Hindawi Publishing Corporation
Evidence-Based Complementary and Alternative Medicine
Volume 2011, Article ID 362517, 10 pages
Homeopathic Doses of
the Behavior of Mice in Response to Novel Environments
Paolo Bellavite,1Paolo Magnani,1Elisabetta Zanolin,2and Anita Conforti3
1Department of Morphological Biomedical Sciences (Chemistry and Microscopy Section), University of Verona, Verona 37134, Italy
2Department of Medicine and Public Health (Biomedical Statistics Section), University of Verona, Verona, Italy
3Department of Medicine and Public Health (Medical Pharmacology Section), University of Verona, Verona, Italy
Correspondence should be addressed to Paolo Bellavite, firstname.lastname@example.org
Received 24 April 2009; Accepted 17 August 2009
Copyright © 2011 Paolo Bellavite et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Gelsemium sempervirens is used in homeopathy for treating patients with anxiety related symptoms, however there have been
few experimental studies evaluating its pharmacological activity. We have investigated the eﬀects of homeopathic doses of G.
sempervirens on mice, using validated behavioral models. Centesimal (CH) dilutions/dynamizations of G. sempervirens,the
reference drug diazepam (1 mg/kg body weight) or a placebo (solvent vehicle) were intraperitoneally delivered to groups of mice of
CD1 strain during 8 days, then the eﬀects were assessed by the Light-Dark (LD) choice test and by the Open-Field (OF) exploration
test, in a fully blind manner. In the LD test, the mean time spent in the illuminated area by control and placebo-treated animals was
15.98%, for mice treated with diazepam it increased to 19.91% (P=.047), while with G.sempervirens 5 CH it was 18.11% (P=.341,
non-signiﬁcant). The number of transitions between the two compartments increased with diazepam from 6.19 to 9.64 (P<.001)
but not with G.Sempervirens. In the OF test, G.sempervirens 5 CH signiﬁcantly increased the time spent and the distance traveled
in the central zone (P=.009 and P=.003, resp.), while diazepam had no eﬀect on these OF test parameters. In a subsequent series
of experiments, G.sempervirens 7 and 30 CH also signiﬁcantly improved the behavioral responses of mice in the OF test (P<
.01 for all tested variables). Neither dilutions of G.sempervirens aﬀected the total distance traveled, indicating that the behavioral
eﬀect was not due to unspeciﬁc changes in locomotor activity. In conclusion, homeopathic doses of G.sempervirens inﬂuence
the emotional responses of mice to novel environments, suggesting an improvement in exploratory behavior and a diminution of
thigmotaxis or neophobia.
In homeopathic Materia Medica,Gelsemium sempervirens
(Loganaceae) is described as a remedy for a variety of
neurological and behavioral symptoms including general
prostration, drowsiness, tiredness, mental apathy, lack of
muscular coordination and discomfort when confronted
with novelty or unfamiliar situations; these symptoms
are alleviated by motion and aggravated by emotion and
excitement [1–3]. However there have been only a few exper-
imental studies supporting this assertion . Therefore,
there is scope for testing this compound in homeopathic
dilutions on validated behavioral models of emotional state
and anxiety. G. sempervirens is a twining vine native to warm
temperate and tropical America, from Guatemala north
to the southeastern USA. All parts of this plant contain
the toxic strychnine-related alkaloids gelsemine, gelseminine
and sempervirine . Neurological signs characterized by
marked progressive weakness and convulsions culminating
in death have been observed in goats after ingestion of G.
sempervirens and ensuing toxicosis . At pharmacological
doses, G.sempervirens has been reported to show sedative,
analgesic and anti-seizure properties [7–9]. In an early
report , ultra-low doses of G.sempervirens in mice were
found to counter the behavioral eﬀects of the anxiogenic
compound RO 15-3505 (inverse agonist of benzodiazepines).
More recently, Bousta et al. report that in some but not
all experimental conditions, homeopathic dilutions of G.
sempervirens reduce stress-induced behavioral alterations
of mice in the staircase and light-dark (LD) tests .
However, these results consist of reversing the eﬀects of
severe stress (conditioned paradigm), and the ﬁndings
vary widely depending on the dose administered and test
2 Evidence-Based Complementary and Alternative Medicine
In this article we report a series of experiments testing
the possible eﬀects on mouse behavioral responses to G.
sempervirens in homeopathic dilutions/dynamizations. Two
validated tests on animal models, namely the LD choice
test and the open-ﬁeld (OF) test, were used in order
to acquire various behavioral parameters widely used in
neuropsychopharmacology for drug screening [12–14]. We
performed the study on the unconditioned responses, using
ethologically based paradigms which involve the sponta-
neous reactions to non-painful stimuli. The LD test is
based on the innate aversion of rodents to brightly lit area:
control mice placed in the large, bright, section will rapidly
move into the dark chamber and a classic anxiolytic eﬀect
manifests itself as an increase in the percentage of time
spent in the light area and in the number of transitions
between the two compartments . The OF test involves
placing an animal in an unknown environment in order to
observe a number of behavior patterns. The tendency to stay
on the periphery of the ﬁeld is known as thigmotaxis and
is often considered an index of highly emotional behavior
[16,17]. Conversely, the percentage of time spent in the
central zone and the distance traveled in the central zone
are considered indicative of exploratory behavior and should
reﬂect a decrease in anxiety.
In homeopathy, a wide range of dosages are used. Such
formulations are better described as homeopathic dilutions
(also called “dynamizations”, or “potencies”), whose prepa-
ration procedure involves successive steps of dilution—
10×or Decimal (D), 100×or Centesimal (CH)—followed
by strong manual or mechanical succussion. Homeopathic
dilutions/dynamizations such as 5 CH are expected to
contain ultra-low concentrations of the purported active
ingredient, but in amounts still within the molecular range
since the dilution factor from the ﬁrst ethanolic extract
(mother tincture) is 1010. In this work we chose to start the
investigation with a series of experiments testing the eﬀects
of the 5 CH dilution/dynamization, which is currently used
not only in homeopathic therapy but is also in the domain
of “conventional” pharmacology and of current scientiﬁc
wisdom. Since homeopathy also makes use of high dilu-
tions/dinamizations, in an ensuing series of experiments, we
investigated the eﬀects of further dilutions/dynamizations,
namely G. sempervirens 7 and 30 CH.
