See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/232166465

Rosmarinic acid content in basil plants grown in vitro and in hydroponics

Article  in  Central European Journal of Biology · December 2011

DOI: 10.2478/s11535-011-0057-1

CITATIONS READS

26 614

6 authors, including:

Claudia Kiferle Mariella Lucchesini

Scuola Superiore Sant'Anna Università di Pisa
19 PUBLICATIONS   166 CITATIONS    38 PUBLICATIONS   323 CITATIONS   

SEE PROFILE SEE PROFILE

Anna Mensuali Rita Maggini

Scuola Superiore Sant'Anna Università di Pisa
82 PUBLICATIONS   1,004 CITATIONS    46 PUBLICATIONS   543 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Aquaponic system for marine fish and eel production View project

La coltivazione del giaggiolo in Toscana View project

All content following this page was uploaded by Anna Mensuali on 17 May 2014.

The user has requested enhancement of the downloaded file.

 Cent. Eur. J. Biol.• 6(6) • 2011 • 946-957
DOI: 10.2478/s11535-011-0057-1

Central European Journal of Biology

Rosmarinic acid content in basil plants grown

in vitro and in hydroponics
Research Article

Claudia Kiferle1,*, Mariella Lucchesini1, Anna Mensuali-Sodi2, Rita Maggini1, Andrea Raffaelli3,
Alberto Pardossi1
1
Department of Crop Biology, University of Pisa,
56124 Pisa, Italy
2
Sant’ Anna School of Advanced Studies,
56100 Pisa, Italy
3
Institute for the Chemistry of Organo Metallic Compounds (ICCOM),
National Council of Research (CNR),
56126 Pisa, Italy
Received 19 January 2011; Accepted 29 May 2011

Abstract: The accumulation of selected caffeic acid derivatives (CADs), in particular rosmarinic acid (RA), was investigated in different
tissues (leaves, roots and plantlet shoots) of sweet basil (Ocimum basilicum L.) plants grown either in vitro or in hydroponic culture
(floating system) under greenhouse conditions. Two cultivars with green leaves (Genovese and Superbo) and one with purple leaves
(Dark Opal) were tested. The content of CADs in HCl-methanol extracts was determined by HPLC. LC-MS and LC-MS-MS were
used to confirm the identification of the metabolites of interest. Apart from rosmarinic acid (RA) and a methylated form of this
substance, no other CADs were detected at significant level in any of the analyzed samples. The content of RA ranged approximately
from 4 to 63 mg/g DW, depending on the growing system. The highest RA content was found during the in vitro multiplication,
in the acclimatized plants and in the roots of hydroponically-grown seedlings at full bloom. In vitro, 6-benzyladenine reduced the
accumulation of RA in purple-leaf Dark Opal cultivar, but an opposite effect of this growth regulator was observed in the green-leaf
genotypes. Our findings suggest the possibility to scale-up RA production by means of in vitro or hydroponic culture of sweet basil.

Keywords: 6-benzylaminopurine • Cytokinin • Floating system • Leaves • Micropropagation phase • Ocimum basilicum • Physiological stage
• Roots • Soilless culture.
© Versita Sp. z o.o.

1. Introduction Medicinal and herbal plants, including basil, are

Sweet basil (Ocimum basilicum L.) in the Lamiaceae
family is one of the most important herbs and is widely
cultivated worldwide [1]. Basil leaves are largely employed
as a flavoring agent for food. For instance, sweet basil
is the main ingredient of the well-known ‘pesto’ sauce
and in Italy the cultivation of this species increased
considerably in the last years due to the growing demand
from the food industry [2]. Along with other species in the
Ocimum genus, sweet basil is used for pharmaceutical
and cosmetic preparations due to the high content of
essential oils [1,3] and rosmarinic acid [4,5]. Rosmarinic
acid (RA) is a caffeic acid ester with several important
biological properties, including antioxidant, antibacterial,
antiviral and anti-inflammatory activities [6].
typically cultivated in open fields, resulting in year-to-
year variability in the biomass production as well as in
the content of secondary metabolites, which are both
affected by factors such as weather, soil fertility, growing
practices and the presence of pests and diseases [7].
Therefore, there is an increasing interest for the artificial
cultivation of these crops, either in vitro or in vivo
(i.e. greenhouse hydroponic culture), where growing
conditions can be strictly controlled to consistently
produce high-standard plant material all-year round
[8,9]. Sweet basil can be grown hydroponically, which
offers several advantages over traditional soil culture,
such as higher yield per unit ground area and better
quality standards of harvested biomass. Hydroponically-
grown plant material is clean and easy to process due

