Phyochemrsrry, Vol. 49, No. 3, pp. 697-702, 1998
Elsev~~ Science Ltd. All rnghts reserved
Printed in Great Britain
003l-9422/98/~see front matter
EFFECTS OF CINNAMIC ACID ON POLYPHENOL PRODUCTION
FRANCOISE FONS,~ ANNICK GARGADENNEC,? ALAIN GUEIFFIER,~ JEAN LOUIS ROUSSEL~ and
CLAUDE ANDARY* I’
TLaboratoire de Botanique, Phytochimie et Mycologic, Facultt de Pharmacie, 15 Avenue C. Flahault 34060
Montpellier, Cedex 2 France and SLaboratoire de Chimie Therapeutique, Facultt de Pharmacie, 31 Avenue Monge 37200
Tours, Cedex France
in revised form 4 March 1998)
Key Word Index-Plantago
culture; roots; caffeic acid
glycoside esters; plantamoside; verbascoside; (E)-cinnamic acid.
contains two main caffeic acid glycoside esters, plantamoside and
verbascoside. These two polyphenols were investigated in the aerial and underground parts of
ribworts. For the first time, it is reported that, whatever the age of this plant, plantamoside and verbascoside
are concentrated in the roots with plantamoside levels double those of verbascoside. When
transferred into a medium containing lop3 M (E)-cinnamic acid, this chemical stress induced a slow degener-
ation of the initial roots. These were superseded by neoroots whose morphology was atypical during the first
eight days following their appearance. In the initial roots, (E)-cinnamic acid induced a temporary appearance
of two cinnamic acid derivatives (NCD), but did not change the plantamoside and verbascoside levels. In the
neoroots, high NCD levels were detected for only eight days. After the large decrease of these NCD, plan-
tamoside and verbascoside appeared and increased. These NCDs have been identified as glucoside esters of
ferulic and p-coumaric acids. These two compounds, which are absent from the traditional chemical profile of
ribwort, probably arose from a (E)-cinnamic acid detoxification pathway. 0 1998 Elsevier Science Ltd. All
In folk medicine, the aerial parts of
are used as an anti-inflammatory, anti-
bacterial, healing, diuretic and anti-asthmatic remedy
without toxicity [l-5]. For this reason, many authors
have studied the chemical composition of
leaves and isolated several active components
that could explain the numerous folk uses of this plant.
contains iridoids (catalpol, aucubin,
asperuloside) with laxative and diuretic activities [&
lo] and flavonoids (apigenin 7-O-glucoside, scu-
tellarein) that are anti-inflammatory [9-l 11. Caffeic
acid glycoside esters (CGEs), i.e. plantamoside
(=plantamajoside) and verbascoside, as well as the
minor lavandulifolioside, and isoverbascoside
14, 151, antifungal [16,17], antiviral
 and antioxidant [17, 19-211 activities and are
selective inhibitors of aldose reductase [17, 221, 5-
lipoxygenase  and protein kinase C [23, 241.
culture has never been used to modify
*Author to whom correspondence should be addressed
the CGE profile of
Therefore, in this
study, the time course of verbascoside and plan-
tamoside content was observed in the leaves and roots
In order to obtain
new pharmacologically active compounds, this plant
was cultured in a medium containing (E)-cinnamic
acid which has been shown to be integrated into the
polyphenol pathway [25, 261. Plantamoside and ver-
bascoside contents were then investigated. The tem-
porary changes in the
are reported and discussed below.
plantamoside and verbascoside contents
in the normal plant
Dried aerial and underground parts of
were investigated for caffeic acid derivatives using
TLC, every eight days from day 21 to day 70. Quan-
titative evaluation of plantamoside (P) and ver-
bascoside (V) was carried out using HPLC.
Whatever the age of the plants, the quantities of P
in the leaves were negligible (Fig. I), never exceeding
FONS et 01.
Fig. 1. Time course of P and V contents
in the roots and leaves of
4.04mgg-’ dry weight. The levels of V were higher
than those of P and oscillated around an average value
of 8.71 mgg-’ dry weight from day 21 to day 70. P
content in the roots (Fig. 1) was always higher than V
content and the two curves fluctuated in the same way,
although the variations of the first component were
more marked. P and V levels reached a first maximum
around day 30 (43.9f2.5 and 32.3f2.2mggp’ dry
weight, respectively) and a second at day 56
(57.0 f 25.3 and 25.4 f 1.5 mg gg
dry weight, respec-
tively). In general, levels oscillated around average
values of 42.9 and 22.6mgg-’ dry weight, respec-
tively. Cumulated P and V content was always low in
the leaves (average value, 9.74mgg-’ dry weight). On
the other hand, cumulated P and V content in the
roots was age-dependent and much higher than in the
leaves (30.9 f 2.4 to 82.5 f 26.7 mg gg
dry weight on
day 42 and day 70 respectively, average value
58.0mgg-’ dry weight).