2.1. Subjects and Maintenance. Male 4-5-week-old mice
of the speciﬁed inbred strains (CD1 or C57BL/J6) were
purchased from Harlan Laboratories (Udine, I) and allowed
distributed 4/cage in standard polycarbonate cages (369 ×
156 ×132 mm) with water and food available ad libitum,in
an air-conditioned room (temperature 22 ±2◦C, humidity
∼55 ±5%). Lights were left switched on between 7 a.m. and
7 p.m. Cages were cleaned and bottles were ﬁlled with fresh
tap water thrice a week, excluding on the days of testing.
The sequential order in which mice were injected and then
tested was arranged so that the mean interval time between
injection and testing was the same for all the experimental
groups. G.sempervirens and the control vehicle (placebo)
were administered intraperitoneally (i.p.) each day at 9 a.m.
for 9 consecutive days, including on the 2 last days devoted
to behavioral testing. Mice belonging to the benzodiazepine
standard treatment group were also injected with the control
solution for 9 days, but on the testing days the control
solution was supplemented with diazepam (see below).
Apart from the drug injections and testing, the animals
were not subjected to pain or other forms of emotional or
physical stress. The animals were used only once in any given
test to avoid the confounding eﬀects of learning and habit-
uation. After completing the tests, the mice were put back
into their original home cages and then sacriﬁced using CO2-
saturated atmosphere. All the tests were conducted at the
Faculty of Medicine, Verona University, Italy, in accordance
with the guidelines approved by the European Community
Council Directive of November 24, 1986 (law86/609/CEE).
The testing procedures were approved by the Animal Ethical
Committee of the Verona University and by the Italian Health
Ministry (authorization n. 4 Tit. VII/5, February 16, 2007).
2.2. Drugs. Gelsemium sempervirens in the 4, 6 and 29
CH dilutions/dynamizations was prepared by Boiron Lab-
oratories (Lyon, F) from the mother tincture of plant
roots, according to the French Pharmacopoeia and
supplied as a hydroalcoholic (30% ethanol, v/v) solution.
The mother tincture was assayed to determine its gelsemine
content, which proved to be 0.021%; since the molecular
weight of this compound is 322.41, its concentration in
the mother tincture used for preparing the homeopathic
dilutions/dynamizations was 6.5×10−4moles/l. The control
solution (placebo) used in this investigation was provided
by Boiron Laboratories, using the same hydroalcoholic
solvent stock (30% ethanol, v/v) employed to prepare the G.
sempervirens dilutions. On the morning of the ﬁrst day of
drug delivery, before starting the treatment, 0.4 ml samples
of the solutions were added to 39.6 ml of distilled sterile and
apyrogenic water in a sterile 50 ml Falcon plastic tube, closed
with a plastic cap and the dilution was manually succussed
with 20 strong vertical strokes to obtain the operative
dilutions 5, 7 and 30 CH, with ﬁnal ethanol concentration
lowered to 0.3% (v/v). Since each dilution step involves a 100
×decrease in concentration, the “theoretical” concentrations
of gelsemine in the 5, 7 and 30 CH solutions were 6.5×10−14,
6.5×10−18 and 6.5×10−64 moles/l, respectively. The operators
who performed the injections and the behavioral tests were
totally unaware (“blind”) of the treatment group to which
the animals had been assigned. Blinding of all procedures
of injection and testing was performed by randomizing and
coding the tubes containing the drug and control solutions.
The person who coded the solutions did not belong to
the research group and signed a conﬁdentiality agreement
form; the codes were recorded on a sheet that was kept
sealed inside an envelope until all the tests and calculations
were completed. The coded solutions were distributed in
15 ml sterile Falcon plastic tubes (7.5 ml/tube), wrapped in
aluminum foil and stored at +4◦C until the day of use. Before
administering the contents, each tube was manually shaken
with 20 strokes. On the days of the behavioral tests, the mice
Evidence-Based Complementary and Alternative Medicine 3
were injected 60 min before the start of the tests, which
were performed between 11 a.m. and 3 p.m. Each day, 0.3 ml
of the remedy or control placebo were administered by i.p.
injection using 1 ml insulin syringes. The benzodiazepine
standard drug was diazepam (Vali u m ,Roche)(5mg/ml),
diluted by an independent investigator to ensure blinding
of operators with respect to group assignment, in the same
control solution and administered intraperitoneally at the
ﬁnal dose of 1 mg/kg body weight (b.w.).
2.3. Behavior Assessment. A video-tracking camera (GZ-
MG135, JVC, Japan) and a software program (“Smart” VTS
system from PanLab, Barcelona, E) were used to record the
sessions automatically. The camera viewed four test arenas,
each one of which, in turn, was divided by the software into
two diﬀerent zones according to the test to be performed
(LD or OF). All the sessions were recorded and stored as
DVD. The video outcome signals were converted through the
image processor into binary images in such a manner that the
animal was tracked as a black spot with a white background.
The movement of the spot was recorded to track the position
of the animals, the time they stayed in diﬀerent zones and
the distance traveled. The number of transitions between
light and dark compartments in the LD test was evaluated
on the video by an operator who was unaware of the group
assignment of the mice.
The experiments were performed on each group of
animals in the following order: LD choice test on day one
(8th day of drug administration) and OF exploration test
on the following day (9th day of drug administration). Mice
were tested individually in each test arena and the operators
stayed outside the testing room during the recording of the
experimental session. Just before testing, the animals were
allowed to acclimate to the room inside their cages for 3 min
after being moved from their usual housing area.
The LD test arena consisted of two interconnected
compartments that diﬀered in size and color: a white open
square (30 cm ×30 cm) and a black covered compartment
(30 cm ×15 cm). Both compartments had walls that were
25 cm high. A small opening (4 ×4 cm) allowed the mice
to freely move from the illuminated to the dark chamber and
vice versa. The white ﬁeld was brightly lit at 200 lux and the
mice were left to explore the space for a 5 min testing period.
The OF test arena consisted of a 50 ×50 cm wooden
platform, painted black and enclosed by 25 cm high walls.
