Sedative and sleep-enhancing properties of linarin, a flavonoid-isolated from Valeriana officinalis

Abstract

We have recently reported the presence of the anxiolytic flavone 6-methylapigenin (MA) and of the sedative and sleep-enhancing flavanone glycoside 2S (-) hesperidin (HN) in Valeriana officinalis and Valeriana wallichii. MA, in turn, was able to potentiate the sleep-inducing properties of HN. The present paper reports the identification in V. officinalis of the flavone glycoside linarin (LN) and the discovery that it has, like HN, sedative and sleep-enhancing properties that are potentiated by simultaneous administration of valerenic acid (VA). These effects should be taken into account when considering the pharmacological actions of valeriana extracts.

Full text

Sedative and sleep-enhancing properties of linarin,
a flavonoid-isolated from Valeriana officinalis
Sebastia
´n Ferna
´ndez, Cristina Wasowski, Alejandro C. Paladini, Mariel Marder *
Instituto de Quı
´mica y Fisicoquı
´mica Biolo
´gicas, Facultad de Farmacia y Bioquı
´mica, Junı
´n 956 (1113), Buenos Aires, Argentina
Received 25 August 2003; received in revised form 25 November 2003; accepted 4 December 2003
Abstract
We have recently reported the presence of the anxiolytic flavone 6-methylapigenin (MA) and of the sedative and sleep-enhancing
flavanone glycoside 2S ( ) hesperidin (HN) in Valeriana officinalis and Valeriana wallichii. MA, in turn, was able to potentiate the sleep-
inducing properties of HN.
The present paper reports the identification in V. officinalis of the flavone glycoside linarin (LN) and the discovery that it has, like HN,
sedative and sleep-enhancing properties that are potentiated by simultaneous administration of valerenic acid (VA).
These effects should be taken into account when considering the pharmacological actions of valeriana extracts.
D2003 Elsevier Inc. All rights reserved.
Keywords: Valeriana; Flavonoids; Linarin; Valerenic acid; CNS; Sedation; Sleep-enhancing properties; Potentiation
1. Introduction
The genus Valeriana belongs to the Valerianaceae and
contains about 250 species. The three most important
species in herbal medicine that are used as mild sedatives
are Valeriana officinalis L. s.l.,Valeriana wallichii DC., and
Valeriana edulis Nutt.ExTorr.andGrayssp.procera
(H.B.K.) F.G. Meyer.
The use of extracts of the valeriana roots and rhizomes to
cause sedation and relieve sleep problems dates back to the
18th century (Madaus, 1976), but the exact composition of
the preparations used was often not clear.
In the search for the active substances of valeriana, many
compounds have been isolated and identified during the last
120 years, but it is as yet uncertain which of them are
responsible for the recorded actions (Bos et al., 1996;
Houghton, 1999).
The most popular compounds, in this connection, are the
epoxy iridoids named valepotriates, their decomposition
products, the baldrinals, and the nonvolatile terpenoids
grouped as valerenic acid (VA) derivatives as well as some
other members of the essential oil (Fig. 1) (Houghton,
1999).
However, several facts have cast doubts on the rele-
vance of these compounds to explain valeriana extracts
effects. The principal of them are as follows: (a) the central
depressant action of valepotriates, valeranone, and of the
essential oil of valeriana could not be demonstrated by a
reduction of the glucose turnover in rat brain (Ho¨lzl,
1997); (b) the sedative potency of these compounds is
rather low (>30 mg/kg, in mice) (Ho¨ lzl, 1997); (c) the
valepotriates rapidly decompose if water is present and the
resulting baldrinals are chemically reactive and may form
polymers (Bos et al., 1996), hence both valepotriates and
baldrinals disappear rapidly from the extracts; and (d) the
roots and rhizomes of different valeriana species show
large differences with regards to their constituents: V.
officinalis mainly contains VA and derivatives as well as
valepotriates and other constituents, whereas V. wallichii
and V. edulis do not contain VA and its derivatives
(Houghton, 1999).
In vitro mutagenic effects have been described for the
valepotriates and their decomposition products (Houghton,
1999). Hence, it would be prudent to prefer valeriana
preparations, which are devoid of potentially hazardous
valepotriates or baldrinals.
