Physiology & Behavior 101 (2010) 124–131

Contents lists available at ScienceDirect

Physiology & Behavior

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p h b

Repeated social defeat stress induces chronic hyperthermia in rats

Sota Hayashida a,⁎, Takakazu Oka a,b, Takashi Mera a, Sadatoshi Tsuji a
a
Division of Psychosomatic Medicine, Department of Neurology, University of Occupational and Environmental Health, Iseigaoka 1-1, Yahatanishi-ku, Kitakyushu, 807-8555, Japan
b
Department of Psychosomatic Medicine, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan

a r t i c l e i n f o a b s t r a c t

Article history: Psychological stressors are known to increase core body temperature (Tc) in laboratory animals. Such single
Received 19 January 2010 stress-induced hyperthermic responses are typically monophasic, as Tc returns to baseline within several
Received in revised form 16 April 2010 hours. However, studies on the effects of repeated psychological stress on Tc are limited. Therefore, we
Accepted 25 April 2010
measured Tc changes in male Wistar rats after they were subjected to 4 social defeat periods (each period
consisting of 7 daily 1 h stress exposures during the light cycle followed by a stress-free day). We also
Keywords:
assessed affective-like behavioral changes by elevated plus maze and forced swim tests.
Repeated social defeat stress
Chronic hyperthermia
In the stressed rats, the first social defeat experience induced a robust increase in Tc (+ 1.3 °C). However, the
Stress-induced hyperthermia Tc of these rats was not different from control animals during the subsequent dark period. In comparison,
Psychogenic fever after 4 periods of social defeat, stressed rats showed a small but significantly higher (+0.2–0.3 °C) Tc versus
Depression control rats during both light and dark periods. Stressed rats did not show increased anxiety-like behavior
versus control rats as assessed by the elevated plus maze test. However, in the forced swim test, the
immobility time of stressed rats was significantly longer versus control rats, suggesting an increase in
depression-like behavior. Furthermore, hyperthermia and depression-like behavior were still observed
8 days after cessation of the final social defeat session. These results suggest that repeated social defeat stress
induces a chronic hyperthermia in rats that is associated with behavior resembling depression but not
anxiety.
© 2010 Elsevier Inc. All rights reserved.

1. Introduction events dramatically increased Tc in an episodic fashion [9,10], while

Psychological stressors such as being placing onto an open field or
having cage mates removed increases core body temperature (Tc) in
laboratory animals [1,2]. For example, when four mice are removed
one by one every 2 min from a cage of five group-housed mice, Tc in
the remaining mouse shows an abrupt increase (up to 1.5 °C) within
20 min, and gradually returns to baseline levels after 60–120 min [1].
Such stress-induced hyperthermia is induced by different mechan-
isms of inflammation-induced fever [1,3,4] and is attributed to
increased anxiety and arousal since substances that have anxiolytic
properties can diminish the magnitude of this hyperthermia [2,5]. In
most cases, a single stress-induced hyperthermic response is mono-
phasic, i.e., it starts abruptly just after exposure to the stressor and
returns to baseline within several hours [1].
Psychological stress-induced hyperthermic responses are also
observed in humans. In addition to normal emotion-based hyper-
thermic phenomena found in healthy subjects [6,7], numerous case
reports have been published on elevated Tc levels possibly caused by
psychological stress [8–12]. In some patients, emotionally salient
⁎ Corresponding author. Tel.: + 81 93 691 7438; fax: + 81 93 693 9842.
E-mail address: hsota@med.uoeh-u.ac.jp (S. Hayashida).
others develop a persistently increased Tc lasting weeks or months in
psychosocially stressful situations [8,12].
To obtain a better understanding of stress-related hyperthermic
phenomena, it is important to elucidate both the long-lasting effects
of repetitive or chronic stress on Tc as well as the temporary effects of
acute stress. However, such studies are limited and their findings are
inconsistent [13–21] Several studies have reported that repeated
social defeat stress, i.e. repeatedly placing animals into the home
cages of dominant residents at fixed time intervals, increases Tc in the
subordinate intruder animals during the light phase [17] resulting in
diminished amplitudes of diurnal changes in Tc [13,20,22]. The
elevated Tc in the light phase can be observed even after one or two
social defeats, and lasts for several days [19–21]. In contrast, other
studies observed that animals subjected to repeated social defeat
increased Tc during the dark phase [16] or even across 24 h [15]. Such
inconsistent results might be due to variations in the intensity and
frequency of social defeat sessions [20] or might be affected by how
animals cope with stressors [21].
In the present study, we observed the effects of repeated social
defeat stress on Tc and investigated if these effects differed according
to the number of defeat experiences. We also assessed behavioral
changes that could indicate altered affective-like behavior using
elevated plus maze and forced swim tests.

