Plant Physiology and Biochemistry 43 (2005) 241–248

www.elsevier.com/locate/plaphy

Effect of water and temperature stress on the content

of active constituents of Hypericum brasiliense Choisy
Ilka Nacif de Abreu, Paulo Mazzafera *
Departamento de Fisiologia Vegetal, Instituto de Biologia, CP 6109, Universidade Estadual de Campinas (UNICAMP), 13083-862 Campinas, SP, Brazil
Received 1 December 2004; accepted 31 January 2005

Available online 09 March 2005

Abstract

Hypericum brasiliense is a medicinal herb containing several compounds with important pharmacological activity. In this study, we
investigated the effects of water stress (waterlogging and drought) and temperature (low and high, constant and alternate) on the content of
betulinic acid and phenolic compounds (quercetin, rutin, 1,5-dihydroxyxanthone, isouliginosin B) in this species. In general, the water stress
increased the levels of all of the compounds analyzed, particularly some of the phenolic compounds. On the other hand, the responses to
alternating temperatures varied according to the compound. The results for plants kept in growth chambers indicated that low light intensity
might have influenced the levels of the compounds. There was also a reallocation of carbon, with water-stressed plants showing a reduction in
growth while the levels of the compounds increased. In the temperature treatments, such an increase was evident only for the phenolic
compounds.
© 2005 Elsevier SAS. All rights reserved.

Keywords: Hypericum brasiliense; Water stress; temperature; Betulinic acid; Phenolics

1. Introduction cies may eventually become a domesticated plant and serve

Hypericum brasiliense (Guttiferae), an herb found in south-
ern and southeastern Brazil, contains substances, including
xanthones, acylfloroglucinols, quercetin, rutin, anthocyanins
and betulinic acid, that have important pharmacological
actions [1]. Studies in vitro have shown that 1,5-
dihydroxyxanthone, isolated from H. brasiliense, inhibits
monoamine oxidases A and B [2], and more recent studies
have shown that this species may also have analgesic proper-
ties [3]. Other biological actions have been attributed to sub-
stances found in H. brasiliense extracts, including antioxi-
dant and anticancer activities and the inhibition of HIV-
1 replication [4–6]. H. brasiliense does not contain hypericin
[7] which, together with pseudohypericin, has been impli-
cated in the anti-depressive action of H. perforatum [8]. How-
ever, recent reports have questioned this role of hypericin [9].
Hypericin may also cause photosensitivity and induce skin
allergies [10]. Because of the similarities in the chemical com-
positions of H. perforatum and H. brasiliense, the latter spe-
* Corresponding author. Fax: +55-19-3788-6210.
E-mail address: pmazza@unicamp.br (P. Mazzafera).
0981-9428/$ - see front matter © 2005 Elsevier SAS. All rights reserved.
doi:10.1016/j.plaphy.2005.01.020
as an alternative to H. perforatum for medicinal use.
In recent years, there has been a dramatic increase in the
sale of products based on or derived from naturally occurring
compounds. Cragg et al. [11] reported that of 520 drugs
approved by the Food and Drug Administration (FDA) in the
United States over a 12-year period (1983–1994), 30% were
natural products or their derivatives. Concomitant with the
exploration of plants to isolate and identify new molecules of
pharmacological importance, there has been an increase in
the use of plant materials in the form of liquid extracts and
powders. Although these products are sold in various coun-
tries under appropriate licenses from the relevant health insti-
tutions responsible for controlling such sales, most of these
compounds lack a detailed chemical characterization [12],
and few or no biological assays have been done to ensure that
the products are indeed active against the target for which
they are sold.
Several factors may affect the levels of the active prin-
ciples of plant material during collection, processing and stor-
age. Indeed, biotic and abiotic stresses exert a considerable
influence on the levels of secondary metabolites in plants [13].
242 I. Nacif de Abreu, P. Mazzafera / Plant Physiology and Biochemistry 43 (2005) 241–248
Hypericum is a genus of about 400 species widespread
throughout the world [14] and although it has been shown
that several of them contain compounds with important phar-
macological application most of the studies have been car-
ried out with H. perforatum [14–17]. Regarding studies on
the effect of biotic and abiotic stresses on plants of this genus,
again the few ones carried out so far were focused on H. per-
foratum [18–20]. They showed that the composition of phy-
tochemicals may be affected by such stresses. Other recent
and important reports have been published on the phenolic
composition and antioxidant activity of extracts from plants
and in vitro cultures of Hypericum androsaemum and H. per-
foratum [21–24].
In a previous report we described the isolation of the trit-
erpene betulinic acid and of the phenolics isouliginosin B,
1,5-dihydroxyxanthone, quercetin and rutin from H. brasil-
iense and investigated their distribution during plant growth
[1]. The aim of this study was therefore to investigate the
effect of water and temperature stresses on the levels of these
bioactive substances in H. brasiliense.
2. Materials and methods
2.1. Plant material
Two- and three-month-old plants (15 and 25 cm high,
respectively) obtained from seeds collected in the municipal-
ity of Ibiuna, in the State of São Paulo, Brazil were used in
the temperature and water stress experiments, respectively.
