquinone formation and normal growth are mutually exclusive, cells were cultured in
medium containing a different auxin, NAA. At a concentration

Figure 1. Accumulation of anthraquinones in cell cultures of Morinda.

Before the onset of the experiment, cells were grown in the presence
of 5 mol/L (solid symbols) or 0.1 mol/L (open triangles) 2,4-D. At the
start of the experiment (t = 0), the cells were transferred to medium
with 5 mol/L 2,4-D (squares), 4.5 mol/L NAA (circles), or without
auxin (triangles) and the accumulation of anthraquinones was monitored.

Figure 2. DAHP synthase activity in Morinda cultures synthesising or

not synthesising anthraquinones. Cells precultured on medium with
0.1 mol/L 2,4-D were transferred at t = 0 to the same medium (squares, no anthraquinone biosynthesis) or to a medium without auxin (circles, anthraquinone biosynthesis induced). Cells were harvested at
intervals and the DAHP synthase activity was determined. Results are
given as means SE (n = 3). The data in Fig. 2 and Fig. 3 are from the
same experiment.

of 4.5 mol/L, NAA led to the same growth rate as 2,4-D, but
moderate amounts of anthraquinones were accumulated (up
to 15 mol/g fw) (Fig. 1). These cultures were stable and could
be subcultured indefinitely without any change in growth rate
or anthraquinone level.
As the presence of 2,4-D caused such a long lag-phase
before anthraquinone biosynthesis was switched on, the concentration was lowered from 5 mol/L to 0.1 mol/L. The cells
kept growing at the same rate and accumulated only minimal
amounts of anthraquinones after this drastic reduction. However, anthraquinone production started within 3 days when
the cells were transferred from this medium to a medium without auxin. Thus, lowering the 2,4-D concentration reduced the
lag time of 3 4 weeks to only a few days (Fig. 1).
The large accumulation of anthraquinones upon removal of
2,4-D probably caused a change in the partitioning of chorismate in the cells. Therefore, the activities of DAHP synthase,
ICS and CM, enzymes involved in chorismate production and
consumption, were measured under both anthraquinone-producing and non-producing conditions. The data on enzyme
activities and anthraquinone accumulation (Figs. 2 3) are
from the same experiment. Whether or not anthraquinones
are synthesised, the activity of DAHP synthase was initially
low, increased during the first half of the growth period, and
attained a maximum after about 6 8 days (Fig. 2). DAHP synthase activity declined sharply during the second half of the
growth period in the anthraquinone-producing cells, but remained high in the non-producing cultures. A comparison of
the data in Figs. 2 3 does not give any indication that a large
production of anthraquinones (Fig. 3) requires an extra increase in DAHP synthase activity. Activities of CM were constantly high ( > 250 pkat/mg protein) during the entire culture

Anthraquinone biosynthesis in Morinda

Figure 3. ICS activity and anthraquinone accumulation in Morinda cell

cultures. Cells were grown in the presence (squares) or absence (circles) of 0.1 mol/L 2,4-D. Samples were taken at intervals and analysed for ICS activity (solid symbols) and anthraquinone content
(open symbols). Results are given as means SE (n = 3).

611

Figure 5. Inhibiton of growth, ICS activity and anthraquinone accumulation in Morinda cell cultures by glyphosate. Cells were cultured at
various concentrations of glyphosate for 7 d in the absence of auxin.
Then the cells were harvested and analysed for fresh weight (squares), ICS activity (triangles), and anthraquinone content. (circles).
Results are given as means SE (n = 3).

Figure 4. Inhibition of ICS activity by auxins in cell cultures of Morinda.

Cells were cultured for 10 d in the absence of auxin and then exposed
to various concentrations of 2,4-D (squares) or NAA (circles) for 24 h
after which the cells were harvested. A culture without auxin served
as a control. ICS activities were determined, expressed as percentages of the control value, and presented as means SE (n = 3).

