Voluntary exercise in C57 mice is anxiolytic across several measures of anxiety

Behavioural Brain Research 197 (2009) 31–40

Contents lists available at ScienceDirect

Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr

Research report

Voluntary exercise in C57 mice is anxiolytic across several measures of anxiety Jasmin N. Salam, James H. Fox, Ezra M. DeTroy, Michele H. Guignon, Dana F. Wohl, William A. Falls ∗ Department of Psychology, University of Vermont, 2 Colchester Avenue, Burlington, VT 05405, United States

a r t i c l e

i n f o

Article history: Received 17 June 2008 Received in revised form 21 July 2008 Accepted 24 July 2008 Available online 3 August 2008 Keywords: Exercise Anxiety Acoustic startle Open field Social interaction Stress-induced hyperthermia Light-enhanced startle

a b s t r a c t Voluntary wheel running in rodents is associated with a number of adaptive behavioral and physiological effects including improved learning, reduction in stress-associated behaviors, neurogenesis, angiogenesis, increases in neurotrophic factors, and changes in several signaling molecules. Exercise has also been reported to reduce anxiety-like behaviors. However, other studies have failed to find an anxiolytic effect of exercise. The inconsistencies in the literature may contribute to the scarcity of data examining the physiological correlates of the anxiolytic effect of exercise. Here we show that 2 weeks of voluntary exercise in male C57 mice is associated with reduced anxiety as measured with acoustic startle, stressinduced hyperthermia, social interaction, light-enhanced startle, and some, but not all, measures in the open field. A great deal is known about the neural circuits underlying anxiety. Given the consistency of the anxiolytic effect of voluntary exercise across several measures, it is now possible to begin a systematic analysis of the physiological basis of the anxiolytic effect of exercise. © 2008 Elsevier B.V. All rights reserved.

1. Introduction A number of human studies support the idea that physical exercise can reduce the signs and symptoms of anxiety [7,10,33,38,46,51,63,70,77] and specifically benefit the treatment of PTSD, panic disorder and phobia [24,93]. Despite the importance of these findings, relatively little is known about the physiological basis of the anxiolytic effect of exercise. This is particularly surprising in light of the fact that in animals exercise is associated with a variety of adaptations in the brain including neurogenesis and increased neuronal survival [8,25,96–99], angiogenesis [5,55,87], increased vascular flow [87], increased expression of neurotrophins [21,37,73,74,100,101], changes in gene expression [91] and signaling molecules [84], and changes in serotonin [40,42], norepinephrine [22] and GABA [18]. However, few of these effects have been directly linked to anxiolytic effects of exercise. This lack of understanding of the physiological basis of the anxiolytic effect of exercise may be due in part to the inconsistent effects that exercise has in animal models of anxiety. In studies allowing animal voluntary access to a running wheel there are reports of an anxiolytic effect of voluntary exercise [4,17,18,21,42,43], no effect of exercise [79] or increased anxiety following exercise [9,95]. While these inconsistencies may be due to any number of experimental variables, they point to the need for continued assessment of

the putative anxiolytic effects of exercise in animal models. We have shown that voluntary exercise in mice is associated with lower acoustic startle amplitude [32]. Specifically, C57BL/6J mice given free access to a running wheel for 2 weeks showed lower acoustic startle amplitude than mice not given access to a running wheel. Higher acoustic startle amplitude is often observed in clinically anxious individuals [57,68] and in rodents, treatments that increase anxiety (e.g., bright lights, anxiogenic drugs) increase startle amplitude [107] while treatments that reduce anxiety (e.g., anxiolytic drugs) decrease startle amplitude. Therefore, the reduction in acoustic startle amplitude in exercising mice [32] may be consistent with the reports indicating that voluntary exercise in mice is anxiolytic [4,17,18,21,42,43]. However, given the inconsistent effects of voluntary exercise across experiments, we sought to examine the generality of the putative anxiolytic effect of voluntary exercise in mice by examining anxiety across a battery of tests [9] that includes acoustic startle. To this end, in separate experiments, mice were given free access to either a functioning running wheel or a non-functioning (i.e., locked) running wheel for 2 weeks. They were then tested for acoustic startle amplitude, open field behavior, stress-induced hyperthermia, social interaction and light-enhanced startle. 2. Materials and methods 2.1. Animals

∗ Corresponding author. Tel.: +1 802 656 5748; fax: +1 802 656 8783. E-mail address: [email protected] (W.A. Falls). 0166-4328/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.bbr.2008.07.036

Eight weeks old, male C57BL6/J mice were obtained from Jackson Laboratories in Bar Harbor, Maine. Mice were housed in groups of four in standard acrylic cages

32

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40

(24 cm (W) × 45 cm (D) × 20 cm (H)) located in an Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) approved conventional animal facility. Mice were maintained on a 12 h light/dark cycle (lights on at 07:00 h) with food and water available at all times. A 7-day acclimation period was given to mice before introduction of the running wheels. All procedures were approved by the University of Vermont Animal Care and Use Committee. 2.2. Voluntary wheel running Mice were given ad lib access to a running wheel (Superpet mini run-a-round, measuring 11.4 cm in diameter) for 2 weeks prior to the start of behavioral testing. For half of the cages, the wheels were locked preventing running (non-exercising control) and for the remaining cages the wheels were functional. 2.3. Behavioral testing All behavioral testing occurred during the light cycle and between the hours of 9:00 and 15:00. All procedures except prepulse inhibition of startle were carried out in naive groups of mice to eliminate the possibility of carry over effects between tests of anxiety [71]. Prepulse inhibition was examined in mice following the open field test. 2.3.1. Acoustic startle Acoustic startle amplitude is a sensitive measure of fear and anxiety [107]. In rodents, treatments that increase anxiety (e.g., bright lights, anxiogenic drugs) increase startle amplitude [107] while treatments that reduce anxiety (e.g., anxiolytic drugs) decrease startle amplitude [13]. We have previously shown in a between subjects design [32] that mice given 2 weeks of access to a running wheel show lower startle amplitude than non running controls. In order to determine whether running reduces startle amplitude, as opposed to startle amplitude increasing in non-running controls, we ran a within subjects design in which mice were tested for acoustic startle amplitude both before and after 2 weeks of voluntary exercise. The startle tests were conducted in eight sound attenuating cubicles measuring 58 cm (W) × 32 cm (D) × 55 cm (H). Each cubicle was lined with black, sound absorbing foam with no internal source of light. Each cubicle contained a stabilimeter device consisting of a load cell platform onto which the behavioral chamber was mounted (MED-ASR-PRO1, Med Associates, Georgia, VT). The chamber was constructed of clear acrylic, cylindrical in shape, 12.5 cm in length, with an inner diameter of 5 cm. The floor of the chamber consisted of a removable grid composed of 6 steel rods 3.2 mm in diameter, and spaced 6.4 mm apart. Startle responses were transduced by the load cell, amplified, and digitized over a range of 0–4096 units. Startle amplitude was defined as the largest peak to trough value within 100 ms after the onset of the startle stimulus. Startle stimuli were 20 ms bursts of white noise provided through a Radio Shack Supertweeter located 10 cm behind the behavioral chamber. Data collection and the control and sequencing of all stimuli were controlled by Med-Associates startle reflex hardware and software. In the first startle experiment mice had access to a locked (n = 16) or functioning (n = 16) running wheel for 2 weeks. Running distance and time spent running were recorded at 12 h intervals (at 7:00 and 19:00 h) over the 2 weeks. Mice were then tested for acoustic startle on each of three consecutive days (i.e., days 15–17). Mice continued to have free access to the running wheels on the test days. On each of the three test days mice were transported to the lab in the home cage and placed individually in the startle apparatus. After a 5 min acclimation period, mice were presented with the first of 30 startle stimulus alone trials. Ten stimuli of each intensity level (95, 100, and 105 dB) were presented in a pseudo-random order (the constraint being that each intensity occur within each block of three trials) with a mean inter-trial interval (ITI) of 60 s. A total of three startle sessions were used to obtain stable measure of acoustic startle amplitude. Mean startle amplitude was computed for each mouse for each startle stimulus intensity within each startle test (i.e., for each of the three pre and post running tests). The second startle experiment used a within subjects design to examine the effect of wheel running on acoustic startle. All mice had access to locked running wheels for 2 weeks prior to the first of three tests for acoustic startle. Groups were then randomly divided into exercising (n = 16) and non-exercising controls (n = 16). For the exercising groups the wheels were then unlocked and for the non-exercising groups the wheels remained locked. Two weeks later the mice were again tested for acoustic startle on each of three consecutive days. The third startle experiment examined prepulse inhibition of startle across a range of prepulse stimulus frequencies in an effort to determine whether the ability to process auditory stimuli is altered in exercising mice. Changes in the ability to process auditory stimuli could contribute to lower startle amplitude. To behaviorally assess the ability to process auditory stimuli, we examined prepulse inhibition (PPI) of startle in exercising and non-exercising mice. PPI is the degree to which the acoustic startle response is reduced when the startle-eliciting stimulus is preceded by a brief non-startle eliciting stimulus [53,56]. PPI is thought to reflect the ability to “gate” sensory information [36,88,89] and has been used as a means of behaviorally assessing hearing in mice [108–112]. For example, mice with high frequency sen-