The plants were grown in plastic pots (700 ml) containing
soil and sand (2:1, v/v) and received a complete nutritive solu-
tion twice a week [25]. After the treatments described below,
the shoots were collected and immediately used for the extrac-
tions. The roots were also used for extraction but were first
washed in running tap water and then blotted dry.
2.2. Experiment with temperature stress
Two sets of plants were used to study the effects of tem-
perature. One set of plants was divided in three groups and
the first group maintained at ambient temperature (green-
house), the second at ambient temperature during the day
(greenhouse) and 10 °C at night (growth chamber, in the dark;
TA10N), and the third at ambient temperature during the day
(greenhouse) and 30 °C at night (growth chamber, in the dark;
TA30N). Control plants were maintained continuously in
greenhouse where the temperature varied from 20 to 32 °C.
During the day the greenhouse temperature varied from 25 to
32 °C. Another set of plants was also divided in three groups
which were maintained constantly in growth chambers (16 h
of light) at 17, 25 or 36 °C (T17DN, T25DN and T36DN,
respectively). The photosynthetic active radiation inside the
chambers was 60 µmol photons cm–2 s–1. For this set of plants,
control plants were maintained continuously in a growth
chamber at 25 °C (T25DN).
2.3. Experiment with water stress
Control plants were irrigated once a day. To produce
hypoxia, a set of pots containing plants were saturated with
water and then placed in other pots (3 l) filled with water. The
water was replenished whenever necessary. Drought stress
was applied gradually in other set of plants. After they had
been excessively watered, the excess water was drained and
the pots were weighed. The next day, the plants were weighed
and half of the weight loss was restored as water (1 g = 1 ml).
After 15 days of drought stress, symptoms of wilting were
seen in some plants in late morning and the experiment was
terminated. The hypoxia experiment was also terminated at
the same time.
2.4. Characterization of the stresses
In the temperature experiment, stress was characterized
by measuring the quantum efficiency of photosystem II-PSII
[26] using a fluorometer (MINI-PAN, Heinz Walz GmabH).
The determinations were done on fully expanded leaves. In
the water stressed plants the water potential was determined
with a pressure chamber (model 1000, PMS instruments, Cor-
vallis, OR, USA). Since the leaves of H. brasiliense were too
small to attach to the chamber, a lateral branch was used.
Lipid peroxidation was determined in the water stressed plants
by measuring the production of malondialdehyde (MDA) [27].
2.5. Extraction of compounds
Both experiments were terminated 15 days after initiating
the stresses. Biochemical analyses were done using shoots
and roots (water stress experiment) and shoots (temperature
stress experiment). At harvest the fresh weights were obtained
and the material subjected to extraction with petrol, dichlo-
romethane or methanol (0.5 g/2 × 10 ml—72 h each extrac-
tion). The tissues were previously ground in mortar with a
pestle using a small volume of organic solvent and then trans-
ferred to sealed cap tubes for extraction. The tubes remained
under gentle agitation during the extraction period. The
extracts were dried under vacuum and solubilized with 2 ml
of solvent. The material extracted with petrol was used to
determine the isouliginosin B and betulinic acid, in dichlo-
romethane to determine 1,5-dihydroxyxanthone, and in
methanol to determine rutin and quercetin.
2.6. Analysis of compounds by HPLC/DAD
Isouliginosin B was determined in petrol extracts [1] and
quantified by HPLC (Shimadzu) using a C 18 column
(microsorb-MVTM, 25 cm × 4 mm, 5 µm, Varian) equili-
brated with 0.5% acetic acid in water (solvent A) and a gra-
dient (0–20 min, 10% B; 20–40 min, 50% B; 40–50 min,
60% B; 50–60 min, 100% B) of 0.5% acetic acid in acetoni-
trile (solvent B). The flow rate was 0.6 ml min–1 and the com-
pounds eluted from the column were detected with a diode
 I. Nacif de Abreu, P. Mazzafera / Plant Physiology and Biochemistry 43 (2005) 241–248
array detector at 284 nm. 1,5-Dihydroxyxanthone was deter-
mined in dichloromethylene extracts and quercetin and rutin
were determined in methanolic extracts [1]. These three com-
pounds were analyzed by reversed-phase high performance
liquid chromatography (RP-HPLC) using a Supelcosil—LC18
column (25 cm × 4.6 mm, 5 µm, Supelco) and a diode array
detector at 254 nm. Solvents A (0.5% acetic acid) and B
(methanol) were used to separate the compounds with a gra-
dient: 0–5 min, 30% B; 5–20 min, 80% B; 20–30 min, 100%
B. The flow rate was 1.0 ml min–1. Total soluble phenolic
compounds were measured [28] in the same methanolic
extracts used to quantify quercetin and rutin. Petrol was used
to extract betulinic acid [1] which was then quantified by
RP-HPLC. A C18 reversed-phase microsorb-MVTM column
(25 cm × 4 mm, 5 µm, Varian) was used to separate the com-
pounds isocratically (acetonitrile/water, pH 3.0, with phos-
phoric acid, 9:1, v/v, flow rate 1 ml min–1). Detection was
done with a diode array detector at 210 nm.
Standards of 1,5-dihydroxyxanthone, betulinic acid and
isouliginosin B purified from H. brasiliense [1] were used in
the HPLC analyses to determine the concentrations of the
different compounds. Pure quercetin and rutin purchased from
Sigma were used as standards. Phenol was used as a standard
for the total phenolic content.
