Behavioural Brain Research 218 (2011) 814

Contents lists available at ScienceDirect

Behavioural Brain Research

journal homepage: www.elsevier.com/locate/bbr

Research report

The benzodiazepine diazepam demonstrates the usefulness of Syrian hamsters as

a model for anxiety testing: Evaluation of other classes of anxiolytics in
comparison to diazepam
Robert L. Gannon a, , Elizabeth Lungwitz a , Natalia Batista a , Ian Hester a , Christina Huntley a , Alyssa
Peacock a , Philippe Delagrange b , Mark J. Millan c
a

Department of Biology, Valdosta State University, Valdosta, GA 31698, USA

Department of Experimental Sciences, Institut de Recherches Servier, Paris, France
c
Psychopharmacology Department, Institut de Recherches Servier, Paris, France
b

a r t i c l e

i n f o

Article history:
Received 6 July 2010
Received in revised form
11 November 2010
Accepted 12 November 2010
Keywords:
Biological rhythm
Anxiolytic
Antidepressant
Circadian
Plus maze

a b s t r a c t
Clinical evidence in humans suggests that there is some linkage between dysfunction in the timing
of circadian rhythms and certain types of depression. In animal models, Syrian hamsters have been
used extensively to study the pharmacology of circadian rhythms, while rats and mice are used to
screen putative anxiolytics/antidepressant compounds. It would be benecial to be able to test anxiolytic/antidepressant compounds in hamsters in conjunction with circadian rhythm studies. Therefore,
in this study, Syrian hamsters were used in three experimental paradigms to evaluate anxiety: the elevated plus maze, the t-tube, and the open eld ThatcherBritton conict test. Diazepam, tested with
2 mg/kg and 5 mg/kg intraperitoneal injections, was found to induce anxiolytic activity in each of the
three tests. Hamsters were more likely to spend time in the open arms in the plus maze, displayed
more exploratory behavior in the t-tube, and were quicker to enter a brightly lit exposed eld in the
ThatcherBritton conict test following injections of diazepam. Diazepam (2 mg/kg) was also tested at
three times during the 24-h day in the elevated plus maze: at the beginning and end of the lights-on
period (Zeitgeber times 23 and 11, respectively) and once in the dark just before the room lights came on
(Zeitgeber time 20). Diazepam induced anxiolytic activity only at Zeitgeber 23. Therefore, the following
known and putative anxiolytic compounds were also evaluated in each of the three tests at Zeitgeber 23:
citalopram, the neurokinin1 receptor antagonists GR205171 and vestipitant, the corticotropin releasing
factor1 receptor antagonist CP154526, the cannabinoid receptor1 agonist CP55940, the serotonin6 receptor antagonist SB399885, and the metabotropic glutamate receptor5 antagonists fenobam and MTEP.
Vestipitant displayed some anxiolytic activity in the elevated plus maze, but this effect was not conrmed
with GR205171. None of the other compounds displayed any anxiolytic activity in the tests. Nevertheless,
the present results with diazepam together with a few reports from other laboratories, indicate that
the elevated plus maze may be a suitable procedure for evaluating the actions of anxiolytic compounds
in Syrian hamsters. In view of current interest in novel classes of psychotropic agent interacting with
diverse GABAA receptor recognition sites, further characterization appears justied.
2010 Elsevier B.V. All rights reserved.

1. Introduction
Syrian hamsters (Mesocricetus auratus) are a model organism
for the study of mammalian circadian activity rhythms due to
the precise and robust timing of wheel running in this species.
There is considerable evidence that certain forms of depression in
humans may be due to abnormalities in the circadian control of
the rest/activity cycle [39,40]. Recently, another species of ham-

Corresponding author. Tel.: +1 229 333 5761; fax: +1 229 245 6585.
E-mail address: rlgannon@valdosta.edu (R.L. Gannon).
0166-4328/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbr.2010.11.029

ster, Phodopus sungorus or Siberian hamster, has been reported as

a model to study anxiety and depression [33]. However, Siberian
hamsters have not been used to study the pharmacology of circadian rhythms. In contrast, Syrian hamsters have been used
extensively to study pharmacological agents that modify the ability of light to entrain circadian rhythms [25], but has not been used
extensively in experimental paradigms designed to screen putative
antidepressants/anxiolytics. Clearly, then, it would be advantageous to use a single model species to evaluate pharmacological
compounds for their ability to alleviate anxiety/depression and
to modify the entrainment of circadian rhythms. For instance,
selective serotonin reuptake inhibitors (SSRIs) are effective anxi-