sorineuronal hearing loss show poor PPI to high frequency auditory prepulse stimuli, but not low frequency prepulse stimuli. Exercising (n = 12) and non-exercising (n = 14) mice were given 30 startleeliciting noise bursts (105 dB SPL) 20 msec in duration at a 1 min inter-trial interval. On prepulse trials, the prepulse stimulus preceded the startle stimulus by 100 ms. The prepulse was a 20 ms, 70 dB pure tone or either 4, 8, 12 or 16 kHz. Four tone frequencies were used in order to span the range of mouse hearing. Twenty-five of the 30 startle stimuli were preceded by a prepulse (5 at each frequency). The remaining 5 trials were startle stimulus alone trials. PPI was calculated by dividing startle amplitude on prepulse trials by startle amplitude on startle stimulus alone trials. Thus, 100% indicates that the prepulse had no effect on startle amplitude whereas 0% indicates that the prepulse completely inhibited startle. 2.3.2. Open field In the open field test anxiety is generally associated with overall lower levels of activity, thigmotaxis and decreased exploratory behavior [1,11]. Open field is a common test for anxiety and is sensitive to both anxiolytic and anxiogenic drugs. Given its wide use and clear ethological validity, we assessed whether 2 weeks of access to a running wheel would be associated with reduced anxiety-related behavior in the open field. The open field consisted of a square, opaque acrylic container (42 cm × 42 cm × 25 cm) located in a dimly lit room. The floor of the container was divided into 9 cm × 14 cm squares. Open field behavior was recorded using a digital video monitoring system (MED-VFC-NIR-M, Med Associates, Georgia, VT) that quantified overall activity by accumulating the change in contrasting video pixels over the recording session at a sample rate of 60 Hz. The digital video also served as a record of behavior in the open field and was used to score the time spent in center of the open field, the number of center crossings, the frequency of grooming, rearing, and the number of attempts to escape the open field which were defined as jumps against the side wall. All scoring of behavior was done offline by an observer blind to the animal’s group assignment. Following 2 weeks of access to a running wheel, mice (n = 12 exercising and n = 14 non-exercising) were transported to the lab in their home cage. A single mouse was placed in the open field and its behavior recorded for 6 min. Following this, the mouse was removed and returned to its home cage. The open field was thoroughly cleaned and dried prior to running the next mouse. 2.3.3. Rota-rod In order to assess motor behavior in exercising and non-exercising mice, mice (n = 16 exercising and n = 16 non-exercising) were placed on a rota-rod (ENV-576, Med Associates, Georgia, VT) programmed to accelerate from 4 to 40 rpm over a 5 min period. Each mouse was given three consecutive trials with a 10 min inter-trial interval. 2.3.4. Stress-induced hyperthermia In mice, core body temperature appears to be a sensitive measure of stress and anticipatory anxiety [75,114]. Borsini et al. [6] observed that among group housed mice, mice taken last from the cage had higher core body temperatures than mice taken first. The increase in temperature has been likened to an ‘emotional fever’ and is thought to be related to stress and anticipatory anxiety associated with handling cage mates and so has been termed stress-induced hyperthermia (SIH) [75]. Consistent with this interpretation, anxiolytic drugs decrease this SIH [6,45,65] while anxiogenic drugs increase SIH [6]. In an extension of this effect, Van Der Heyden [94] showed that within singly housed mice, repeating the rectal temperature measurement at 10 min intervals led to a robust hyperthermia that reached asymptote within 30 min. This effect was also blocked by anxiolytic drugs [76,94] and has been offered as a method for examining anticipatory anxiety in individual mice. To examine whether 2 weeks of voluntary exercise would decrease SIH, groups of mice (n = 15 exercising and n = 15 non-exercising) were individually housed 24 h prior to temperature measurement. Mice from exercising groups continued to have access to a functioning running wheel while individually housed. Core body temperature was measured with a Thermalert Monitoring Thermometer (PhysiTemp TH-5, Clifton, New Jersey) equipped with a mouse rectal temperature probe (PhysiTemp RET-3, 0.16 mm tip diameter). Temperature measurements were made in the colony room to decrease potential effects of transport stress. For temperature measurement a mouse was placed into a small acrylic restraining tube (4 cm × 12 cm) and lightly restrained by holding the tail. The temperature probe was lubricated with peanut oil and then inserted 5 mm into the rectum and held in place for 30 s. At 30 s, the temperature was recorded to the nearest 0.1 ◦ C. The mouse was then placed back into its home cage (still individually housed). Ten minutes later, the mouse’s temperature was again assessed. SIH was defined as an increase in core body temperature from the first temperature measurement to the second measurement. 2.3.5. Social interaction The social interaction test is an ethologically relevant test in which the time spent by pairs of rodents in social interaction (sniffing, grooming or following one another) is taken as a measure of emotionality [29,31]. Rodents that make more frequent social contacts or spend more time contacting are thought to be less anxious. Consistent with this, anxiolytic drugs increase the amount of social interaction [28] and anxiogenic drugs decrease the amount of social interaction [30].