From day 21 until the end of the experiment (day
71) the average ratio P (roots)/P (leaves + roots) was
close to 0.93 (Fig. 2) which confirmed that P was
heavily stocked in the roots and that only very small
amounts are found in the aerial parts. The average
ratio V (roots)/V (leaves + roots) was 0.73 and showed
that the level of V in the roots was always higher than
in the leaves, although the values were more dispersed
than the P values, oscillating between 0.55 and 0.90
from day 2 1 to day 7 1. All these results indicate that
P is the major CGE in the roots of ribwort and they
highlight the importance of P and V storage in the
which has not previously been
Fig. 2. Ratio: CGEs (roots)/CGEs (leaves + roots).
Effects of cinnamic acid on polyphknol production in
reported. For this reason, only the roots containing
much higher levels of CGEs than the leaves were ana-
lysed during the experiment with (E)-cinnamic acid.
Time course of P and V contents in the roots of
fed with (E)-cinnamic acid.
for 75 days following ger-
mination were transferred into MS culture medium
containing (E)-cinnamic acid. P and V levels in the
roots of these plants were then evaluated over 41 days
and compared with the control. (E)-cinnamic acid was
added at the minimum toxic concentration for the
plants (lO_jM on day 0). This concentration was
determined during a preliminary toxicity study on
whole plants (F. Fons, unpublished results). It was
chosen because it did not prevent the plant from living
and growing throughout the experiment, but induced
morphological and chemical profile modifications in
the roots. Although this acid was only ever detected
at negligible concentrations in these plants, their roots
gradually degenerated, darkened and, 41 days later,
withered completely. At the same time, new roots
(neoroots) appeared at day 8 and gradually super-
seded the initial ones. From day 8 to day 16, the
neoroots showed an atypical morphology while being
anisodiametric and thicker than the usual roots.
Thereafter, the neoroots became morphologically nor-
mal (isodiametic and thin). Hence the interest in com-
paring their chemical profiles with the control
throughout this period.
P and V contents were 63.6+ 13.3 and
29.3 f 12.5 mgg-’ dry weight respectively in the con-
trol roots on day 0 (Fig. 3). Both levels decreased
slightly after the plants were transferred, but no major
differences were noted between the contents of
stressed and control plants. Our results showed that
(E)-cinnamic acid was not integrated into the P and
It was also of considerable interest to study the P
and V contents in the neoroots, which appeared at
day 8 and which constituted almost the total mass of
the roots on day 41 in plants fed with (E)-cinnamic
acid. These neoroots were analysed from day 13
because, until this day, their biomass was too low. P
and V were absent in the neoroots from day 13 to day
19 and then increased up to 14.2& 5.2 and
20.5f8.3 mgg-’ dry weight respectively on day 41
The most striking chemical modifications detected
during the course of this experiment involved the tem-
porary appearance of two cinnamic acid derivatives
detected by TLC and HPLC analysis. Their con-
centrations from day 0 to day 41 were estimated from
peak areas as a fraction of their maximum con-
centrations, attained on day 11 (Fig. 4). These NCD,
which appeared on day 2 in the initial roots of
fed with (E)-cinnamic acid, were com-
pletely absent from the roots of the control plants.
The cumulated ratio of NCD increased in the initial
roots during the first week of the experiment
(0.27+0.00 on day 2, 0.87kO.17 on day 6) to reach
high concentrations throughout the second week
(highest value, 1.00+0.19 on day 11) and then
decreased from the third week to the end of the experi-
ment (0.13 f0.10 on day 41). These NCD were also
detected in neoroots at high levels from day 13
(0.75kO.53 and 0.95f0.59 on day 13 and 16 respec-
tively) but levels fell suddenly on day 19 (0.13 + 0.07).
From day 19 to the end of the experiment, low levels
were detected (0.04 on day 41). It is very interesting
to note that from day 19, when NCD levels in the
roots fell dramatically, P and V appeared and attained
the levels found in the controls. At the same time, the
morphology of these roots became normal.
