- Email: [email protected]
Variability factors in the expression of stress-induced behavioural sensitisation
Behavioural Brain Research 132 (2002) 69 – 76 www.elsevier.com/locate/bbr
Research report
Variability factors in the expression of stress-induced behavioural sensitisation Rianne Stam a,*, Teus-Jan van Laar a, Louis M.A. Akkermans b, Victor M. Wiegant a a
Di6isions of Pharmacology and Anatomy, Rudolf Magnus Institute for Neurosciences, Uni6ersity Medical Center Utrecht, P.O. Box 85060, 3508 AB Utrecht, The Netherlands b Di6ision of Surgery, Rudolf Magnus Institute for Neurosciences, Uni6ersity Medical Center Utrecht, P.O. Box 85060, 3508 AB Utrecht, The Netherlands Received 4 July 2001; received in revised form 13 September 2001; accepted 13 September 2001
Abstract Altered behavioural and physiological responsivity following a short session of foot shocks in the rat has proven to be a stable and clinically relevant model of stress-induced sensitisation. However, a number of key factors influencing effect size or direction have not previously been reported. Rats underwent a single, 15-min session of foot shocks and were exposed to a variety of novel stressful challenges 1 or 2 weeks later. Sensitised behavioural responses (increased immobility) in preshocked rats remained present over 3 days of repeated exposure to noise stress. In mild novel challenges (open field, empty cage), behavioural sensitisation and defecation was most clearly expressed at the beginning of the dark phase (evening). Higher-arousal challenges (prod, noise) caused increased behavioural inhibition in preshocked rats at all three time points (morning, afternoon, evening). Female preshocked rats showed a different pattern of behavioural and defecation sensitisation than preshocked males. The robustness of the model makes it suitable for further investigations into the mechanisms and vulnerability factors involved in the long-term consequences of stress. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Stress; Sensitisation; Rat; Behaviour
1. Introduction Increasing attention in recent years has focused on long-lasting sensitisation-like phenomena following intense but relatively short-lasting stressors (for a recent review, see Ref. [13]). The neurobiological principles involved differ from those in adaptive and dysregulated responsivity after chronic stress, and are thought to be relevant to post-traumatic psychiatric disturbances in humans and their physiological correlates [19]. One of the more widely used and well-characterised animal models of stress-induced sensitisation uses a short-lasting exposure to a grid cage in which electric currents are repeatedly, and usually unpredictably, applied [12]. * Corresponding author. Tel.: + 31-30-2538830; fax: +31-302539032. E-mail address: [email protected] (R. Stam).
Rats exposed to a single session of repeated foot shocks show sensitised behavioural responsivity in a variety of novel stressful challenges that develops over time [17,18]. The general pattern of behavioural sensitisation is stable after 7 days, though individual behavioural components may still show small increases after that [18]. Altered behavioural responsivity can be accompanied by neuroendocrine [16] and autonomic [2,14] hyperresponsivity, although their relative expression depends on the testing conditions used. It has been assumed that the expression of sensitisation depends to a high degree on the novelty of the subsequent challenge [18], but it is unclear whether this is a general phenomenon or test-dependent. While there is evidence for circadian variability in the expression of post-traumatic changes in humans [20], information on optimal time-points for the expression of shock-induced sensiti-
0166-4328/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 3 2 8 ( 0 1 ) 0 0 3 8 7 - 4
70
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
sation in rats is not yet available. Likewise, few studies have addressed sex-differences in the development and expression of stress-sensitisation, although there is clear evidence for gender-differences in vulnerability for stress-related psychiatric and physiological abnormalities [1,10]. The present study sought to address these gaps in our knowledge by systematically investigating the contribution of these three factors (repeated testing, circadian and sex variability) on the expression of foot shock-induced behavioural sensitisation.