0031-9384/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.physbeh.2010.04.027
 S. Hayashida et al. / Physiology & Behavior 101 (2010) 124–131
2. Methods
2.1. Animals
Male Wistar rats (Kyudo Co. Ltd, Kumamoto, Japan) were used as
intruders. Rats were 6 weeks old and weighed 170–180 g upon arrival.
After arrival, they were housed individually in plastic cages (l × w × h:
40 × 25 × 20 cm) with wood chip bedding. Standard rodent food pellets
and water were available ad libitum. They were housed in a
temperature (24 ± 1 °C)- and humidity (60 ± 10%)-controlled room
under 12 h/12 h light–dark cycle conditions (lights on 0700–1900 h).
After surgery, their body weights decreased slightly and then
recovered to 170–180 g by the start of the experiment. Male Long
Evans rats (SLC, Inc., Shizuoka, Japan) weighing 400–500 g were used
as residents. They were housed in a separate room but under the same
general conditions. All experimental procedures were performed in
accordance with the guidelines on the use and care of laboratory
animals as set out by the Physiological Society of Japan (as contained
in the International Guiding Principles for Biomedical Research
Involving Animals [CIOMS, 1984] and Guide for the Care and Use
of Laboratory Animals [U. S. Department of Health, Education, and
Welfare Publication NIH 85-23, 1985]) and under control of the
Ethics Committee of Animal Care and Experimentation, University
of Occupational and Environmental Health, Japan (approval number
AE 08-008, -028). The number of animals used and their suffering
were kept to a minimum.
2.2. Surgery and monitoring of core body temperature (Tc)
Tc and physical activity were measured using a telemetry system
(Data Science Inc., St Paul, MN, USA). A battery-operated transmitter
(TA10TA-F40) was implanted into the peritoneal cavity of each rat
under sodium pentobarbitone anaesthesia (50 mg/kg, i.p.) more than
1 week before the experiment began. After surgery, rats were housed
individually. The Tc and activity signals were received by an antenna
below the rat cage and relayed to a signal processor (Dataquest A.R.
T.™ System, Data Sciences International, St. Paul, MN, USA) connected
Fig. 1. Experimental design and parameters measured in experiments 1 and 2. ○: elevated plus maze test; □: forced swim test.
125
to an LG computer. Tc and physical activity monitoring began at least
24 h before the first day of social defeat (to determine baseline
values), and both measures were recorded every minute throughout
the experiment.
2.3. Experimental protocols
2.3.1. Experiment 1: effect of single and repeated social defeat stress on Tc
After surgery, Wistar rats were divided into two groups (social
defeat stress and control) and their Tc and physical activity were
monitored for 40 days. A modified resident–intruder confrontation
procedure [14,23] was used to apply a daily social defeat stress in the
stress group. Each intruder male Wistar rat was transferred from its
home cage and placed into the cage of an unfamiliar male Long–Evans
resident rat. Resident rat cages were moved to the experimental room
5 min before the encounter. When the intruder was placed into the
resident's cage, in most cases, the intruder was attacked by the
resident and defeated within the first 5 min. As soon as the intruder
showed clear freezing behavior and submissive posture, it was
separated from the resident by being transferred into a small wire-
mesh compartment (w × h: 25 × 20 cm) within the resident's cage for
the duration of the encounter. Thus, the intruder was protected from
direct physical contact, but remained in olfactory, visual, and auditory
contact with the resident. After 1 h, intruders were returned to their
home cages. To avoid individual differences in defeat intensity, each
day the intruders were confronted with a different resident male
Long–Evans rat. These stress exposures were performed between 1000
and 1200 h.
Animals from the stress group were subjected to daily social defeat
on days 1–7, 9–15, 17–23, and 25–31. Tc and physical activity were
recorded on days 1 (Day 1), 8 (Phase 1), 16 (Phase 2), 24 (Phase 3),
and 32 (Phase 4). After phase 4, they were housed in their home cages
without exposure to social defeat for another 7 days (days 33–39) and
the Tc and physical activity were recorded again on day 40 (Phase 5)
(Fig. 1). Their body weights were measured at 1200 h during each
phase. Food intake (24 h) was also measured at 1200 h during each
phase. Control rats were transferred and placed into an empty cage for
126 S. Hayashida et al. / Physiology & Behavior 101 (2010) 124–131