White light (100 lux) was present in the room. The 10 min
test began with the subject being placed inside the OF arena,
in a corner near the walls. Using the video-tracking system,
the arena was virtually divided to deﬁne a square central zone
occupying 25% of the total area. The tendency to enter the
central zone and to travel inside it, instead of running along
the walls or staying in the corners, is considered to be a sign
of a decrease in the anxiety induced by a novel environment
(neophobia) and of an increase in exploratory attitude,
typical of mice that consider the experimental setting to be
familiar. Thigmotaxis is particularly apparent upon the ﬁrst
exposure to a novel space, and helps the animal to deﬁne the
boundaries of an unfamiliar environment. In unconditioned
behavior tests in which an animal is placed for the ﬁrst time
in an arena, it also reﬂects novelty induced anxiety, general
activity, exploratory behavior and decision-making .
2.4. Statistics. In the ﬁrst series of experiments, the data from
ﬁve experiments performed on three treatment groups of
mice, which received either the 5 CH dilution or diazepam
or a control placebo (total of ﬁve experiments =72 mice per
treatment group) were pooled and analyzed. In the second
series of experiments, the data from one experiment per-
formed on three treatment groups of mice, which received
either the 5 or the 7 CH dilution or the placebo (total =16
mice per treatment group) and those from two experiments
performed with four treatment groups of mice, which
received either the 5 CH or the 7 CH or the 30 CH dilutions
or the placebo (total =32 mice per treatment group) were
pooled and analyzed. Extreme behavioral responses were
considered as outlier data and excluded when their value
exceeded 2 standard deviations (SDs). This occurred in very
few cases (see Results section). When the distribution of
data was normal (data from OF tests), the groups were
compared by two-way analysis of variance (ANOVA SPSS,
version 11 for Windows, Chicago, IL), using the treatment
group and the experiments as the factors. The latter factor
was included in the ANOVA analysis because the control
values of activity varied considerably between experiments.
Post-hoc t-tests were performed assuming equal variances
with least signiﬁcant diﬀerence (LSD) corrections to adjust
for multiple comparisons. When the distribution was not
normal (data from LD tests), the groups were compared by
the non-parametric Kruskal-Wallis analysis of variance test,
followed by the Mann–Whitney test for two independent
samples to determine whether the activities of the control
and drug-treated groups diﬀered from each other.
The eﬀect of the drugs on the various behavioral
parameters was also calculated as a percentage with respect
to the control values (taken as 100%) for each experiment,
according to the formula:
Activity of each drug-treated animal/mean activity of
control animals ×100.
This allowed the eﬀects of the various dilutions tested in
the two series of experiments, standardized as a percentage
of their internal control values for untreated animals, to be
compared and statistically evaluated.
3.1. LD Choice. Table 1 reports the mouse behavior data in
the LD test, in terms of the two main parameters: percentage
of time spent in the illuminated area and the number of
transitions between the two compartments. The mean time
spent by control (placebo-treated) animals in the illuminated
area was 15.98%, for mice treated with G. sempervirens
5 CH it was 18.11% (non-signiﬁcant), while diazepam
instead signiﬁcantly increased the mean time spent in the
illuminated compartment to 19.91% (P=.047). The number
of transitions between the two compartments increased little
in the G.sempervirens 5 CH-treated group, while in the
diazepam-treated group the values rose from a mean of 6.19
(control) to a mean of 9.64 (P<.001).
4 Evidence-Based Complementary and Alternative Medicine
Tab le 1: Eﬀects of G. sempervirens 5 CH and diazepam on the behavioral parameters in the LD.
Parameter Control G. sempervirens (5 CH) Diazepam
nMean SEM nMean SEM PanMean SEM Pa
Percentage of time in light∗71 15.98 1.29 69 18.11 1.42 .341 69 19.91 1.39 .047
Number of transitions∗∗ 71 6.19 0.58 70 6.40 0.56 .699 70 9.64 0.64 <.001
aMann—Whitney test of treated group versus control.
∗Kruskal Wallis for groups P=.159.
∗∗Kruskal Wallis for groups P<.001.
Tab le 2: Eﬀects of G. sempervirens 5 CH and diazepam on the behavioral parameters in the OF test.
Parameter Control G. sempervirens (5 CH) Diazepam
nMean SEM nMean SEM PanMean SEM Pa
Percentage of time in center∗69 6.34 0.41 70 7.86 0.41 .009 69 6.21 0.48 .812
Distance in center∗∗ 68 589.9 37.8 70 738.9 38.0 .003 68 585.1 49.5 .925
Total distance a∗∗∗ 70 5181.5 166.2 71 5371.4 168.1 .306 71 5321.7 211.1 .449
aPost-hoc analysis of treated group versus control.
Two-way ANOVA for groups: ∗P<.001, ∗∗P=.001, ∗∗∗ P=.647.
For experiments ∗P<.001, ∗∗P=.001, ∗∗∗P<.001.
3.2. OF Exploration. Table 2 reports the results for the OF
tests. The placebo-treated control animals spent 6.34% of the
time in the center of the OF arena. This value is considerably
<25% one would expect in the case of a random choice,
indicating that the central zone was quite uncomfortable for
the mice. In the G.sempervirens 5 CH-treated group, the time
spent in the central zone of the arena increased to 7.86%
(P=.009), while the action of diazepam in this test was
totally ineﬀective. The distance traveled by mice in the central
zone of the arena was also positively inﬂuenced by treatment
with G.sempervirens 5 CH, with an overall mean increase
from 589.9 to 738.9 cm (P=.003). During the OF test the
total distance traveled by the mice in the entire arena was
also analyzed. No signiﬁcant eﬀect was found in drug-treated
versus placebo-treated animals, indicating that the observed
diﬀerences in time and distance spent in the central zone
were not due to changes in the general, unspeciﬁc, locomotor
activity of the mice.
The weight of the animals before and after the treatment
was similar in the three treatment groups, suggesting no
diﬀerences in feeding, metabolism or growth attributable
to drug eﬀects (data not shown). There was found to be
a signiﬁcant diﬀerence between the mean values for the
ﬁve experiments (see notes to Tables 1and 2), which was
taken into account to control for its possible confounding
eﬀects. Finally, no signiﬁcant interactions between groups
and experiments were noted.