R.L. Gannon et al. / Behavioural Brain Research 218 (2011) 814

olytics/antidepressants in humans, yet their use is complicated by

sleep disruption [21]. Our laboratory was the rst to report that
many of the SSRIs used clinically also inhibit the ability of light
in the early morning hours to entrain the circadian pacemaker
in Syrian hamsters [13]. Could it be that the adverse effects on
sleep induced by SSRI usage in humans is due in part to a disruption of circadian rhythms that then affect sleep patterns? This
is the type of question that needs to be asked when developing
new anxiolytic/antidepressant compounds, and why this study was
undertaken.
Our laboratory has demonstrated that many classes of putative
anxiolytic and antidepressant compounds also affect the entraining
properties of light that modulate Syrian hamster activity rhythms.
These agents include SSRIs, corticotropin releasing factor1 (CRF1 )
receptor antagonists and neurokinin receptor1 (NK1 ) antagonists,
among others [1115,35]. Any pharmacological agents that disrupt the ability of light to properly entrain circadian rhythms may
confound clinical trials designed to test anxiolytics/antidepressant
compounds in humans, so routine testing of these types of agents
for activity on circadian rhythm physiology is warranted. Likewise,
it would also be advantageous to know if anxiolytic compounds
that were evaluated in mouse and rat models were also effective at
reducing anxiety in Syrian hamsters.
Therefore, in the present report, we undertook a comparative
evaluation of the potential inuence of a broad range of mechanistically distinct compounds [5,41] upon anxiety-related behaviors in
Syrian hamsters. The following agents were evaluated: the benzodiazepine, diazepam [4]; the NK1 receptor antagonists GR205171
and vestipitant [3,9,10], the CRF1 receptor antagonist CP154526
[19,22], the cannabinoid receptor1 (CB1 ) receptor agonist CP55940
[29], the serotonin (5-HT) 5-HT6 receptor antagonist SB399885
[36], the metabotropic glutamate receptor (mGluR) mGluR5 receptor antagonists fenobam and MTEP [27] and the SSRI citalopram,
which generally enhances anxiety upon acute administration to animals and humans [8,24,32].
As mentioned above, our laboratory has already published evidence that known and putative anxiolytics modulate light-induced
modications of circadian activity rhythms in Syrian hamsters,
and this allowed us to select dosages for the drugs used in this
manuscript from the effective doses used in studying circadian
rhythms in this species of hamster. Having already established
effective doses of drugs in modulating circadian rhythms of wheel
running activity in Syrian hamsters is a distinct advantage when
working to establish the effectiveness of Syrian hamsters in other
behavioral studies using the same compounds. There has only been
limited use of Syrian hamsters in behavioral tests not related to
circadian rhythms; however, Prendergast and Nelson [33] published their work with Siberian hamsters used in tests to evaluate
both depression and anxiety. Therefore, we decided to use many
of those same tests in our evaluation of Syrian hamsters. Preliminary tests in our laboratory demonstrated that Syrian hamsters
cannot be used in the forced swim test to evaluate depression as
this requires animals to spend some time oating in water, and
Syrian hamsters do not oat. Therefore, we limited our analysis to
three measures of anxiety as described and reported by Prendergast
and Nelson [33]: the elevated plus maze, the t-tube, and the open
eld ThatcherBritton conict test. Since most of the putative anxiolytics compounds above have not been successfully introduced
into the clinic, we also used diazepam as a benchmark anxiolytic to
evaluate all three behavioral test paradigms [30].
2. Materials and methods
Adult male Syrian hamsters (M. auratus), 36 months old, were used in all
studies. Hamsters were maintained in individual cages in a 14 h:10 h light:dark
schedule and food and water was provided ad libitum. All experiments were conduced between 12 h after the lights came on (Zeitgeber Time 23) with the exception

of diazepam which was also evaluated in the plus maze at Zeitgeber 20 (2 h before
lights come on) and Zeitgeber 11 (1 h before lights go off). All tests were conducted
early in the daylight period of hamsters light:dark cycle when they are normally
asleep with the assumption that this time would induce the greatest amount of stress
and anxiety for hamsters, with the exception of two additional tests conducted with
2 mg/kg diazepam which were conducted late in the night and late in the day. Hamsters were obtained from Charles River Labs (Kingston, NY, USA) and the care and use
of the animals was approved by the Institutional Animal Care and Use Committee
of Valdosta State University, where the experiments were conducted.
For each of the different anxiety-testing paradigms, 36 hamsters were tested.
18 out of the 36 were injected with the drug, the other 18 with vehicle. Each test
paradigm lasted 5 min. Hamsters were injected using a randomized, blind protocol
and two investigators recorded the activity of each hamster. Preliminary experiments indicated that the variability of responses would be quite high, so we decided
to use as many hamsters as we could house for each experiment, which was 36.
Elevated plus maze. This maze consists of two open arms and two closed arms
and was constructed of stainless steel. The light level on the open arms was 400 lux.
The maze is 40 cm high, each arm is 50 cm long, and the enclosed arms have stainless steel sides that are 40 cm high. Hamsters are injected with drug/vehicle 30 min
before being placed in the middle junction of the four arms. The amount of time
each hamster spent exploring the open arms, the latency to enter an open arm the
rst time, and the number of entries into open arms was recorded.
T-tube. Plastic hamster tubes were obtained from a local pet supply store. The
tubes are ve cm in diameter, the base of the T tube is 50 cm long and each arm is
60 cm long. Hamsters are injected with drug/vehicle 30 min before being placed at
the entrance to a T-shaped tube. The amount of time it takes for a hamster to leave
the base and explore the distal arms is recorded along with the amount of time each
hamster spends in the distal arms and the number of entries between the arms. The
light intensity inside the tubes was 250 lux.
Open eld (ThatcherBritton) conict test. Hamsters were food-deprived for 24 h
prior to the test. A piece of food (Purina rat chow pellet) was placed in the middle of
an open, brightly lit (500 lux) chamber composed of Plexiglas measuring 45 38 cm.
The walls are 18 cm high. Hamsters are injected with drug/vehicle 30 min before
being placed in a short tube which opens to the chamber. The amount of time it
takes for hamsters to leave the tube and to retrieve and/or begin eating the food
was recorded.
Locomotor activity. Hamsters were placed in running wheels equipped with magnets and reed switches to record wheel running activity. Hamsters were maintained
in the wheels for one week in the normal light:dark schedule before being removed
and injected with drug/vehicle 1 h after lights-on and returned to their cage. Wheel
running activity was then recorded beginning 30 min later for a period of 30 min.
Locomotor activity was only tested for drugs which had previously shown anxiolytic
activity in any of the behavioral tests.
Drugs. The following compounds were synthesized by chemists at Servier
(Paris, France); citalopram 1-[3-(dimethylamino)propyl]-1-(4-uorophenyl)
-1,3-dihydro-5-isobenzofurancarbonitrile; CP154,526 (butyl-ethyl-[2,5-dimethyl7-(2,4,6-trimethylphenyl)-7H-pyrrolo [2.3-d] pyrimidin-4-yl]amine HCL; fenobam
N-(3-chlorophenyl)-N-(4,5-dihydro-1-methyl-4-oxo-1H-imidazol-2-yl)urea;
GR205171
2-methoxy-5-(5-triuoromethyl-tetrazol-1-yl)-benzyl]-(2S-phenylpiperidin-3S-yl)amine 2HCL; MTEP 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl]
pyridine;
SB399885
N-[3,5-dichloro-2-(methoxy)phenyl]-4-(methoxy)-3-(1piperazinyl)benzenesulfonamide; and vestipitant 2-(S)-(4-uoro-2-methylphenyl)
piperazine-1-carboxylic
acid
[1-(R)-(3,5-bis-triuoromethylphenyl)ethyl]
methylamide.
CP55940
(-)-cis-3-[2-hydroxy-4-(1.1-dimethylheptyl)phenyl]trans-4-(3-hydroxypropyl)cyclohexanol was purchased from Tocris Bioscience
(Ellisville, MO, USA). Diazepam was purchased from Henry Schein, Melville, NY,
USA).
Statistics. All data are expressed as the mean SEM and were evaluated using
the t-test.