* E-mail: ckiferle@agr.unipi.it
946
C. Kiferle et al.
to minimal contamination from pollutants, pests and
pathogens [10]. Floating systems is a hydroponic
technique that is increasingly used for greenhouse
production of fresh-cut leafy vegetables, including basil
[11], and has also been utilized for the cultivation of
medicinal plants [12,13].
Cell and tissue culture, or in vitro approaches,
constitute an alternative to conventional agriculture
for the production of high-value plant metabolites,
because they provide an opportunity to regulate plant
biosynthetic pathways in a controlled environment
[9,14]. The possible application of plant tissue culture
for the production of bioactive compounds has been
demonstrated in several species [9,14], although at
present only very few substances are produced in vitro
on a commercial scale [15]. Different basil species
were found to accumulate larger quantities of RA in
cell, callus, hairy roots and shoot cultures than in vivo
[16-19]. The total phenolic and RA levels were higher
in sweet basil grown hydroponically than in soil-grown
plants [20]
Compared to in vitro culture, hydroponics offers the
advantages of a higher rate of biomass production per
unit area and less expensive growing structures [21,22],
although the production cost per unit weight of the
metabolites of interest is not necessarily lower in vivo
than in vitro [23].
In this work, the accumulation of caffeic acid
derivatives (CADs), in particular of RA, was studied in
three cultivars of sweet basil, with green (Genovese and
Superbo) or purple (Dark Opal) leaves, grown either
in vitro or in hydroponic culture. To our knowledge,
no paper has been published on the influence of
micropropagation phase on the accumulation of these
metabolites in basil.
2. Experimental Procedures
2.1 Plant material
The plants were originated from seeds purchased from
SAIS (Cesena, Italy) and grown at University of Pisa
either in vitro or in hydroponics (floating system).
2.2 In vitro experiments
An original protocol for the micropropagation was
developed previously. Nodal segments were excised,
at the time of flowering, from basil seedlings grown
in floating systems under greenhouse conditions. To
prevent pathogen contamination, the mother plants
were sprayed weekly with Benomyl (1.0 g L−1; Du Pont
Agricultural products, Wilmington, Delaware, USA) in
the last three weeks before the collection of explants.
The explants were reduced in size, washed in current
water for 30 min, surface-sterilized by soaking them
under agitation in a 15% aqueous solution of NaOCl (8%
active chlorine) with a few drops of Tween 20TM (Sigma-
Aldrich, Milan, Italy) for 15 min and rinsed three times
with sterile deionized water. The explants were then cut
in single 1-cm long nodal segments under a laminar flow
cabinet. Each explant was placed horizontally in 30-ml
polycarbonate UryTM vials (PBI, Milano, Italy) with 5 ml of
MS (Murashige and Skoog) [24] with 30 g L-1 of sucrose,
300 mg L-1 of reduced gluthatione (GSH) and 7 g L-1
agar. 2-(N-morpholino) ethanesulfonic acid (MES)
was added (500 mg L-1) to stabilize the pH, which was
adjusted to 5.8 before autoclaving at 121°C for 15 min.
MES concentration (2.3 mmol L-1) was much lower than
the levels that were found to be toxic to callus culture
(10 mmol L-1) [25] or to affect secondary metabolism
in Catharanthus roseus hairy roots (50 mmol L-1) [26].
During shoot initiation, 0.25 mg L-1 6-benzylaminopurine
(BA) was added to the basal medium and the explants
were placed in a climatic chamber at 25±1°C with a
16 h photoperiod and an irradiance of 70 mmol s-1 m-2
from cool fluorescent tubes. In the multiplication phase,
nodal explants were sub-cultured every four weeks
using PCCV25TM boxes (TQPL Co., New Milton, United
Kingdom) with 50 mL of solid growth medium (eight
explants per box). Root formation in nodal explants
(1 cm long, excised from shoots at the end of the
multiplication phase) was induced in PCCV25TM boxes
containing 100 mL of perlite soaked with 50 mL of half-
strength MS nutrient solution without growth regulators
(eight explants per box).
Finally, well rooted plantlets were acclimatized in
4-cm height LinfaboxTM boxes (Micropoli, Milano, Italy)
containing 200 mL of sterilized peat-perlite mixture (1:1,
v:v); five plantlets were transplanted in each box. The
containers were wrapped with plastic and incubated in a
growth chamber at 20±3°C at 16 h of photoperiod with
100 mmol s-1 m-2 PAR (from HPS lamps). Plastic covers
were partially opened after one week and completely
removed the subsequent week. Acclimatization
concluded after four weeks, when the surviving (ex vitro)
plants were transferred to a greenhouse.
We conducted two experiments using this
previously-described protocol. In the first experiment,
growth and tissue concentration of CADs and pigments
(anthocyanins and chlorophyll) were determined in
shoot explants of cv. Genovese at the end of each
micropropagation phase. In the second experiment, we
investigated the effect of BA concentration (0.1, 0.25,
0.5 and 1 mg L-1) on the rate of shoot proliferation and
the content of CADs and pigments in all genotypes at
the end of the multiplication phase.
947
 Rosmarinic acid content in basil plants grown in vitro and in hydroponics
2.3 Hydroponic culture
Seeds were germinated in rockwool tray plugs in a
growth chamber (25±1°C; 250 mmol s-1 m-2 PAR;
12 h photoperiod) and seedlings were transferred at
the second-leaf stage to a glasshouse under natural
temperature and light conditions. Two weeks after
sowing, the plants were transferred to 12 separate
hydroponic systems, each consisting of a polystyrene
plug tray floating in a plastic tank with stagnant nutrient
solution (300 L m-2), which was continuously aerated
(oxygen content >6.0 mg L-1). Crop density was 40
plants per tank, corresponding roughly to 160 plants m-2
(ground area). The nutrient solution contained the
following concentrations of macro- and micro-nutrients:
4.0 mol m-3 N-NO3, 1.0 mol m-3 N-NH4, 0.5 mol m-3
P-H2PO4, 2.5 mol m-3 K, 3.0 mol m-3 Ca, 1.0 mol m-3 Mg,
0.5 mol m-3 S-SO4, 40.0 mmol m-3 Fe, 40.0 mmol m-3 B,
5.7 mmol m-3 Cu, 17.6 mmol m-3 Zn, 10.0 mmol m-3 Mn,
1.0 mmol m-3 Mo. Electrical conductivity (EC) of nutrient
solution oscillated between 1.55-1.80 dS m-1 and pH was
maintained between 5.5 and 7.0 by frequent adjustment
with diluted sulphuric acid.
Climatic parameters were continuously monitored
by a weather station located inside the greenhouse.
The minimum and ventilation air temperature were 16
and 27°C, respectively; maximum temperature reached
up to 30–32°C during daytime in early summer. Daily
global radiation and mean air temperature averaged,
respectively, 12.0 MJ m-2 and 25.1°C.
Plants were sampled during the vegetative and
flowering stages (two and four weeks after transplanting,
respectively) for growth and chemical analysis. Plant
height and the fresh and dry weight of leaves, stems
and roots were measured in each sample consisting of
one individual plant. All the roots and the leaves (apart
from a few basal leaves, if senescent) were sampled
and a representative aliquot was extracted for the
determination of CADs and pigments.
2.4 Phytochemical analysis
Plant samples from in vitro or hydroponic cultures