5 min, separated into the wire-mesh compartment for 55 min, and
then returned to their home cage between 1000 and 1200 h instead of
being placed into the resident rat cage.
2.3.2. Experiment 2: effect of repeated social defeat stress on elevated
plus maze and forced swim tests
To assess if repeated social stress affects anxiety- and depression-like
behavior, elevated plus maze and forced swim tests were conducted.
Stressed rats were exposed to chronic, daily social defeat stress as in
experiment 1. Control rats were also handled as in experiment 1.
Here, rats were divided into five groups to test different levels of
social defeat stress. In group 1, behavioral tests were conducted on
day 8 after rats were exposed to daily stress on days 1–7. In group 2,
tests were conducted on day 16 after rats were exposed to stress on
days 1–7 and 9–15. In group 3, tests were conducted on day 24 after
rats were exposed to stress on days 1–7, 9–15, and 17–23. In group 4,
tests were conducted on day 32 after rats were exposed to stress on
days 1–7, 9–15, 17–23, and 25–31. In group 5, tests were conducted
on day 40 after rats were exposed to stress on days 1–7, 9–15, 17–23,
25–31, and allowed a 7-day recovery period.
2.3.2.1. Elevated plus maze test. Elevated plus maze testing was
performed using an automated elevated plus maze system (SMART,
panlab s. l. Barcelona, Spain). It was elevated to a height of 50 cm with
two open (50 × 10 cm) and two enclosed arms (50 × 10 × 40 cm), and
was arranged so that the arms of the same type were opposite to each
other, all extending from a common central platform (10 × 10 cm).
The maze was made from black Plexiglas. The experiments were
performed under dim light conditions, with the light intensity on the
central platform bring 10 lx. Each rat performed a 5-min trial, and the
maze was cleaned between subjects. Automated recordings were
made by a PC running the SMART program. The percentage of time
spent in the open arms of the elevated plus maze apparatus and the
numbers of entries into open and closed arms were recorded for the 5-
min period. This test was performed between 1900 and 2000 h under
dim light conditions.
2.3.2.2. Forced swim test. The forced swim testing procedures followed
the method described by Porsolt et al. [24]. Two days after elevated
plus maze testing, animals were individually placed into glass
cylinders (40 cm height; 18 cm diameter) containing 18 cm of water

Fig. 2. A: Diurnal Tc changes in rats exposed to a single social defeat between 10 a.m. and 12 a.m. (■, n = 6) or control rats (○, n = 6). Data represent mean ± SEM. **p b 0.01, *p b 0.05
compared with control. B: Maximal Tc of rats exposed to 1, 7, 14, 21, and 28 social defeat sessions (n = 6 each) during the light period. Data represent mean ± SEM.
 S. Hayashida et al. / Physiology & Behavior 101 (2010) 124–131
at 23 °C (pre-test measure). After 15 min, they were transferred to a
30 °C drying environment for 30 min. Twenty-four hours later, the
animals were returned to the cylinders for 5 min (test measure), and
this session was recorded with a video camera. The cylinder was
cleaned and filled with fresh water between rats. An experimenter
who was unaware of the treatment received by the animals observed
the videotapes, and immobility time was measured. A rat was
considered immobile when floating and making no movements
except those that were necessary to keep its nostrils above the
water surface. This test was performed between 1000 and 1200 h.
2.4. Statistics
All data are presented as means ± standard error of the mean
(SEM). In experiment 1, Tc was averaged every 30 min and is shown in
Figs. 2 and 3. The effects of stress on Tc during both light and dark
periods were analyzed using a two-way repeated measures analysis of
Fig. 3. A, B: Diurnal Tc changes in rats after exposure to repeated social defeat (■, n = 6) or control rats (○, n = 6) during phases 3 (A) and 5 (B). Data represent mean ± SEM. *p b 0.05
compared with control.
127
variance (ANOVA), followed by Tukey's multiple-comparison post-
hoc test (Figs. 2A, 3). That is, two-way repeated measures ANOVAs
[stress (social defeat stress or control) × time (repeated measure)] on
data from day 1 and phases 1, 2, 3, 4, and 5 were conducted for the
light period (0700–1900 h) and the dark period (1900–0700 h). To
compare the effects of single vs. repeated (7, 14, 21, and 28) social
defeat stress on maximal Tc, one-way ANOVA was conducted (Fig. 2B).
Tc for each phase was also averaged every 3 h and shown in Table 1.
Differences in Tc between social defeat stress and control groups were
analyzed by Student's t-test. The effects of stress on physical activity in
the light period, dark period, and total 24-hour period were analyzed
using a two-way repeated measures ANOVA (Fig. 4). The effects of
stress on body weight and food intake were also analyzed using two-
way repeated measures ANOVA.
In experiment 2, time spent in the open arms on the elevated plus
maze and immobility time in the forced swim tests were analyzed
using two-way ANOVA [stress (social defeat stress or control) × time
128 S. Hayashida et al. / Physiology & Behavior 101 (2010) 124–131

Table 1
Average Tc's in rats exposed to social defeat (n = 6) or control rats (n = 6) across phases 1–5. Tc was averaged every 3 h. Data represent mean ± SEM. **p b 0.01, *p b 0.05 compared
with control.

Dark period Light period Dark period

Hour 1–4 4–7 7–10 10–13 13–16 16–19 19–22 22–1

Phase 1
Stress 37.83 ± 0.07 37.69 ± 0.05 37.30 ± 0.05** 37.16 ± 0.05 37.20 ± 0.04 37.23 ± 0.05 37.80 ± 0.05 37.84 ± 0.03
Control 37.76 ± 0.06 37.61 ± 0.12 37.03 ± 0.05 36.96 ± 0.09 37.00 ± 0.14 37.14 ± 0.06 37.65 ± 0.05 37.78 ± 0.06