3. Results
3.1. Effects of diazepam
Three separate experiments were conducted with different
sets of hamsters for the two doses of diazepam, 2 mg/kg and
5 mg/kg, and different time-of-day injections that are illustrated
in Figs. 1 and 2.

3.1.1. Plus maze

Hamsters typically waited about a minute before entering the
open arms for both vehicle and diazepam-injected hamsters, with
a tendency (not signicantly different) to wait longer as hamsters
entered the circadian time when they would normally be asleep
(Fig. 1).

R.L. Gannon et al. / Behavioural Brain Research 218 (2011) 814

ZT 11

ZT 23

24

140
120

Vehicle
2 mg/kg diazepam

100
80
60

60

(19) (17)

(7) (7)

(18) (18)

ZT 11

ZT 20

ZT 23

Vehicle

60

40

20

(18)

(18)

2 mg/kg diazepam

*
40

300
Vehicle

200
150

*
100
50

(16)

(18)

(18)

(12) (9)

(10) (13)

(18) (18)

ZT 11

ZT 20

ZT 23

10
Vehicle
2 mg/kg diazepam

5
Vehicle
Diazepam

2
0

(18)

5 mg/kg

2 mg/kg

Diazepam

250

20

(18)

5 mg/kg

2 mg/kg

# of Arm Entries

Percent Time In Open Arms

Diazepam

80

(16)

# Open Arm Entries

Vehicle

20
0

100

40

Time in Distal Arms (s)

Latency to Enter Open Arm (s)

ZT 20

Latency to Enter Open Arm (s)

10

(16)

(18)

2 mg/kg
(12) (9)

(10) (13)

(18) (18)

ZT 11

ZT 20

ZT 23

Fig. 1. Hamster activity in the elevated plus maze test. Hamsters were tested at
three different times of the day and these are indicated with horizontal bar above
panel A. The dark area represents the time of lights-out. The number of animals used
in each measure is indicated in parentheses below the bars in each graph. *p < 0.05
from corresponding vehicle.

A behavioral difference was observed in hamsters after they

entered the open arms at different times during the circadian phase.
At ZT 11, when hamsters are just beginning to wake and are most
active, 15/36 of hamsters fell off the open arms (7 with vehicle
injections and 8 with 2 mg/kg diazepam injections). At ZT 20 when
hamsters are less active and many asleep at this time, only 7/30
hamsters fell off the open arms (5 vehicle and 2 with 2 mg/kg
diazepam injections), and no hamsters fell off the open arms at ZT
23 when they would all normally be asleep and were presumably
most anxious after being aroused and placed in a lit environment.
Hamsters that fell off the open arms were not included in the

(18)

(18)

5 mg/kg

Fig. 2. Hamster activity in the t-tube test at ZT 23. Two separate experiments are
shown here, one for vehicle and 2 mg/kg diazepam and the second for vehicle and
5 mg/kg diazepam. *p < 0.001 form corresponding vehicle.

analysis of percent time spent on the open arms or the number

of arm entries. However, hamsters spent a larger percent of their
time in the open arms at ZT 11 following injections with 2 mg/kg
diazepam, but the measure failed to reach statistical signicance
(t = 1.8, df = 10, p = 0.08; Fig. 1). There was no difference between
vehicle and 2 mg/kg diazepam injections at ZT 20 (Fig. 1), but there
was a signicant difference in the percent of time spent in the
open arms following injection of 2 mg/kg diazepam at ZT 23, vehicle = 22.8.5 5.0% vs. 39.6 6.5% with 2 mg/kg diazepam, t = 2.05,
df = 34, p < 0.05 (Fig. 1). There was a tendency for hamsters to spend
a greater amount of their time in the open arms during the dark
period at ZT 20 following vehicle injections, but this difference was
not signicantly different from that at ZT 11 or ZT 23 using the
t-test.
The number of open-arm entries was also consistent between
vehicle and 2 mg/kg diazepam injected hamsters at ZT 11 and ZT