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40

33

To examine whether 2 weeks of voluntary exercise would increase social interaction, pairs of mice (n = 12 pairs of exercising and n = 12 pairs of non-exercising mice) were placed in a shallow rectangular observation box (42 cm × 42 cm × 25 cm) located in a dimly lit room. The box was novel to each pair of mice. Each mouse of a pair was taken from a separate cage but both mice of a pair were from either exercising or non-exercising groups. Social interaction was video taped and scored offline by an observer blind to the animal’s group assignment. The frequency of contacts (sniffing, following, grooming, climbing on) and the amount of time contacting between the two mice were measured over a 5 min observation period. Care was taken to clean the observation box between each pair of mice. 2.3.6. Light-enhanced startle Exposure to high levels of illumination enhances acoustic startle amplitude in both rats and mice. Given the aversive quality of bright lights in rodents, Walker and Davis have argued that light-enhanced startle (LES) reflects unconditioned anxietylike response to high levels of illumination. Consistent with this interpretation, LES is attenuated by anxiolytic drugs [103,105] and by lesions of the bed nucleus of the stria terminalis [104] an area known to play an important role in anxiety-related behaviors [3,14,26,47,78,80,86,102,104,107]. To examine whether voluntary exercise would affect LES, mice (n = 20 exercising and n = 20 non-exercising) were tested for acoustic startle on each of three consecutive days (i.e., days 15–17 of wheel access) as described above. These tests were meant to provide internal replication of the reduction of acoustic startle by prior exercise. Mice were then tested for LES in two phases. Mice were placed individually in the startle apparatus and after a 5 min acclimation period, mice were presented with the first of 30 startle stimulus alone trials. Ten stimuli of each intensity level (95, 100, and 105 dB) were presented in a pseudo-random order (the constraint being that each intensity occur within each block of three trials) with a mean inter-trial interval (ITI) of 60 s. This constituted phase I. Mice remained in the startle apparatus, and after 5 min stimulus free period, were given an additional 30 startle stimulus alone trials (10 each at 95, 100, and 105 dB). This constituted phase II. Each mouse was tested twice using this procedure (days 18 and 19 of exercise). For one of these tests the startle apparatus was dark for both phases I and II (i.e., dark–dark) [103,105]. On the alternate day the chamber was illuminated during phase II (i.e., dark–light). Illumination was provided by an 8 W fluorescent bulb located 10 cm above and behind the behavioral chamber. The ordering of dark–dark and dark–light tests was counterbalanced across groups. Mean startle amplitude was computed for each mouse and for each test phase. The percent change in startle from phases I to II was then computed for the dark–dark and dark–light tests. LES was defined as greater percent change in the presence of the light than in the dark. During the experiment one mouse from the exercising group was euthanized due to illness leaving 19 mice in the exercising group. 2.3.7. Statistics Student t-tests were used for analysis of between group data in the absence of a repeated measure. Analysis of variance was used for all other comparisons with group (exercising and non-exercising) as a between subject factor. Significant interactions were followed up with lower order ANOVAs and Student t-tests. A chi square test for independence was used to examine escape attempts in the open field experiment. For all comparisons ˛ = 0.05.

3. Results 3.1. Wheel running and body weight C57BL/6 mice run approximately 5 km/day [50]. In order to quantify wheel running in our group housed mice, we modified running wheels to accept standard bicycle odometers (Enduro 8, Cateye Inc., Boulder, CO). The total distance run and elapsed time were recorded for each cage of 4 mice (n = 4 cages) at 7:00 and 19:00 h each day (corresponding to lights on and light off in the colony room). Body weight was taken for each mouse every 3 days and compared to a similar cohort of mice with locked running wheels. Average running distance and elapsed time for the 4 cages of mice are shown in Fig. 1. The 4 mice in each cage gradually increased their running distance over the first 7 days. Consistent with previous reports, over the 14 days the groups of 4 mice ran an average of 16.5 km/day (±0.7 km). At day 14, the groups of 4 mice ran an average of 18.3 km (±0.7 km), 17.6 km during the dark cycle alone (Fig. 1A). The amount of time the groups of 4 mice ran was also stable over the 14 days averaging 578 min (±19.9 min) during the dark cycle and 58.7 min (±10.6 min) during the light cycle (Fig. 1B). Because mice are group housed, we sought to examine the pro-

Fig. 1. Average running distance, running time and body weight in group housed C57BL/6J mice. Mice were housed in groups of 4 and the running data represent 4 cages of mice. Groups of mice ran an average of 16.5 km/day (±0.7 km). At day 14 the groups of mice ran an average of 18.3 km (±0.7 km), 17.6 km during the dark cycle alone (A). The amount of time the groups of mice run was also stable over the 14 days averaging 578 min (±19.9 min) during the dark cycle and 58.7 min (±10.6 min) during the light cycle (B). Body weight increased for both exercising and non-exercising mice (C). There was no difference in body weight between the two groups.

portion of time individual mice run using near infra-red video recording. Mice were video taped for a 6 h prior during the dark cycle between days 11 and 14. Individual mice were identified with unique tail markings and a time sampling procedure was used to identify which mouse was on the wheel. The interval between samples was 3 min. On average, each mouse within a cage was observed running on the wheel in 25% of the samples (±1.5%, range 16–34%). Thus, on average, all 4 mice in a group cage run on the wheel an equivalent proportion of the total time. Consistent with previous reports [4,20,50] there were no body weight differences between exercising and non-exercising groups (F(5,30) = 1.18, p = 0.286). Fig. 1C shows the body weights sampled across the 14 days. While both groups of mice gained weight over the 14 days (F(5,30) = 84.36, p < 0.000), there was no difference in weight gain between exercising and non-exercising mice (F(5,30) = 1.44, p = 0.212).

34

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40

Fig. 2. Two weeks of voluntary exercise is associated with lower acoustic startle amplitude. Acoustic startle was assessed on each of three consecutive days following 2 weeks voluntary access to a functioning (exercising) or non-functioning (nonexercising) running wheel. Exercising mice showed lower acoustic startle amplitude across the three acoustic startle tests (F(1,30) = 8.93, p = 0.006).

3.2. Acoustic startle Mice given free access to a running wheel showed lower acoustic startle amplitude across the three tests (Fig. 2; F(1,30) = 8.93, p = 0.006). There was no effect of startle test F(2,60) = 1.91, p = 0.157) and no exercise group by startle test interaction, F(2,60) = 1.191, p = 0.312. In some circumstances acoustic startle amplitude is negatively correlated with motor activity. Thus it is possible that the lower acoustic startle amplitude in exercising mice is a consequence of increased motor activity in the chamber. We examined this by sampling the cage for the 100 ms period prior to the onset of the startle stimulus. There were no differences in motor activity between exercising and non-exercising mice (Table 1; F < 1). Activity decreased over the three acoustic startle tests (F(2,60) = 9.12, p = < 0.000) perhaps as a result of habituation to the apparatus or freezing induced by repeated startle tests [64], but this effect was similar in exercising and non-exercising mice (F(2,60) = 1.62, p = 0.205). To extend these findings and to examine whether voluntary exercise indeed acts to reduce startle amplitude, we used a within subjects design and examined startle amplitudes before and after 2 weeks of voluntary exercise. Naive mice were given 2 weeks of access to a locked running wheel and then tested for acoustic startle. The wheels were then unlocked for half of the mice. Two weeks later the mice were tested for acoustic startle a second time. After 2 weeks of access to an unlocked running wheel, startle amplitude was reduced from the pre-exercising levels (Post-Test, exercising group) but remained relatively constant in mice in

Fig. 3. Two weeks of voluntary exercise reduces startle amplitude. Acoustic startle amplitude was assessed before and after 2 weeks voluntary access to a functioning (exercising) or non-functioning (non-exercising) running wheel. Startle amplitude in the non-exercising group was unchanged across the 2 weeks (F < 1) whereas startle amplitude in the exercising group was reduced from their pre-exercising levels (F(1,30) = 20.93, p = 0.000).

which the wheels were locked throughout (Fig. 3; Post-Test, non-exercising group; group by test interaction F(1,30) = 12.01, p = 0.002). Simple effects tests showed that startle amplitude in the non-exercising group was unchanged across the 2 weeks (F < 1) whereas startle amplitude in the exercising group was reduced from their pre-exercising levels (F(1,30) = 20.93, p = 0.000). Thus exercise is associated with a reduction in startle amplitude. The reduction in acoustic startle by voluntary exercise was not associated with differences in motor activity (Table 1) or body weight (Fig. 1C). Nor does it appear to be associated with a change

Table 1 Pre-startle stimulus activity does not differ between exercising and non-exercising mice Startle test

1

2

3

Exercising Non-exercising

196.7 (22.5) 238.3 (38.5)

172.3 (31.3) 212.7 (31.8)

164.8(22.2) 168.5 (19.1)

Numbers in parentheses represented the standard error of the mean.