The NCD were isolated and purified in order to
determine their structure. Total acid hydrolysis of
both components released a sugar which was ident-
ified as glucose by TLC and ClinistixN test (Ames).
Alkaline hydrolysis of both esters produced two mol-
ecules identified, using TLC and HPLC, as (E)-p-
coumaric acid and (E)-ferulic acid. Traces of the two
(Z)-isomers of each compound were also detected in
the crude extracts of plant roots using HPLC after
UV-exposure. The amounts of these two NCDs were
too small to allow full structural determination. The
(E)-cinnamic acid introduced into the culture medium
at 10m3 M can be considered as con-
stituting a toxic chemical stress for this plant. This
acid induced a progressive withering of the initial
roots, which were superseded by neoroots. These neo-
roots showed disturbed morphology and chemical
profile for eight days (no CGEs and high levels of
NCD). We can suppose that both cinnamic derivatives
(NCD), appearing temporarily during the experiment,
arose from a short detoxication pathway which was
the most efficient way to protect
high levels of cinnamic acid. Following this, the dis-
appearance of these detoxication molecules allowed
the normal chemical and morphological root profiles
to be re-established.
Plant material and growth conditions.
L. (a gift from Dr. S. Puech,
Institut de Botanique, Montpellier, France) were ster-
ilized in 70% EtOH for 30sec, then in a commercial
solution of NaClO (3.6%) for 20min. The seeds were
then rinsed ( x 3) in sterile HZ0 and cultured in Mur-
ashige and Skoog’s (MS) medium  with 1Ogll’
agar under alternating 12hr light/dark. During the
growth experiments, the seedlings were grown without
transfer to fresh medium, in order to avoid stress-
induced variations in CGE content and were analysed
from day 21 to day 70. For the experiment with cin-
10 15 20 2s 30 35 40
Fig. 3. Time course of root CGEs contents during the experiment with trans-cinnamic acid, (a) CGEs contents in the roots
of control plants, (b) CGEs contents in the roots of stressed plants.
acid, after a first transfer into MS medium at CGEs extraction was carried out under the same
day 35, 75-day-old plants were subcultured in MS
medium containing 10-3M (E)-cinnamic acid dis- conditions for both experiments. 50 seedlings were
solved in 0.1% DMSO and the experiment run for a used for each extraction. Each experiment was carried
further 41 days. out in triplicate and the average value calculated.
Error values were not integrated into graphs to sim-
Effects of cinnamic acid on polyphtnol production in
0 I 0 2 4 6 8
11 13 16 19 23 28 31 34 41
Fig. 4. Time course of NCD ratio in both initial roots and neoroots during the experiment with (E)-cinnamic acid.
plify their appearance. Dried tissues (roots, leaves and
further neoroots) were separately weighed, ground in
a mortar, extracted at room temperature in 70%
MeOH (3 x 50ml/g dry weight) with magnetic agi-
tation (15 min), sonication (15 min) and then filtered.
The combined extracts were coned. to dryness under
vacuum and then analysed using TLC and HPLC.
The crude extracts were taken up into MeOH
(1 ml/lOOmg of dry organ) and submitted to TLC
on cellulose plates using: EtOAc-MeOH-H,O-C,H,,
(10:2:1.5:1) or 2% HOAc. CGEs were detected by UV
fluorescence after spraying with 1% 2-aminoethyl-
phenylborinate in MeOH. The MeOH extracts of
roots, leaves and neoroots used for TLC were diluted
10 times in 70% MeOH, filtered with a Millipore filter
(0.45 pm) and injected (20 pl) into an HPLC system
comprising a Kromasil RP-18 column (250 x 4.6 mm,
5 pm); mobile phase: MeCN-H,PO, (1:4) (pH 2.6);
flow rate 1 mlmin-‘. Quantitative analysis was per-
formed at 335nm and compared with plantamoside
(Rt:6.6 min) and verbascoside (Rt:8.8 min) standard
solutions isolated from
Extraction and iden@cation of NCD
Dried roots of
fed with cinnamic acid
were extracted twice with 70% MeOH for 90min at
room temperature. The crude extract was coned under
vacuum and subjected to CC on polyamide with 30%
EtOH as eluent. A second purification was carried out
on a cellulose column with Hz0 to yield two NCDs.