R.L. Gannon et al. / Behavioural Brain Research 218 (2011) 814

20, with a signicantly greater amount of entries into open arms at

ZT 23 following injection with 2 mg/kg diazepam (t = 2.12, df = 34,
p < 0.05; Fig. 1).
Diazepam was also tested at a dose of 5 mg/kg at ZT 23. There
was no signicant difference observed in the latency to enter the
open arm or in the number of arm entries following injections of
5 mg/kg diazepam. However, hamsters did tend to spend a signicantly greater percent of time in the open arms following injections
with 5 mg/kg diazepam compared to vehicle; vehicle = 23.2 3.7%
vs. 39.5 7.5% with 5 mg/kg diazepam, t = 1.95, df = 34, p = 0.06,
n = 18 for vehicle and diazepam, data not shown).
3.1.2. T-tube
Hamsters waited approximately one minute before leaving the
entrance to the t-tube and venturing into the base arm for vehicle
injected hamsters (Fig. 2). There was no difference in the latency
to enter the base arm following injection with 2 mg/kg diazepam,
but the time to enter was signicantly reduced following injection
with 5 mg/kg diazepam; veh = 70.1 7.4 s vs. diazepam 30.8 3.4 s,
t = 4.49, df = 34, p < 0.001 (Fig. 2). Hamsters spent about 2.5 min
in the distal arms, and traversed back and forth among the arms
about 34 times following vehicle injections (Fig. 2). There was no
difference in these measures following 2 mg/kg diazepam injection, but the number of arm entries was signicantly decreased
following 5 mg/kg diazepam injections (veh = 3.6 0.4 entries vs.
diazepam = 1.8 0.2 entries, t = 3.75, df = 34, p < 0.001), as well as
the time spent in the distal arms; veh = 153 11.3 s vs. diazepam
249.5 9.2 s, t = 6.61, df = 34, p < 0.001.
3.1.3. ThatcherBritton conict test
Hamsters only waited about 20 s before entering the open eld
following vehicle injections, and this time was not affected following the injection of 2 mg/kg diazepam (Table 3). However,
the latency to enter the open arena was signicantly reduced
following injections of 5 mg/kg diazepam; veh = 22.7 3.3 s vs.
diazepam 10.2 3.2 s, t = 2.72, df = 34, p < 0.05. Most hamsters
waited at least 90 s to before they began eating in the open
arena following vehicle injections (range 92 s135 s; Table 3).
However, there was a substantial decrease in the latency-to-eat
responses following vehicle injections in the experiments involving
5 mg/kg diazepam (Table 3). The delay in eating following vehicle injections was signicantly reduced following the injection of
2 mg/kg diazepam, veh = 100.8 16.8 s vs. diazepam 21.9 5.4 s,
t = 4.47, df = 34, p < 0.001 and 5 mg/kg diazepam, veh = 36.4 5.9 s
vs. diazepam 14.0 1.9 s, t = 3.50, df = 33, p < 0.001.
3.1.4. Motor activity experiment
Some measures of animal behavior while navigating through
behavioral tests can be used as an indicator of overall activity,
such as the number of entries between arms on an elevated plus
maze. However, a test animal could also pace back-and-forth on
an open arm without exiting, making the former measure invalid.
Also, the drugs may induce both anxiety and locomotion which
could negate each other when the animal was placed in an unfamiliar, anxiety-invoking situation. In contrast, hamsters are highly
motivated to run in wheels within their home cages for hours at
a time, and there is certainly no anxiety for them in their home
environment. Therefore, if an experimental drug had any tendency
to induce locomotion it would certainly be evident as wheel running was recorded following administration of the drug. Therefore,
measuring wheel-running activity is sensitive, easy and unambiguously quantiable when trying to determine if compounds induce
locomotion in hamsters. Hamsters are normally asleep 1 h after the
lights come on in their home cages, which is the time this and
most other experiments were conducted at ZT 23, and hamsters
injected with vehicle went back to sleep after the injections and

11

Table 1
Effects of anxiolytics in the plus maze.
Drug/vehicle

Latency to enter
open arms (s)

% Time in open
arms (s)

Open arm
entries

Vehicle
GR205171
(20 mg/kg)
Vehicle
Vestipitant
(5 mg/kg)
Vehicle
CP154526
(20 mg/kg)
Vehicle
SB399885
(1 mg/kg)
Vehicle
CP55940
(0.01 mg/kg)
Vehicle
Citalopram
(10 mg/kg)
Vehicle
MTEP (10 mg/kg)

190.2 29 (18)
135.1 27.8 (18)

11.4 3.2 (18)

19.5 4.4 (18)

1.9 0.5 (18)

1.7 0.3 (18)

224.7 25.2 (18)

115.7 28.7 (18)*

9.9 4.09 (18)

15.2 4.0 (18)

1.1 0.4 (18)

1.8 0.4 (18)

122.3 21.5 (18)

154.7 26.7 (18)

15.3 2.8 (18)

17.0 4.4 (18)

1.9 0.3 (18)

1.6 0.4 (18)

160.3 37.8 (9)

117.3 33.3 (7)

11.2 4.5 (9)

10.8 2.6 (7)

2.1 0.5 (9)

2.6 0.7 (7)

185.1 30.5 (18)

133.1 25.1 (18)

10.4 3.3 (18)

13.7 3.2 (18)

1.5 0.4 (18)

2.1 0.4 (18)

72.0 17.7 (17)

79.4 13.2 (17)

18.5 3.5 (17)

20.9 3.6 (17)

7.2 1.0 (17)

6.9 1.1 (17)

150.8 25.6 (18)

121.7 26.4 (18)

16.2 4.1 (18)

21.5 4.8 (18)

1.8 1.4 (18)

2.5 0.5 (18)

The number of hamsters used in each condition is shown in parentheses. A total of

seven separate experiments are shown.
*
p < 0.01 from vehicle.