period (data not shown). Switching anthraquinone biosynthesis on or off had no detectable effect on the activity of CM.
The relation between ICS and anthraquinone synthesis was
further investigated in inhibition experiments. To study the
sensitivity of ICS activity to inhibition by 2,4-D and NAA, the
auxins were added to anthraquinone-producing cells in a
concentration range from 0 to 10 mol/L. Enzyme activity was
reduced to 50 % by 0.01 mol/L 2,4-D. Concentrations of 2,4D exceeding 1 mol/L completely shut down ICS activity (Fig.
4). NAA was also inhibitory to ICS activity, but the Morinda
cultures were about 30 times less sensitive to NAA than to
2,4-D.
To examine the effect of chorismate limitation on anthraquinone production, cells were exposed in a medium without
auxin to sublethal doses of glyphosate, an inhibitor of choris-

Figure 6. Accumulation of shikimate in Morinda cells exposed to glyphosate. Cultures were grown in the presence (upper panel) or
absence (lower panel) of 0.5 mol/L glyphosate. The cells were
extracted after 14 d and the extract was analysed by HPLC. The shikimate peak is marked 1.

612

Marc Stalman et al.

Discussion

Figure 7. Effect of glyphosate on growth, ICS activity and anthraquinone accumulation in Rubia cultures. Cells were cultured at various
concentrations of glyphosate for 10 d in a medium containing 25 mol/
L 2,4-D. The cells were harvested and fresh weight (squares), ICS
activity (triangles) and anthraquinone content (circles) were determined. Results are given as means SE (n = 3).

mate formation (Pinto et al. 1988). After an incubation of 7 d,

glyphosate had reduced both growth and anthraquinone biosynthesis but the inhibition of the latter was more severe (Fig.
5). Both processes may be negatively affected by lack of
chorismate. That the amount of chorismate was reduced in
the cell by inhibition of the conversion of shikimate into chorismate was concluded from the large accumulation of shikimate in glyphosate-treated cells (Fig. 6). However, anthraquinone formation was additionally impaired by a reduction of
ICS activity (Fig. 5). The negative effect of glyphosate on ICS
activity was not caused by direct inhibition of the enzyme by
glyphosate because in vitro activity of ICS was not reduced
by the herbicide (data not shown).
The correlation between ICS and anthraquinone accumulation was found to exist even in the presence of glyphosate.
Anthraquinones were accumulated as long as ICS was active, even when chorismate levels were insufficient to maintain
normal growth as is the case at 0.1 mmol/L glyphosate
(Fig. 5).
To check whether the observed relation between ICS activity and anthraquinone synthesis also exists in other members
of the Rubiaceae, cell cultures of Rubia tinctorum were used.
Rubia cultures were less sensitive to 2,4-D with respect to inhibition of anthraquinone synthesis. As with Morinda cells, removal of 2,4-D induced accumulation of anthraquinones
(data not shown). When treated with glyphosate, Rubia cells
were less sensitive to glyphosate than Morinda cells. High
glyphosate concentrations inhibited growth, but led to an increase in ICS activity and anthraquinone production even in
the presence of 2,4-D (Fig. 7). This response was the opposite of the situation in Morinda, but the relationship between
ICS activity and anthraquinone synthesis was identical.