Fig. 4. Two weeks of voluntary exercise was not associated with altered prepulse inhibition of startle. Exercising and non-exercising mice tested for PPI using prepulses of 4, 8, 12 or 16 kHz. Exercising mice had lower startle amplitude than non-exercising mice (no prepulse trials, Mexercising = 260.8, Mnon-exercising = 797.9; t (24) = 3.11, p = 0.005). There was no effect of group, prepulse frequency or group by prepulse interaction (F’s < 1).

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40

in the ability to process auditory stimuli (Fig. 4) as exercising and non-exercising mice had similar levels of PPI across a range or prepulse frequencies. On startle stimulus alone (i.e., no prepulse) trials, exercising mice once again had lower startle amplitude than non-exercising mice (Mexercising = 260.8, Mnon-exercising = 797.9; t (24) = 3.11, p = 0.005). However, both exercising and non-exercising mice showed equivalent prepulse inhibition of startle across the 4 prepulse frequencies (all main effects and interaction; F’s < 1) suggesting that the lower startle amplitude in exercising mice is not associated with changes in the ability to process auditory stimuli [108–112]. 3.3. Open field Mice given free access to a running wheel spent more time in the center of the open field, (t (24) = 2.50, p = 0.04), crossed the center more often (t(24) = 2.11, p = 0.02), and were less likely

35

to attempt to escape from the open field (2 (1, N = 26) = 17.33, p < 0.000) than non-exercising mice (Fig. 5). The number of rears did not differ between the groups (data not shown). Together, these data are consistent with the hypothesis that exercise is associated with a reduction in anxiety. However, exercising mice had more episodes of grooming (t(24) = 3.03, p = 0.01), and were less active in the open field (Fig. 6; F(1,22) = 5.93, p = 0.023) which are often interpreted as indicating increased anxiety [58–60,90]. Grooming is a common rodent behavior that is sensitive to levels of stress. High anxiety is associated with increased bouts of grooming with short durations whereas decreased anxiety is associated with less frequent, longer bouts of grooming. In the present study we observed that exercising mice showed more bouts of grooming, however we did not measure grooming duration. Thus, on the face of it, increased episodes may be indicative of increased anxiety in exercising mice. However, as argued by Kalueff and Tuohimaa

Fig. 5. Two weeks of voluntary exercise is associated with an increased number of open field center crossings, increased time spent in the center of the open field, more grooming and fewer escape attempts.

36

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40

Fig. 6. Exercising mice were less active in the open field (F(1,22) = 5.93, p = 0.023).

[58–60], given the complexity of grooming behavior, any further interpretation may require a more detailed ethological analysis. In a recent series of experiments, Duman et al. [21] suggest that exercise-induced fatigue may contribute to decreased open field activity levels in exercising mice. In their experiment, exercising mice had lower home cage activity and reduced activity in the open field when testing was carried out shortly into the light cycle (i.e., shortly after peak running time). Although we found no differences in pre-startle activity (see above), we set out to examine whether exercising mice were more fatigued using a common measure of motor activity, the rota-rod task. If exercising mice are fatigued, then they might perform more poorly on this task. 3.4. Rota-rod Both exercising and non-exercising mice improved their performance over trials (Fig. 7; F(2,60) = 12.381, p < 0.000) but

Fig. 8. Two weeks of voluntary exercise is associated with a reduction in stressinduced hyperthermia (SIH). Core body temperature was assessed twice with a 10 min interval between measures. Shown is the change in core body temperature. There was no difference between the groups in their initial core body temperature (Mexercising = 35.64, Mnon-exercising = 35.59; t < 1). Non-exercising mice show a larger increase in core body temperature.

importantly, exercising mice had longer latencies to fall off the rota-rod (F(1,30) = 5.76, p = 0.023) indicative of better motor performance. Thus, it does not appear that exercising mice are more fatigued at the time of testing. 3.5. Stress-induced hyperthermia Mice given free access to a running wheel showed less SIH (Fig. 8; t (28) = 2.96, p = 0.004). There was no difference between the groups in their initial core body temperature (Mexercising = 35.64, Mnon-exercising = 35.59; t < 1) and both groups showed a significant increase in temperature from time 1 to time 2 (t (14) = 5.40, p < 0.000, t (14) = 8.57, p < 0.000, for groups exercising and nonexercising respectively). Thus, exercise does not eliminate SIH but reduces its magnitude. 3.6. Social interaction Mice given free access to a running wheel made more social contacts (t (22) = 2.23, p = 0.036) and spent more time contacting (t (22) = 5.12, p < 0.000) than non-exercising mice (Fig. 9). We did not observe any instances of bighting or fighting among the pairs of mice in either group. Increased social contacts and time spent contacting is indicative of reduced anxiety. 3.7. Light-enhanced startle

Fig. 7. Two weeks of voluntary exercise is associated with improved performance on the accelerating rota-rod task. Exercising mice had longer latencies to fall off the rota-rod (F(1,30) = 5.76, p = 0.023). The fact that exercising mice show better rota-rod performance suggests that their decreased open field activity is not the consequence of exercise-induced fatigue.

Exercising mice did not show LES (Fig. 10). ANOVA revealed a significant effect of group (F(1,37) = 6.46, p = 0.015) and test (F(1,37) = 7.93, p = 0.008). The test by group interaction was not significant (F(1,37) = 1.67, p = 0.204). However, a priori t-tests were conducted to evaluate LES within each group. In non-exercising mice, but not exercising mice, the increase in startle from phases I to II was larger in the presence of the light (t(19) = 4.08, p < 0.000; t(18) = 1.48, p = 0.156; non-exercising and exercising groups respectively). Thus, only the non-exercising mice showed LES. Because LES is thought to reflect an unconditioned anxiety-like response to high levels of illumination [103,105], these data are consistent with the hypothesis that voluntary exercise reduces anxiety. Lastly, as shown in Experiment 1, startle amplitude was lower in mice given

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40

37

Fig. 9. Two weeks of voluntary exercise is associated with an increase in social interaction. Exercising mice show more social contacts and spend more time contacting than non-exercising mice.

free access to a running wheel (average of the three test session; Mexercising = 378, Mnon-exercising = 806.5; F(1,37) = 21.47, p < 0.000). 4. Discussion Exercise is associated with a reduction in anxiety in humans [23,34,70,72,81,83]. However, in rodents, the effects of exercise are much less consistent with reports of an anxiolytic effect of voluntary exercise [4,17,18,21,42,43], no effect of exercise [79] or increased anxiety following exercise [9,95]. The inconsistency in this relatively small group of studies is likely due to any number of procedural differences including species (rat or mouse), sex, housing conditions (e.g., single versus grouped housed) and anxiety model. For this reason we set out to examine the putative anxiolytic effects of exercise across several rodent models. We have shown