were rapidly washed in tap water, rinsed in deionised
water and gently dried with a towel. A sub-sample of
approximately 0.5 g FW was frozen in liquid nitrogen
and stored at -80°C before laboratory analysis,
which was performed within a few weeks following
sampling.
The concentration of CADs in each sample was
determined as reported by Maggini et al. [13]. Briefly,
frozen tissue was ground in a mortar with 5 ml
extraction solvent (MeOH:H2O:HCl 70:29:1 v/v) and
sieved into plastic tubes. Samples were shaken for 4 h
on a magnetic stirrer and then centrifuged for 8 min at
948
5000 rpm. Following centrifugation, the supernatant was
collected in plastic tubes and stored overnight at -20°C.
The pellet was extracted again with 5 ml of solvent and
the two supernatants were combined for analysis. All
extracts were filtered with Chromafil® 0.45 μm cellulose
mixed esters membrane, 25 mm diameter syringe filters
(Macherey-Nagel, Düren, Germany) prior to HPLC
separations.
HPLC grade solvents and the following chemically-
pure standards were used: chlorogenic acid, caffeic
acid, ferulic acid, t-cinnamic acid, p-coumaric acid
(Sigma–Aldrich, Milano, Italy) and RA (Extrasynthese
S.A., Genay, France). HPLC analytical equipment
included a Jasco (Tokyo, Japan) PU-2089 four-solvent
low-pressure gradient pump and a UV-2077 UV/Vis
multichannel detector. Analyses were performed
using a Macherey–Nagel C18 250/4.6 Nucleosil®
100-5 column, at a flow rate of 1 ml min-1, equipped
with a guard column, using acetonitrile (solvent A) and
aqueous 0.1% phosphoric acid (solvent B) as elution
solvents. The gradient elution was programmed as
follows: 0.0-0.4 min, B 95%; 0.4-0.5 min, B 95-85%;
0.5-10 min, B 85-80%; 10-20 min, B 80-60%; 20-21 min,
B 60-5%; 21-25 min, B 5%; 25-26 min, B 5-95%;
26-30 min, B 95%. Detection was made at four
wavelengths: 325 nm, 280 nm, 300 nm and 350 nm.
The injection volume was 20 μL and the analyses
were performed at room temperature (23–29°C). The
substances of interest were identified by comparing
their retention times with those of reference standards
and quantified on the basis of the integrated peak area,
as compared with a standard curve.
The content of CADs was expressed per gram DW
on the basis of the dry matter content determined in an
aliquot of each sample after desiccation in a ventilated
oven at 85°C (to reach a given weight). The detection
limit of the analytical protocol was on the order of
0.05 mg g-1 DW.
Peak identification was accomplished by LC-MS and
LC-MS-MS using a PE Sciex API 365 triple quadrupole
mass spectrometer LC-MS System (Concord, ON,
Canada), equipped with a Turbospray source and
coupled to a 200 Series HPLC system with quaternary
pump and autosampler. The separations were carried
out with the same column used for HPLC analysis. In
order to improve the analytical sensitivity, phosphoric
acid was replaced by formic acid because the former
produces a strong suppression effect under electrospray
ionization conditions [27]; this modification did not
alter the elution pattern. The identification of selected
CADs was confirmed by comparison with authentic
standards showing the same retention times and
MS-MS fragmentation patterns.
C. Kiferle et al.
2.5 Pigments
Total content of chlorophyll and anthocyanins was
determined spectrophotometrically and expressed on a
DW basis.
Leaf samples (20 mg) were extracted with 2 mL
ethanol 95% (v/v) overnight at 4°C in the dark. The
extracts were centrifuged (3 min at 5000 rpm) and
subsequently analyzed at 666.2 nm and 654.6 nm.
Chlorophyll concentrations were calculated according
to Lichtenthaler’s [28]. For anthocyanin analysis,
methanol 80% (v/v) containing 1.2 M HCl was used
for extraction and the concentration of cyanidin-3-
glucoside equivalents was determined by measuring the
absorbance of the extracts at 535 nm.
2.6 Statistical analysis
The experimental design was completely randomized.
Data were subjected to a one-way or two-way analysis
of variance (ANOVA) and the means were separated
using the Tukey’s test. Each experiment was repeated
two to three times with similar results and those from a
representative run are reported in this paper.
3. Results
3.1 Quantification of CADs in basil tissues
Preliminary experiments were conducted in order to
validate the HPLC method adopted for this study and to
identify the major CADs contained in frozen, non-dried,
or oven-dried (70°C) roots and leaves of sweet basil
plants grown in hydroponic culture and/or in vitro.
Among the six CADs of interest, only RA was
found in considerable concentrations in all analyzed
samples, while the other metabolites were present
in trace quantities (e.g. ferulic and caffeic acid) or
below the detection limits (0.05 mg g-1 DW). The
typical chromatogram of methanolic extract of basil
sample showed two main peaks, one of which was
identified as RA by means of LC-MS analysis while the
other corresponded to a methylated derivative of this
substance (Figure 1). LC-MS-MS analysis suggested
that methylation was present on the caffeic acid moiety.
Frozen, non-dried samples of all types of tissues
contained invariably much more RA and the methylated
derivative compared to the dried ones (data not shown).
Therefore, in the experiments with both in vitro and
in vivo plants, only frozen, non-dried samples were
analyzed.
Only trace amounts of chlorogenic acid, caffeic acid,
ferulic acid, t-cinnamic acid and p-coumaric acid were
detected in the analyzed samples, while the methylated
form of RA was detected at significant concentrations in
all samples, although a quantitative determination of this
compound was not performed.
3.2 In vitro culture
In all tissue culture experiments, the plantlets
developed normally and exhibited well-formed shoots
without hyperhydric symptoms in each phase of
micropropagation (Figure 2). All ex vitro plants survived
and adapted well to greenhouse cultivation.
Total dry mass of in vitro plantlets (on average,
50.34 mg each) and the length of newly-formed
shoots (1.05 cm) were similar at the end of induction
and multiplication phase, whereas shoot number was
higher in the latter stage (Table 1). Dry mass and
shoot formation were lower during rooting than in any
other culture phase (Table 1). Acclimatized plantlets
exhibited the highest dry biomass of all stages sampled.
Rosmarinic acid was detected in in vitro plantlets during
all phases of micropropagation (Figure 3). The highest
RA content (43.21 mg g-1 DW) was found in acclimatized
plantlets, although the highest production of RA under
genuine in vitro conditions was detected in the explants
at the multiplication phase.
In the second experiment, which was carried out with
all three cultivars, we demonstrated that the proliferation
rate of explants could be increased by modulating the BA
concentration in the medium. No clear effects of the tested
BA concentrations (0.10, 0.25, 0.50 and 1.00 mg L-1)
were found on the number and the length of regenerated
Figure 1. Typical HPLC chromatogram of pure standard of rosmarinic
acid (RA, top) and HCl-methanolic extract of in vitro shoots
of sweet basil (O. basilicum L.) (bottom). The peak marked
with asterisk in the bottom chromatogram was identified
by LC-MS as a methylated form of RA.
949
 Rosmarinic acid content in basil plants grown in vitro and in hydroponics