0091-3057/$ – see front matter D2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.pbb.2003.12.003
* Corresponding author. Tel.: +54-11-4962-5506; fax: +54-11-4962-
5457.
E-mail address: mmarder@qb.ffyb.uba.ar (M. Marder).
www.elsevier.com/locate/pharmbiochembeh
Pharmacology, Biochemistry and Behavior 77 (2004) 399 – 404

A promising approach to detect new active substances in
valeriana extracts consists in searching for the presence of
ligands for the principal brain receptors predominantly
associated to anxiolytic, sedative, and/or sleep-enhancing
properties (Marder and Paladini, 2002). Unfortunately, stud-
ies of this kind have, in general, given inconclusive results
quite probably due to the use of crude extracts in the assays
(Ho¨lzl and Godau, 1989; Mennini et al., 1993).Bodesheim
and Ho¨lzl (1997) found that the lignan (+) hydroxypinor-
esinol (Fig. 1), present in valeriana extracts, is a medium-
low affinity ligand for the 5-HT receptor but its in vivo
effects were not investigated.
In our laboratory, we have applied the ‘‘ligand-searching
approach’’ using, as far as possible, purified extracts and were
able to report the presence of 6-methylapigenin (MA) (Fig. 1)
in V. wallichii and V. officinalis and to prove that it is a
benzodiazepine binding site (BDZ-bs) ligand (Wasowski et
al., 2002). We have also made the first report of the presence
of 2S ( ) hesperidin (HN) (Fig. 1) in V. wallichii and in V.
officinalis and found that it has sedative and sleep-enhancing
properties. MA, in turn, had anxiolytic activity and was able
to potentiate the sleep-enhancing properties of HN (Marder et
al., 2003).
The present paper describes the first identification of the
flavonoid glycoside linarin (LN) (Fig. 1) in V. officinalis, the
discovery of its sedative and sleep-enhancing properties in
mice, and the potentiation of these effects by simultaneous
administration with VA. The effective doses of these com-
pounds are commeasurable with their concentrations in the
plant extracts and with the doses used in folkloric medicine.
2. Methods
2.1. Plant material
V. officinalis L. (Valerianaceae) was obtained from a local
commercial source. Its identification was done at the Botany
Museum of the School of Pharmacy in Buenos Aires, where
the voucher specimen 10392 BAF was deposited.
2.2. General procedures
Spectroscopic measurements were done as follows:
NMR on a 300-MHz Bruker apparatus with the sample
dissolved in DMSO-d
6
; UV-Vis in a Shimadzu 160A
spectrophotometer with methanolic solutions; and EIMS
on a Shimadzu Mass-Spectrometer QP-5000 at 70 eV with
direct probe inlet.
2.3. Subjects
Adult male Wistar rats weighing 250 g were used for
biochemical experiments. Adult male Swiss mice weighing
25– 30 g were used for pharmacological assays. Animals
were housed in a controlled environment, with free access to
food and water and maintained on a 12:12-h day/night cycle.
Housing, handling, and experimental procedures com-
plied with the National Institutes of Health Guide for Care
and Use of Laboratory Animals (Publication No. 85-23,
revised 1985).
Behavioral experiments were conducted from 10:00 a.m.
to 2:00 p.m.
2.4. Biochemical experiments
Binding of
3
H-flunitrazepam (
3
H-FNZ) (84.5 Ci/mmol;
New England Nuclear, NEN) to BDZ-bs in washed crude
synaptosomal membranes from rat cerebral cortex was
carried out as described by Marder et al. (2003).
2.5. Behavioral experiments
2.5.1. Holeboard test
This assay was conducted in a walled acrylic arena of
60 60 cm square floor and 30 cm high walls, with four
equally spaced holes in the floor, 2 cm in diameter each. The
Fig. 1. Molecular structures of active compounds in valeriana including
valtrate (as an example of valepotriates); baldrinal; valeranone; (+)
hydroxipinoresinol, valerenic acid; 6-methylapigenin; hesperidin, and
linarin.