Fig. 10. Two weeks of voluntary exercise is associated with a decrease in lightenhanced startle. Startle amplitude was assessed in two consecutive phases on each of 2 days. On one day phases I and II testing were both conducted in the dark. On the alternate day phase 1 testing was in the dark and phase II testing was in the presence of high illumination. Shown is the percent change in startle from phases I to II for dark–dark and dark–light tests. Non-exercising mice show an enhancement of startle in high illumination. Exercising mice do not.

that 2 weeks of voluntary exercise in group housed male C57BL/6J mice is consistently associated with reduced anxiety-like behavior as measured with acoustic startle, open field, stress-induced hyperthermia, social interaction and light-enhanced startle. The fact that 2 weeks of exercise was sufficient to produce an anxiolytic effect is interesting because in other studies rodents may run for up to 6 weeks. Unpublished data from our lab suggests that 1 week, but not 3 days, of exercise is sufficient to reduce startle amplitude and the reduction in startle amplitude persists as long as the mice are allowed to run (up to 12 weeks). One unexpected finding was that exercising mice showed lower activity in the open field. This effect was first reported by Duman et al. [21] and was attributed to exercise-induced fatigue. While Duman et al. provided strong support for their fatigue hypothesis [21], we do not believe that fatigue plays a role in the decreased open field activity in the present experiment. We found that exercising mice performed better than non-exercising mice on the accelerating rota-rod, a task that would presumably be sensitive to fatigue. Moreover, we found no differences in home cage activity or in pre-startle activity between exercising and non-exercising mice. Activity in the open field is affected by many variables [9] and decreased activity may not always reflect increased anxiety. For example, it has been suggested that increased activity in the open field, reflected in behaviors such as rapid running and jumping, may be positively associated with defensive or panic behaviors [16,18,54]. Thus, lower activity in the open field may reflect an attenuation of defensive, panic-like behavior (e.g., darting and jumping) elicited by the open field [16,18,19,54]. Consistent with this, the activity differences were confined to the open field and we observed that more non-exercising mice made attempts to escape the open field. The exercise-induced reduction in acoustic startle amplitude is particularly impressive because acoustic startle may be a very useful model system for examining the neural correlates of the anxiolytic effects of exercise. The acoustic startle response is a sensitive measure of anxiety [14,107]: drugs and environmental conditions known to increase anxiety increase acoustic startle [44,66,103] whereas drugs and environmental conditions known to decrease anxiety decrease acoustic startle [15,61,82]. Moreover, acoustic startle is easily and objectively quantified [12,106]

38

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40

and a great deal is known about the pharmacology and neural circuitry for acoustic startle and its modulation by anxiety [14,67,107,113]. Any number of adaptive changes in the brain could contribute to the anxiolytic effect of voluntary exercise [18,19,22,43,85]. However, given the important role of serotonin (5-HT) in anxiety (see [48] for review) and in the etiology and treatment of anxiety disorders, it is likely that changes in central 5-HT functioning play some role in the anxiolytic effects of exercise [41,42]. Activation of 5-HT2 receptors produces anxiogenic responses across a range of models including those used in the present experiment [2,32,49,62] and mice lacking 5-HT2 receptors show an anxiolytic phenotype [52]. Thus, voluntary exercise may reduce anxiety by decreasing the function at post-synaptic 5-HT2 receptors. Consistent with this, we have recently shown that 2 weeks of voluntary exercise blunted the startle-enhancing effect of the anxiogenic 5-HT2B/C agonist meta-chlorophenylpiperazine (mCPP) [32]. Greenwood et al. [41–43] have provided further evidence that voluntary exercise alters central 5-HT function. Six weeks of voluntary exercise in rats was associated with an up-regulation of mRNA for 5-HT1A somatodentritic autoreceptors in the dorsal raphe nucleus (DRN) [41]. If the up-regulation of mRNA for 5-HT1A somatodentritic autoreceptors results in an up-regulation of receptor protein, additional 5-HT1A autoreceptors would decrease DRN activity by enhancing autoinhibition of DRN cell firing. This, in turn, should decrease 5HT release in DRN projection areas that are known to play a role in anxiety-related behaviors. Consistent with this, Dishman et al. [19] have reported that voluntary exercise in rats was associated with decreased shock-induced elevation in the 5-HT metabolite 5-hydroxyindole acetic acid in the hippocampus and amygdala suggesting that exercise decreased 5-HT function in these DRN targets associated with anxiety-like behavior. A number of studies suggest that the central extended amygdala, which includes the central nucleus of the amygdala (CeA) and bed nucleus of the stria terminalis (BNST), plays an important role in fear and anxiety [14,86,92,107]. Lesion or inactivation of the BNST interferes with anxiety-like responding to intracerebroventricular corticotropin-releasing factor (CRF) [66], bright lights (LES) [104], uncontrollable shock [47], and exposure to predator odor [26]. Interestingly, manipulation of 5-HT in the BNST affects anxiety-like behavior. For example, direct intra-BNST infusion of the anxiogenic 5-HT2B/C agonist meta-chlorophenylpiperazine (mCPP) [27,35,39] increases acoustic startle (unpublished results) whereas direct infusions of the 5-HT1-like/7 agonist 5-carboxytryptamine (5-CT) produces an anxiolytic-like effect on acoustic startle [69]. The fact that voluntary exercise is anxiolytic across a range of rodent models including those affected by manipulation of the BNST, raises the interesting possibility that voluntary exercise affects anxiety through changes in the BNST. Studies are underway to examine whether voluntary exercise affects 5-HT function in the BNST. Acknowledgements This work was supported by the Department of Psychology, University of Vermont. The authors wish to thank John Green and Jom Hammack for their comments on the manuscript. References [1] Archer J. Tests for emotionality in rats and mice: a review. Animal Behaviour 1973;21:205–35. [2] Bagdy G, Graf M, Anheuer ZE, Modos EA, Kantor S. Anxiety-like effects induced by acute fluoxetine, sertraline or m-CPP treatment are reversed by pretreatment with the 5-HT2C receptor antagonist SB-242084 but not the 5-HT1A receptor antagonist WAY-100635. International Journal of Neuropsychopharmacology 2001;4:399–408.