2. Materials and methods
2.1. Animals and sensitisation procedure Experimental procedures were approved by the Commission on Laboratory Animal Experiments of the Faculty of Medicine, Utrecht University. Ten week old male and female Wistar rats (U:WU, Central Animal Laboratory, Utrecht University) were housed individually in macrolon cages measuring 20× 30 × 15 cm (l× w × h) and received standard laboratory chow and tap water ad libitum. Ambient temperature was kept at 21 °C and lighting regulated (on at 07:00, off at 19:00 h). All rats were briefly handled on 4 successive days in week 1. In week 2, between 15:00 and 17:30 h, half the rats were placed in a metal grid cage (32× 30 × 38 cm l ×w × h) and underwent a 15-min session of scrambled electric foot shocks (10×6 s, 0.5 mA, with randomised intervals of 20– 210 s; preshocked), the other half serving as non-shocked controls.
2.2. No6el stressful challenges 2.2.1. Experiment 1: effects of repeated testing Seven days after foot shocks, between 15:00 and 17:00 h in accordance with earlier studies on behaviour and colonic motility, male preshocked or control rats (n= 8 per group) were taken to an adjacent room and placed in an empty perspex cage with stainless steel floor (25×25×35 cm, l ×w × h) where they were exposed to a 85 dB broad spectrum noise. After 5 min the noise was switched off and the rats observed for another 5 min. Behaviour was recorded on video and any fecal boli produced were counted and the cage thoroughly cleaned with detergent and rinsed with water. The noise test was repeated on day 8 and 9 after foot shocks following identical procedures. Light intensity in the test cage was 270 lux. 2.2.2. Experiment 2: effects of time of day Male preshocked and control rats (n =24 per group) were each divided in three subgroups of n= 8. On the
four successive challenge days, one control and one preshocked group was always challenged between 09:00 and 11:00 h (light; morning), one control and preshocked group between 15:00 and 17:00 h (light; afternoon), and one control and preshocked group between 21:00 and 23:00 h (dark; evening). At the latter time point, animals were observed in red light. Seven days after foot shocks, rats were picked up from their home cages, taken to an adjacent room and placed in the center of a black polyester circular arena (‘open field’, diameter 130 cm, height of wall 30 cm). Traveled distance during 5 min was recorded automatically via a digitised image using Ethovision tracking software (Noldus Information Systems, Wageningen, The Netherlands). Eight days after foot shocks, rats were picked up from their home cages, taken to an adjacent room and placed in an empty perspex cage with stainless steel floor (25× 25× 35 cm, l × w× h) for 5 min. Nine days after foot shocks, rats were picked up from their home cages, taken to an adjacent room and placed in the empty perspex cage where an electrified prod was inserted that delivered a 2 mA current on touch and removed after 5 min. Ten days after foot shocks, rats were picked up from their home cages, taken to an adjacent room and placed in the empty perspex cage where they were exposed to a 85 dB broad spectrum noise. After 5 min the noise was switched off and the rats observed for another 5 min. On each challenge day, behaviour was recorded on video and any fecal boli produced were counted and the cage thoroughly cleaned with detergent and rinsed with water. Light intensity in morning and afternoon periods was 280 lux (open field) or 400 lux (empty cage, prod and noise tests). Light intensity in the evening was 6–8 lux.
2.2.3. Experiment 3: effects of sex Noise exposure was part of a larger pilot study into the subsequent effects of colonic distention, and experimental conditions therefore varied slightly from experiment 1 and 2. The control group was exposed to the shock cage without receiving any shocks, and testing was conducted 14 days later at a slightly later time in the afternoon. Male and female rats were subdivided in control and preshocked groups of n= 10 each. Fourteen days after foot shocks, between 16:45 and 17:45 h, rats were picked up from their home cages, taken to an adjacent room and placed in a perspex cage with stainless steel floor (25× 25×35 cm, l ×w× h) where they were exposed to a 85 dB broad spectrum noise. After 5 min the noise was switched off and the rats observed for another 5 min. Behaviour was recorded on video. Male and female rats were tested on alternate days so as not to confound behavioural reactivity by olfactory
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
cues. Light intensity at the level of the test cage was 270 lux.