Figure 2. Micropropagation of sweet basil (O. basilicum L., cv. Genovese): in vivo (hydroponic) cultivation of mother plants (A); shoot bud
induction on MS + 0.25 mg L-1BA (B); shoot proliferation on MS + 0.25 mg L-1BA (C); rooting of nodal segments on ½ MS (D);
acclimatized plants (E).

Phase Shoot number Shoot length (cm) Plantlet height(cm) Fresh weight (g) Dry weight (mg)

Induction 1.88b 0.96bc 1.80c 0.598a 50.68b

Multiplication 2.57a
1.14ab
2.52b
0.602a
50.00b

Rooting 1.09c 1.41a 1.65c 0.169b 11.50c

Acclimatization 1.20c 0.55c 3.80a 0.742a 82.77a

Table 1. Growth parameters of sweet basil (O. basilicum L., cv. Genovese) in vitro plantlets at the end of each micropropagation phase. Each
phase lasted four weeks. Mean values of 15 replicates, each consisting of one plantlet. One-way ANOVA was performed. Values followed
by different letter differs significantly (P≤0.001).

shoots, and on the total dry weight of individual plantlets
(data not shown), which averaged 2.14 shoot per
explants, 1.35 cm and 48.16 mg, respectively. However,
a strong genotype-dependent effect of BA was observed
on RA production. In green-leaf cultivars, the highest
RA contents were observed in the plantlets grown with
1.0 mg L-1 BA in the medium, while in red-leaf Dark Opal
the RA concentration decreased with increasing BA
level (Figure 4).
The content of anthocyanins was much lower in
the green-leaf cultivars (0.61 mg g-1 DW, on average)
than in Dark Opal (on average, 8.59 mg g-1 DW) and,
in the latter cultivar, increased with BA concentration
(Figure 5). The chlorophyll content was not significantly
affected by plant genotype nor BA level, averaging
9.92 μg g-1 DW.
3.3 Hydroponic culture
In the floating systems, plants grew vigorously and
flowered abundantly within one month after planting
(Figure 6). Fresh and/or dry matter (Table 2) accumulation
in shoots, leaves and roots of hydroponically-grown

950
C. Kiferle et al.
Figure 3. The content of rosmarinic acid (RA) in vitro plantlets of
sweet basil (O. basilicum L. cv. Genovese) sampled at
the end of each micropropagation phase. Each phase
lasted four weeks: Induction (I); Multiplication (M);
Rooting (R); Acclimatization (A). Mean values of four
replicates, each consisting one single plantlet. One way
ANOVA was performed. Values followed by different
letters differ significantly (P≤0.05).
Figure 4. The effect of BA concentrations in the growing medium
on the content of rosmarinic acid (RA) in different
cultivars (Genovese, Dark Opal and Superbo) of
sweet basil (O.  basilicum L.) in vitro plantlets at the
end of multiplication phase, which lasted four weeks.
Mean values of four replicates, each consisting of one
single plantlet. The effects of BA concentration (A) and
genotype, and their interaction (A x B), were significant
at P≤0.001 according to ANOVA. Values followed by
different letters differ significantly (P≤0.05).
plants was much higher at the flowering stage compared
to the vegetative stage. Results indicate that four to five
crops per year could be performed with an estimated
annual leaf fresh biomass production of 30 kg m-2. The
root to shoot DW ratio increased approximately two-
fold at the flowering stage in all the cultivars. At the
vegetative stage, there were no significant differences
among the cultivars in terms of plant height and total
dry mass. However, the red-leaf Dark Opal plants were
much smaller than the green-leaf genotypes (Genovese
and Superbo) when sampled at the flowering stage
(Table 2).
Figure 5. The effect of BA concentrations in the growing medium
on the content of anthocyanins (Cy-3-Glc equivalents,
measured spectrophotometrically) in the shoots of sweet
basil (O.  basilicum L. cv. Dark Opal) grown in vitro and
sampled at the end of the multiplication phase, which lasted
four weeks. Mean values of four replicates, each consisting
of one single plantlet. The effects of BA concentration (A)
and genotype, and their interaction (A x B), were significant
at P≤0.001 according to ANOVA. Values followed by
different letters differ significantly (P≤0.05).
RA content increased at the flowering stage in both
leaves and roots of Genovese and Superbo plants
compared to the vegetative stage, with levels reaching
28.88 mg g-1 DW (Figure 7). In contrast, no difference in
RA levels between developmental stages were observed
in Dark Opal plants, which on average contained much
less RA than the other genotypes (4.87 vs. 13.60 mg g-1
DW in the leaves and 9.29 vs. 21.23 mg g-1 DW in the
roots). A significant difference in RA level between
Genovese and Superbo was found only in the leaves of
blooming plants, which was considerably higher in the
latter cultivar (Figure 7). The level of RA was higher in
the roots (on average, 17.64 mg g-1 DW ) than in the
leaves (10.69 mg g-1 DW) regardless of the cultivar and
the physiological stage, with the exception of flowering
Superbo (Figure 7). There were no differences among
the cultivars in chlorophyll content, which averaged
16.65 μg g-1 DW. However, the leaves of Dark Opal
seedlings contained more anthocyanins than those of
other cultivars (on average, 13.15 mg g-1 DW compared
to 0.89 mg g-1 DW).
4. Discussion
4.1 CADs quantification
Many authors found several CADs in basil tissues
although, in many cases, RA was the most abundant
[4,5]. In contrast, in our samples only RA was present
at concentrations well above the detection limit of the
analytical method used in this study; all other CADs
were found in trace quantities regardless of growing
system and plant tissue. In addition, we observed a
951
 Rosmarinic acid content in basil plants grown in vitro and in hydroponics

Figure 6. Hydroponic cultivation (floating system) of different cultivars (Genovese, Dark Opal and Superbo) of sweet basil (O. basilicum L.):
plants at full bloom, which typically occurred four weeks after planting (A); the typical leaves of cv. Genovese (B), cv. Dark Opal (C),
cv. Superbo (D); the typical inflorescences of the cultivars with purple leaves (cv. Dark Opal; E) or green leaves (cv. Genovese and
Superbo; F).