3.3. Study of a Diﬀerent Mice Strain. We also performed
one experiment in which the G. sempervirens 5 CH dilu-
tion/dynamization was tested on a diﬀerent strain of mouse,
C57BL. This yielded results qualitatively similar to those
for the CD1 mice in both test models, conﬁrming an
increase in the time spent and distance traveled in the center
of the OF induced by G. sempervirens 5CH.However,
due to the variability of responses between animals, the
diﬀerences between control- and G. sempervirens-treated
groups (e.g., 6.11 ±0.80 and 8.83 ±1.38% time spent in the
center, respectively) in this individual experiment were not
statistically signiﬁcant (P=.068).
3.4. Even High Dilutions Work Better than Placebo. Follow-
ing these positive results obtained with the 5 CH dilu-
tion/dynamization, we performed three further experiments,
using the same protocol to also test higher dilutions, namely
G. sempervirens 7 and 30 CH, in addition to 5 CH and
a control placebo. Diazepam was not included here to
avoid needlessly sacriﬁcing animals, since the results for this
control drug were suﬃciently clear from the ﬁrst series of
experiments. The 5 and 7 CH dilutions/dynamizations of
G.sempervirens were ineﬀective in the LD test; the 30 CH
dilution slightly increased the values of the LD parameters,
especially the number of transitions, but these eﬀects were
not statistically signiﬁcant (data not shown). In the OF test
(Table 3), the percentage time spent in the center of the
arena and the distance traveled in the center were increased
by all the G.sempervirens dilutions/dynamizations, with
highly statistically signiﬁcant results using the 7 and the 30
CH dilutions/dynamizations (P<.01). The eﬀect of the 5
CH dilution/dynamization was quantitatively similar to that
previously described in Ta b l e 2 , but in these experiments it
did not reach the statistical signiﬁcance threshold, due to
the lower size of subject population and to high inter-subject
data variability (SEM). The maximum and most signiﬁcant
eﬀect was obtained in the OF (distance traveled in the center)
with the 30 CH dilution/dynamization (P=.002). In this
series of experiments, too, the tested dilutions were found
to have no signiﬁcant eﬀect on the unspeciﬁc locomotor
activity, as indicated by the total distance traveled in the
3.5. Summary of All the Experimental Data. A summary of
the main results of this investigation is given in Figure 1.
The mean percentage values for all the mice of the treated
Evidence-Based Complementary and Alternative Medicine 5
Tab le 3: Eﬀects of G. sempervirens 5, 7 and 30 CH on the behavioral parameters in the OF testa.
Parameter Control G. sempervirens (5 CH) G. sempervirens (7 CH) G. sempervirens (30 CH)
nMean SEM nMean SEM PbnMean SEM PbnMean SEM Pb
time in center∗47 5.98 0.51 48 7.09 0.61 .169 47 8.19 0.52 .007 32 8.60 0.80 .004
center∗∗ 45 501.1 42.6 48 618.5 46.6 .089 47 702.6 41.3 .004 32 744.0 75.1 .002
distance∗∗∗ 47 4892.9 167.6 47 5180.3 146.4 .183 46 5248.7 171.7 .102 31 5158.1 235.4 .378
aData are pooled from three experiments (control, 5 CH and 7 CH) and from two experiments (30 CH).
bPost-hoc analysis of treated group versus control.
Two-way ANOVA for groups: ∗P=.012, ∗∗ P=.008, ∗∗∗ P=.371. For experiments ∗P=.075, ∗∗ P=.867, ∗∗∗ P<.001.
groups conﬁrms the increase due to diazepam but not due
to G.sempervirens dilutions in the LD paradigm (panels (a)
and (b)) and the signiﬁcant increase due to all the tested
homeopathic dilutions/dynamizations of G.sempervirens in
the OF-tested variables (panels (c) and (d)).
In the ﬁeld of psychopathology, animal models have become
an invaluable tool for analyzing the mechanisms of various
disorders, and have aided in developing and predicting
therapeutic responses to pharmacological agents such as
benzodiazepines. In this work we investigated the eﬀects
of Gs.sempervirens on the non-conditioned behavior of
mice, using two widely validated models that allow the
expression of various “symptoms” of emotional responses.
We used ethological models in which the animal is inﬂuenced
by emotional states of fear, curiosity or anxiety induced
under diﬀerent test paradigms without further conditioning,
thus allowing for comprehensive “behavioral proﬁling”.
Traditional diﬃculties in accepting these models stem from
the argument that there is no conclusive evidence that what
occurs in animals is equivalent to what occurs in humans.
On the other hand, most conventional drugs and, recently,
several homeopathic medicines have been tested in animal
models, whose main advantages are that they allow multiple
testing under controlled conditions and easier access to
studying the mechanism of drug(s) activity.
4.1. Diﬀerent Symptoms Modulation According to the Model.
Most behavioral procedures for studying the pharmacology
of anxiety use models involving non-conditioned behavior,
that are usually based on novelty induced variations in
exploratory activity. Considering the LD model, diazepam
was active, as expected, signiﬁcantly increasing the time spent
in the light arena and the number of transitions. In the same
assays, G.sempervirens wasmuchlessactivethanthecon-
ventional anxiolytic drug. Our data, showing small, but not
signiﬁcant, eﬀects of G.sempervirens in the LD model, are in
agreement with those of Bousta et al.  who reported some
anxiolytic-like eﬀects in mice stressed by repeated electric
shock, but no eﬀects in normal mice. Diﬀerences between
the type and severity of external stressors or the animal
strain or in the experimental setup might account for the
high variability of results reported in diﬀerent experimental
conditions and by diﬀerent laboratories [21,22]. It has been
noted that the extent to which an anxiolytic compound can
facilitate exploratory activity depends on its baseline level in
the control group . Since our experimental setting did
not involve prior exposure to stress, it is conceivable that in
those conditions the response of the mice to G.sempervirens
was slight because the level of basal anxiety, assessed by LD
choice, was low.
In the OF model, all dilutions/dynamizations of G.
sempervirens changed the behavior of the mice, reducing
thigmotaxis, an anxiety-like behavioral response. The eﬀects
of G. sempervirens on the tendency of animals to stay
and move in the center of the OF arena rather than in
the periphery was consistent in the diﬀerent experiments
and highly statistically signiﬁcant. Thigmotaxis is a very
conspicuous behavior in anxious animals, and plays a role
in the formation of avoidance behavior and cognition. It is
therefore a primordial behavior, with a genetic basis, that
is ecologically important and used by both animals and
humans for exploring the environment.