2.3. Beha6iour analysis and statistics Behaviour analysis of taped sessions was performed blindly by an experienced observer. For empty cage and noise exposure, mean duration of the following six mutually exclusive components was calculated: locomotion (movement of whole body), rearing (upright posture), grooming, immobility and attention (sitting while head and whiskers are moving). For the electrified prod test, in addition to locomotion, rearing, grooming, immobility and attention as defined above, the set of mutually exclusive behavioural components included prod-directed behaviour (‘prod’, a combination of approaching, sniffing or otherwise exploring the prod). Effects of repeated testing and preshock treatment on behaviour in experiment 1 were assessed with multivariate analysis of variance for repeated measures. Differences in open field travelled distance and duration of behavioural components between preshocked and control groups in experiment 2 were evaluated with unpaired Student’s t-tests. Effects of preshock treatment and sex on behaviour in the noise test in experiment 2 were tested with two-factor univariate analysis of variance. Defecation differences (number of boli) were analysed with the non-parametric Mann– Whitney Utest (experiment 1 and 2) or Kruskal– Wallis analysis of variance (experiment 3). SPSS for WINDOWS 9.0 was used for all statistical analyses and significance levels set at P= 0.05.
71
3.2. Experiment 2: effects of time of day Open field traveled distance on day 7 showed trends toward a decrease in preshocked rats on all three time points, but the decreased activity was significant only in the evening group. This appeared mainly due to greater traveled distance in control rats in the dark period (Table 1). A significant increase in defecation, too, was found only in the evening period (Table 1). Effects of preshock treatment on behaviour in the empty cage on day 8 were clearest in the evening, with significantly decreased locomotion, rearing and attention and increased immobility (Fig. 2). Rats tested in the morning also showed less locomotion and more immobility, but no behavioural differences were found in the afternoon. In contrast, in the prod test on day 9, a very clear increase in immobility was seen in preshocked rats at all three time points, with a reduction in more active behaviours (locomotion, rearing, grooming, attention; Fig. 2). During the noise on period on day 10, preshocked rats showed increased immobility in the morning period and decreased locomotion in the evening (Fig. 3). During the following noise off period, immobility in preshocked rats was greater than that in controls in the afternoon and evening, and attention decreased in the afternoon. In the morning period, preshocked rats only showed decreased grooming behaviour (Fig. 3). No significant differences in defecation were found in empty cage, prod or noise tests (results not shown).
3.3. Experiment 3: effects of sex 3. Results
3.1. Experiment 1: effects of repeated testing Rats generally showed somewhat more active behaviour during the 5-min noise on period and more immobility during the 5-min noise off period (Fig. 1). In the noise on period, there was a small but significant decrease in the duration of locomotion over the 3 testing days (F(2, 28)= 7.64, P =0.002) as well as a small increase in grooming (F(2, 28)=4.33, P =0.023). Overall immobility was increased in the preshocked group compared to controls (F(1, 14)=5.73, P= 0.031). In the noise off period, immobility decreased over the 3 testing days (F(2, 28)= 5.25, P =0.012). The days × preshocked interaction did not reach significance (P=0.086), showing that the increased immobility response due to the previous foot shock experience (F(1, 14)=79.4, P B 0.001) did not habituate after repeated testing. No significant effects of test day or preshock treatment were found on defecation (results not shown).