S. Ferna
´ndez et al. / Pharmacology, Biochemistry and Behavior 77 (2004) 399–404400

holes housed an infrared light emitting diode. The interrup-
tion of the light beam by an exploring mouse during at least
100 ms triggers a counting device that records, in a
computer, the number of head dips and the time head
dipping. The mice were placed singly at the center of the
board, facing away from the observer and the number and
time of holes explored, as well as the number of rearings, in
a 5-min session were recorded. After each trial, the appara-
tus was wiped clean to remove traces of the previous assay.
A decrease in the number of head dips, the time spent head
dipping, and/or the number of rearings reveal a sedative
behavior (File and Pellow, 1985).
2.5.2. Sodium thiopental-induced sleeping time assay
A subhypnotic dose of sodium thiopental (35 mg/kg) was
intraperitoneally injected to mice 20 min after a similar
injection of vehicle or the drug. Sleeping time was deter-
mined as the interval between the loss and the recovery of
the righting reflex (Ferrini et al., 1974).
2.6. Drugs or extract solutions and injection procedures
The drugs and extracts used to perform the pharmacolog-
ical tests were as follows: dried fractions obtained from V.
officinalis as described in Fig. 2;VA(Fig. 1), kindly provided
by Dr. R. Bos; LN (Fig. 1) isolated by us from V. officinalis or
the genuine drug obtained from Extrasynthese, Genay,
France; and MA (Fig. 1) isolated by us from V. wallichii as
described in Wasowski et al. (2002). These drugs were
dissolved by the sequential addition of dimethylsulfoxide
up to a final concentration of 10%, ethanol up to a final
concentration of 10%, and saline to complete 100% volume.
Sodium thiopental (Fada, Biochemie Gesellschaft, Kundl/
Tirol, Austria) was dissolved in saline. The rodents were
intraperitoneally injected 20 min before performing the tests.
The volume of intraperitoneal injections was 0.15 ml/30 g of
body weight. The potentiating effects were tested by coad-
ministering VA and MA, LN and MA, LN and VA and LN,
and MA plus VA as indicated in Table 2. In each session, a
control group-receiving vehicle was tested in parallel with
those animals receiving drug treatment.
2.7. Statistical analyses
Data obtained from the holeboard test were subjected to
one-way ANOVA. Post hoc comparisons between individual
treatments and controls were made using Dunnett’s multiple
comparisons test. Dunn’s multiple comparison test was used
after Kruskal– Wallis test (nonparametric ANOVA) when
sleeping times were compared. Significance was reported
starting at the .05 level.
2.8. Isolation and identification of LN from ethanolic
extracts of V. officinalis
Dry V. officinalis roots and rhizomes were submitted to
the extraction and fractionation scheme shown in Fig. 2.
Powdered dry roots and rhizomes (100 g) of V. officinalis
were suspended in 500 ml of 70% ethanol and the mixture
was kept 2.5 h at 37 jC, with stirring. The filtrate was
concentrated to 1/3 of the original volume to eliminate most
of the ethanol and extracted with an equal volume of
petroleum ether, which was discarded. The aqueous phase
slightly concentrated to eliminate the remained petroleum
ether was extracted three times with an equal volume of
amyl alcohol and the alcohol phase (AP) was evaporated to
dryness. The solid residue was chromatographed in a silica
gel column (4.5 20 cm), prepared from a suspension of
silica H (Sigma, USA) in chloroform, which was eluted with
20% methanol in chloroform. The fractions showing the
presence of compounds active in the sleep-enhancing assay
were pooled. After partial evaporation of the solvent, the
pool deposits a whitish precipitate that is collected by
filtration and purified by recrystallization from ethanol.
Yield: 0.4 mg per g of valeriana powder. This material
permitted the identification of (7-[[6-O-(6-deoxy-a-L-man-
nopyranosyl)-h-D-glucopyranosyl]oxy]-5-hydroxy-2-(4-
methoxyphenyl)-4H-benzopyran-4-one or acacetin-h-ruti-
noside or LN (Fig. 1). From the mother solution remaining
after precipitation of LN, a pure sample of HN could also be
isolated, as described by Marder et al. (2003).
Fig. 2. Flow sheet of the V. officinalis fractionation scheme.
S. Ferna
´ndez et al. / Pharmacology, Biochemistry and Behavior 77 (2004) 399–404 401

LN was purified by crystallization from ethanol water
and identified by UV,
1
H-NMR, and mass spectroscopy.