Phase 2
Stress 37.76 ± 0.04 37.66 ± 0.04 37.15 ± 0.05 37.20 ± 0.13 37.11 ± 0.04 37.22 ± 0.06 37.79 ± 0.04 37.75 ± 0.05
Control 37.72 ± 0.05 37.64 ± 0.07 36.96 ± 0.08 37.00 ± 0.10 37.00 ± 0.08 37.07 ± 0.05 37.52 ± 0.04 37.74 ± 0.08

Phase 3
Stress 37.65 ± 0.03 37.39 ± 0.09 36.98 ± 0.06 37.26 ± 0.08** 37.06 ± 0.08 37.15 ± 0.07 37.73 ± 0.05 37.70 ± 0.05
Control 37.75 ± 0.09 37.53 ± 0.16 36.91 ± 0.07 36.82 ± 0.07 36.92 ± 0.10 37.07 ± 0.09 37.55 ± 0.05 37.80 ± 0.07

Phase 4
Stress 37.61 ± 0.05 37.40 ± 0.07 36.95 ± 0.09 37.32 ± 0.06** 37.10 ± 0.06 37.23 ± 0.07 37.74 ± 0.07* 37.69 ± 0.03
Control 37.70 ± 0.10 37.53 ± 0.11 36.83 ± 0.04 36.85 ± 0.05 37.00 ± 0.11 37.03 ± 0.05 37.46 ± 0.07 37.75 ± 0.10

Phase 5
Stress 37.84 ± 0.03** 37.65 ± 0.05 37.05 ± 0.09** 37.02 ± 0.04** 37.03 ± 0.04 37.18 ± 0.07 37.70 ± 0.06* 37.81 ± 0.05
Control 37.55 ± 0.07 37.44 ± 0.13 36.76 ± 0.08 36.75 ± 0.09 36.90 ± 0.08 37.11 ± 0.08 37.48 ± 0.05 37.67 ± 0.13

(groups 1, 2, 3, 4, and 5)], followed by Tukey's multiple-comparison residents were compared among rats that were exposed to 1, 7, 14, 21,
post-hoc tests. p b 0.05 was considered statistically significant for all and 28 social defeat experiences. However, the maximal Tc's were not
analyses. different among these animals (Fig. 2B).

3. Results
3.1. Effect of a single social defeat stress on Tc
At baseline, Tc's were not different between the two groups in either
the light or the dark period (data not shown). To compare the effects
of a single exposure to social defeat stress on Tc, diurnal changes in Tc
of stressed and control animals are shown in Fig. 2A. During the light
period, there was a significant effect of stress exposure (p b 0.05), with
defeated rats having an overall higher Tc compared to controls. There
was also a significant time effect (p b 0.05), and a significant stress × time
interaction (p b 0.05) according to post-hoc tests indicating higher Tc
between 1030 and 1400 h in stressed rats compared to controls. Tc
during the dark period did not differ between the two groups.
To investigate if the social defeat stress-induced increase in Tc
differs by repetition, the maximal increases in Tc during exposure to
3.2. Effect of repeated social defeat stress on Tc and general activity
during each phase
3.2.1. Tc
Fig. 3 shows diurnal changes in Tc of rats in the stress and control
groups during phases 3 (Fig. 3A) and 5 (Fig. 3B). In the light periods
of all phases, there were significant effects of defeat stress (p b 0.05)
with defeated rats having an overall higher Tc compared to controls.
There was also a significant time effect (p b 0.05) and a significant
stress × time interaction (p b 0.05). As shown in Fig. 3A, post-hoc tests
indicated that stressed rats had significantly higher Tc's between 1000
and 1200 h compared to controls during phase 3. In addition, post-hoc
tests indicated that stressed rats in phase 5 had significantly higher
Tc's at 0730, 0800 and 1200 h compared to controls (Fig. 3B). In the
dark period of phases 4 and 5, there was a significant stress effect
(p b 0.05) and time effect (p b 0.05). There was also a significant
stress × time interaction (p b 0.05). As shown in Fig. 3B post-hoc tests
indicated that stressed rats in phase 5 had significantly higher Tc at
0100 and 2030 h during the dark cycle when compared to controls.
Table 1 shows Tc's averaged every 3 h for both social defeat stress
and control groups. During phase 1, stressed animals displayed a
significantly higher Tc (+ 0.27 °C) between 0700 and 1000 h versus
control (p b 0.01). During phase 3, stressed animals showed a signi-
ficantly higher Tc (+0.44 °C) versus control rats between 1000 and
1300 h (p b 0.01). In phase 4, stressed rats showed a significantly
higher Tc versus control animals between 1000 and 1300 h (+0.47 °C,
p b 0.01) and also between 1900 and 2200 h (+0.28 °C, p b 0.05).
Finally, during phase 5, the stress group showed small but signi-
ficantly higher Tc's (+0.22–0.29 °C) versus control between 0100 and
0400 h (p b 0.01), 0700–1000 h (p b 0.01), 1000–1300 h (p b 0.01), and
1900–2200 h (p b 0.05).