Total acid hydrolysis of NCD: each NCD was dis-
solved in H20, 3M HCl was added (1: 1) and the mix-
ture heated to 100” maintained for 5 hr. The cooled
solution was extracted 3 times with an equal volume
of EtOAc. The aq. layer was evaporated to dryness
and the residue taken up in 50% MeOH. The solution
was analysed using Clinistix (Ames) or precoated sil-
ica gel TLC plates, with two solvent systems: EtOAc-
H,O-MeOH-HOAc (13:3:3:5) and EtOAc-iso-PrOH-
H,O (2:7:1). Developed plates were sprayed with
naphthoresorcine-H,P0,, anisaldehyde-HOAc or tri-
phenyltetrazolium reagent and heated to 100”. D-glu-
cose was identified by direct comparison with a stan-
dard. Alkaline hydrolysis was carried out on an aq.
solution of NCD and 1 M NaOH (1: 1) heated for
45min at 50”. The solution was then acidified with
Dowex resin SOW, filtered and then analysed using
TLC (cellulose plates developed with 2% HOAc using
1% 2-aminoethyldiphenylborinate in MeOH or
diazotized p-nitroaniline acid as reagents). The same
solution was analysed using HPLC (conditions were
the same as those used for CGE analysis, see above)
and compared with standards of p-coumaric acid
(Rt: 10.3 min) and ferulic acid (Rt: 12.0 min).
are very grateful to Dr S.
Rapior and Dr M. P. Bonzom for their advice during
this work. We would like to thank Mr N. Wynn for
1. Rombi, M.,
Cent plantes mtdicinales.
Van Hellemont, J.,
Compendium de Phyto-
Association Pharmaceutique Belge,
Prtcis de Phytotherapie.
Maloine, Paris, 1992.
Guide de traitement par les plantes m&d-
Heures de France, Paris, 1976.
Long, C., Moulis, C., Stanislas, E. and Fourastt,
Journal de Pharmacie de Belgique, 1995, 50,
El-Naggar, L. J. and Beal, J. L.,
Products, 1980, 43, 649.
Handjieva, N. and Saadi, H.,
Zeitschrift fur Nat-
urforschung, 1991,46, 963.
Harborne, J. B. and Baxter, H.,
Taylor and Francis, London, 1995.
Harborne, J. B. Baxter, H. and Moss, G. P.,
tionary of Plant Toxins.
J. Wiley, Chichester,
Haznagy, A., Toth, G. and Bula, E.,
Murai, M., Tamayama, Y. and Nishibe, S.,
Medica, 1995,61, 479.
Andary, C., Motte-Florae, M. E., Gargadennec,
A., Wylde, R. and Heitz, A.,
et Phytotherapie, 1988,22, 17.
Debrauwer, L., Maillard, C., Babadjamian, A.,
Vidal-Ollivier, E., Laget, M., Salmona, G. and
Pharmaceutics Acta Helvetica,
Ravn, H. and Brimer, L.,
Andary, C., Roussel, J. L., Motte, M. E., Rascal,
J. P. and Privat, G.,
1981, 2, 119.
Andary, C., In
Scalbert. INRA, Paris, 1993, p. 237.
Cometa, F., Tomassini, L., Nicoletti, M. and
Fitoterapia, 1993, 64, 195.
Wang, P., Kang, J., Zheng, R., Yang, Z., Lu, J.,
Gao, J. and Jia, Z.,
Li, J., Wang, P. F., Zheng, R. L., Liu, Z. M. and
Jia, Z. J.,
Planta Medica, 1993, 59, 3 15.
Born, M., Carrupt, P. A., Zini, R., Bree, F., Til-
lement, J. P., Hostettmann, K. and Testa, B.,
vetica Chimica Acta, 1996, 79,
Andary, C., In
Proceedings 13th International
Conference of the Groupe Polyphenols.
her, 1986, p. 15.
Herbert, J. M., Maffrand, J. P., Taoubi, K., Aug-
ereau, J. M., Fouraste, I. and Gleye, J.,
of Natural Products, 1991, 54,
Saracoglu, I., Inoue, M., Calis, I. and Ogihara,
Biological Pharmaceutical Bulletin, 1995, 18,
Ushiyama, M., Kumagai, S. and Furuya, T.,
Phytochemistry, 1989,28, 3335.
Edwards, R., Mavandad, M. and Dixon, R. A.,
Phytochemistry, 1990, 29,
Murashige, T. and Skoog, F.,