Height Shoot FW Shoot DW Leaf FW Leaf DW Root DW

Cultivar
(cm) (g) (g) (g) (g) (g)

Vegetative stage (2 weeks after transplanting)

Genovese 25.80 c
13.037b 0.787c 8,407b 0.517b 0.058d

Dark Opal 22.00c 9.918b 0.611c 6,373b 0.379b 0.024d

Superbo 27.70 c
13.863 b
0.862 c
9,195 b
0.537 b
0.053d

Full bloom (4 weeks after transplanting)

Genovese 83.59 a
67.138a 7.516a 41,752a 3.835 a 1.047a

Dark Opal 53.65b 60.667a 4.128b 39,446a 3.260a 0.362b

Superbo 80.60 a
72.138 a
8.158 a
47,392 a
4.365 a
0.952a

Significance

Cultivar (A) *** * * NS NS ***

Developing stage (B) *** *** *** *** *** ***

AxB ** NS NS NS NS ***

Table 2. Growth parameters of different cultivars (Genovese, Dark Opal and Superbo) of sweet basil (O. basilicum L.) plants grown hydroponically
and sampled during vegetative stage and at full bloom. The shoots included inflorescences when sampled at flowering stage. Mean value
of 15 replicates, each consisting of one individual plant. Values followed by different letters differ significantly (P≤0.05). Two-way ANOVA
was performed (*P≤0.05; **P≤0.01; ***P≤0.001; NS=not significant).