2.7. Data analysis and experimental design
In both experiments, each treatment consisted of three rep-
licates in a random statistical design. The data were initially
compared by analysis of variance (ANOVA) and differences
between means were detected using the Tukey test. Values of
P < 0.05 indicated significance.
3. Results
3.1. Temperature stress
The photosynthetic quantum efficiency of PSII was deter-
mined in order to evaluate the onset of temperature stress.
Significant changes were seen only in plants maintained in
the greenhouse during the day and transferred to 10 °C at
night (TA10N), which showed a decrease in photosynthetic
efficiency (Table 1).
Table 1
PSII quantum efficiency and contents of total soluble phenol, isouliginosin B, 1,5-dihydroxyxanthone, rutin, quercetin and betulinic acid either in the shoots of
H. brasiliense plants grown in a greenhouse and transferred or not (control) to 10 °C (TA10N) and 30 °C (TA30N) during the night, or in the shoots of plants
grown in a growth chamber at a constant temperature of 25 °C (control), 17 °C (T17DN) or 36 °C (T36DN)
Greenhouse/growth chamber
Control
PSII efficiency 0.8a
Total soluble phenols (mg g–1) 41.1b
1.5-Dihydroxyxanthone (µg g–1) 28.3c
Rutin (µg g–1) 93b
Quercetin (µg g–1) 23.7 b
Betulinic acid (mg g–1) 6.2a
Different letters indicate a significant difference (P ≤ 0.05, Tukey test) among the treatments of plants grown in a greenhouse or in a growth chamber.
243
There was an increase in the levels of total soluble phe-
nols and of all the phenolic compounds analyzed in plants
maintained at TA10N (Table 1). Although to a less extend,
1,5-dihydroxyxanthone also increased in plants kept at 30 °C
during the night (TA30N). In contrast, rutin and quercetin
decreased with this treatment. Betunilic acid was reduced in
TA30N. Plants that were kept in growth chambers at 17 °C
(T17DN) and 36 °C (T36DN) had higher levels for all of the
compounds analyzed, except for quercetin at 17 °C, indicat-
ing that, in addition to variations in temperature, a low light
intensity could also influence the levels of these compounds.
Low (TA10N) or high (TA30N) temperatures during the
night affected the accumulation of plant mass, as well as the
total content of phenolic compounds (sum of isouliginosin B,
1,5-dihydroxyxanthone, rutin and quercetin) and betulinic
acid in the shoot (Fig. 1). While the total amount of betulinic
acid in the shoot followed the pattern of variation seen for the
accumulation of fresh mass, the changes in the phenolic com-
pounds showed the opposite pattern, i.e. a decrease in mass
was accompanied by an increase in these compounds. Except
for betulinic acid in plants at T17DN, the same responses
were observed with constant temperatures.
3.2. Water stress
Water stress was monitored by measuring the MDA levels
and the water potential in the leaves (Table 2). Plants with a
water deficit had a lower water potential, indicating that stress
had developed at the time of the biochemical evaluations.
MDA did not increase significantly in the shoots and roots of
these plants. The development of hypoxia stress was con-
firmed by membrane lipid peroxidation since the highest
MDA values were seen in the shoots and roots of these plants.
In this experiment, analysis of the roots also provided data
on the content of isouliginosin B (Table 2). This compound is
found only in the roots of H. brasiliense [1]. Compared to
control plants, there was generally an increase in the levels of
phenolic compounds in water-stressed plants, as demon-
strated by the content of total soluble phenolic compounds.
An exception to this was the roots of plants under hypoxia,
which showed the same level observed in the control plants.
This same pattern of accumulation was seen for 1,5-
dihydroxyxanthone. In contrast, isouliginosin B increased
Growth chamber
TA10N TA30N T36DN
0.6b 0.73a 0.7a 0.7a 0.7a
60.3a 31.8b 13.8b 41.1a 33.8a
629.7a 112.2b 0.5c 8b 38.4a
184.4 a
73.1c 46.7c 163.2b 340.2a
173.8 a
2.3c 23.3a 19.8b 25.3a
5.0a 3.7b 4.4b 8.2a 9.0a
244 I. Nacif de Abreu, P. Mazzafera / Plant Physiology and Biochemistry 43 (2005) 241–248

Fig. 1. Fresh mass (g shoot–1) and the levels of phenolic compounds (isouliginosin B, 1,5-dihydroxyxanthone, rutin, quercetin) and betulinic acid (all in
mg shoot–1) in the shoots of H. brasiliense plants grown in a greenhouse and transferred or not (control) to 10 °C (TA10 N) and 30 °C (TA30 N) during the night,
or in the shoots of plants grown in a growth chamber at a constant temperature of 25 °C (control), 17 °C (T17DN) or 36 °C (T36DN). Different letters indicate
a significant difference (P ≤ 0.05, Tukey test) among the treatments of plants grown in a greenhouse or growth chamber.