The ﬁnding of a drug-induced increase in the distance
traveled in the center of the arena, but not in the total
distance traveled, suggests that the eﬀects of G.sempervirens
in our model system are anxiolytic rather than sedative
or otherwise aﬀecting animal locomotion. Since we used a
control vehicle as a placebo in all experiments, the eﬀects
cannot be attributed to the very low dose of ethanol
present in the ﬁnal operating solutions, unlike in certain
previous studies where the use of alcohol as vehicle for
homeopathic preparations might have been able to produce
some background eﬀects in behavioral tests . A further
methodological issue that reinforces our ﬁndings is the
systematic use of blinding, a procedure rarely employed in
animal research, but which is worthwhile especially when
testing hypotheses that are apparently in contrast with
conventional scientiﬁc wisdom, such as the very existence of
biological activity of solutions diluted beyond the Avogadro
Since the reference benzodiazepine drug was inactive
in the OF test, the results indicate that, at least in our
experimental conditions, OF and LD explore diﬀerent
emotional responses, with diﬀerent sensitivity to drugs and
neurological mechanisms. This suggestion is in agreement
6 Evidence-Based Complementary and Alternative Medicine
LD test: Time spent in light
Gels. 5 CH
Gels. 7 CH
Gels. 30 CH
Gels. 5 CH
Gels. 7 CH
Gels. 30 CH
LD test: Number of transitions
Gels. 5 CH
Gels. 7 CH
Gels. 30 CH
OF test: Time spent in centre
Percent of mean control values ±S.E.M.
Gels. 5 CH
Gels. 7 CH
Gels. 30 CH
OF test: Distance traveled in centre
Figure 1: Summary of the eﬀects of G.sempervirens 5, 7 and 30 CH, and of diazepam 1 mg/Kg bw on the main parameters of the LD
test (Panels a and b) and of the OF test (Panels c and d), expressed as percent of the placebo-treated control values. In each experiment
the functional activity of mice was calculated as a percentage with reference to the mean value of the placebo group (control vehicle). The
values from eight experiments (control, placebo and 5 CH), from ﬁve experiments (diazepam), from three experiments (7 CH) and from
two experiments (30 CH) were pooled and analyzed by two-way ANOVA or by the Mann-Whitney test as described in Section 1.∗P<.01,
with reports showing that anxiolytic treatments do not by
themselves increase exploration in the central zone of the
OF, but that they do decrease the stress-induced inhibition
of exploration behavior . It has been previously reported
that the index of thigmotaxis, assessed by the OF test in
CD1 mice, was not decreased by diazepam (≤5mg/kg).
Others reported anxiolytic eﬀects of diazepam in the OF
paradigm in rat [26,27], but in mice 1 mg/kg [16,28]or
1.5 mg/kg  diazepam did not show signiﬁcant eﬀects
on the time spent in central area of OF. To the best of
Evidence-Based Complementary and Alternative Medicine 7
Control of anxiety-like
symptoms (in LD paradigm)
Control of other emotional
responses (in OF paradigm)
Figure 2: Hypothetical mechanism of action of G.sempervirens. Benzodiazepines act by enhancing the inhibitory eﬀects of the
neurotransmitter GABA. The neurosteroid allopregnanolone is produced from progesterone and acts both on the GABA receptor—
at a binding site diﬀerent from benzodiazepines—and on other receptors including the serotonine (5HT) and nicotinic-acetylcholine
receptor, thus reducing the impulse generation in postsynaptic neurons. By ﬁne tuning inhibitory transmission through glycinergic system
and allopregnanolone synthesis [19,20], G.sempervirens 5 CH may help the nervous system adapt to adverse situations. Due to the
multicomponent nature of G.sempervirens and to our ﬁnding of preferential eﬀects on OF paradigm even at high dilutions/dynamizations,
the existence of other neurosteroid-independent mechanisms may be hypothesized.
our knowledge open ﬁeld behaviors of mice were reduced
by diazepam at doses which may be within the sedative–
hypnotic range [16,28]. These conﬂicting drives could be
inﬂuenced in various ways by drugs and small diﬀerences in
the experimental conditions may have marked eﬀects on the
outcomes. Ethological models present individual diﬀerences
and variable behavioral baseline levels and this requires
strong care for variable parameters linked to environment,
handling and testing .
4.2. Doses and Dilutions. Due to the complexity of the
experimental setting, and in particular the need to use many
animals in each group, in this study we started with a single
dose (potency) of G. sempervirens,namely5CH,whichwas
initially chosen because its dilution is suﬃciently high to be
representative of the homeopathic formulations commonly
used in humans, but also suﬃciently low to contain at least a
few molecules of the purported active principle(s). We made
the assumption that, given a theoretical concentration of the
active principle in the 5 CH solution of 6.5×10−14 moles/l,
its activity on the behavior of mice could be consistent with
a molecular paradigm and with the existence of highly sen-
sitive receptors for this drug in the animals’ central nervous
system, in agreement with other recent evidence [19,20].
However, the results of the second series of experiments
indicate that the drug may be active also at ultra-high
dilutions/dynamizations. From our data (Figure 1) the 30
CH dilution/dynamization appears to be the most active in
the OF variables, and these apparently paradoxical results
require further conﬁrmation. Indeed, at 30 CH, there is no
active principle in this dilution/dynamization and further
basic evidence will be necessary to make this observation
acceptable. There is growing evidence in the literature that
rodents are responsive to very high dilutions/dynamizations
in immunological [30,31] and behavioral [11,32] models. If
conﬁrmed, these eﬀects on mouse behavior would be much
more coherent with the homeopathic paradigm and with
traditional homeopathic medicine than initially expected.
A number of observations—coming from several
research ﬁelds [11,24,31–34]—suggest that biologically
8 Evidence-Based Complementary and Alternative Medicine
active compounds may indeed have high-dilution eﬀects
which mimics those of lower dilutions (higher doses): in
homeopathy there does not exist linearity or proportionality
between molecular concentration of active principles and
therapeutic eﬀect [33–41].