952
C. Kiferle et al.
Figure 7. The content of rosmarinic acid (RA) in leaves
and roots of different cultivars (Genovese,
Dark Opal and Superbo) of sweet basil
(O. basilicum L.) plants grown hydroponically and
sampled during vegetative stage and at full bloom.
Vegetative and flowering plants were sampled two and
four weeks after transplanting, respectively. Mean values
of four replicates, each consisting of one single plant.
The effects of cultivar (A) and physiological stage (B),
and their interaction (A x B), were significant at P≤0.001
according to ANOVA. Values followed by different letters
differ significantly (P≤0.05).
peak in all samples that LC-MS analysis attributed to a
methylated form of RA. In many samples, the peak area
corresponding to this compound was even higher than
that of RA (data not shown).
Methyl rosmarinate (MeRA) has been detected
in several plants species, particularly within the
Lamiaceae [29,30]. However, to our knowledge no
paper has been published reporting the presence of
MeRA in basil tissues. In the present study, LC-MS-MS
analysis excluded the correspondence of the unknown
peak with MeRA because the methylation was present
on the caffeic acid moiety. Similar to our work, a later-
eluting RA methylated derivative was found in basil
cell suspensions by Strazzer et al. [31]. Since RA
methylation can occur both in vivo and during extraction
with methanol, Strazzer et al. [31] quantified RA in
samples by considering both the area of the RA peak
and that of its methylated form.
In vitro studies showed that the antioxidant capacity
of RA alkyl derivatives could be higher than RA [32],
and further work is in progress to more thoroughly
characterize the methylated form of RA and to verify
whether its production is a result of genuine biosynthesis
or artifactual methylation during methanol extraction.
The high levels of RA detected in our study
(approximately, from 4 to 60 mg g-1 DW), were within
those reported in the literature for Sweet basil tissues,
which range from less than 0.1 mg g-1 DW [17,20] to
nearly 100 mg g-1 DW [4]. This big variability is probably
a consequence of differences of plant genotype,
growing conditions as well as of the method used for
the quantification. The observed difference in RA levels
is likely a result of differences among plant genotypes,
growing conditions, as well as of the method used for
RA quantification.
In our preliminary work, RA content was much lower in
samples that had been desiccated at 70°C. Sample heating
was found to reduce the root content of echinacoside in
Echinacea angustifolia [33] and Echinacea pallida [34] due
to volatilization or thermal decomposition. Consequently,
only frozen, not dried, samples were processed and
analyzed in our reported experiments.
4.2 In vitro culture
In experiment 1, the proliferation aptitude of explants,
as indicated by the number and length of shoots at the
end of multiplication stage, was lower compared with
those found in similar studies with sweet basil [35,36].
Nevertheless, our protocol resulted in biomass production
on the order of 50 mg DW per plantlet (Table 1), which
was nearly 50-fold higher than that observed by Kintzios
et al. [17] in nodal segments cultured in a bioreactor.
The lowest shoot dry weight was found in rooted
plantlets (Table 1). Root formation may have occurred
at the expense of shoot development, as found in other
plant species grown in vitro [e.g. 37].
Ex vitro plantlets acclimatized successfully, with
considerable growth (1.56 mg per plantlet per day)
measured in the four weeks following the transfer out
of in vitro vessels, although newly-formed lateral shoots
were shorter compared to those developed in the in vitro
phases (Table 1). Ex vitro conditions in combination
with the absence of cytokinin application likely restored
apical dominance, enhancing plant elongation and
reducing the development of lateral shoots.
The highest RA content (43.21 mg g-1 DW)
was found at the end of acclimatization (Figure 3).
Increased RA accumulation at this stage was likely a
response to the stress induced by ex vitro conditions
[38] which stimulated secondary metabolism. It is well
known that the synthesis of CADs is enhanced by
both biotic and abiotic stress [39].The accumulation of
considerable amounts of biomass (Table 1) and the high
953
 Rosmarinic acid content in basil plants grown in vitro and in hydroponics
RA content (30.45 mg g-1 DW; Figure 3) indicates the
feasibility of in vitro shoot culture for the production of
RA in sweet basil.
In experiment 2, we did not observe any significant
effect of BA on shoot proliferation. In contrast, previous
studies testing BA concentrations similar to those used in
our study have demonstrated a strict correlation between
shoot proliferation and the level of BA in the medium for
O. basilicum [36,40]. The influence of cytokines on shoot
initiation in in vitro culture is affected by many factors,
including the growing protocol and plant genotype [41],
which may explain the differences between our findings
and those reported by other authors.
The hormone composition of the growth medium
may affect the synthesis of plant secondary
metabolites in vitro [14,42]. For instance, addition
of cytokines to the culture medium enhanced the
accumulation of alkamide and CADs in Echinacea
angustifolia [42]. In this work, we observed the most
significant differences in RA content among the three
cultivars at the highest BA concentration (Figure 4).
Previous research has also shown a large variability in
RA accumulation of different basil cultivars in culture
[43]. Increasing BA concentrations resulted in an
increase of RA content in green-leaf cultivars, which
coincides with previous findings (Figure 4). Rady
and Nazif [18] observed that RA content increased
in in vitro shoots of Ocimum americanum L. with BA
concentrations up to 1 mg L-1. In contrast to green-
leaf basil genotypes, RA content decreased in purple-
leaf Dark Opal at BA concentrations above 0.5 mg L-1.
Anthocyanin accumulation increased in response to
BA addition in Dark Opal plantlets (Figure 5), which
agrees with previous findings in other species such as
Catharanthus roseus, Celosia argentea and Cordyline
terminalis [44]. The accumulation of anthocyanins
was not affected by BA in green-leaf cultivars (data
not shown).
Regardless of genotype, the concentration of RA
in O. basilicum shoot tissues (40 to 60 mg g-1 DW)
exceeded the levels previously reported by Kintzios
et al. [16,17] and by Moschopoulou and Kintzios
[45], who used different in vitro systems (e.g. callus,
cell suspension and immobilized cells) and growth
regulators (e.g. NAA, 2,4D ). These findings suggest
that the pattern of RA accumulation in vitro is not
predictable. On the other hand, increased synthesis
of secondary metabolites is often associated with
cell or tissue differentiation, such as aggregation of
cell suspensions or the formation of more complex
structures such as adventitious roots or shoots [46],
which may explain the accumulation of high amounts
of RA in cultured shoots in our study.
954
4.3 Hydroponic culture
The floating system was found to be an efficient
approach for producing large amounts of plant material
that can be easily processed on an industrial scale
(Table 2) in agreement with previous findings [10,11].
One of the advantages of the hydroponic growth
system is that it facilitates harvesting of root tissue,
which accounted for 10% of total dry mass (Table 2) and
had significantly higher RA concentrations than shoot
tissues (Figure 7). Previous studies on members of
the Lamiaceae also reported higher RA content in the
roots compared to the leaves [19,47]. The antimicrobial