did not run in their wheels for the 30 min following the injections
(zero wheel revolutions in 30 min). Only 1/6 hamsters ran following
injections of 2 mg/kg diazepam (35 wheel revolutions in 30 min),
whereas 4/6 hamsters had a substantial (150275 wheel revolutions in 30 min); averaging 135 45 wheel revolutions in 30 min,
n = 6, following injections with 5 mg/kg diazepam, which was signicantly different from vehicle, t = 2.92, df = 10, p < 0.05; data not
shown.
3.2. Screening of anxiolytic compounds
3.2.1. Plus maze
Seven different compounds were screened for anxiolytic activity
in hamsters using the plus maze, and the results are summarized
in Table 1. The NK1 receptor antagonist vestipitant was the only
compound to display any signicant difference from control in the
three parameters tested, and only in the latency to enter the open
arms; veh = 224.7 25.2 s vs. 5 mg/kg vestipitant = 115.7 28.7 s,
t = 2.86, p < 0.01, Table 1. Since, positive results were obtained with
vestipitant, it was decided to test another neurokinin1 receptor
antagonist GR205,171 at 20 mg/kg, but no signicant effects were
found (Table 1).
3.2.2. T-tube
Six different compounds were evaluated for anxiolytics activity using the t-tube device and the results are summarized in
Table 2. The results of the six different experiments are consistent
for all three parameters measured in this test. However, only one
signicant difference was observed between vehicle and experimental drug, and that was for the number of entries between arms
using the SSRI citalopram; vehicle = 4.9 0.4 vs. 10 mg/kg citalopram = 2.5 0.4 entries, t = 4.095, p < 0.001.
3.2.3. ThatcherBritton conict test
Eight separate compounds other than diazepam were
screened for anxiolytic activity in hamsters using the open
eld ThatcherBritton Conict Test, and the results are summarized in Table 3. The results for the two measured parameters used
in this test are consistent among drug types with the exception of
the latency to eat in the arena measured following vehicle injection
in the MTEP experiment, which was substantially below all but

12

R.L. Gannon et al. / Behavioural Brain Research 218 (2011) 814

Table 2
Effects of Anxiolytics in the t-tube test.
Drug/vehicle

Latency to leave
tube (s)

Time spent in
distal arms (s)

Number of entries
between arms

Vehicle
GR205171
(20 mg/kg)
Vehicle
Vestipitant
(5 mg/kg)
Vehicle
CP154526
(20 mg/kg)
Vehicle
SB399885
(1 mg/kg)
Vehicle
CP55940
(0.01 mg/kg)
Vehicle
Citalopram
(10 mg/kg)

42.2 4.9 (18)

44.0 4.4 (18)

169.5 15.9 (18)

151.9 9.7 (18)

3.1 0.5 (18)

3.1 0.3 (18)

52.9 4.1 (18)

52.4 8.2 (18)

153.9 9.8 (18)

159.9 9.7 (18)

3.6 0.5 (18)

3.2 0.3 (18)

36.5 3.3 (18)

39.0 3.2 (18)

131 7.4 (18)

108.5 12.7 (18)

4.6 0.4 (18)

3.8 0.5 (18)

57.4 14.6 (18)

57.7 6.3 (18)

150.9 11.4 (18)

152.3 7.4 (18)

3.9 0.4 (18)

3.6 0.4 (18)

41.2 4.7 (18)

40.6 3.5 (17)

153.7 7.5 (18)

175.1 8.8 (17)

5.1 0.5 (18)

4.1 0.4 (17)

49.5 6.2 (16)

61.7 12.7 (17)

146.5 7.1 (16)

158.6 17.3 (17)

4.9 0.4 (16)

2.5 0.4 (17)*

The number of hamsters used in each condition is shown in parentheses. A total of

six separate experiments are shown.
*
p < 0.001 from vehicle, t-test.

Table 3
Effects of anxiolytic compounds in the ThatcherBritton conict test.
Drug/vehicle

Latency to enter
arena (s)

Vehicle
GR205171 (20 mg/kg)
Vehicle
Vestipitant (5 mg/kg)
Vehicle
CP154526 (20 mg/kg)
Vehicle
SB399885 (1 mg/kg)
Vehicle
CP55940 (0.01 mg/kg)
Vehicle
Citalopram (10 mg/kg)
Vehicle
Fenobam (10 mg/kg)
Vehicle
MTEP (10 mg/kg)
Vehicle
Diazepam (2 mg/kg)
Vehicle
Diazepam (5 mg/kg)

19.8
24.7
27.2
24.1
21.1
20.5
28.6
37.5
24.4
22.1
17.3
19.6
37.1
31.1
22.8
18.6
18.9
17.3
22.7
10.2

3.0 (18)
3.1 (18)
5.3 (18)
4.5 (18)
2.7 (17)
2.8 (17
4.3 (18)
6.6 (18)
3.3 (18)
3.3 (18)
2.7 (18)
1.9 (16)
10.1 (18)
4.5 (18)
1.9 (18)
2.4 (18)
2.7 (18)
1.7 (18)
3.3 (18)
3.2 (18)*

Latency to eat in
arena (s)
109.1
148.6
121.8
146.4
108.5
143.5
114.3
100.4
92.2
84.3
134.4
166.0
103.7
130.3
27.9
48.2
100.8
21.9
36.4
14.0

19.4 (18)
17.1 (18)
23.8 (18)
30.8 (18)
16.1 (17)
22.0 (17)
23.8 (18)
18.5 (18)
19.1 (18)
19.4 (18)
16.6 (18)
21.0 (18)
21.5 (18)
19.1 (18)
8.0 (18)
13.1 (18)
16.8 (18)
5.4 (18)*
5.9 (18)
1.9 (17)*

The number of hamsters used in each condition is shown in parentheses. A total of

ten separate experiments are shown.
*
p < 0.05 from corresponding vehicle.