So far there is no satisfactory or uniting theoretical
explanation for these observations, but recent evidence
seems to point to organization of the solvent water on
a mesoscopic scale: the nano-heterogenous structure of
water can be determined by interactive phenomena such as
coherence [42–44], epitaxy [45,46], temperature-pressure
processes during strong agitation and formation of colloidal
nanobubbles containing gaseous inclusions of oxygen, nitro-
gen, carbon dioxide, silica and possibly the remedy source
material [46–51]. These unusual properties of high dilutions,
which merit further investigation, are potentially relevant
not just to homeopathic pharmaceutical practice, but also to
basic research into cell sensitivity to regulation.
4.3. Hypothetical Mechanisms. A hypothetical diagram
showing the possible action mechanism(s) of G.sempervirens
is reported in Figure 2. A possible target of G.sempervirens
action at neurological level has been identiﬁed by recent
studies [19,20], showing that in rat central nervous system
(spinal cord and hippocampus) extremely low doses of this
compound (10−10 M) and of its active principle gelsemine
enhance the enzymatic production of the neurosteroid
allopregnanolone (5a,3a-tetrahydroprogesterone), an active
stimulator of GABAa and 5HT receptors and therefore of
inhibitory signaling in the central nervous system. This
eﬀect could be due to a speciﬁc interaction at the level
of the glycinergic receptors since this was antagonized by
strychnine. However, since we noted diﬀerences in the action
of conventional and homeopathic anxiolytic-like eﬀects, it
is highly conceivable that the multicomponent nature of the
active principles of G.sempervirens make this compound
able to interact with further receptor or gene expression
systems, which in turn modulate the behavior in novel
environments at more subtle and complex levels.
Beneﬁcial, therapeutical eﬀects of extremely low doses of
agents which are toxic at high doses are deﬁned hormesis.
Recently Calabrese provided a broad range of examples of
neurobiological processes and behavioral eﬀects, including
anxiety models, that exhibit non-linear, biphasic (“reverse-
U”) dose responses to drugs and stressors [52,53]. According
to some authors, hormesis is not directly related to classical
explanations of the mechanism of action of homeopathic
medicines when utilized in the molecular range of dilu-
tions/dynamizations (approximately <9 CH) [45,55,56].
The development of interventions that activate hormetic
signaling pathways in neurons is a promising new approach
for the prevention and treatment of a range of neurological
Taken together, our data suggest that the eﬀects of G.
sempervirens at homeopathic dilutions on unconditioned
mouse behavioral responses concern the exploratory attitude
of the animals, a parameter that is better brought out by
the OF test. Mouse behavior in the OF test is aﬀected
by various factors such as individual testing (since mice
are social animals), neophobia (emotional response to a
novel environment) and agoraphobia (being in an exposed
setting from which there is no easy means of escape). In
our experimental conditions G.sempervirens showed positive
eﬀects on these parameters, and the eﬀect observed in
the OF test ﬁts very well with the classical archetype of
the G.sempervirens.Materia Medica, according to which
individuals who respond to this medicine are characterized
by experiencing strong discomfort when confronted with
novelty or unfamiliar situations.
In conclusion, homeopathic doses of G.sempervirens
positively inﬂuence the emotional responses of mice to novel
environments, suggesting an improvement in exploratory
behavior and a decrease in thigmotaxis or neophobia.
The study provides a basic knowledge of the pharmaco-
logical eﬀects of G.sempervirens and can implement the
clinical research on the anxiolytic eﬀect of this drug on
humans, which are based only on empirical knowledge
but not on fundamental evidence supplied by controlled
Grants from Laboratoires Boiron s.r.l. (Milan, Italy) to
Verona University and from the Ministry of University and
Scientiﬁc and Technologic Research.
Recently the same group showed that following small
methodological changes a signiﬁcant anxiolytic-like eﬀect of
Gelsemium s. is detected also in the Light-dark test (Magnani
P, Conforti A, Zanolin E, Marzotto M, Bellavite P. Dose-
eﬀect study of Gelsemium sempervirens in high dilutions
on anxiety-related responses in mice. Psychopharmacology
The authors wish to thank Prof. Cristiano Chiamulera, Prof.
Andrea Sbarbati, Prof. Paolo Fabene, Dr Naoual Boujedaini
and Dr Luigi Marrari for their advice and cooperation, and
Dr Lara Magro and Dr Ilaria Meneghelli for carrying out the
coding of the samples for blind testing. The authors have no
conﬂicts of interest.
 W. Boericke, MateriaMedicawithRepertory,Boericke&Tafel,
Santa Rosa, Calif, USA, 9th edition, 1927.
 J. Barbancey, Pratique Hom´
eopathique en Psycho-Pathologie.
Tome I I , Editions Similia, Paris, France, 1987.
 M. Guermonprez, Hom´
eopathie, Principles: Clinique—Tech-
niques, CEDH, Paris, France, 2006.
Evidence-Based Complementary and Alternative Medicine 9
 K. Pilkington, G. Kirkwood, H. Rampes, P. Fisher, and J.
Richardson, “Homeopathy for anxiety and anxiety disorders:
asystematicreviewoftheresearch,”Homeopathy, vol. 95, no.
3, pp. 151–162, 2006.
sempervirens,” Journal of Natural Products,vol.50,no.2,pp.
 L. J. Thompson, K. Frazier, S. Stiver, and E. Styer, “Multi-
ple animal intoxications associated with Carolina Jessamine
(Gelsemium sempervirens) ingestions,” Veterinary and Human
Tox i c o lo g y , vol. 44, no. 5, pp. 272–273, 2002.
 J. Valnet, Phytoth´
erapie, Maloine, Paris, Rance, 1992.
 D. Demarque, J. Jouanny, B. Poitevin, and Y. Saint-Jean,
Pharmacologie Et Mati`
CEDH, Lyon, France, 2nd edition, 1995.
 O. Peredery and M. A. Persinger, “Herbal treatment fol-
lowing post-seizure induction in rat by lithium pilocarpine:
scutellaria lateriﬂora (Skullcap), Gelsemium sempervirens
(Gelsemium) and Datura stramonium (Jimson Weed) may
prevent development of spontaneous seizures,” Phytotherapy
Research, vol. 18, no. 9, pp. 700–705, 2004.