3.2.2. Locomotor activity

Fig. 4. Effects of repeated social defeat stress on immobility time during forced swim
testing in rats exposed to repeated social defeat (■, n = 6) or control rats (□, n = 6).
Stressed rats in groups 1, 2, 3 and 4 had been exposed to 7, 14, 21, and 28 social defeat
sessions, respectively. Stressed rats in group 5 were tested 8 days after the final (28th)
defeat session. Data are shown as a time (seconds) spent immobile during 5 min test.
Data represent mean ± SEM. *p b 0.05 compared with control.
The effects of repeated social stress on locomotor activity were
assessed by measuring activity counts during the light and dark
periods, as well as total activity counts (activity during both light and
dark periods) at phases 1–5. Total activity counts were not different
between stressed rats and controls. Furthermore, activity counts
 S. Hayashida et al. / Physiology & Behavior 101 (2010) 124–131
measured during either the light period or the dark period alone were
not different between the two groups (data not shown).
3.2.3. Body weight and food intake
Body weight and food intake over 24 h were measured at each
phase and compared between stress and control groups. However,
these measures were not different between the two groups at any
phase (data not shown).
3.3. Effects of repeated social defeat stress on elevated plus maze and
forced swim tests
3.3.1. Elevated plus maze test
The percentage of time spent in the open arms of the elevated plus
maze was compared between stressed and control rats in groups 1–5.
The percentage of time spent in the open arms was not different
between stressed rats and controls. In addition, the number of entries
into open or closed arms was not different between stressed rats and
controls (data not shown).
3.3.2. Forced swim test
In Fig. 4, the immobility time in forced swim tests was compared
between stressed and control rats in groups 1–5. There was a
significant overall effect of stress (p b 0.05) in the defeated rats, which
demonstrated longer immobility times compared to controls. There
was also a significant time effect (p b 0.05), as well as a significant
stress × time interaction (p b 0.05). Post-hoc tests indicated that the
immobility time of stressed rats was significantly longer than that of
control rats in groups 4 and 5 (p b 0.05, respectively).
4. Discussion
The present study demonstrated that socially defeated rats
exhibited higher Tc's versus control animals during the light period
after 7, 14, and 21 defeat sessions. As the number of defeated
experiences increased further (up to 28 defeats), stressed rats began
to show higher Tc's during the dark period as well, even 8 days after
cessation of stress. Coinciding with the appearance of the 24-hour
hyperthermic state, the immobility time of stressed rats in forced
swim tests became longer than that seen in control rats. These
findings suggest that repeated social defeat stress induces chronic
hyperthermia in rats and that the appearance of chronic hyperthermia
is associated with the development of depression-like behavior.
Experiment 1 showed that the first social defeat experience, which
started at 10 a.m., induced a robust increase in Tc (+1.3 °C). The Tc's of
stressed rats returned to control levels within 4 h, and Tc's during the
dark period were not different from controls. This social stress-induced
hyperthermic response neither habituated nor sensitized with repeated
defeats during our observation period (total of 28 defeats) as was
reported previously [14–17,22,25]. At phase 1 (a stress-free day after 7
defeats), stressed rats showed higher Tc's versus control rats between 7
and 10 a.m. In comparison, at phase 3, they showed a higher Tc versus
controls not between 7 and 10 a.m., but between 10 a.m. and 1 p.m., the
time when they were normally exposed to dominant residents. Thus,
these hyperthermias may represent anticipatory and conditioned
responses [17,26], with the magnitude of their conditioned hyperther-
mias being smaller (∼0.5 °C) than that seen during actual social stress
exposures. It is interesting that possible anticipatory hyperthermia (at
7–10 a.m.) developed in phase 1 but disappeared thereafter, whereas
hyperthermia at 10 a.m.–1 p.m. did not develop until phase 3. In any
case, the hyperthermic state of the stressed rats was restricted to the
light period in these phases (≤ 21 defeats). In comparison, by phases 4
and 5 the Tc of stressed rats was elevated over control rats in both light
and dark periods. In phase 5, stressed rats showed small (0.2–0.3 °C) but
significantly higher Tc's at 7–10 p.m., 1–4 a.m., and 7 a.m.–1 p.m. 8 days
after their last stress exposure.
129
Two previous studies have shown that Tc's in stressed mice during
the dark period were higher than in control mice [15] or during the
pre-stress condition [16] using a modified social defeat stress
paradigm. However, in these studies, defeated mice were kept
separate in the same cage opposite a dominant animal throughout
both light and dark periods. Chronic elevation of Tc's may depend on
being continuously housed opposite the dominant animal since Tc's
rapidly returned to control levels when the animals were single-
housed during a recovery phase [15]. Most studies have shown that
repeated social defeat stress increased Tc in subordinated animals
solely during the light period [13,17,22]. Even a single or double
exposure to social defeat was reported to increase the light period Tc
[19–21]. It is possible that these studies did not observe changes in Tc
during the dark period because they measured the effects of only 16 or
fewer social defeat trials on Tc. In the present study, we did not
observe increased dark cycle Tc's in stressed rats until phase 4 (after
28 trials). The present study is the first to show that extensive
repetition of social defeat stress induces even more severe changes in
Tc than previously reported. That is, social defeat stress eventually
increases Tc in the active, dark period in addition to resting Tc's in the
light period, and this 24-hour hyperthermic state lasts even after
the termination of stress.
In experiment 2, stressed rats did not display higher anxiety-like
behavior versus control rats as assessed by elevated plus maze testing.
In contrast, stressed rats in groups 4 and 5 showed a significantly
longer immobility time versus control rats according to forced swim