2.9. Identification of LN
All the spectra of valeriana LN and authentic LN (extra-
syntheses) were identical:
1
H NMR y
H
12.90 (1H, s, OH-5),
8.05 (2H, d, J = 8.81 Hz, H-2Vand H-6V), 7.14 (2H, d,
J = 9.10 Hz, H-3Vand H-5V), 6.94 (1H, s, H-3), 6.78 (1H,
d, J = 2.06 Hz, H-8), 6.44 (1H, d, J = 2.06 Hz, H-6), 5.05
(1H, d, J = 7.04 Hz, H-1 glucosyl), 4.44 (1H, d, J = 6.16 Hz,
H-1 rhamnosyl), 3.86 (3H, s, OCH
3
-4V), 3.09– 3.46 (m, H-
sugars), and 1.07 (3H, d, J = 6.16 Hz, CH
3
-rhamnosyl); UV
E
max
(methanol): 327.5, 269.0, and 210.0 nm.
2.10. Identification of acacetin in LN
To further characterize the aglycone moiety of the
isolated compound, a sample of valeriana LN was hydro-
lyzed in boiling 1 M HCl for 1 h. The aglycone, acacetin,
was extracted from the hydrolysate with ethyl ether. The
spectroscopic analyses were as follows:
1
H-NMR y
H
12.91
(1H, s, OH-5), 10.84 (1H, s, OH-7), 8.03 (2H, d, J = 8.80
Hz, H-2Vand H-6V), 7.10 (2H, d, J =9.10 Hz, H-3Vand H-
5V), 6.86 (1H, s, H-3), 6.49 (1H, d, J = 2.05 Hz, H-8), 6.18
(1H, d, J = 2.05 Hz, H-6), 3.84 (3H, s, CH
3
-4V). EMS m/z:
284 (M
+
), 269, 256, 241, and 152. UV E
max
(methanol):
328.5, 268.0, and 213.0 nm. These values were identical to
those in the literature (Harborne, 1994).
A sample from the hydrolysate aqueous phase remain-
ing after the ethyl ether extraction was submitted to thin
layer chromatography on silica gel polyester sheets, with
254 nm fluorescent indicator (Sigma), developed with
butanol/acetic acid/ethyl ether/water (9:6:3:1 v/v) as the
solvent and stained using a general indicator for sugars
(aniline/diphenylamine/acetone/phosphoric acid). This
chromatography was performed alongside authentic sugar
standards. The only sugars detected were glucose and
rhamnose.
3. Results
3.1. Biochemical experiments
LN at concentrations up to 100 AM did not displace
3
H-
FNZ binding to BDZ-bs present in synaptosomal mem-
Fig. 3. Effects of LN on sedative behavior. Mean FS.E.M. of number of
head dips (open bars, left scale, in numbers) or time spent head dipping
(grey bars, left scale, in seconds) and number of rearings (closed bars, right
scale) registered in a 5-min session in the hole board test. * P< .05, **
P< .01, significantly different from vehicle; Dunnett’s multiple comparison
test after ANOVA. Number of animals per group ranged between 7 and 19.
Table 1
Effectiveness of VA, MA, LN, and AP on sodium thiopental-induced
sleeping time in mice
Sample Dose
(mg/kg)
nSleeping time median,
interquartile range (s)
VEH – 18 0 (0/0)
VA up to 15 15 0 (0/0)
MA up to 10 13 0 (0/0)
LN 4 9 0 (0/0)
7 7 1710 (1161/1800) * *
14 11 900 (300/1800) *
AP 200
y
10 585 (180/1275) *
Median (interquartile range) of sleeping time of mice measured in a sodium
thiopental-induced sleep test after 20 min of an intraperitoneal injection of
vehicle (VEH), linarin (LN 4, 7, and 14 mg/kg), valerenic acid (VA, up to
15 mg/kg), 6-methylapigenin (MA, up to 10 mg/kg), or the amyl alcohol
phase (AP) (see Fig. 1). The sleeping time was measured as the time spent
between disappearance and reappearance of righting reflex (see Methods);
n= number of mice.
*P< .05, significantly different from vehicle; Dunn’s multiple
comparison test after Kruskal – Wallis test (nonparametric ANOVA).