 J. Guillemain, A. Rousseau, P. Dorfman, and M. Tetau,
“Recherche en psychopharmacologie,” Cahiers de Biother, vol.
103, pp. 53–66, 1989.
 D. Bousta, R. Soulimani, I. Jarmouni et al., “Neurotropic,
immunological and gastric eﬀects of low doses of Atropa bel-
ladonna L., Gelsemium sempervirens L. and Poumon histamine
in stressed mice,” Journal of Ethnopharmacology, vol. 74, no. 3,
pp. 205–215, 2001.
 G. Griebel, C. Belzung, R. Misslin, and E. Vogel, “The free-
exploratory paradigm: an eﬀective method for measuring
neophobic behaviour in mice and testing potential neophobia-
reducing drugs,” Behavioural Pharmacology,vol.4,no.6,pp.
 Y. Cl´
ement, C. Joubert, C. Kopp et al., “Anxiety in mice: a
principal component analysis study,” Neural Plasticity, vol.
2007, Article ID 35457, 2007.
 M. Bourin, B. Petit-Demouli`
ere, B. Nic Dhonnchadha, and
oet, “Animal models of anxiety in mice,” Fundamental
and Clinical Pharmacology, vol. 21, no. 6, pp. 567–574, 2007.
 M. Bourin and M. Hasco¨
et, “The mouse light/dark box test,”
European Journal of Pharmacology, vol. 463, no. 1–3, pp. 55–
 P. Simon, R. Dupuis, and J. Costentin, “Thigmotaxis as
an index of anxiety in mice. Inﬂuence of dopaminergic
transmissions,” Behavioural Brain Research, vol. 61, no. 1, pp.
 P. K. Lepp¨
anen, N. Ravaja, S. B´
eatrice, and M. Ewalds-
Kvist, “Prepartum and postpartum open-ﬁeld behavior and
maternal responsiveness in mice bidirectionally selected for
open-ﬁeld thigmotaxis,” Journal of General Psychology, vol.
135, no. 1, pp. 37–53, 2008.
 Anonymous, Pharmacop´
eopathique Franc¸aise, Saint-
Denis Cedex, 10th edition, 2002.
 C. Venard, N. Boujedaini, P. Belon, A. G. Mensah-Nyagan,
and C. Patte-Mensah, “Pharmacological modulators of the
glycinergic system regulate allopregnanolone biosynthesis in
the rat spinal cord,” in Fourth International Meeting Sterids
and Nervous System, G. C. Panzica and S. Gotti, Eds., p. 262,
Instituto Cajal, Madrid, Spain, 2007.
 C. Venard, N. Boujedaini, P. Belon, A. G. Mensah-Nyagan,
and C. Patte-Mensah, “Regulation of neurosteroid allopreg-
nanolone biosynthesis in the rat spinal cord by glycine and the
alkaloidal analogs strychnine and gelsemine,” Neuroscience,
vol. 153, no. 1, pp. 154–161, 2008.
 C. Belzung and G. Griebel, “Measuring normal and patho-
logical anxiety-like behaviour in mice: a review,” Behavioural
Brain Research, vol. 125, no. 1-2, pp. 141–149, 2001.
 L. Prut and C. Belzung, “The open ﬁeld as a paradigm to
measure the eﬀects of drugs on anxiety-like behaviors: a
review,” European Journal of Pharmacology, vol. 463, no. 1–3,
pp. 3–33, 2003.
 M. Hasco¨
et, M. Bourin, and B. A. Nic Dhonnchadha, “The
mouse light-dark paradigm: a review,” Progress in Neuro-
Psychopharmacology and Biological Psychiatry,vol.25,no.1,
pp. 141–166, 2001.
 S. A. Gordinho Pinto, E. Bohland, C. de Paula Coelho, M. S.
Furquim de Azevedo Morgulis, and L. V. Bonamin, “An animal
model for the study of Chamomilla in stress and depression:
pilot study,” Homeopathy, vol. 97, no. 3, pp. 141–144, 2008.
 A. Ennaceur, S. Michalikova, R. van Rensburg, and P. L.
Chazot, “Are benzodiazepines really anxiolytic?. Evidence
Research, vol. 188, no. 1, pp. 136–153, 2008.
 M. Nazar, M. Jessa, and A. Płtaznik, “Benzodiazepine-
GABA(A) receptor complex ligands in two models of anxiety,”
Journal of Neural Transmission, vol. 104, no. 6-7, pp. 733–746,
 W. X. Kong, S. W. Chen, Y. L. Li et al., “Eﬀects of taurine
on rat behaviors in three anxiety models,” Pharmacology
Biochemistry and Behavior, vol. 83, no. 2, pp. 271–276, 2006.
 J. N. Crawley, “Exploratory behavior models of anxiety in
mice,” Neuroscience and Biobehavioral Reviews, vol. 9, no. 1,
pp. 37–44, 1985.
 E. Choleris, A. W. Thomas, M. Kavaliers, and F. S. Prato,
“A detailed ethological analysis of the mouse open ﬁeld test:
eﬀects of diazepam, chlordiazepoxide and an extremely low
frequency pulsed magnetic ﬁeld,” Neuroscience and Biobehav-
ioral Reviews, vol. 25, no. 3, pp. 235–260, 2001.
 M. Bastide, “Immunological examples on ultra high dilution
research,” in Ultra High Dilution,P.C.EndlerandJ.Schulte,
Eds., pp. 27–33, Kluwer Academic Publishers, Dordrecht, The
 P. Bellavite, R. Ortolani, and A. Conforti, “Immunology and
homeopathy. 3. Experimental studies on animal models,”
Evidence-Based Complementary and Alternative Medicine, vol.
3, no. 2, pp. 171–186, 2006.
 M. P. da Silva Rocha, F. M. Soares, L. C. Martini, and L. V.
Bonamin, “Behavior of rats treated with Rhus toxicodendron
200 CH,” International Journal of High Dilution Research, vol.
7, pp. 3–6, 2008.
 P. Bellavite, A. Conforti, F. Pontarollo, and R. Ortolani,
“Immunology and homeopathy. 2. Cells of the immune
system and inﬂammation,” Evidence-Based Complementary
and Alternative Medicine, vol. 3, no. 1, pp. 13–24, 2006.