Table 2
Water potential and levels of MDA, total soluble phenol, isouliginosin B, 1,5-dihydroxyxanthone, rutin, quercetin and betulinic acid in the shoots and roots of
H. brasiliense plants under hypoxia and drought stress
Shoot Root
Control Hypoxia Drought Control Hypoxia Drought
Water potential (MPa) –1.19b –1.25b –2.97a
MDA (mmol g–1) 3.0b 8.6a 3.4b 0.2b 1.1a 0.8ab
Total soluble phenols (mg g–1) 31.5b 57.3a 56.9a 5.2b 4.5b 8.4a
Isouliginosin B (µg g–1) 11.4c 39.1b 478.2a
1,5-Dihydroxyxanthone (µg g–1) 82.5c 306.4a 260.8b 13.6b 11.9b 89.6a
Rutin (mg g–1) 0.3c 2.3a 1.5b
Quercetin (mg g–1) 1.4b 0.4c 2.2a
Betulinic acid (mg g–1) 2.3b 4.5a 3.7a 1.3b 2.7a 2.6a
Different letters indicate a significant difference (P ≤ 0.05, Tukey test) among the treatments for each plant organ.

both in the roots of drought-stressed plants and in the roots of 4. Discussion

plants under hypoxia. However, in drought-stressed plants,
the level of this acylfloroglucinol was 12-fold higher than
those in plants under hypoxia and 43-fold higher than control
plants. 1,5-Dihydroxyxanthone levels increased in the shoots
of stressed plants and in the roots of drought-stressed plants.
Rutin and quercetin were not found in the roots of H. brasil-
iense. Rutin increased in response to drought and hypoxia
stress but quercetin increased only under a water deficit. Betu-
linic acid increased in the shoots and roots of plants in both
treatments.
Since the accumulation of fresh mass was reduced by water
stress, there was an increase in the content of the phenolic
compounds analyzed (Fig. 2). However, contrary to the
temperature experiment, in this case betulinic acid followed
the same pattern and increased with the reduction in plant
mass.
Plants growing at low temperatures may show a loss of
vigor and a reduction in growth, probably because of a
decrease in photosynthesis [29]. Plants exposed to low tem-
peratures usually experience oxidative stress caused by reac-
tive oxygen species (ROS) which are very reactive and can
damage cells by oxidizing lipids, proteins and nucleic acids.
The simultaneous occurrence of low temperature stress and a
high incidence of light apparently increases the tissue dam-
age by causing long-term reductions in photosynthetic effi-
ciency [30].
In H. brasiliense plants subjected to low and high tem-
peratures, stress was monitored by determining the photosyn-
thetic efficiency, which decreased only in plants maintained
at TA10N. Although 17 °C was chosen as a low temperature,
the photosynthetic efficiency of plants at T17DN did not
change. Such an apparent discrepancy can be explained by
 I. Nacif de Abreu, P. Mazzafera / Plant Physiology and Biochemistry 43 (2005) 241–248
Fig. 2. Fresh mass (g shoot–1) and the levels of phenolic compounds (isouliginosin B, 1,5-dihydroxyxanthone, rutin, quercetin) and betulinic acid (all in
mg plant–1) in entire plants (shoot + root) of H. brasiliense under hypoxia and drought stress. Different letters indicate significant differences (P ≤ 0.05, Tukey
test).
the fact that 17 °C may not be a chilling temperature for H.
brasiliense. In addition, during the day, plants in the TA10N
group were exposed to natural day light and high tempera-
tures (28–32 °C) in the greenhouse, and this tended to aggra-
vate the effects of the night temperature.
Several studies have shown that extracts from Hypericum
species have antioxidant activity [16,31]. Recently a direct
relationship with this property was established with fla-
vonoids and caffeoylquinic acids extracted from H. perfora-
tum [32].
Although photosynthesis showed alterations only in the
TA10N treatment, increased levels of all of the analyzed sub-
stances were observed at all of the temperatures studied, indi-
cating that light intensity and quality may play a role in this
variation. In addition, control plants maintained under green-
house conditions had higher contents of the compounds stud-
ied (except for quercetin) than did control plants in the growth
chamber. Grace et al. [33] reported a higher chlorogenic acid
content in the leaves of Mahonia repens under high light con-
ditions during the winter than in the leaves of plants under
high light during the summer. However, the content was simi-
lar in shaded leaves from both seasons. These authors also
observed a linear correlation between the phenolic content
and the antioxidant activity of leaf extracts.
Although no parameter indicative of ROS production was
measured in the temperature experiments, the variations seen
in the content of phenolic compounds could be partly related
to the presence of these species. Indirect evidence that fla-
vonols (the flavonoid class to which quercetin and rutin
belong) may confer protection against ROS has been obtained
in experiments with A. thaliana mutants in which UV/blue
light-absorbing flavonols had a strong protective function
against excessive visible light [34].
Wang and Zheng [35] cultivated strawberry plants under
different combinations of night and day temperatures
(12/18 °C, 12/25 °C, 22/25 °C, 22/30 °C) and examined the
composition of phenolic compounds, as well as the antioxi-
dant capacity of the juice. The highest night and day tempera-
tures resulted in the highest production of phenolics and of
245
ROS. Plants grown at 12/18 °C generally had the lowest antho-
cyanin, flavonol (quercetin 3-glucoside and quercetin
3-glucoronide) and phenolic acid contents. H. brasiliense
responded differently with quercetin. A marked increase
(eightfold) was seen only in plants of the TA10N group
(Table 1). In contrast, rutin (quercetin 3-rhamnoglucosyl)
increased in the T17DN (~threefold) and T36DN (~seven-
fold) groups. As with quercetin, rutin decreased in the TA30N
group, possibly because of the strong temperature stress to
which these plants were subjected. This decrease also applied
to the levels of total soluble phenolics.