Morinda cells regulate anthraquinone biosynthesis by modulating the activity of ICS, the first enzyme of the secondary
pathway. The auxins 2,4-D and NAA both repress anthraquinone formation and ICS in a correlated fashion, but 2,4-D is
approximately 30 times more inhibitory than NAA. The activities of the enzymes DAHP synthase and CM are of little or no
regulatory importance in anthraquinone synthesis. The relationship between ICS activity and anthraquinone biosynthesis
exists also in cell cultures of Rubia (Van Tegelen et al. 1999).
Only when the ICS activity is very high is the correlation between enzyme activity and anthraquinone accumulation less
pronounced.
The biosynthesis of anthraquinones in Morinda is very sensitive to 2,4-D. The auxin inhibits anthraquinone accumulation
not only during the period of exposure, but also after its removal (Fig. 1). The length of the lag period before anthraquinone formation starts depends on the 2,4-D concentration to
which the cells were exposed in the growth medium during
preculturing, and ranges from a few days at 0.1 mol/L to several weeks at 5 mol/L. For this reason, the delay is interpreted as a carry-over effect of the growth regulator.
ICS activity and anthraquinone accumulation are correlated, which makes ICS a potential regulatory enzyme in anthraquinone synthesis. Several lines of evidence lead to this
conclusion. Anthraquinone production is always preceded by
an increase in ICS activity, whereas the activity of this enzyme
is absent in cultures that do not produce anthaquinones (Fig.
3). Addition of low concentrations of 2,4-D leads to a rapid
decrease in ICS activity. NAA allows for a modest anthraquinone accumulation (Fig. 1) and is about 30 times less effective as an inhibitor of ICS than 2,4-D (Fig. 4). The pivotal role
of ICS in anthraquinone formation combined with the difference in auxin sensitivity explains why NAA cultures remain
stable while still accumulating anthraquinones. The low enzyme activity left in NAA cultures is sufficient to maintain a
constant anthraquinone concentration, while at the same time
potentially harmful effects on growth from anthraquinone
overproduction are avoided. The toxicity of anthraquinones at
high concentration is suggested by the sudden death of cells
cultured in the absence of auxins after 14 days. However,
other possibilities such as exhaustion of essential medium
components are not excluded.
The rapid decline in ICS activity upon administration of 2,4D (Fig. 4) suggests that ICS is a rather unstable enzyme. The
fall in enzyme activity is preceded by a strong reduction in the
level of the ICS mRNA (unpublished results). The instability of
the enzyme and of the corresponding mRNA indicates that
the activity of the enzyme is continuously under genetic control and that 2,4-D modifies expression of the ICS gene.
The control of anthraquinone biosynthesis by ICS is not absolute. In a culture without 2,4-D, ICS activity drops in the final
days to less than half the maximal activity, but the decline has

Anthraquinone biosynthesis in Morinda

only a moderate effect on anthraquinone accumulation (Fig.
3). This would mean that the ICS activity is not a rate-limiting
factor during the whole culture period. In the second half,
precursor availability may be limiting as is predicted by the
overflow model (Jensen 1986, Poulsen and Verpoorte 1991).
CM and DAHP synthase do not show a correlation between
enzyme activity and anthraquinone biosynthesis. The activity
of CM does not change after induction of anthraquinone biosynthesis. The high level apparently does not prevent chorismate from being channeled towards anthraquinone formation
as soon as ICS becomes active. Thus, the chorismate concentration is unlikely to be a limiting factor at the onset of
anthraquinone biosynthesis. This idea is supported by the
finding that the activity of DAHP synthase does not increase if
anthraquinone synthesis is switched on by withdrawal of 2,4D. Limitation of chorismate biosynthesis by glyphosate causes a rise in DAHP synthase activity, a process which is genetically regulated (Pinto et al. 1988, Herrmann and Weaver
1999). The absence of such an increase in our experiments
indicates that the intake of carbon into the shikimate pathway
is high enough to supply the cells with sufficient chorismate
whether or not anthraquinones are synthesised.
Artificially lowering the flow of precursors through the shikimate pathway by glyphosate results in a reduction of anthraquinone accumulation due to the inhibiting effect of glyphosate on chorismate formation (Amrhein et al. 1980) and on
ICS activity. That ICS plays a regulatory role in anthraquinone
biosynthesis even under chorismate-limiting conditions is indicated by the observation that glyphosate leads to an increase in ICS activity and anthraquinone accumulation in Rubia cells. Even under the extreme condition that the chorismate concentration is insufficient to maintain normal growth,
the relationship between the rate of anthraquinone accumulation and ICS activity remains. There is no easy explanation for
the phenomenon that Morinda and Rubia cells react differently to glyphosate with respect to anthraquinone biosynthesis.
In conclusion, the data indicate that ICS plays an important
role in the regulation of anthraquinone synthesis by controlling the flux of precursors into the pathway leading to anthraquinone formation. However, the activity of the enzyme may
become so high that it no longer is rate-limiting. Measurements of enzyme activities in vitro do not always reflect the
actual flow of precursors in vivo. Labeling experiments with
13
C-labeled substrates in which the flux of precursors through
the chorismate branching point and, therefore, the partitioning of this important precursor, are determined would be a
valuable approach to confirm the importance of ICS in anthraquinone biosynthesis.