** P< .01, signif icantly different from vehicle; Dunn’s multiple
comparison test after Kruskal – Wallis test (nonparametric ANOVA).
y
200 mg is the dry residue of the AP obtained from 4 g of V. officinalis.
Fig. 4. Potentiation of LN sedative action by valerenic acid. Mean FS.E.M.
of number of head dips (open bars, left scale) or time spent head dipping in
seconds (grey bars, left scale) and number of rearings (close bars, right
scale) registered in a 5-min session in the hole board test performed 20 min
after the intraperitoneal injection of vehicle (VEH), linarin (LN, 4 mg/kg),
valerenic acid (VA, 5 mg/kg), or both drugs coinjected (LN 4 mg/kg + VA 5
mg/kg). * P< .05, ** P< .01, significantly different from vehicle; Dunnett’s
multiple comparison test after ANOVA. Number of animals per group
ranged between 7 and 19.
S. Ferna
´ndez et al. / Pharmacology, Biochemistry and Behavior 77 (2004) 399–404402

branes of rat cerebral cortex. Its aglycone, acacetin, caused a
reduction of approximately 30% in the binding only at a
concentration of 100 AM.
3.2. Sedative and sleep-enhancing effects of LN
The sedative action of LN, determined in the holeboard
test, is shown in Fig. 3. Doses of 4 and 7 mg/kg ip of LN
significantly reduced the number of rearings performed in
the holeboard test ( P< .05 and .01, respectively). No signif-
icant differences were observed in the exploration of holes.
The effect of LN on sleep is shown in Table 1. LN, at
doses of 7 or 14 mg/kg ip, significantly augmented the
sleeping time induced by thiopental ( P< .01 and .05,
respectively).
The dry residue of the amyl AP obtained from 4 g of V.
officinalis administered per kg of mice (200 mg dry residue/
kg) increased significantly the sodium thiopental-induced
sleeping time ( P< .05; Table 1).
Moreover, in Table 1, it is also shown that MA (up to 10
mg/kg) and VA (up to 15 mg/kg) are devoid of sleep-
enhancing capacity in mice.
3.3. Sleep-enhancing and sedative actions of LN coadmi-
nistered with VA and/or 6-methyl apigenin
As shown in Fig. 3 and Table 1, an intraperitoneal
injection of LN at a dose of 4 mg/kg had a moderate sedative
effect but did not increase the sleeping time induced by
sodium thiopental. On the other hand, an intraperitoneal
administration of VA at a dose of 5 mg/kg was not sedative
as measured in the holeboard test and did not increase the
sodium thiopental-induced sleeping time (Fig. 4 and Table 1,
respectively). However, the coadministration of both sub-
stances at these doses had sedative and sleep-enhancing
effects as evidenced by the remarkable reduction in the
exploration of holes ( P< .01), the time mice spent head
dipping ( P< .05) and the number of their rearings ( P< .01)
as assayed in the holeboard test (Fig. 4), and also produced a
striking increase in the sleeping time induced by sodium
thiopental ( P< .01; Table 2). In contrast, the coadministra-
tion of LN (4 mg/kg) and MA (1 mg/kg) or VA (5 mg/kg) and
MA (1 mg/kg) showed no effect (Table 2).
When LN (4 mg/kg), VA (5 mg/kg), and MA (1 mg/kg)
were coinjected, a significant increase in the sleeping time
induced by sodium thiopental was observed ( P< .01; Table
2). Nevertheless, this effect was not significantly different
from the one observed when LN and VA were coadminis-
tered (Table 2).
4. Discussion
LN was first identified in V. wallichii by Thies (1968) in
the form of its isovaleryl ester, but its pharmacological
properties were not explored. More recently, it has been
demonstrated that LN, isolated from the leaves of Buddleia
cordata, exerts central analgesic properties and is responsi-
ble for the antipyretic activity of this plant (Martı
´nez-
Va
´zquez et al., 1996). The same authors later found that
extracts of this species and LN itself exerted anti-inflam-
matory effects (Martı
´nez-Va
´zquez et al., 1998).
The present paper reports the identification of LN in V.
officinalis demonstrating its sedative and sleep-enhancing
properties in mice, as shown in Fig. 3 and Table 1.