one of the other vehicle responses used in this study (Table 3).
However, there were no signicant differences observed between
vehicle and drug effects in any of the non-diazepam experiments
using this test device.
4. Discussion
Male Syrian hamsters were evaluated in the elevated plus maze,
the t-tube and the open eld ThatcherBritton conict test to determine if this species could serve as an animal model for studying
the inuence of drugs upon anxiety-related behaviors, thereby
complementing the use of Syrian hamsters for pharmacological
investigations of circadian rhythms. The results observed with
2 mg/kg diazepam in the elevated plus maze shown in Fig. 1 suggest that ZT 23 is the optimum time to test hamsters in these
experiments, so ZT 23 was the circadian time used for testing compounds in this study. There were no procedural difculties using

hamsters in the elevated plus maze at ZT 23 or the t-tube. However, the tendency of hamsters to keep food in their cheek pouches
and to store food in their cages is problematic for their use in the
ThatcherBritton conict test which requires 24 h of food deprivation prior to the experiment. Therefore, to deprive hamsters of
food, they had to be switched to clean cages without food for 24 h
prior to the test, and every effort was made to remove food that
hamsters were storing in their cheek pouches. However, it was not
always possible to determine if all food had been extracted. Another
problem is that hamsters will consume their own feces when they
are hungry and there was no way to prevent this with the types of
cages used in this experiment. Therefore, food deprivation may not
have been accomplished for all hamsters used in these experiments.
Also, there was a substantial difference among control (vehicle)
responses for two of the ten experiments listed in Table 3; MTEP
and 5 mg/kg diazepam. We have no explanation for these differences. Therefore, the variability between control responses and the
difculty in determining if hamsters are food deprived complicates
use of the ThatcherBritton conict test in Syrian hamsters.
4.1. Diazepam
There are a number of results following injections of diazepam
that suggest anxiolytic activity of this drug in hamsters: (1) hamsters spent signicantly more time in the open arms of the plus
maze following injections of 2 mg/kg and 5 mg/kg diazepam as
compared to controls (Fig. 1; Section 3.1); (2) hamsters exhibited a decreased latency to enter into the t-tube and spent a
greater amount of time in the distal arms of the t-tube following injections of 5 mg/kg diazepam as compared to controls
(Fig. 2); (3) hamsters were quicker to enter into the open arena in
the ThatcherBritton conict test following injections of 5 mg/kg
diazepam as compared to controls (Table 3); and (4) hamsters
exhibited a signicantly reduced latency to begin eating in the open
arena of the ThatcherBritton conict test following injections of
2 mg/kg and 5 mg/kg diazepam as compared to controls (Table 3).
However, another benzodiazepine, triazolam, is known to increase
locomotor activity in hamsters [26,37], and this was conrmed
in this study for doses of 5 mg/kg, but not 2 mg/kg, diazepam as
described in Section 3.1 using the very accurate method of wheel
running activity. Therefore, the reduced latency to enter the t-tube
and the open arena in the ThatcherBritton conict test following injections of 5 mg/kg diazepam could be the result of increased
locomotor tendencies rather than any reduced anxiety in hamsters. However, other results argue against a diazepam-induced
increase in locomotor activity in hamsters. First, in the elevated
plus maze, 5 mg/kg diazepam did not cause hamsters to reduce
the latency at which they entered the open arms, or increase the
number of times they entered into the open arms (Fig. 1). Second,
in the t-tube, 5 mg/kg diazepam signicantly reduced the number
of entries between the distal arms compared to controls (Fig. 2).
Therefore, hamsters did not display all of the behaviors that would
be expected if diazepam was simply increasing locomotor activity, but rather that some anxiolytic activity was apparent following
diazepam injections, particularly in the plus maze and the t-tube.
In a previous report from another laboratory, hamsters also
spent about 33% more time in the open arms following injections
of 2 mg/kg diazepam at ZT 8ZT 12 [38], and we observed a smaller
effect of around 18% at ZT 11 (Fig. 1B). Hamsters also spent more
time in the open arms during the dark period in the report by Yannielli et al. [38], and we saw a similar (although not statistically
signicant) effect herein when you examine the vehicle responses
in Fig. 1B. A more recent study likewise reported anxiolytic activity
in the Syrian hamster using 2 mg/kg diazepam in the elevated plus
maze, although the time-of-day the experiments were conducted
was not reported [23]. Altogether then, there are two previous

R.L. Gannon et al. / Behavioural Brain Research 218 (2011) 814

reports and this study that conrms that Syrian hamsters do exhibit
anxiolysis following injection of 2 mg/kg diazepam and that the elevated plus maze can be used to demonstrate anxiolytic activity of
experimental compounds in Syrian hamsters.
4.2. NK1 receptor antagonist
This report also detected signicant effects with the NK1 receptor antagonist vestipitant in reducing the latency of hamsters to
enter the open arms of the plus maze (Table 1). However, this
effect was not conrmed with another NK1 receptor antagonist
GR205171, although the results did trend in the correct direction
(Table 1). NK1 receptor antagonists have shown the potential for
development as anxiolytics [9]. However, GR205171 had no effect
when tested on rats in the elevated plus maze [34], and to our
knowledge this is the only report of vestipitant usage in the elevated plus maze. Therefore, the lack of conrmatory effect with
GR205171 is not surprising and our results with vestipitant suggest that additional testing with other NK1 receptor antagonists is
warranted.
4.3. Cannabinoid receptor1 agonist
Syrian hamsters have also been used to demonstrate anxiolytic
activity following injections of an inhibitor of fatty-acid amide
hydrolase which is presumed to increase endocannabinoid concentrations [23]. However, in our study, we did not detect any
anxiolytic activity following injections of the CB1 receptor agonist
CP55940 in the plus maze, although, again, the trends were in the
appropriate direction. As can be seen by the data reported in Table 1,
the variability of the responses within each group of hamsters was
quite high, and that variability prevented the results from being statistically signicant. Renement of behavioral testing techniques
and more extensive dosing regimens using CP55940 may reveal
anxiolytic effects of this compound in future studies.
4.4. CRF1 , 5-HT6 and mGluR5 receptor antagonists
The CRF1 receptor antagonist CP154526 did not display any anxiolytic activity in our studies (Table 1). CP154526, at doses similar
to ours (up to 30 mg/kg) did not display any activity in the elevated
plus maze using the mouse [18] or rat [22]. Therefore, this report
conrms in a third species that this compound does not display
anxiolytic activity when using the elevated plus maze. The same
cannot be said however for our results with the 5-HT6 receptor
antagonist SB399885 as this compound at doses of 13 mg/kg did
display anxiolytic activity in the elevated plus maze using rats [36]
while we saw no effect in the elevated plus maze with 1 mg/kg
SB399885, although the data did trend in the appropriate direction (Table 1). Perhaps more extensive testing of this compound in
hamsters would reveal a signicant anxiolytic effect in this species.
Finally, the mGluR5 receptor antagonist MTEP at 10 mg/kg did not
elicit any anxiolytics activity in the elevated plus maze in our study
(Table 1), nor did it in rats with 5 mg/kg as reported by Pietraszek
et al. [31]. Therefore, again, MTEP does not appear to have any efcacy in the elevated plus maze. Finally, we only tested citalopram
(10 mg/kg) following acute injections, and were expecting to see an
anxiogenic, rather than an anxiolytic response on the plus maze,
as was reported with a similar dose in rats by Griebel et al. [16].
However, we observed absolutely no effect with citalopram in our
experiment (Table 1). Altogether, then, the lack of positive results
we observed with the test compounds in the elevated plus maze
is not unexpected as these compounds have displayed minimal
efcacy in other laboratories and species.
The t-tube and open eld ThatcherBritton conict tests did not
reveal any anxiolytic activity with any of the compounds tested in