Although rutin is derived from quercetin, adequate discus-
sion of our data are hampered by the lack of information on
rutin biosynthesis. A recent report showed that plants of Cra-
taegus laevigata and Crataegus monogyna subjected to
drought, flooding and cold showed completely different
responses in terms of the amount of quercetin and rutin pro-
duced [36]. Whereas drought decreased the levels of querce-
tin in the first species, an increase was seen in the second. In
contrast, in both species, rutin increased with drought. Hence,
even among related species, variations in quercetin and rutin
are not necessarily correlated.
Benzophenones are intermediates in the biosynthesis of
xanthones in plants and the basic structure of benzophenone
is derived from benzoic acid and three malonyl-CoA mol-
ecules [37]. Therefore, it would be expected that any factor
capable of increasing the levels of phenolic compounds would
also increase the amount of xanthone present in plants.
Contrary to quercetin and rutin, the biosynthesis of 1,5-
dihydroxyxanthone was less affected by the combination of
high night (TA30N) and day (greenhouse) temperatures since
only a fourfold increase was observed with this temperature
combination. Among plants maintained at a constant tempera-
ture in growth chambers there was a marked increase at 36 °C,
from 0.5 µg g–1 (control) to 38.4 µg g–1 (T36DN). Despite
this increase, the concentration of 1,5-dihydroxyxanthone in
the T36DN plants was still markedly below the level reached
in plants maintained under natural light (TA10N, TA30N and
respective control). These results indicated that variations in
246 I. Nacif de Abreu, P. Mazzafera / Plant Physiology and Biochemistry 43 (2005) 241–248
temperature and in the intensity and duration of light ac-
counted for the changes in the levels of phenolic compounds.
The photosynthetic active radiation in the greenhouse on a
sunny day was approximately 1500 µmol photons m–2 s–1
whereas in the chamber it was as low as 60 µmol photons
m–2 s–1.
Levya et al. [38] studied the regulation of the biosynthesis
of anthocyanins in A. thaliana plants exposed to low tem-
perature and found that the process was light-dependent. The
expression of phenylalamine ammonia-lyase and chalcone
synthase, which are key enzymes of the phenylpropanoid
pathway, increased only when low temperature-stressed plants
were illuminated.
The biosynthesis of phenolic compounds is under genetic
control and is strongly influenced by biotic and abiotic stimuli
such as pathogens, light conditions (including ultraviolet
light), temperature and herbicides [13]. As shown here,
drought generally increased the levels of phenolic com-
pounds in the shoots and roots of H. brasiliense. Increases
were also observed with hypoxia, except for 1,5-dihydroxy-
xanthone in the roots and quercetin in the shoots, both of
which remained unchanged. Increases in the levels of pheno-
lics and of the enzymes involved in their biosynthesis in
response to drought and hypoxia have also been reported in
other plants [39,40].
Similarly to our observation, plants of H. perforatum sub-
jected to a brief drought stress showed an increase of querce-
tin and rutin [19]. However, this increase was lower than we
observed for H. brasiliense, what might related with the stress
intensity.
Plants that were subjected to drought showed a lower water
potential when compared with control plants and with plants
stressed by hypoxia (Table 1). However, the manner in which
water was withheld resulted in the gradual development of
stress, with no accumulation of MDA in these plants. PSII is
capable of reorganizing, depending on the way stress is
imposed [41], and rapid MDA formation has been observed
in plants subjected to sudden withdrawal of water [42].
The plants under hypoxia showed a marked increase in the
MDA content of their roots and shoots, suggesting lipid per-
oxidation. Similar results were observed with other plant spe-
cies under low oxygen availability [43]. Hence, the increase
in phenolic compounds seen in H. brasiliense plants stressed
by drought and hypoxia might also represent an antioxidant
response to ROS production.
In the temperature and water stress experiments, the accu-
mulation of phenolic compounds was negatively correlated
with the accumulation of plant mass (Figs. 1 and 2). The cost
of adaptation (fitness) in plant–insect interactions has been
extensively discussed in terms of the reallocation of the car-
bon used for growth to the production of secondary metabo-
lites that confer protection [44]. In relation to drought toler-
ance, the efficiency in the use of water has been suggested to
be a target of natural selection since a higher efficiency is
expected to confer a fitness advantage in drought conditions
[45].
The data reported here indicate that a water deficit may
also be associated with an increase in secondary metabolites
through the reallocation of the assimilated carbon as plant
growth is progressively reduced. The physiological rel-
evance of this phenomenon in terms of the cost to fitness is
unclear, although for phenolics with an antioxidant capacity
it may be related to a mechanism to counteract the deleteri-
ous effects of ROS. This relationship has been suggested for
hypoxia [43].
The concentration of betulinic acid increased in drought
and hypoxia stressed plants of H. brasiliense, as well as in
plants maintained at a constant temperature of 17 °C (T17DN)
or 36 °C (T36DN) in growth chambers with a low light inten-
sity. The effects of drought and waterlogging stress on the
levels of monoterpenes have been studied in needles of Picea
abies [46]. Whereas an increase was observed in drought-
stressed plants, there were no changes in waterlogged plants.