 S. Chirumbolo, M. Brizzi, R. Ortolani, A. Vella, and P.
Bellavite, “Inhibition of CD203c membrane up-regulation in
human basophils by high dilutions of histamine: a controlled
replication study,” Inﬂammation Research, vol. 58, no. 11, pp.
 E. Belougne-Malfatti, O. Aguejouf, F. Doutremepuich, P.
Belon, and C. Doutremepuich, “Combination of two doses of
acetyl salicylic acid: experimental study of arterial thrombo-
sis,” Thrombosis Research, vol. 90, no. 5, pp. 215–221, 1998.
10 Evidence-Based Complementary and Alternative Medicine
 V. Brown and M. Ennis, “Flow-cytometric analysis of basophil
activation: inhibition by histamine at conventional and home-
opathic concentrations,” Inﬂammation Research, vol. 50, no. 2,
pp. S47–S48, 2001.
 D. Marotta, A. Marini, K. Banaudha, S. V. M. Maharaj,
and W. B. Jonas, “Nonlinear eﬀects of glutamate and KCl
on glutamate toxicity in cultured rat cerebellar neurons,”
International Journal of Neuroscience, vol. 113, no. 4, pp. 491–
 P. Belon, J. Cumps, M. Ennis et al., “Histamine dilutions
modulate basophil activation,” Inﬂammation Research, vol. 53,
no. 5, pp. 181–188, 2004.
 F. X. Eizayaga, O. Aguejouf, P. Belon, and C. Doutremepuich,
“Platelet aggregation in portal hypertension and its mod-
iﬁcation by ultra-low doses of aspirin,” Pathophysiolo gy of
Haemostasis and Thrombosis, vol. 34, no. 1, pp. 29–34, 2005.
 G. Ruiz-Vega and G. Estevez-Delgado, “Non-linearity mod-
eling of ultra-dilutions: the histamine disturbances case,” in
Signals and Images. Contributions and Contradictions about
High Dilution Research, L. V. Bonamin, Ed., pp. 67–82,
Springer Science, Dordrecht, The Netherlands, 2008.
 C. M. Witt, M. Bluth, H. Albrecht, T. E. R. Weißhuhn, S.
Baumgartner, and S. N. Willich, “The in vitro evidence for an
eﬀect of high homeopathic potencies—a systematic review of
the literature,” Complementary Therapies in Medicine, vol. 15,
no. 2, pp. 128–138, 2007.
 R. Arani, I. Bono, E. Del Giudice, and G. Preparata, “QED
coherence and the thermodynamics of water,” International
JournalofModernPhysicsB, vol. 9, no. 15, pp. 1813–1841,
 E. Del Giudice and G. Preparata, “Coherence electrodynamics
in water,” in Fundamental Research in Ultrahigh Dilution
and Homeopathy, J. Schulte and C. Endler, Eds., pp. 89–100,
Kluwer, Dordrecht, The Netherlands, 1998.
 C. W. Smith, “Quanta and coherence eﬀects in water and living
systems,” Journal of Alternative and Complementary Medicine,
vol. 10, no. 1, pp. 69–78, 2004.
 D. Mastrangelo, “Hormesis, epitaxy, the structure of liquid
water, and the science of homeopathy,” Medical Science
Monitor, vol. 13, no. 1, pp. SR1–SR8, 2007.
 M. L. Rao, R. Roy, I. R. Bell, and R. Hoover, “The deﬁning
role of structure (including epitaxy) in the plausibility of
homeopathy,” Homeopathy, vol. 96, no. 3, pp. 175–182, 2007.
 J.-L. Demangeat, P. Gries, B. Poitevin et al., “Low-ﬁeld NMR
water proton longitudinal relaxation in ultrahighly diluted
aqueous solutions of silica-lactose prepared in glass material
for pharmaceutical use,” Applied Magnetic Resonance, vol. 26,
no. 4, pp. 465–481, 2004.
 J.-L. Demangeat, “NMR water proton relaxation in unheated
and heated ultrahigh aqueous dilutions of histamine: evidence
for an air-dependent supramolecular organization of water,”
Journal of Molecular Liquids, vol. 144, no. 1-2, pp. 32–39, 2009.
 V. Elia and M. Niccoli, “New physico-chemical properties
of extremely diluted aqueous solutions,” Journal of Thermal
Analysis and Calorimetry, vol. 75, no. 3, pp. 815–836, 2004.
 V. Elia, E. Napoli, and R. Germano, “The ’memory of
water’: an almost deciphered enigma. Dissipative structures in
extremely dilute aqueous solutions,” Homeopathy, vol. 96, no.
3, pp. 163–169, 2007.
 M. F. Chaplin, “The memory of water: an overview,” Home-
opathy, vol. 96, no. 3, pp. 143–150, 2007.
 E. J. Calabrese, “U-shaped dose response in behavioral
pharmacology: historical foundations,” Critical Reviews in
Tox i c o lo g y , vol. 38, no. 7, pp. 591–598, 2008.
 E. J. Calabrese, “An assessment of anxiolytic drug screening
tests: hormetic dose responses predominate,” Critical Reviews
in Toxicology, vol. 38, no. 6, pp. 489–542, 2008.
 M. Oberbaum, N. Samuels, S. R. Singer, and E. J. Calabrese,
“Hormesis is not homeopathy (multiple letters),” To xi c o lo g y
and Applied Pharmacology, vol. 206, no. 3, pp. 365–366, 2005.
 S. Bernardini and A. Dei, “Hormesis may provide a cen-
tral concept for homeopathy development,” To x ic o l og y a nd
Applied Pharmacology, vol. 211, no. 1, pp. 84–85, 2006.
 P. Bellavite, R. Ortolani, F. Pontarollo, G. Pitari, and A.
Conforti, “Immunology and homeopathy. 5. The rationale of
the ’Simile’,” Evidence-Based Complementary and Alternative
Medicine, vol. 4, no. 2, pp. 149–163, 2007.
 M. P. Mattson, “Awareness of hormesis will enhance future
research in basic and applied neuroscience,” Critical Reviews
in Toxicology, vol. 38, no. 7, pp. 633–639, 2008.