tests. As increased immobility in this test indicates behavioral despair,
a characteristic of depression in humans [27,28], it suggests that rats
displayed more depression-like behavior after 28 defeat sessions.
Moreover, this affective state lasted for 8 days after the final defeat
experience. Considering stressed rats also displayed a 24-hour
hyperthermic state in phases 4 (after 28 defeats) and 5 (8 days after
recovery), this study suggests that repetitive exposure to psychosocial
stress results in hyperthermia and co-morbid depression-like behav-
ior. This finding is in agreement with human studies demonstrating
that depressed patients have higher Tc's compared to healthy subjects
[29,30] and that circadian Tc rhythms are disrupted in depressed
patients [31]. In addition to immobility time in the forced swim test,
there exist other methods for assessing depressive-like states, such as
exploratory activity in an open field [32] or preference for a sucrose
solution [27]. A previous study reported that a single social defeat
could reduce activity in an open field, which implied a deficit in
exploratory motivation [32]. Another study also showed that
exploratory activity was reduced by one week of daily social defeat,
whereas sucrose preference was decreased only after three weeks of
defeat, implying the generation of an anhedonic state [27]. Consid-
ering our observed increased immobility in the forced swim test in
phase 4 of the current study, the effect of social defeat stress on
behavioral changes indicative of depression may differ depending on
which behavioral assay is employed.
In our study, body weight and food intake were not different
between the stress and control groups. This finding is not necessarily
in agreement with previous studies that indicated blunted increases
in body weight and reduced food intake in the stress group compared
to controls. [27,32,33]. The reason for such a discrepancy is unclear.
One of the possible reasons may be due to the difference in time
periods when animals were exposed to social defeat. In our
experiment, rats were placed into the resident animal's cage during
the light period when rats are less active, whereas in the previous
studies, animals were exposed to residents in the dark period when
animals are generally more active [19–21,27,32,33]. Thus, the effects
of social defeat on weight and food intake in our experiment might
have been attenuated compared to other studies.
Previous studies have shown that a single or double social defeat
affects Tc, food intake, or activity in an open field, and that these
effects last for several days [19–21,32–34]. Therefore, one may
130 S. Hayashida et al. / Physiology & Behavior 101 (2010) 124–131
speculate that the present findings are due to the long-lasting effects
of a single defeat. This is possible in regards to the elevated Tc's
observed in the light period. However, elevated Tc's in the dark period
and increased immobility in forced swim tests may not be due to the
long-lasting effect of a single stress, but to repeated stress effects since
neither the Tc's nor immobility times of stressed rats were different
from control rats in either phase 1 or group 1.
From the present study alone, it is not clear how repeated social
defeat stress induces hyperthermia. Chronic hyperthermia in our
animals was unlikely due to increased activity since the total amount
of physical activity in stressed rats was not different from that in
control rats at any phase. This is also supported by a previous study
showing that Tc increases after social defeat were in fact inversely
correlated with changes in activity [19]. One of the possible
mechanisms for our observations is a disturbance in brain serotonin
(5-HT) systems since central 5-HT circuitry plays an important role in
thermoregulation. For example, intracerebroventricular administra-
tion of 5-HT reduces Tc [35]. Pharmacological studies have revealed
that activation of 5-HT1A receptors induces hypothermia, and that
activation of 5-HT2A or 5-HT2C receptors induces hyperthermia, in rats
and mice [36–38]. Disturbances in 5-HT neurotransmission have also
been implicated in the pathogenesis of major depressive disorder [39]
and there have been some reports indicating that social defeat
experiences alter 5-HT neurotransmission [40], 5-HT mediated
neuroendocrine responses [41], and 5-HT receptor binding. In
summary, social defeat blunts 5-HT1A receptor-mediated adrenocor-
tical activation [41]. After social defeat, binding to 5-HT1A receptors
was reduced in the hippocampus, while binding to 5-HT2 receptors
was increased in the cerebral cortex [42]. It is not known yet whether
repeated social defeat stress causes such changes in brain nuclei that
are crucial for thermoregulation such as the preoptic area of the
hypothalamus or the raphe pallidus nucleus. To further examine the
5-HT hypothesis, our lab is currently determining whether chronic
administration of selective serotonin reuptake inhibitors will modu-
late chronic hyperthermia.
In conclusion, the present study suggests that repeated social
defeat stress induces chronic hyperthermia in rats, and that de-
velopment of chronic hyperthermia is associated with depression-
like behavior. The repeated social defeat stress paradigm may be
a useful animal model to understand the mechanisms of psychogenic
fever and hyperthermia associated with depressive disorders in
humans.
Acknowledgements
We thank Dr. Tomoko Soga (Brain Research Institute Monash
Sunway, School of Medicine and Health Sciences, Monash University
Sunway campus) for her technical advice. This study was supported in
part by a Grant-in-Aid for Scientific Research from the Japan Society for
the Promotion of Science (numbers 20590711 and 22590671 to T.O.).
References
[1] Oka T, Oka K, Kobayashi T, Sugimoto Y, Ichikawa A, Ushikubi F, et al. Characteristics
of thermoregulatory and febrile responses in mice deficient in prostaglandin EP1
and EP3 receptors. J Physiol 2003;551:945–54.
[2] Adriaan Bouwknecht J, Olivier B, Paylor R. The stress-induced hyperthermia
paradigm as physiological animal model for anxiety: a review of pharmacological
and genetic studies in the mouse. Neurosci Biobehav Rev 2007;31:41–59.
[3] Saha S, Engström L, Mackerlova L, Jakobsson PJ, Blomqvist A. Impaired febrile
responses to immune challenge in mice deficient in microsomal prostaglandin E