Drought stress induces the biosynthesis of isopentenyl pyro-
phosphate, the primary precursor of terpenes in plants, with a
consequent increase in abscisic acid, carotenoids and xantho-
phylls [47,48].
Light intensity induces qualitative changes in the terpenes
of Sitka spruce [49]. In addition to high and low tempera-
tures, which may activate different mechanisms of terpene
accumulation, light also appears to be a key factor in control-
ling the production of betulinic acid in H. brasiliense since
an increase in the content of this acid was seen in growth
chambers at low (17 °C) and high (36 °C) temperatures.
As with phenolics, the increase in betulinic acid content
may be a response to ROS generation. Several recent reports
have shown that isoprene is formed to protect plants against
ROS produced by environmental stresses. Heat stress has been
the most investigated [50], although photoxidative stress has
been also studied [51]. The levels of carotenoids, which are
tetraterpenes, are increased to scavenge ROS induced by high
light [52]. In addition, high light intensity associated with
low temperature (8 ± 2 °C) differentially induces two dis-
tinct isopentenyl diphosphate isomerases [53]. This enzyme
is responsible for the isomerization of isopentenyl diphos-
phate to dimethylallyl diphosphate to form the isoprene unit,
the basic compound for the synthesis of all terpenes.
In conclusion, our results show that environmental fac-
tors, particularly temperature, have a marked influence on the
content of pharmacologically active secondary metabolites
in H. brasiliense. However, there is a variation depending on
the stress and the analyzed compound. The increase in the
contents of the compounds analyzed agrees with data in the
literature indicating that this is probably a response to the
generation of ROS. A similar physiological response may also
occur under field conditions and the final content of these
and other compounds will probably reflect on the effective-
ness of a natural product sold in the market. These results
also show that it is possible to increase the content of phar-
macologically desirable compounds by manipulating agricul-
tural techniques such as irrigation or using photobioreactor
systems [36]. Alternatively, hydroponic systems might also
be used [54].
 I. Nacif de Abreu, P. Mazzafera / Plant Physiology and Biochemistry 43 (2005) 241–248
Acknowledgements
The authors thank Arthur Germano Fett Neto and Ladaslav
Sodek for critical reading of the manuscript. This work was
supported by CNPq and FAPESP, Brazil.
References
[1] I.N. Abreu, A.L.M. Porto, A.J. Marsaioli, P. Mazzafera, Distribution
of bioactive substances from Hypericum brasiliense during plant
growth, Plant Sci. 167 (2004) 949–954.
[2] L. Rocha, A. Marston, M. Kaplan, H. Stoecklievans, U. Thull,
B. Testa, et al., An antifungal c-pyrone and xanthones with monoam-
ine oxidase inhibitory activity from Hypericum brasiliense, Phy-
tochemistry 36 (1994) 1381–1385.
[3] F.R. Mendes, R. Mattei, E.A. Carlini, Activity of Hypericum brasil-
iense and Hypericum cordatum on the central nervous system in
rodents, Fitoterapia 73 (2002) 462–471.
[4] D.O. Kim, K.W. Lee, H.J. Lee, C.Y. Lee, Vitamin C equivalent
antioxidant capacity (VCEAC) of phenolic phytochemicals, J. Agr.
Food Chem. 50 (2002) 3713–3717.
[5] T. Akihisa, J. Ogihara, J. Kato, K. Yasukawa, M. Ukiya, S. Yamanou-
chi, et al., Inhibitory effects of triterpenoids and sterols on human
immunodeficiency virus-1 reverse transcriptase, Lipids 5 (2001) 507–
512.
[6] Y.Q. Wei, X. Zhao, Y. Kariya, H. Fukata, K. Teshigawara, A. Uchida,
Induction of apoptosis by quercetin: involvement of heat shock pro-
tein, Cancer Res. 54 (1994) 4952–4957.
[7] A.B.F. Ferraz, S. Bordignon, D. Mans, A. Schmitt, A.P. Ravazzolo,
G.L. Von Poser, Screening for the presence of hypericins in southern
Brazilian species of Hypericum (Guttiferae), Pharm. Biol. 40 (2002)
294–297.
[8] O. Suzuki, Y. Katsumata, M. Oya, Inhibition of monoamine oxidase
by hypericin, Planta Med. 50 (1984) 272–274.
[9] W.E. Muller, M. Rolli, C. Schafer, U. Hafner, Effects of Hypericum
extract (LI 60) in biochemical models of antidepressant activity,
Pharmacopsychiatry 30 (1997) 102–107.
[10] C.M. Schempp, K.A. Muller, B. Winghofer, E. Schopf, J.C. Simon,
Saint John’s wort in dermatology, Hautarzt 53 (2002) 316.
[11] G.M. Cragg, J.T. Baker, R.P. Borris, B. Carte, G.A. Cordell, D.D. Soe-
jarto, Interactions with source countries. Guidelines for members of
the American Society of Pharmacognosy, J. Nat. Prod. 60 (1997)
654–655.
[12] G.A. Cordell, Biodiversity and drug discovery—a symbiotic relation-
ship, Phytochemistry 55 (2000) 463–480.
[13] R.A. Dixon, N. Paiva, Stress-induced phenylpropanoid metabolism,
Plant Cell 7 (1995) 1085–1097.