13

Tables 2 and 3. To our knowledge, Syrian hamsters have not previously been used in a t-tube paradigm to evaluate anxiolytics, and
aside from the results presented in Fig. 2 with diazepam (some
of which may be activity-related), there is no indication from this
study that the t-tube is an effective model for Syrian hamsters. Similarly, as described above, the ThatcherBritton conict test may not
be very useful to screen anxiolytics in this species since it is so difcult to control food hoarding by hamsters and the baseline control
values can vary considerably (Table 3).
5. Conclusion
The present results, together with two previous reports on
diazepam and one study of cannabinoids [23,38], suggest that
the elevated plus maze is a useful procedure for characterizing
the inuence of putative anxiolytic/antidepressant agents upon
anxiety-related behaviors in Syrian hamsters. Furthermore, the
plus maze may be of particular utility in that the present data discourage the use of the t-tube or the open eld conict test in Syrian
hamsters at least at the time of day used in this study. Thus,
additional work is warranted to further characterize the inuence
of potential psychotropic agents upon anxious behaviors in Syrian hamsters, both in general (novel classes of agent not evaluated
herein) and in particular as regards drugs exerting chronobiologic
effects and/or acting via benzodiazepine-related mechanisms (Alo
et al. [1], Millan, 2009). The relevance of such studies is underlined
by the continuing interest in (1), the inuence of benzodiazepines
upon anxious behavior and biological rhythms both in animal
models and in patients [1,2,6,7,41] and (2), intensive studies of
agents acting at specic modulatory sites on GABAA and GABAB
receptors for the treatment of anxiodepressive states and other
psychiatric disorders [2,17,20,28,41].
Acknowledgements
Elizabeth Lungwitz, Natalia Batista and Alyssa Peacock were
undergraduate students in the Department of Biology at Valdosta
State University. Supported by NSF IOS 0549980 (RLG) and the
Institut de Recherches Servier.
References
[1] Al R, Avolio E, Di Vito A, Carelli A, Facciolo RM, Canonaco Mi. Distinct subunit
variations of the hypothalamic GABAA receptor triplets () are linked to
hibernating state in hamsters. BMC Neurosci 2010;6:11.
[2] Atack JR. Subtype-selective GABA(A) receptor modulation yields a novel pharmacological prole: the design and development of TPA023. Adv Pharmacol
2009;57:13785.
[3] Brocco M, Dekeyne A, Mannoury la Cour C, Touzard M, Girardon S, Veiga S, et al.
Cellular and behavioural prole of the novel, selective neurokinin1 receptor
antagonist, vestipitant: a comparison to other agents. Eur Neuropsychopharmacol 2008;18:72950.
[4] Childress SJ, Gluckmann MI. 1,4-benzodiazepines. J Pharm Sci 1964;53:57790.
[5] Christmas D, Hood S, Nutt D. Potential novel anxiolytic drugs. Curr Pharm Des
2008;14:353446.
[6] Cloos JM, Ferreira V. Current use of benzodiazepines in anxiety disorders. Curr
Opin Psychiatry 2009;22:905.
[7] Da Settimo F, Taliani S, Trincavelli ML, Montali M, Martini C. GABAA/Bz receptor
subtypes as targets for selective drugs. Curr Med Chem 2007;14:2680701.
[8] Dekeyne A, Denorme B, Monneyron S, Millan MJ. Citalopram reduces social
interaction in rats by activation of serotonin (5-HT)(2C) receptors. Neuropharmacology 2000;39:11147.
[9] Duffy RA. Potential therapeutic targets for neurokinin-1 receptor antagonists.
Expert Opin Emerg Drugs 2004;9:921.
[10] Ebner K, Sartori SB, Singewald N. Tachykinin receptors as therapeutic targets
in stress-related disorders. Curr Pharm Des 2009;15:164774.
[11] Gannon RL, Millan MJ. The selective tachykinin neurokinin1 (NK1 ) receptor antagonist, GR 205,171, stereospecically inhibits light-induced phase
advances of hamster circadian activity rhythms. Eur J Pharm 2005;527:8693.
[12] Gannon RL, Millan MJ. The corticotropin-releasing factor (CRF)1 receptor antagonists CP154,526 and DMP695 inhibit light-induced phase advances of hamster
circadian activity rhythms. Brain Res 2006;1083:96102.