[3] Bangasser DA, Santollo J, Shors TJ. The bed nucleus of the stria terminalis is critically involved in enhancing associative learning after stressful experience. Behavioral Neuroscience 2005;119:1459–66. [4] Binder E, Droste SK, Ohl F, Reul JM. Regular voluntary exercise reduces anxietyrelated behaviour and impulsiveness in mice. Behavioural Brain Research 2004;155:197–206. [5] Black JE, Isaacs KR, Anderson BJ, Alcantara AA, Greenough WT. Learning causes synaptogenesis, whereas motor activity causes angiogenesis, in cerebellar cortex of adult rats. Proceedings of the National Academy of Sciences of the United States of America 1990;87:5568–72. [6] Borsini F, Lecci A, Volterra G, Meli A. A model to measure anticipatory anxiety in mice? Psychopharmacology 1989;98:207–11. [7] Brandon J, Loftin J. Relationship of fitness to depression, state and trait anxiety, internal health locus of control, and self control. Perceptual and Motor Skills 1991;73:563–8. [8] Brown J, Cooper-Kuhn CM, Kempermann G, Van Praag H, Winkler J, Gage FH, et al. Enriched environment and physical activity stimulate hippocampal but not olfactory bulb neurogenesis. European Journal of Neuroscience 2003;17:2042–6. [9] Burghardt PR, Fulk LJ, Hand GA, Wilson MA. The effects of chronic treadmill and wheel running on behavior in rats. Brain Research 2004;1019:84–96. [10] Craft L, Perna F. The benefits of exercise for the clinically depressed. Primary Care Companion to the Journal of Clinical Psychiatry 2004;6:104–11. [11] Crawley JN. Exploratory behavior models of anxiety in mice. Neuroscience & Biobehavioral Reviews 1985;9:37–44. [12] Davis M. In: Eaton RC, editor. The mamalian startle response in neural mechanisms of startle behavior. New York: Plenum Publishing; 1984. p. 287–351. [13] Davis M. Pharmacological analysis of fear-potentiated startle. Brazilian Journal of Medical and Biological Research 1993;26:235–60. [14] Davis M. Neural systems involved in fear and anxiety measured with fearpotentiated startle. American Psychologist 2006;61:741–56. [15] Davis M, Falls WA, Campeau S, Kim M. Fear-potentiated startle: a neural and pharmacological analysis. Behavioral Brain Research 1993:58. [16] de Bortoli VC, Nogueira RL, Zangrossi Jr H. Effects of fluoxetine and buspirone on the panicolytic-like response induced by the activation of 5-HT1A and 5HT2A receptors in the rat dorsal periaqueductal gray. Psychopharmacology 2006;183:422–8. [17] Dishman RK. Brain monoamines, exercise, and behavioral stress: animal models. Medicine & Science in Sports & Exercise 1997;29:63–74. [18] Dishman RK, Dunn AL, Youngstedt SD, Davis JM, Burgess ML, Wilson SP, et al. Increased open field locomotion and decreased striatal GABAa binding after activity wheel running. Physiology and Behavior 1996;60:699–705. [19] Dishman RK, Renner KJ, Youngstedt SD, Reigle TG, Bunnell BN, Burke KA, et al. Activity wheel running reduces escape latency and alters brain monoamine levels after footshock. Brain Research Bulletin 1997;42:399–406. [20] Droste SK, Gesing A, Ulbricht S, Muller M, Linthorst A, Reul JM. Effects of longterm voluntary exercise on the mouse hypothalamic-pituitary-adrenal axis. Endocrinology 2003;144:3012–23. [21] Duman CH, Schlesinger L, Russel D, Duman RS. Voluntary exercise produces antidepressant and anxiolytic behavioral effects in mice. Brain Research 2008;1199:148–58. [22] Dunn AL, Reigle TG, Youngstedt SD, Armstrong RB, Dishman RK. Brain norepinephrine and metabolites after treadmill training and wheel running. Medicine & Science in Sports & Exercise 1996;28:204–9. [23] Dunn AL, Trivedi MH, O’Neal HA. Physical activity dose-response effects on outcomes of depression and anxiety. Medicine & Science in Sports & Exercise 2001;33:609–10 (S587–97; discussion). [24] Ernst C, Olson AK, Pinel JPJ, Lam RL, Christie BR. Antidepressant effects of exercise: Evidence for an adult-neurogenesis hypothesis? Journal of Psychiatry & Neuroscience 2006;31:84–92. [25] Farmer J, Zhao X, van Praag H, Wodtke K, Gage FH, Christie BR. Effects of voluntary exercise on synaptic plasticity and gene expression in the dentate gyrus of adult male Sprague–Dawley rats in vivo. Neuroscience 2004;124:71–9. [26] Fendt M, Endres T, Apfelbach R. Temporary inactivation of the bed nucleus of the stria terminalis but not of the amygdala blocks freezing induced by trimethylthiazoline, a component of fox feces. Journal of Neuroscience 2003;23:23–8. [27] Feuchtl A, Bagli M, Stephan R, Frahnert C, Kolsch H, Kuhn KU, et al. Pharmacokinetics of m-chlorophenylpiperazine after intravenous and oral administration in healthy male volunteers: implication for the pharmacodynamic profile. Pharmacopsychiatry 2004;37:180–8. [28] File SE. The use of social interaction as a method for detecting anxiolytic activity of chlordiazepoxide-like drugs. Journal of Neuroscience Methods 1980;2:219–38. [29] File SE, Hyde JR. Can social interaction be used to measure anxiety? British Journal of Pharmacology 1978;62:19–24. [30] File SE, Pellow S, Braestrup C. Effects of the beta-carboline FG 7142 in the social interaction test of anxiety and the holeboard: correlations between behaviour and plasma concentrations. Pharmacology, Biochemistry & Behavior 1985;22:941–4. [31] File SE, Seth P. A Review of 25 years of the social interaction test. European Journal of Pharmacology 2003;463:35–53. [32] Fox JH, Hammack SE, Falls WA. Exercise is associated with reduction in the anxiogenic effect of mCPP on acoustic startle. Behavioral Neuroscience 2008;122:943–8.