properties of RA may represent a constitutive defense
of the plants against microbial infection, playing an
important role in the root-pathogen interaction [19].
Shoot and root concentrations of RA ranged between
4 and 29 mg g-1 DW, and were similar to levels reported
in previous studies [4,43]. The highest levels of RA were
detected at full bloom in both leaves and roots (Figure 7).
Juliani et al. [5] reported that leaf concentration of RA
in basil increased during flowering with respect to the
vegetative stage, although to a lesser extent in Dark
Opal plants than in the green-leaf genotypes under
investigation. Similarly, the highest content and the best
composition of essential oils are generally found in basil
plants at full bloom [3]. In contrast, in other Lamiaceae,
such as rosemary [48], spearmint and peppermint [49],
leaf RA content was lower at the flowering stage than at
the vegetative stage.
Contrasting results are reported in the literature
regarding RA content in basil cultivars with different leaf
color. In our study, leaf and root concentrations of RA
in the purple-leaf cultivar Dark Opal were lower than in
the green-leaf cultivars (Figure 7), in agreement with
Javanmardi et al. [4]. Opposite results were reported by
Juliani et al. [5] and Nguyen and Niemeyer [43], who
found higher RA levels in Dark Opal plants than in other
cultivars with green leaves.
4.4 In vitro vs. hydroponic culture
In vitro grown shoots, especially those of the Dark
Opal cultivar, contained much higher levels of RA
(Figures 3, 4) than did the plants in hydroponic culture
(Figure 7). In experiments with various Ocimum
species, tissues or cells cultured in vitro accumulated
RA to levels that were 3- to 11-fold higher than in donor
plants [16,50]. Secondary metabolism may be altered
by artificial environmental conditions provided by in vitro
culture and by the composition of the growth medium
[14,51].
The lower content of both chlorophylls and
anthocyanins in basil tissues grown in vitro compared
to hydroponically-grown plants was likely a result of
C. Kiferle et al.
the presence of sugars in the medium as well as of
the reduced irradiance inside the cultivation vessels.
Exogenous supply of sucrose reduced chlorophyll
content in Solanum tuberosum [52]. Anthocyanins are
able to reduce photo-oxidative injury in leaves and their
concentration is generally reduced when the plants are
grown under low irradiance [53]. Moreover, the reduction
of gas exchange in cultivation vessels could lead to an
accumulation of ethylene and CO2, which stimulates the
degradation of photosynthetic pigments [54].
Observing the RA content of plants cultured both
in vivo and in vitro, it stands out a negative relation
between the accumulation of the metabolite and
anthocyanins. When in vivo and in vitro plants, green
and purple-leaf genotypes or the Dark Opal plantlets
cultured in vitro at different BA concentrations in the
basal medium are compared in terms of RA and
anthocyanins content, it emerges a negative relation
between the accumulation of these metabolites. A
common substrate of RA and anthocyanins pathways
is 4-coumaroyl-CoA [55] and a competition between
their individual biosynthesis could be hypothesized, as
also suggested by the higher RA content in green- vs.
red- varieties of Perilla frutescens [56]. However, in
cell suspensions of Dark Opal basil, Strazzer et al. [31]
References
[1] Makri O., Kintzios S, Ocimum sp. (Basil): botany,
cultivation, pharmaceutical properties, and
biotechnology, J. Herbs. Spices Med. Plants, 2007,
13, 123-150
[2] De Masi L., Silviero P., Esposito C., Castaldo D.,
Siano F., Laratta B., Assessment of agronomic,
chemical and genetic variability in common basil
(Ocimum basilicum L.), Eur. Food Res. Technol.,
2006, 223, 273-281
[3] Zheljazkov V.D., Cantrell C.L., Tekwani B., Khan
S.I., Content, Composition, and Bioactivity of
the Essential Oils of Three Basil Genotypes as
a Function of Harvesting, J. Agric. Food Chem.,
2008, 56, 380-385
[4] Javanmardi J., Khalighi A., Kashi A., Bais H.P.,
Vivanco J.M., Chemical characterization of basil
(Ocimum basilicum L.) found in local accessions
and used in traditional medicines in Iran, J. Agric.
Food Chem., 2002, 50, 5878-5883
[5] Juliani H.R., Koroch A.R., Simon J.E., Basil: A new
source of rosmarinic acid. In: Ho C.T., Simon J.E.,
Shahidi F., Shao Y., (Eds), Dietary Supplements,
American Chemical Society Symposium Series
987, American Chemical Society, Washington,
D.C. USA, 2008, 129-143
found a positive correlation between the content of RA
and anthocyanins.
In conclusion, the highest RA concentrations
were observed during the in vitro multiplication, in the
acclimatized plants and in the roots of hydroponically-
grown seedlings at full bloom. In vitro, BA reduced the
accumulation of RA in purple-leaf Dark Opal cultivar, but
increased RA accumulation in the green-leaf genotypes.
Our findings suggest RA production can be improved
by utilizing in vitro or hydroponic culture of sweet basil.
The differences in RA and anthocyanin content
observed in vivo and in vitro suggest a negative relationship
between the accumulation of these two compounds. Work
is in progress to elucidate the possible interaction between
the synthesis of these metabolites in basil tissues as well
as the nature and the origin (genuine in vivo biosynthesis
or artifact during the methanol extraction) of the methylated
form of RA detected in our samples.
Acknowledgements
This work was partially supported by 2008-PRIN project
titled “Plant cultures as source of endocrine interfering
metabolites” (scientific resp. A. Mensuali).
[6] Petersen M., Simmonds M.S.J., Molecules of
interest Rosmarinic acid, Phytochemistry, 2003,
62, 121-125
[7] Orcutt D.M., Nilsen E.T., The physiology of plants
under stress soil and biotic factors, John Wiley and
Sons Inc., New York , 2000
[8] Canter P.H., Thomas H., Ernst E., Bringing
medicinal plants into cultivation: opportunities and
challenges for biotechnology, Trends Biotech.,
2005, 23, 180-185
[9] Karuppusamy S., A review on trends in production
of secondary metabolites from higher plants by in
vitro tissue, organ and cell cultures, J. Med. Plants
Res., 2009, 3, 1222-1239
[10] Pardossi A., Malorgio F., Incrocci L., Tognoni F.,
Hydroponic technologies for greenhouse crops, In: Dris
R., (Ed.), Crops: Quality, Growth and Biotechnology,
WFL Publisher, Helsinki, 2006, 23, 360-378
[11] Miceli A., Moncada A., Vetrano F., D’Anna F.,
First results on yield and quality response of basil
(Ocimum basilicum L.) grown in a floating system,
Acta Hort., 2003, 609, 377-381
[12] Dorais M., Papadopoulos A.P., Luo X., Leonhart
S., Gosselin A., Pedneault K., et al., Soilless
greenhouse production of medicinal plants in North
955
 Rosmarinic acid content in basil plants grown in vitro and in hydroponics
Eastern Canada, Acta Hort. (ISHS), 2001, 554,
297-304
[13] Maggini R., Raffaelli A., Annaheim K.A., Tozzini
L., Pacifici S., Guidi L., et al., Effect of post-
harvest handling and extraction on the content of
echinacoside and cynarin in the root tissues of
Echinacea angustifolia DC., J. Food Agric. Environ.
2010, 8, 266-271
[14] Matkowski A., Plant in vitro for the production of
antioxidants – A review, Biotech. Adv., 2008, 26,
548-560
[15] Weathers P.S., Towler M.J., Xu J., Bench to batch:
advances in plant cell culture for producing useful
products., Appl. Microbiol. Biotechnol., 2010, 85,
1339-1351
[16] Kintzios S., Makri O., Panagiotopoulos E., Scapeti
M., In vitro rosmarinic acid accumulation in sweet
basil (Ocimum basilicum L.), Biotechnol. Lett.,
2003, 25, 405-408
[17] Kintzios S., Kollias H., Straitouris E., Makri
O., Scale-up micropropagation of sweet basil