Notwithstanding, LN and its aglycone acacetin are not
ligands for the BDZ-bs in brain; hence, the mechanism of
LN depressor actions is still unclear.
Hendriks et al. (1985) showed that VA had nonspe-
cific central depressant effects following its intraperitone-
al administration in mice. At doses above 100 mg/kg
body weight, effects were found in a rotarod and in a
traction tests. Higher doses were toxic. Spontaneous
locomotor activity of mice was reduced by VA at a dose
of 50 mg/kg, and a prolongation of the barbiturate-
induced sleeping test was found as well (Hendriks et
al., 1985). It was shown also that pure VA-antagonized
picrotoxin induced convulsions in mice at 12.5 and 25
mg/kg ip (Hiller and Zetler, 1996). VA was assumed to
be the most important active component in valeriana.
This hypothesis is not supported today by the well-
known fact, among others, that VA is only present in
V. officinalis and not in other active species widely used
like V. wallichii and V. edulis. We demonstrate here that
Table 2
Effectiveness of coadministration of LN, VA, and/or MA on sodium
thiopental-induced sleeping time in mice
Sample Dose
(mg/kg)
nSleeping time median,
interquartile range (s)
VEH – 18 0 (0/0)
VA 5
+ 6 0 (0/0)
MA 1
LN 4
+ 6 0 (0/240)
MA 1
LN 4
+ 15 960 (219/1650) * *
VA 5
LN 4
+
VA 5 9 300 (60/660) * *
+
MA 1
Median (interquartile range) of sleeping time of mice measured in a sodium
thiopental-induced sleep test after 20 min of an intraperitoneal injection of
vehicle (VEH) or the coinjection of valerenic acid and 6-methylapigenin
(VA 5 mg/kg + MA 1 mg/kg), linarin and 6-methylapigenin (LN 4 mg/
kg + MA 1 mg/kg), linarin and valerenic acid (LN 4 mg/kg+ VA 5 mg/kg),
and linarin and valerenic acid plus 6-methylapigenin (LN 4 mg/kg + VA 5
mg/kg + MA 1 mg/kg). The sleeping time was measured as the time spent
between disappearance and reappearance of righting reflex (see Methods);
n= number of mice.
** P< .01, significantly different from vehicle; Dunn’s multiple
comparison test after Kruskal – Wallis test (nonparametric ANOVA).
S. Ferna
´ndez et al. / Pharmacology, Biochemistry and Behavior 77 (2004) 399–404 403

VA per se has no in vivo effects in mice at low doses
(up to 15 mg/kg; Table 1).
The content of VA in the subterranean parts of various
subspecies of V. officinalis ranges between 0.3 and 3 mg/g
(Bos et al., 1996). Considering that the doses of various
pharmaceutical forms of valeriana are usually equivalent
to no more than 3 g of crude drug per day, the doses of
VA administered in this way are not therapeutically
significant unless we take into account the dramatically
potentiating effect in sedation and sleep induction that is
evident when VA and LN are acting together (Fig. 4 and
Table 2).
We have already described the presence of HN and MA
in V. wallichii and V. officinalis and demonstrated that MA
was able to potentiate the sleep-enhancing properties of HN
(Marder et al., 2003). In contrast, LN sleep-enhancing effect
is not potentiated by its coinjection with a dose of MA that
causes a clear anxiolytic effect (Table 2).
We propose that the sedative and hypnotic effects of V.
officinalis extracts may be attributed to the presence of LN,
HN, MA, and VA plus the potentiating effects produced by
their combinations, as is shown in Table 2.
The results in this paper establish the existence in
valeriana of flavonoid glycosides with sedative and sleep-
enhancing properties and demonstrate the existence of
potentiating effects in its extracts. The suspected presence
of synergic effects in valeriana (Hobbs, 1989) has been
substantiated by these findings and brought to the fore for
future clarification of the mechanisms involved (William-
son, 2001). We consider also that the correct standardization
of valeriana formulations, derived nutraceuticals, or other
medicaments has now to be done in terms of the above
mentioned compounds.
Acknowledgements
The authors are indebted to Drs. H Viola and JH Medina
for helpful discussions of the manuscript. VA was kindly
provided by Dr. R Bos (Rijksuniversiteit, Groningen, The
Netherlands). Grants were received from the International
Foundation for Science, Sweden, and from the National
Research Council and the University of Buenos Aires from
Argentina.