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40 [33] Fox K. The influence of physical activity on mental well being. Public Health Nutrition 1999;2:411–8. [34] Fox KR. The influence of physical activity on mental well-being. Public Health Nutrition 1999;2:411–8. [35] Gatch MB. Discriminative stimulus effects of m-chlorophenylpiperazine as a model of the role of serotonin receptors in anxiety. Life Sciences 2003;73:1347–67. [36] Geyer MA, Krebs-Thomson K, Braff D, Swerdlow N. Pharmacological studies of prepulse inhibition models of sensorimotor gating deficits in schizophrenia: a decade in review. Psychopharmacology 2001;156:117–54. [37] Gomez-Pinilla F, Ying Z, Roy RR, Molteni R, Edgerton VR. Voluntary exercise induces a BDNF-mediated mechanism that promotes neuroplasticity. Journal of Neurophysiology 2002;88:2187–95. [38] Goodwin RD. Association between physical activity and mental disorders among adults in the United States. Preventive Medicine 2003;36: 698–703. [39] Graeff FG, Garcia-Leal C, Del-Ben CM, Guimaraes FS. Does the panic attack activate the hypothalamic-pituitary-adrenal axis? Anais Da Academia Brasileira de Ciencias 2005;77:477–91. [40] Greenwood BN, Foley TE, Burhans D, Maier SF, Fleshner M. The consequences of uncontrollable stress are sensitive to duration of prior wheel running. Brain Research 2005;1033:164–78. [41] Greenwood BN, Foley TE, Day H, Burhans D, Brooks L, Campaeu S, et al. Wheel running alters serotonin (5-HT) transporter, 5-HT(1a), 5-HT(1b), and alpha(1b)-adrenergic receptor mRNA in the rat raphe nuclei. Biological Psychiatry 2004;57:559–68. [42] Greenwood BN, Foley TE, Day HE, Campisi J, Hammack SH, Campeau S, et al. Freewheel running prevents learned helplessness/behavioral depression: role of dorsal raphe serotonergic neurons. Journal of Neuroscience 2003;23:2889–98. [43] Greenwood BN, Strong PV, Brooks L, Fleshner M. Anxiety-like behaviors produced by acute fluoxetine administration in male Fischer 344 rats are prevented by prior exercise. Psychopharmacology 2008;199:209–22. [44] Grillon C, Pellowski M, Merikangas KR, Davis M. Darkness facilitates the acoustic startle response in humans. Biolgical Psychiatry 1997;42:453–60. [45] Groenink L, Compaan J, van der Gugten J, Zethof T, van der Heyden J, Olivier B. Stress-induced hyperthermia in mice. Pharmacological and endocrinological aspects. Annals of the New York Academy of Sciences 1995;771:252–6. [46] Guszkowska M. Effects of exercise on anxiety, depression, and mood. Psychiatria Polska 2004;38:611–20. [47] Hammack SE, Richey KJ, Watkins LR, Maier SF. Chemical lesions of the bed nucleus of the stria terminalis blocks the behavioral consequences of uncontrollable stress. Behavioral Neuroscience 2004;118:443–8. [48] Handley SL. 5-Hydroxytryptamine pathways in anxiety and its treatment. Pharmacology & Therapeutics 1995;66:103–48. [49] Handley SL, McBlane JW, Critchley MA, Njung’e K. Multiple serotonin mechanisms in animal models of anxiety: environmental, emotional and cognitive factors. Behavioural Brain Research 1993;58:203–10. [50] Harri M, Lindblom J, Malinen H, Hyttinen M, Lapvetelainen T, Eskola S, et al. Effects of access to a running wheel on behavior of C57BL/6J mice. Lab Animal Science 1999;49:401–5. [51] Harris A, Cronkite R, Moos R. Physical activity, exercise coping, and depression in a 10-year cohort study of depressed patients. Journal of Affective Disorders 2006;93:79–85. [52] Heisler LK, Zhou L, Bajwa P, Hsu J, Tecott LH. Serotonin 5-HT(2C) receptors regulate anxiety-like behavior. Genes, Brain, & Behavior 2007;6:491–6. [53] Hoffman HS, Ison JR. Reflex modification in the domain of startle: I Some empirical findings for how the nervous system processes sensory input. Psychological Review 1980;87:175–89. [54] Hogg S, Michan L, Jessa M. Prediction of anti-panic properties of escitalopram in the dorsal periaqueductal grey model of panic anxiety. Neuropharmacology 2006;51:141–5. [55] Isaacs KR, Anderson BJ, Alcantara AA, Black JE, Greenough WT. Exercise and the brain: angiogenesis in the adult rat cerebellum after vigorous physical activity and motor skill learning. Journal of Cerebral Blood Flow & Metabolism 1992;12:110–9. [56] Ison JR, Hoffman HS. Reflex modification in the domain of startle: II. The anomolous history of a robust and ubiquitous phenomenon. Psychological Bulletin 1983;94:3–17. [57] Jovanovic TKM, Fiallos A, Myers KM, Davis M, Duncan EJ. Fear potentiation and fear inhibition in a human fear-potentiated startle paradigm. Biological Psychiatry 2005;57:1559–64. [58] Kalueff AV, Tuohimaa P. Contrasting grooming phenotypes in C57Bl/6 and 129S1/SvImJ mice. Brain Research 2004;1028:75–82. [59] Kalueff AV, Tuohimaa P. Contrasting grooming phenotypes in three mouse strains markedly different in anxiety and activity (129S1, BALB/c and NMRI). Behavioural Brain Research 2005;160:1–10. [60] Kalueff AV, Tuohimaa P. Mouse grooming microstructure is a reliable anxiety marker bidirectionally sensitive to GABAergic drugs. European Journal of Pharmacology 2005;508:147–53. [61] Koch M, Schmid A, Schnitzler H-U. Pleasure-attentuation of startle is disrupted by lesions of the nucleus accumbens. Neuroreport 1996;7:1442–6. [62] Kshama D, Hrishikeshavan HJ, Shanbhogue R, Munonyedi US. Modulation of baseline behavior in rats by putative serotonergic agents in three ethoexperimental paradigms. Behavioral & Neural Biology 1990;54:234–53.

39

[63] Larun L, Nordheim L, Ekeland E, Hagen K, Heian F. Exercise in prevention and treatment of anxiety and depression among children and young people. In: Cochrane database of systematic reviews; 2006; 3: Art. No.: CD004691. doi:10.1002/14651858.CD004691.pub2. [64] Leaton RN, Cranney J. Potentiation of the acoustic startle response by a conditioned stimulus paired with acoustic startle stimulus in rats. Journal of Experimental Psychology: Animal Behavior Processes 1990;16:279– 87. [65] Lecci A, Borsini F, Volterra G, Meli A. Pharmacological validation of a novel animal model of anticipatory anxiety in mice. Psychopharmacology 1990;101:255–61. [66] Lee Y, Davis M. Role of the hippocampus, the bed nucleus of the stria terminalis, and the amygdala in the excitatory effect of corticotropin releasing hormone on the acoustic startle reflex. The Journal of Neuroscience 1997;17:6434–46. [67] Lee Y, Lopez DE, Meloni EG, Davis M. A primary acoustic startle pathway: obligatory role of the cochlear root neurons and the nucleus reticularis pontis caudalis. The Journal of Neuroscience 1996;16:3775–89. [68] Lemke MRFC, Wendorff T, Fritzer G, Rupp Z, Tetzlaff S. Modulation of involuntary and voluntary behavior following emotional stimuli in healthy subjects. Progress in Neuro-psychopharmacology & Biological Psychiatry 2005;29:69–76. [69] Levita L, Hammack SH, Mania I, Li X-Y, Davis M, Rainnie DG. 5Hydroxytryptamine 1A- like receptor activation in the bed nucleus of the stria terminalis: elctrophysiological and behaviroal studies. Neuroscience 2004;128:583–96. [70] Manger TA, Motta RW. The impact of an exercise program on posttraumatic stress disorder, anxiety, and depression. International Journal of Emergency Mental Health 2005;7:49–57. [71] McIlwain KL, Merriweather MY, Yuva-Paylor LA, Paylor R. The use of behavioral test batteries: effects of training history. Physiology & Behavior 2001;73:705–17. [72] Morgan WP, Goldstein SE. Exercise and mental health. In: Spielberger CD, editor. The series in health psychology and behavioral medicine. Washington, DC: Hemisphere Publishing Corporation; 1987. [73] Neeper SA, Gomez-Pinilla F, Choi J, Cotman C. Exercise and brain neurotrophins. Nature 1995;373:109. [74] Neeper SA, Gomez-Pinilla F, Choi J, Cotman CW. Physical activity increases mRNA for brain-derived neurotrophic factor and nerve growth factor in rat brain. Brain Research 1996;726:49–56. [75] Olivier B, Molewijk E, van Oorschot R, van der Poel G, Zethof T, van der Heyden J, et al. New animal models of anxiety. European Neuropsychopharmacology 1994;4:93–102. [76] Olivier B, Zethof TJ, Ronken E, van der Heyden JA. Anxiolytic effects of flesinoxan in the stress-induced hyperthermia paradigm in singly-housed mice are 5-HT1A receptor mediated. European Journal of Pharmacology 1998;342:177–82. [77] Paluska S, Schwenk T. Physical activity and mental health: current concepts. Sports Medicine (Auckland, NZ) 2000;29:167–80. [78] Pego JM, Morgado P, Pinto LG, Cerqueira JJ, Almeida OFX, Sousa N. Dissociation of the morphological correlates of stress-induced anxiety and fear. European Journal of Neuroscience 2008;27:1503–16. [79] Pietropaolo S, Feldon J, Alleva E, Ciruli F, Yee BK. The role of voluntary exercise in enriched rearing: a behavioral analysis. Behavioral Neuroscience 2006;120:787–803. [80] Sahuque LL, Kullberg EF, McGeehan AJ, Kinder JR, Hicks MP, Blanton MG, et al. Anxiogenic and aversive effects of corticotropin-releasing factor (CRF) in the bed nucleus of the stria terminalis in the rat: role of CRF receptor subtypes. Psychopharmacology 2006;186:122–32. [81] Salmon P. Effects of physical exercise on anxiety, depression, and sensitivity to stress: a unifying theory. Clinical Psychology Review 2001;21:33–61. [82] Schweimer J, Fendt M, Schnitzler HU. Effects of clonidine injections into the bed nucleus of the stria terminalis on fear and anxiety behavior in rats. European Journal of Neuroscience 2005;507:117–24. [83] Scully D, Kremer J, Meade MM, Graham R, Dudgeon K. Physical exercise and psychological well being: a critical review. British Journal of Sports Medicine 1998;32:111–20. [84] Shen H, Tong L, Balazs R, Cotman CW. Physical activity elicits sustained activation of the cyclic AMP response element-binding protein and mitogen-activated protein kinase in the rat hippocampus. Neuroscience 2001;107:219–29. [85] Soares J, Holmes PV, Renner KJ, Edwards GL, Bunnell BN, Dishman RK. Brain noradrenergic responses to footshock after chronic activity wheel running. Behavioral Neuroscience 1999;113:558–66. [86] Sullivan GM, Apergis J, Bush DE, Johnson LR, Hou M, Ledoux JE. Lesions in the bed nucleus of the stria terminalis disrupt corticosterone and freezing responses elicited by a contextual but not by a specific cue-conditioned fear stimulus. Neuroscience 2004;128:7–14. [87] Swain RA, Harris AB, Wiener EC, Dutka MV, Morris HD, Theien BE, et al. Prolonged exercise induces angiogenesis and increases cerebral blood volume in primary motor cortex of the rat. Neuroscience 2003;117:1037–46. [88] Swerdow NR, Geyer MA. Using an animal model of deficient sensorimotor gating to study the pathophysiology and new treatment of schizophrenia. Schizophrenia Bulletin 1998;24:285–301. [89] Swerdow NR, Martinez ZA, Hanlon FM, Platten A, Farid M, Auerbach P, et al. Toward understanding the biology of a complex phenotype: rat strain and sub-