sunthase-1. Am J Physiol Regul Intergr Comp Physiol 2005;288:R1100–7.
[4] Vinkers CH, Groenink L, van Bogaert MJ, Westphal KG, Kalkman CJ, van Oorschot R,
et al. Stress-induced hyperthermia and infection-induced fever: two of a kind?
Physiol Behav 2009;98:37–43.
[5] Borsini F, Lecci A, Volterra G, Meli A. A model to measure anticipatory anxiety in
mice? Psychoparmacology 1998;98:207–11.
[6] Marazziti D, Di Muro A, Castrogiovanni P. Psychological stress and body
temperature changes in humans. Physiol Behav 1992;52:393–5.
[7] Briese E. Emotional hyperthermia and performance in humans. Physiol Behav
1995;58:615–8.
[8] Oka T, Oka K. Age and gender differences of psychogenic fever: a review of the
Japanese literature. Biopsychosoc Med 2007;1:11.
[9] McNeil GN, Leighton LH, Elkins AM. Possible psychogenic fever of 103.5 °F in a
patient with borderline personality disorder. Am J Psychiatry 1984;141:896–7.
[10] Timmerman RJ, Thompson J, Noordzij HM, van der Meer JW. Psychogenic periodic
fever. Neth J Med 1992;41:158–60.
[11] Kaneda Y, Tsuji S, Oka T. Age distribution and gender differences in psychogenic
fever patients. Biopsychosoc Med 2009;21:6.
[12] Hiramoto T, Oka T, Yoshihara K, Kubo C. Pyrogenic cytokines did not mediate a
stress interview induced hyperthermic response in a patient with psychogenic
fever: a case report. Psychosom Med 2009:932–6.
[13] Tornatzky W, Miczek KA. Long-term impairment of autonomic circadian rhythms
after brief intermittent social stress. Physiol Behav 1993;53:983–93.
[14] Tornatzky W, Miczek KA. Behavioral and autonomic responses to intermittent
social stress: differential protection by clonidine and metoprolol. Psychopharma-
cology (Berl) 1994;116:346–56.
[15] Keeney AJ, Hogg S, Marsden CA. Alterations in core body temperature, locomotor
activity, and corticosterone following acute and repeated social defeat of male
NMRI mice. Physiol Behav 2001;74:177–84.
[16] Bartolomucci A, Palanza P, Costoli T, Savani E, Laviola G, Parmigiani S, et al. Chronic
psychosocial stress persistently alter autonomic function and physical activity in
mice. Physiol Behav 2003;80:57–67.
[17] Pardon MC, Kendall DA, Pérez-Diaz F, Duxon MS, Marsden CA. Repeated sensory
contact with aggressive mice rapidly leads to an anticipatory increase in core body
temperature and physical activity that precedes the onset of aversive responding.
Eur J NeuroSci 2004;20:1033–50.
[18] Ushijima K, Morikawa T, To H, Higuchi S, Ohdo S. Chronobiological disturbances
with hyperthermia and hypercortisolism induced by chronic mild stress in rats.
Behav Brain Res 2006;173:326–30.
[19] Meerlo P, De Boer SF, Koolhaas JM, Daan S, Van den Hoofdakker RH. Changes in
daily rhythms of body temperature and activity after a single social defeat in rats.
Physiol Behav 1996;59:735–9.
[20] Meerlo P, Van den Hoofdakker RH, Koolhaas JM, Daan S. Stress-induced changes in
circadian rhythms of body temperature and activity in rats are not caused by
pacemaker changes. J Biol Rhythms 1997;12:80–92.
[21] Meerlo P, Sgoifo A, De Boer SF, Koolhaas JM. Long-lasting consequences of a social
conflict in rats: behavior during the interaction predicts subsequent changes in
daily rhythms of heart rate, temperature, and activity. Behav Neurosci 1999;113:
1283–90.
[22] Sgoifo A, Pozzato C, Meerlo P, Costoli T, Manghi M, Still D, et al. Intermittent
exposure to social defeat and open-filed test in rats: acute and long-term effects
on ECG, body temperature and physical activity. Stress 2002;5:23–35.
[23] Miczek KA. Tolerance to the analgesic, but not discriminative stimulus effects of
morphine after brief social defeat in rats. Psychopharmacology (Berl) 1991;104:
181–6.
[24] Porsolt RD, Anton G, Blavet N, Jalfre M. Behavioural despair in rats: a new model
sensitive to antidepressant treatments. Eur J Pharmacol 1978;47:379–91.
[25] Bhatnagar S, Vining C, Iyer V, Kinni V. Change in hypothalamic–pituitary–adrenal
function, body temperature, body weight and food intake with repeated social
stress in rats. J Neuroendocrinol 2006;18:13–24.
[26] Oka T, Oka K, Hori T. Mechanisms and mediators of psychological stress-induced
rise in core temperature. Psychosom Med 2001;63:476–86.
[27] Rygula R, Abumaria N, Flugge G, Fuchs E, Ruther E, Havemann-Reinecke U.
Anhedonia and motivational deficits in rats: impact of chronic social stress. Behav
Brain Res 2005;162:127–34.
[28] Rygula R, Abumaria N, Havemann-Reinecke U, Rüther E, Hiemke C, Zernig G, et al.
Pharmacological validation of a chronic social stress model of depression in rats:
effects of reboxetine, haloperidol and diazepam. Behav Pharmacol 2008;19:183–96.
[29] Avery DH, Wildschiødtz G, Smallwood RG, Martin D, Rafaelsen OJ. REM latency
and core temperature relationships in primary depression. Acta Psychiatr Scand
1986;74:269–80.
[30] Szuba MP, Guze BH, Baxter Jr LR. Electroconvulsive therapy increases circadian
amplitude and lowers core body temperature in depressed subjects. Biol
Psuchiatry 1997;42:1130–7.
[31] Tsujimoto T, Yamada N, Shimoda K, Hanada K, Takahashi S. Circadian rhythms
in depression. Part 2: circadian rhythms in patients with various mental disorders.
J Affect Disord 1990;18:199–210.
[32] Meerlo P, Overkamp GJ, Benning MA, Koolhaas JM, Van den Hoofdakker RH. Long
term changes in open field behaviour following a single social defeat in rats can be
reversed by sleep deprivation. Physiol Behav 1996;60:115–9.
[33] Meerlo P, Overkamp GJ, Daan S, Van den Hoofdakker RH, Koolhaas JM. Changes in
behaviour and body weight following a single or double social defeat in rats. Stress
1996;1:21–32.
[34] Koolhaas JM, Meerlo P, De Boer SF, Strubbe JH, Bohus B. The temporal dynamics of
the stress response. Neurosci Biobehav Rev 1997;21:775–82.
[35] Crawshaw LI. Effects of intracerebral 5-hydroxytryptamine injection on themor-
egulation in rat. Physiol Behav 1972;9:133–40.
[36] Abdel-Fattah AF, Matsumoto K, el-Hady KA, Watanabe H. 5-HT1A and 5-HT2
receptors mediate hypo- and hyperthermic effects of tryptophan in pargyline-
pretreated ras. Pharmacol Biochem Behav 1995;52:379–84.
[37] Mazzola-Pomietto P, Aulakh CS, Wozniak KM, Hill JL, Murphy DL. Evidence that 1-
(2, 5-dimethoxy-4-iodophenyl)-2-aminopropane(DOI)-induced hyperthermia in
rats is mediated by stimulation of 5-HT2A receptors. Psychopharmacology (Berl)
1995;117:193–9.
 S. Hayashida et al. / Physiology & Behavior 101 (2010) 124–131 131