[14] A. Pabuccuoglu, S. Konyalioglu, M. Bas, G.E. Meral, The in vitro
effects of Hypericum species on human leukocyte myeloperoxidase
activity, J. Ethnopharmacol. 87 (2003) 89–92.
[15] R. Dall’Agnol, A. Ferraz, A.P. Bernardi, D. Albring, C. Nor, L. Sar-
mento, et al., Antimicrobial activity of some Hypericum species,
Phytomedicine 10 (2003) 511–516.
[16] M. Couladis, R.B. Badisa, P. Baziou, S.K. Chaudhuri, E. Pilarinou,
E. Verykokidou, et al., Antioxidant and cytotoxic activities of Hyperi-
cum sp on brine shrimps and human cancer cell lines, Phytother. Res.
16 (2002) 719–722.
[17] J. Heilmann, K. Winkelmann, O. Sticher, Studies on the antioxidative
activity of phloroglucinol derivatives isolated from Hypericum spe-
cies, Planta Med. 69 (2003) 202–206.
[18] A.P. Guedes, L.R. Amorim, A. Vicente, M. Fernandes-Ferreira, Varia-
tion of the essential composition in leaves from cultivated plants of
Hypericum androsaemum L, Phytochem. Anal. 15 (2004) 146–151.
247
[19] D.E. Gray, S.G. Pallardy, H.E. Garret, G.E. Rottinghaus, Effect of
acute drought stress and time of harvest on phytochemistry and dry
weight of St. John’s wort leaves and flowers, Planta Med. 69 (2003)
1024–1030.
[20] S.J. Murch, K. Haq, H.P.V. Rupasinghe, P.K. Saxena, Nickel contami-
nation affects growth and secondary metabolite composition of St.
John’s wort (Hypericum perforatum L.), Environ. Exp. Bot. 49 (2003)
251–257.
[21] A.C.P. Dias, R.M. Seabra, P.B. Andrade, F. Ferreres, M. Fernandes-
Ferreira, Xanthone biosynthesis and accumulation in calli and sus-
pended cells of Hypericum androsaemum, Plant Sci. 150 (2000)
93–101.
[22] A.C.P. Dias, R.M. Seabra, P.B. Andrade, F. Ferreres, M. Fernandes-
Ferreira, Xanthone production in calli and suspended cells of Hyperi-
cum perforatum, J. Plant Phisiol. 158 (2001) 821–827.
[23] P. Valentão, E. Fernandes, F. Carvalho, P.B. Andrade, R.M. Seabra,
M.L. Bastos, Antioxidant activity of Hypericum androsaemum
infusion: scavenging activity against superoxide radical, hydroxyl
radical and hypochlorous acid, Biol. Pharm. Bull. 25 (2002) 1320–
1323.
[24] P. Valentão, A. Dias, M. Ferreira, B. Silva, P.B. Andrade, M.L. Bastos,
et al., Variability in phenolic composition of Hypericum androsaeum,
Nat. Prod. Res. 17 (2003) 135–140.
[25] D.R. Hoagland, D.I. Arnon, The water-culture method for growing
plants without soil, Calif. Agric. Exp. Station Circ. 347 (1950) 1–39.
[26] B. Genty, J.M. Briantais, N.R. Baker, The relationship between the
quantum yield of photosynthetic electron-transport and quenching of
chlorophyll fluorescence, Biochim. Biophys. Acta 990 (1989) 87–92.
[27] I. Cakmak, J. Horst, Effect of aluminium on lipid peroxidation, super-
oxide dismutase, catalase, and peroxidase activities in root tips of
soybean (Glycine max), Physiol. Plant. 83 (1991) 463–468.
[28] T. Swain, W.E. Hillis, The phenolic constituents of Prunus domestica.
The quantitative analysis of phenolic constituents, J. Sci. Food Agric.
10 (1959) 63–68.
[29] G. Oqüist, Effects of low-temperature on photosynthesis, Plant Cell
Environ. 6 (1983) 281–300.
[30] N.R. Baker, P.K. Farage, C.M. Stirling, S.P. Long, Photoinhibition of
crop photosynthesis in the field at low temperatures, in: N.R. Baker,
J.R. Bowyer (Eds.), Photoinhibition of photosynthesis—from
molecular mechanisms to the field, Bios Scientific Publishers,
Oxford, 1994, pp. 349–363.
[31] M. Couladis, P. Baziou, E. Verykokidou, A. Loukis, Antioxidant
activity of polyphenols from Hypericum triquetrifolium Turra, Phy-
tother. Res. 16 (2002) 769–770.
[32] B.A. Silva, F. Ferreres, J.O. Malva, A.C.P. Dias, Phytochemical and
antioxidant characterization of Hypericum perforatum alcoholic
extracts, Food Chem. 90 (2005) 157–167.
[33] S.C. Grace, B.A. Logan, W.W. Adams, Seasonal differences in foliar
content of chlorogenic acid, a phenylpropanoid antioxidant, in Maho-
nia repens, Plant Cell Environ. 21 (1998) 513–521.
[34] M. Havaux, K. Kloppstech, The protective functions of carotenoid and
flavonoid pigments against excess visible radiation at chilling tem-
perature investigated in Arabdopsis npq and tt mutants, Planta 213
(2001) 953–966.