613

References
Amrhein N, Deus B, Gehrke P, Steinrcken HC (1980) The site of the
inhibition of the shikimate pathway by glyphosate. Plant Physiol 66:
830 834
Bradford MM (1976) A rapid and sensitive method for quantification of
microgram quantities of protein utilyzing the principle of proteindye binding. Anal Biochem 72: 159163
Cooley WE, Foy CL (1992) Effects of SC-0224 and glyphosate on free
amino acids, soluble protein, and protein synthesis in inflated
duckweed (Lemna gibba). Weed Sci 40: 345 350
Gamborg OL, Miller RA, Ojima V (1968) Nutrient requirements of suspension cultures of soybean root cells. Exp Cell Res 50: 151158
Ganson RJ, DAmato TA, Jensen RA (1986) The two-isozyme system
of 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase in Nicotiana silvestris and other higher plants. Plant Physiol 82: 203
210
Grisch H (1978) A new test for chorismate mutase activity. Anal Biochem 86: 764768
Haderlie LC, Widholm JM, Slife FW (1977) Effect of glyphosate on carrot and tobacco cells. Plant Physiol 60: 40 43
Hagendoorn MJM, Van der Plas LHW, Segers GJ (1994) Accumulation of anthraquinones in Morinda citrifolia cell suspensions. Plant
Cell Tissue Organ Cult 38: 227 234
Haslam E (1996) Aspects of the enzymology of the shikimate pathway.
In: Herz W, Kirby GW, Moore RE, Steglich W, Tamm C (eds) Progress in the Chemistry of Organic Natural Products, vol 69.
Springer-Verlag, Wien pp 157 240
Herrmann KM (1995) The shikimate pathway: early steps in the biosynthesis of aromatic compounds. Plant Cell 7: 907 919
Herrmann KM, Weaver LM (1999) The shikimate pathway. Annu Rev
Plant Physiol Plant Mol Biol 50: 473 503
Jensen RA (1986) The shikimate arogenate/pathway: Link between
carbohydrate metabolism and secondary metabolism. Physiol
Plant 66: 164168
Jensen RA, Nester EW (1966) Regulatory enzymes of aromatic amino
acid biosynthesis in Bacillus subtilis. I. Purification and properties
of 3-deoxy-D-arabino-heptulosonate 7-phosphate synthetase. J
Biol Chem 241: 3365 3372
Pinto JEBP, Dyer WE, Markley JL, Herrmann KM (1986) 3-deoxy-Darabino-heptulosonate 7-phosphate synthase from potato tuber
(Solanum tuberosum L.). Plant Physiol 82: 10401044
Pinto JEBP, Dyer WE, Weller SC, Herrmann KM (1988) Glyphosate induces 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase in
potato (Solanum tuberosum L.) cells grown in suspension culture.
Plant Physiol 87: 891 893
Poulsen C, Verpoorte R (1991) Roles of chorismate mutase, isochorismate synthase and anthranilate synthase in plants. Phytochemistry
30: 377 386
Poulsen C, Van der Heijden R, Verpoorte R (1991) Assay of isochorismate synthase from plant cell cultures by high-performance liquid
chromatography. Phytochemistry 30: 2873 2876
Poulsen C, Verpoorte R (1992) Activities of chorismate utilizing enzymes and of enzymes involved in indole alkaloid biosynthesis in
cell suspension cultures. Plant Physiol Biochem 30: 105113

Acknowledgements. The authors wish to thank Professors L. W. H.

van der Plas and J. G. Woolley, and Dr. M. J. M. Hagendoorn for stimulating discussions and critical reading of the text.

Van der Plas LHW, Eijkelboom C, Hagendoorn MJM (1995) Relation

between primary and secondary metabolism in plant cell suspensions. Plant Cell Tissue Organ Cult 43: 111116

614

Marc Stalman et al.

Van Tegelen LJP (1999) The role of isochorismate synthase in the regulation of anthraquinone biosynthesis. PhD thesis, University of Nijmegen
Van Tegelen LJP, Bongaerts RJM, Croes AF, Verpoorte R, Wullems GJ
(1999) Isochorismate synthase isoforms from elicited cell cultures
of Rubia tinctorum. Phytochemistry 51: 263 269

Zenk MH, El-Shagi H, Schulte U (1975) Anthraquinone production by

cell suspension cultures of Morinda citrifolia. Planta Med, Suppl
1975: 79101
Zenk MH, Schulte U, El-Shagi H (1984) Regulation of anthraquinone
formation by phenoxyacetic acids in Morinda cell cultures. Naturwissenschaften 71: 266