40

[90] [91]

[92] [93]

[94] [95]

[96]

[97] [98]

[99] [100]

[101]

[102]

J.N. Salam et al. / Behavioural Brain Research 197 (2009) 31–40 strain differences in the sensorimotor gating-disruptive effects of dopamine agonists. The Journal of Neuroscience 2000;20:4325–36. Tharp GD, Carson WH. Emotionality changes in rats following chronic exercise. Medicine and Science in Sports 1975;7:123–6. Tong L, Shen H, Perreau VM, Balazs R, Cotman CW. Effects of exercise on gene-expression profile in the rat hippocampus. Neurobiology of Disease 2001;8:1046–56. Toufexis D. Region- and sex-specific modulation of anxiety behaviours in the rat. Journal of Neuroendocrinology 2007;19:461–73. Trivedi M, Greer T, Grannemann B, Chambliss H, Jordan A. Exercise as an augmentation strategy for treatment of major depression. Journal of Psychiatric Practice 2006;12:205–13. Van der Heyden JA, Zethof TJ, Olivier B. Stress-induced hyperthermia in singly housed mice. Physiology & Behavior 1997;62:463–70. Van Hoomissen JD, Holmes PV, Zellner AS, Poudevigne AM, Dishman RK. Effects of B-Adrenoreceptor Blockade During Chronic Exercise on Contextual Fear Conditioning and mRNA for galanin and brain-derived neurotrophic factor. Behavioural Neuroscience 2004;118:1378–90. van Praag H, Christie BR, Sejnowski TJ, Gage FH. Running enhances neurogenesis, learning, and long-term potentiation in mice. Proceedings of the National Academy of Sciences of the United States of America 1999;96:13427– 31. van Praag H, Chunm TS, Gage FH. Exercise enhances learning and hippocampal neurogenesis in aged mice. The Journal of Neuroscience 2005;25:8680–6. van Praag H, Kempermann G, Gage FH. Running increases cell proliferation and neurogenesis in the adult mouse dentate gyrus. Nature Neuroscience 1999;2:266–70. van Praag H, Kempermann G, Gage FH. Neural consequences of environmental enrichment. Nature Reviews Neuroscience 2000;1:191–8. Vaynman S, Ying Z, Gomez-Pinilla F. Exercise induces BDNF and synapsin I to specific hippocampal subfields. Journal of Neuroscience Research 2004;76:356–62. Vaynman S, Ying Z, Gomez-Pinilla F. Hippocampal BDNF mediates the efficacy of exercise on synaptic plasticity and cognition. European Journal of Neuroscience 2004;20:2580–90. Waddell J, Morris RW, Bouton ME. Effects of bed nucleus of the stria terminalis lesions on conditioned anxiety: aversive conditioning with long-duration

[103]

[104]

[105] [106]

[107]

[108]

[109]

[110]

[111]

[112]

[113] [114]

conditional stimuli and reinstatement of extinguished fear. Behavioral Neuroscience 2006;120:324–36. Walker DL, Davis M. Anxiogenic effects of high illumination levels assessed with the acoustic startle paradigm. Biological Psychiatry 1997;42: 461–71. Walker DL, Davis M. Double dissociation between the involvement of the bed nucleus of the stria terminalis and the central nucleus of the amygdala in startle increases produced by conditioned versus unconditioned fear. The Journal of Neuroscience 1997;17:9375–83. Walker DL, Davis M. Light enhanced startle: further pharmacological and behavioral characterization. Psychopharmacology 2002;159:304–10. Walker DL, Davis M. Quantifying fear potentiated startle using absolute versus proportional increase scoring methods: implications for the neurocircuitry of fear and anxiety. Psychopharmacology 2002;164:318–28. Walker DL, Toufexis DJ, Davis M. Role of the bed nucleus of the stria terminalis versus the amygdala in fear, stress, and anxiety. European Journal of Pharmacology 2003;463:199–216. Willott JF, Carlson S. Modification of the acoustic startle response in hearingimpaired C57BL/6J mice: prepulse augmentation and prolongation of prepulse inhibition. Behavioral Neuroscience 1995;109:396–403. Willott JF, Carlson S, Chen H. Prepulse inhibition of the startle response in mice: relationship to hearing loss and auditory system plasticity. Behavioral Neuroscience 1994;108:703–13. Willott JF, Kulig J, Satterfield T. The acoustic startle response in DBA/2 and C57/6 mice: relationship to auditory neuronal response properties and hearing impairment. Hearing Research 1984;16:161–7. Willott JF, Parham K, Hunter K. Response properties of inferior colliculus neurons in middle aged C57 mice with presbycusis. Hearing Research 1988;37:15–28. Willott JF, Turner JG, Carlson S, Ding D, Bross LS, Falls WA. The BALB/c mouse as an animal model for progressive sensorineural hearing loss. Hearing Research 1998;115:162–74. Yeomans JS, Frankland PW. The acoustic startle reflex: neurons and connections. Brain Research Reviews 1996;21:301–14. Zethof TJ, Van der Heyden JA, Tolboom JT, Olivier B. Stress-induced hyperthermia as a putative anxiety model. European Journal of Pharmacology 1995;294:125–35.