References
Bodesheim U, Ho¨lzl J. Isolierungm, Strukturaufkla¨rung und Radioreceptor-
assays von Alkaloiden und Lignan aus Valeriana officinalis L Pharma-
zie 1997;52:386 – 91.
Bos R, Woerdenbag HJ, Hendriks H, Zwaving JH, De Smet PAGM, Tittel
G, Wikstro¨ m HV, Scheffer JJC. Analytical aspects of phytotherapeutic
valerian preparations. Phytochem Anal 1996;7:143 – 51.
Ferrini R, Miragoli G, Taccardi B. Neuro-pharmacological studies on SB
5833, a new psychotherapeutic agent of the benzodiazepine class. Arz-
neim-Forsch (Drug Res.) 1974;24:2029 – 32.
File SE, Pellow S. The effects of triazolobenzodiazepines in two animal tests
of anxiety and in the holeboard. Br J Pharmacol 1985;86:729– 35.
Harborne JB. The Flavonoids. Advances in research since 1994. London:
Chapman & Hall, 1994. p. 448– 9.
Hendriks H, Bos R, Woerdenbag HJ, Koster AS. Central nervous depressant
activity of valerenic acid in the mouse. Planta Med 1985;51:28 – 31.
Hiller KO, Zetler G. Neuropharmacological studies on ethanol extracts of
Valeriana officinalis L.: behavioural and anticonvulsant properties.
Phytother Res 1996;10:145 – 51.
Hobbs C. Valerian: a literature review. HerbalGram 1989;21(Fall):19 – 34.
Ho¨lzl J. The pharmacology and therapeutics of Valeriana. In: Houghton PJ,
editor. Valerian. The Netherlands: Harwood, 1997. p. 55 – 75.
Ho¨lzl J, Godau P. Receptor binding studies with V. officinalis on benzodi-
azepine receptor. Planta Med 1989;55:642.
Houghton PJ. The scientific basis for the reputed activity of Valerian.
J. Pharm Pharmacol 1999;51:505 – 12.
Madaus G. Lehrbuch der biologischen Heilmittel. Hidesheim: Georg Olms
Verlag, 1976. p. 2770 – 7.
Marder M, Paladini AC. GABA
A
-Receptor ligands of flavonoid structure.
Curr Top Med Chem 2002;2:853 – 67.
Marder M, Viola H, Wasowski C, Ferna
´ndez S, Medina JH, Paladini AC. 6-
Methylapigenin and hesperidin: new valeriana flavonoids with activity
on the CNS. Pharmacol Biochem Behav 2003;75:737 – 45.
Martı
´nez-Va
´zquez M, Ramı
´rez Apan TO, Aguilar H, Bye R. Analgesic and
antipyretic activities of an aqueous extract and of the flavone linarin of
Buddelia cordata. Planta Med 1996;62:137 – 40.
Martı
´nez-Va
´zquez M, Ramı
´rez Apan TO, Lastra AL, Bye R. A comparative
study of the analgesic and anti-inflammatory activities of pectolinarin
isolated from Cirsium subcoriaceum and linarin isolated from Buddelia
cordata. Planta Med 1998;64:134 – 7.
Mennini T, Bernasconi P, Bombardelli E, Morazzoni P. In vitro study of the
interaction of extracts and pure compounds from Valeriana officinalis
roots with GABA, benzodiazepine and barbiturate receptors in rat brain.
Fitoterapia 1993;64:291 – 300.
Thies PW. Linarinisovalerianate, ein bischer unbekannte Flavonoid aus V.
wallichii D.C. Planta Med 1968;16:361 – 71.
Wasowski C, Marder M, Viola H, Medina JH, Paladini AC. Isolation and
identification of 6-methylapigenin, a competitive ligand for the brain
GABA
A
receptors, from Valeriana wallichii D.C. Planta Med
2002;68:934 – 6.
Williamson EM. Synergy and other interactions in phytomedicines. Phyto-
medicine 2001;8:401 – 9.
S. Ferna
´ndez et al. / Pharmacology, Biochemistry and Behavior 77 (2004) 399–404404