[38] Aulakh CS, Mazzola-Pomietto P, Murphy DL. Long-term antidepressant treatments
alter 5-HT2A and 5-HT2C receptor-mediated hyperthermia in Fawn-Hooded rats.
Eur J Pharmacol 1995;282:65–70.
[39] Maes M, Meltzer HY. The serotonin hypothesis of major depression. In: Bloom F,
Kupfer DJ, editors. Psychopharmacology: the fourth generation of progress.
NewYork, NY: Raven Press; 1995. p. 933–44.
[40] Keeney A, Jessop DS, Harbus MS, Marsden CA, Hogg S, Blackburn-Munro RE.
Differential effects of acute and chronic social defeat stress on hypothalamic–
pituitary–adrenal axis function and hippocampal serotonin release in mice.
J Neuroendocrinol 2006;18:330–8.
[41] Korte SM, Buwalda B, Meijer O, De Kloet ER, Bohus B. Socially defeated male rats
display a blunted adrenocortical response to a low dose of 8-OH-DPAT. Eur J
Pharmacol 1995;272:45–50.
[42] McKittrick CR, Blanchard DC, Blanchard RJ, McEwen BS, Sakai RR. Serotonin
receptor binding in a colony model of chronic social stress. Biol Psychiatry
1995;37:383–93.