14

R.L. Gannon et al. / Behavioural Brain Research 218 (2011) 814

[13] Gannon RL, Millan MJ. Evaluation of serotonin, noradrenaline and dopamine
reuptake inhibitors on light-induced phase advances in hamster circadian
activity rhythms. Psychopharmacology 2007;195:32532.
[14] Gannon RL, Millan MJ. Positive and negative modulation of circadian activity
rhythms by mGluR5 and mGluR2/3 metabotropic glutamate receptors. Neuropharmacology; in press. doi:10.1016/j.neuropharm.2010.08.022.
[15] Gannon RL, Peglion J-L, Millan MJ. Differential inuence of selective 5-HT5A vs
5-HT1A , 5-HT1B , or 5-HT2C receptor blockade upon light-induced phase shifts in
circadian activity rhythms: interaction studies with citalopram. Eur Neuropsychopharm 2009;19:88797.
[16] Griebel G, Moreau JL, Jenck F, Misslin R, Martin JR. Acute and chronic treatment
with 5-HT reuptake inhibitors differentially modulate emotional responses in
anxiety models in rodents. Psychopharmacology 1994;113:46370.
[17] Herd MB, Belelli D, Lambert JJ. Neurosteroid modulation of synaptic and
extrasynaptic GABA(A) receptors. Pharmacol Ther 2007;116:2034.
[18] Hodgson RA, Higgins GA, Guthrie DH, Lu SX, Pond AJ, Mullins DE, et al.
Comparison of the V1b antagonist, SSR149415, and the CRF1 antagonist, CP154,526, in rodent models of anxiety and depression. Pharmacol Biochem
Behav 2007;86:43140.
[19] Kehne JH. The CRF1 receptor, a novel target for the treatment of depression, anxiety, and stress-related disorders. CNS Neurol Disord Drug Targets
2007;6:16382.
[20] Lewis DA, Cho RY, Carter CS, Eklund K, Forster S, Kelly MA, et al. Subunitselective modulation of GABA type A receptor neurotransmission and cognition
in schizophrenia. Am J Psychiatry 2008;165:158593.
[21] Mayers AG, Baldwin DS. Antidepressants and their effect on sleep. Hum Psychopharmacol Clin Exp 2005;20:53359.
[22] Millan MJ, Brocco M, Gobert A, Dorey G, Casara P, Dekeyne A. Anxiolytic
properties of the selective, non-peptidergic CRF1 antagonists, CP154,526
and DMP695: a comparison to other classes of anxiolytics agent. Neuropsychopharm 2001;25:585600.
[23] Moise AM, Eisenstein SA, Astarita G, Piomelli D, Hohmann AG. An endocannabinoid signaling system modulates anxiety-like behavior in male Syrian
hamsters. Psychopharmacology 2008;200:33346.
[24] Mombereau C, Gur TL, Onksen J, Blendy JA. Differential effects of acute and
repeated citalopram in mouse models of anxiety and depression. Int J Neuropsychopharmacol 2010;13:32134.
[25] Morin LP, Allen CN. The circadian visual system. Brain Res Rev 2006;51:160.
[26] Mrosovsky N, Salmon PA. Triazolam and phase-shifting acceleration reevaluated. Chronobiol Int 1990;7:3541.

[27] Palucha A. Are compounds acting at metabotropic glutamate receptors the

answer to treating depression? Expert Opin Investig Drugs 2006;15:154553.

[28] Partyka A, Kodzinska

A, Szewczyk B, Wieronska
JM, Chojnacka-Wjcik E,
Librowski T, et al. Effects of GABAB receptor ligands in rodent tests of anxietylike behavior. Pharmacol Rep 2007;59:75762.
[29] Patel S, Hillard CJ. Pharmacological evaluation of cannabinoid receptor ligands in a mouse model of anxiety: further evidence for an anxiolytics role for
endogenous cannabinoid signaling. J Pharmacol Exp Ther 2006;318:30411.
[30] Pellow S, Chopin P, File SE, Briley M. Validation of open:closed arm entries in
an elevated plus-maze as a measure of anxiety in the rat. J Neurosci Methods
1985;14:14967.
[31] Pietraszek M, Sukhanov I, Maciejak P, Szyndler J, Gravius A, Wislowska A, et al.
Anxiolytic-like effects of mGlu1 and mGlu5 receptor antagonists in rats. Eur J
Pharmacol 2005;514:2534.
[32] Pollock BG. Citalopram: a comprehensive review. Expert Opin Pharmacother
2001;2:68198.
[33] Prendergast BJ, Nelson RJ. Affective responses to changes in day length
in Siberian hamsters (Phodopus sungorus). Psychoneuroendocrinology
2005;30:43852.
[34] Rupniak NM, Carlson EJ, Webb JK, Harrison T, Porsolt RD, Roux S, et al. Comparison of the phenotype of NK1R/ mice with pharmacological blockade of the
substance P (NK1) receptor in assays for antidepressant and anxiolytics drugs.
Behav Pharmacol 2001;12:497508.
[35] Sanford AE, Castillo E, Gannon RL. Cannabinoids and hamster circadian activity
rhythms. Brain Res 2008;1222:1418.
[36] Wesolowska A, Nikiforuk A. Effects of the brain-penetrant and selective 5-HT6
receptor antagonist SB-399885 in animal models of anxiety and depression.
Neuropharmacology 2007;52:127483.
[37] Van Reeth O, Turek FW. Stimulated activity mediates phase shifts in the
hamster circadian clock induced by dark pulses or benzodiazepines. Nature
1989;339:4951.
[38] Yannielli PC, Kanterewicz BI, Cardinali DP. Daily rhythms in spontaneous
and diazepam-induced anxiolysis in Syurian hamsters. Pharm Biochem Behav
1996;54:6516.
[39] Monteleone P, Maj M. The circadian basis of mood disorders: recent developments and treatment implications. Eur Neuropsychopharm 2008;18:70111.
[40] Germain A, Kupfer DJ. Circadian rhythm disturbances in depression. Hum Psychopharmacol Clin Exp 2008;23:57185.
[41] Millan MJ. The neurobiology and control of anxious states. Prog Neurobiol
2003;70:83244.