(Ocimum basilicum L.) in an airlift bioreactor and
accumulation of rosmarinic acid, Biotechnol. Lett.,
2004, 26, 521-523
[18] Rady M.R., Nazif N.M., Rosmarinic acid content
and RAPD analysis of in vitro regenerated basil
(Ocimum basilicum L.) plants, Fitoterapia, 2005,
76, 525-533
[19] Bais H.P., Walker T.S., Schweizer H.P., Vivanco J.M.,
Root specific elicitation and antimicrobial activity
of rosmarinic acid in hairy root cultures of Ocimum
basilicum, Plant Physiol. Biochem., 2002, 40, 983-995
[20] Sgherri C., Cecconami S., Pinzino C., Navari-Izzo
F., Izzo R., Levels of antioxidants and nutraceuticals
in basil grown in hydroponics and soil, Food Chem.,
2010, 123, 416-422
[21] Ahloowalia B.S., Integration of technology from
lab to land, In: Low cost options for tissue culture
technology in developing countries, Proceedings
of a Technical Meeting, Joint FAO/IAEA Division of
Nuclear Techniques in Food and Agriculture, (26–30
August 2002, Vienna, Austria), IAEA, 2004, 87-89
[22] Montero J.I., Stanghellini C., Castilla N.,
Greenhouse technology for sustainable production
in mild winter climate areas: Trends and needs,
Acta Hort., 2009, 807, 33-44
[23] Chaterjee A., Shukla S., Mishra P., Rastogi A.,
Singh S.P., Prospects of in vitro production of
thebaine in opium poppy (Papaver somniferum L.),
Ind. Crops Prod., 2010, 32, 668-670
[24] Murashige T., Skoog F., A revised medium for rapid
growth and bioassays with tobacco tissue cultures,
Physiol. Plant., 1962, 15, 473-493
956
[25] Parfitt D.E., Almehdi A.A., Bloksberg L.N., Use of
organic buffers in plant tissue-culture systems, Sci.
Hortic., 1988, 36, 157-163
[26] Morgan J.A., Barney C.S., Penn A.H., Shanks J.V.,
Effects of buffered media upon growth and alkaloid
production of Catharanthus roseus hairy roots,
Appl. Microbiol. Biotechnol., 2000, 53, 262-265
[27] Wolfender J.L., HPLC in natural product
analysis: The detection issue, Planta Med.,
2009, 75, 719-734
[28] Lichtenthaler H.K., Chlorophylls and carotenoids:
pigments of photosynthetic membranes,
Meth. Enzymol., 1987, 148, 350-382
[29] Jiang R.W., Lau K.M., Hon P.M., Mak T.C.W., Woo
K.S., Fung K.P., Chemistry and Biological Activities
of Caffeic Acid Derivatives from Salvia miltiorrhiza.,
Curr. Med. Chem., 2005, 12, 237-246
[30] Fecka I., Turek S., Determination of Water-
Soluble Polyphenolic Compounds in Commercial
Herbal Teas from Lamiaceae: Peppermint,
Melissa, and Sage, J. Agric. Food Chem., 2007,
55, 10908-10917
[31] Strazzer P., Guzzo F., Levi M., Correlated
accumulation of anthocyanins and rosmarinic acid
in mechanically stressed red cell suspension of
basil (Ocimum basilicum), J. Plant Physiol., 2011,
168, 288-293
[32] Laguerre M.L., Lopez Giraldo L.J., Lecomte
J., Figueroa-Espinoza M.C., Barea B., Weiss
J., et al., Relationship between hydrophobicity
and antioxidant ability of “phenolipids” in
emulsion: a parabolic effect of the chain length
of rosmarinate esters, J. Agric. Food Chem.,
2010, 58, 2869-2876
[33] Kabganian R., Carrier D.J., Soskansanj S., Drying
of Echinacea angustifolia roots, J. Herbs Spices
Med. Plants, 2003, 10, 11-18
[34] Li T.S.C., Wardle D.A., Effects of root drying
temperature and moisture content on the levels
of active ingredients in Echinacea roots, J. Herbs
Spices Med. Plants, 2001, 8, 15-22
[35] Sahoo Y., Patfnaik S.K., Chand P.K., In vitro
propagation of an aromatic medicinal herb
Ocimum basilicum L. (Sweet basil) by axillary
shoot proliferation, In Vitro Cell. Dev. Biol. Plant.,
1997, 33, 293-296
[36] Begum F., Amin M.N., Azad M.A.K., In vitro rapid
clonal propagation of Ocimum basilicum L., Plant
Tissue Cult., 2002, 12, 27-35
[37] Lucchesini M., Mensuali-Sodi A., Influence of
medium composition and vessel ventilation on in
vitro propagation of Phillyrea latifoglia L., Sci. Hort.,
2004, 100, 117-125
 C. Kiferle et al.
[38] Hazarika B.N., Morpho-physiological disorders in in
vitro culture of plants, Sci. Hortic., 2006, 108, 105-120
[39] Dae K.Y., Kim H.J., Yoon S.H., Induction of
Phenolics and Terpenoids in Edible Plants Using
Plant Stress Responses, In: Hou C.T., Shaw J.F.,
(Eds.), Biocatalysis and Agricultural Biotechnology,
CRC Press, 2009, 249-258
[40] Siddique I., Anis M., An improved plant
regeneration system and ex vitro acclimatization
of Ocimum basilicum L., Acta Physiol. Plant.,
2008, 30, 493-499
[41] Van Staden J., Zazimalova E., George E.F., Plant
growth regulators II: Cytokinins, their analogues
and antagonists, In: George E.F., Hall M.A.,
De Klerk G.J., (Eds.), Plant Propagation by
Tissue Culture, 3rd Ed., Springer, Dordrecht, The
Netherlands, 2008, 205-226
[42] Lucchesini M., Bertoli A., Mensuali-Sodi A., Pistelli
L., Establishment of in vitro tissue cultures from
Echinacea angustifolia D.C. adult plants for the
production of phytochemical compounds, Sci.
Hort., 2009, 122, 484-490
[43] Nguyen P., Niemeyer E., Effects of nitrogen
fertilization on the phenolic composition and
antioxidant properties of basil (Ocimum basilicum
L.), J. Agric. Food Chem., 2008, 56, 8685-8691
[44] Taha H.S., Abd El-Rahman R.A., Fathalla
Abd-El-Kareem M., Aly U.E., Successful
application for enhancement and production
of anthocyantn pigment from calli cultures of
some ornamental plants, Aust. J. Basic Appl.
Sci., 2008, 2, 1148-1156
[45] Moschopoulou G., Kintzios S., Achievement of
thousand-fold accumulation of rosmarinic acid
in immobilized cells of sweet basil (Ocimum
basilicum L.) by ten-fold increase of the volume of
the immobilization matrix, J. Biol. Res., 2011, 15,
59-65
[46] Collin H.A., Secondary product formation in plant
tissue cultures, Plant Growth Regul., 2001, 34, 19-134
[47] Karam N.S., Jawad F.M., Arikat N.A., Shibli R.A.,
Growth and rosmarinic acid accumulation in callus,
cell suspension, and root cultures of wild Salvia
View publication stats
fruticosa, Plant Cell, Tissue and Organ Cult., 2003,
73, 117-121
[48] Papageorgiou V., Mallouchos A., Komaitis M.,
Investigation of the Antioxidant Behavior of Air- and
Freeze-Dried Aromatic Plant Materials in Relation
to Their Phenolic Content and Vegetative Cycle, J.
Agric. Food Chem., 2008, 56, 5743-5752
[49] Fletcher R.S., Slimmon T., Kott L.S., Environmental
factors affecting the accumulation of rosmarinic acid
in Spearmint (Mentha spicata L.) and Peppermint
(Mentha piperita L.), The Open Agric. J., 2010, 4,
10-16
[50] Hakkim F.L., Shankar C.G., Girua S., Chemical
composition and antioxidant property of holy
basil (Ocimum sanctum L.) leaves, stems, and
inflorescence and their in vitro callus cultures., J.
Agric. Food Chem., 2007, 55, 9109-9117
[51] Van Staden J., Fennell C.W., Taylor N.J., Plant
stress in vitro: the role of phytohormones, Acta
Hort., 2004, 725, 55-61
[52] Kovač M., Ravnikar M., Sucrose and jasmonic acid
interact in photosynthetic pigment metabolism and
development of potato (Solanum tuberosum L. cv.
Sante) grown in vitro, Plant Growth Regul., 1998,
24, 101-107
[53] Neill S.O., Gould K.S., Anthocyanins in leaves:
light attenuators or antioxidants, Functional Plant
Biol., 2003, 30, 865-873
[54] Lucchesini M., Monteforti G., Mensuali-Sodi A.,
Serra G., Leaf ultrastructure, photosynthetic rate
and growth of myrtle plantlets under different in
vitro culture conditions, Biol. Plant., 2006, 50,
161-168
[55] Petersen M., Hans J., Matern U., Biosynthesis of
Phenylpropanoids and Related Compounds, In:
Wink M., (Ed.), Annual Plant Reviews: Biochemistry
of Plant Secondary Metabolism, Vol. 40, 2nd Ed. ,
Wiley-Blackwell, Oxford, UK, 2010, 182-230
[56] Yamazaki M., Nakajima J., Yamanashi M.,
Sugiyama M., Makita Y., Springob K., et al.,
Metabolomics and differential gene expression
in anthocyanins chemo-varietal forms of Perilla
frutescens, Phytochemistry, 2003, 62, 987-995
957