[35] S.Y. Wang, W. Zheng, Effect of plant growth temperature on antioxi-
dant capacity in strawberry, J. Agric. Food Chem. 49 (2001) 4977–
4982.
[36] A. Kirakosyan, P. Kaufman, S. Warber, S. Zick, K. Aaronson, S. Bol-
ling, et al., Applied environmental stresses to enhance the levels of
polyphenolics in leaves of hawthorn plants, Physiol. Plant. 121 (2004)
182–186.
[37] S. Peters, W. Schmidt, L. Beerhues, Regioselective oxidative phenol
couplings of 2,3’,4,6-tetrahydroxybenzophenone in cell cultures of
Centaurium erythraea RAFN and Hypericum androsaemum L,
Planta 204 (1998) 64–69.
248 I. Nacif de Abreu, P. Mazzafera / Plant Physiology and Biochemistry 43 (2005) 241–248
[38] A. Levya, J.A. Jarillo, J. Salinas, J.M. Martinezzapater, Low-
temperature induces the accumulation of phenylalanine ammonia-
lyase and chalcone synthase messenger-RNAs of Arabidopsis
thaliana in a light-dependent manner, Plant Physiol. 108 (1995)
39–46.
[39] G. EnglishLoeb, M.J. Stout, S.S. Duffey, Drought stress in tomatoes:
changes in plant chemistry and potential nonlinear consequences for
insect herbivores, Oikos 79 (1997) 456–468.
[40] E.A. Aguilar, D.W. Turner, K. Sivasithamparam, Fusarium oxysporum
f.sp cubense inoculation and hypoxia alter peroxidase and phenylala-
nine ammonia lyase activities in nodal roots of banana cultivars (Musa
sp.) differing in their susceptibility to Fusarium wilt, Aust. J. Bot. 48
(2000) 589–596.
[41] M.T. Giardi, A. Cona, B. Geiken, T. Kucera, J. Masojidek, A.K. Mat-
too, Long-term drought stress induces structural and functional reor-
ganization of photosystem II, Planta 199 (1996) 118–125.
[42] Y. Jiang, B. Huang, Drought and heat stress injury to two cool-season
turfgrasses in relation to antioxidant metabolism and lipid peroxida-
tion, Crop Sci. 41 (2001) 436–442.
[43] W.P. Hurng, C.H. Kao, Effect of flooding on the activities of some
enzymes of activated oxygen metabolism, the levels of antioxidants
and lipid peroxidation in senescing _obacco leaves, Plant Growth
Regul. 14 (1994) 37–44.
[44] M. Heil, I.T. Baldwin, Fitness costs of induced resistance: emerging
experimental support for a slippery concept, Trends Plant Sci. 7
(2002) 61–67.
[45] M.S. Heschel, K. Donohue, N. Hausmann, J. Schmitt, Population
differentiation and natural selection for water-use efficiency in Impa-
tiens capensis (Balsaminaceae), Int. J. Plant Sci. 163 (2002) 907–912.
[46] P. Kainulainen, J. Oksanen, V. Palomaki, J.K. Holopainen, T. Holo-
painen, Effect of drought and waterlogging stress on needle monoter-
penes of Picea abies, Can. J. Bot. 70 (1992) 1613–1616.
[47] B.V. Milborrow, The pathway of biosynthesis of abscisic acid in
vascular plants: a review of the present state of knowledge of ABA
biosynthesis, J. Exp. Bot. 52 (2001) 1145–1164.
[48] S. Turtola, A.M. Manninen, R. Rikala, P. Kainulainen, Drought stress
alters the concentration of wood terpenoids in Scots pine and Norway
spruce seedlings, J. Chem. Ecol. 29 (2003) 1981–1995.
[49] D. Wainhouse, R. Ashburner, G.I. Forrest, R.C. Boswell, The effect of
variation in light and nitrogen on the composition of resin in young
Sitka spruce, Silvae Genet. 49 (2000) 45–49.
[50] E.L. Singsaas, M. Lerdau, K. Winter, T.D. Sharkey, Isoprene increases
thermotolerance of isoprene-emitting species, Plant Physiol. 115
(1997) 1413–1420.
[51] J.G. Zeidler, H.K. Lichtenthaler, H.U. May, F.W. Lichtenthaler, Is
isoprene emitted by plants synthesized via the novel isopentenyl
pyrophosphate pathway? Z. Naturforsch. [C] 52 (1997) 15–23.
[52] K.K. Niyogi, O. Bjorkman, A.R. Grossman, The roles of specific
xanthophylls in photoprotection, in: Proc. Natl. Acad. Sci, États-Unis,
1997, pp. 14162–14167 94.
[53] A. Nakamura, H. Shimada, T. Masuda, H. Ohta, K. Takamiya, Two
distinct isopentenyl diphosphate isomerases in cytosol and plastid are
differentially induced by environmental stress in tobacco, FEBS Lett.
506 (2001) 61–64.
[54] S.J. Murch, H.P.V. Rupasinghe, P.K. Saxena, An in vitro and hydro-
ponic growing system for hypericin, pseudopypericin, and hyperforin
production of St. John’s wort (Hypericum perforatum cv New Stem),
Planta Med. 68 (2002) 1108–1112.