During the noise on period, a significant preshocked× sex interaction indicated that the reduction in locomotion (F(1, 36)=4.25, P=0.047) and increase in immobility (F(1, 36)= 4.54, P= 0.040) caused by the preshock experience were more pronounced in males than in females (Fig. 4). Preshocks reduced the duration of rearing (F(1, 36)= 9.72, P=0.004) and grooming (F(1, 36)= 9.77, P= 0.003) in both males and females. During the noise off period, a significant preshocked× sex interaction appeared to result from an increased locomotion (F(1, 36)= 12.0, P=0.001) and rearing (F(1, 36)=8.80, P=0.005) response in preshocked females compared to the other three experimental groups (Fig. 4). Compared to their respective controls, the immobility response increased in preshocked males but decreased in preshocked females (F(1, 36)= 13.0, P= 0.001). A significant group effect was found for noise-induced defecation ( 2 = 10.1, P= 0.018). Post-hoc analysis indicated that the number of boli was higher only in preshocked compared to control females (mean n= 3 and 7, respectively; critical difference: 13.79).
72
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
4. Discussion Three experiments were conducted to clarify variability factors in the effectiveness of stress-induced sensitisation, using a model foot shock experience. Sensitised
behavioural responsivity remained present during 3 days of repeated testing with a noise challenge. In a relatively mild or less complex novel challenge (open field, empty cage), behavioural sensitisation was expressed most clearly in the evening (dark), while sensiti-
Fig. 1. Duration of five mutually exclusive behavioural components during 5 min after onset of noise (noise on) and 5 min after cessation of noise (noise off), in sequential testing 7, 8 and 9 days after a brief session of foot shocks (prs) or no shocks (con). For significant effects see the results section. locom, locomotion; rear, rearing; groom, grooming; immob, immobility; atten, attention.
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
73
Table 1 Open field traveled distance and defecation, with S.E.M., during 5 min testing, at three different time points, 7 days after foot shocks (prs) or no shocks (con) AM
Con Prs
PM1
PM2
Distance (cm)
Defecation (number)
Distance (cm)
Defecation (number)
Distance (cm)
Defecation (number)
2064 9154 1581 9232
5 4
19359244 16349166
5 6
2314 9158 1692 9 135*
2 6**
* Significantly different from control, P = 0.01. ** PB0.001.
sation was seen on all time points during more intense or complex novel challenges (prod, noise). Female rats showed a different pattern of expression of sensitisation than males. In general, sensitisation of defecation responses was much less consistently expressed than behavioural sensitisation. The results of repeated testing (experiment 1) show that the degree of novelty of the challenge may not be a key factor in the expression of shock-induced behavioural sensitisation. The decreased locomotion and increased grooming over the 3 testing days could indicate some habituation in overall behaviour, but the sensitised immobility response in prehocked rats remained present. Persistent sensitisation after repeated testing has been demonstrated previously, but using a more severe shock protocol [11]. In our hands, both reduced locomotion and increased immobility in controls, and increased locomotion and reduced immobility in preshocked rats after repeated testing was much less pronounced than those reported over a 1-week test interval, where these trends negated differences between preshocked rats and controls [18]. Though inter-trial intervals may play a role in the stability of sensitisation, we have found behavioural sensitisation in another challenge, the light– dark box, to remain stable in three successive challenges with 4-week intervals [3]. Thus, sensitised responsivity to a challenge different from the original foot shock experience may be a more stable phenomenon than previously thought and depend mainly on the perceived aversiveness of the stimulus. The results of experiment 2 show that sensitisation also remains clearly present when different challenges are given at 1-day intervals. Recent findings indicate that a similar principle may hold for the expression of autonomic sensitisation [2]. Interestingly, failure of extinction of generalised hyper arousal and stimulus responsivity is one of the key attributes of post-traumatic changes in humans [4]. While foot shock-induced sensitisation of responsivity in the noise test has been reported for both early light and early dark phases of the circadian cycle [18], differences in expression have not been examined within a single experiment. Our results (experiment 2) show
that the differential expression of behavioural sensitisation at different time points depends on the quality or intensity of the novel stimulus. In the prod test, clear effects of preshock treatment on activity are seen on all three time points; in the noise test more limited effects are seen mainly in afternoon and evening; in the open field and empty cage, differences are clearest in the evening and smaller or absent in the morning and afternoon periods. We have subsequently shown that locomotor differences in open field and empty cage in the early dark phase remain clear-cut in animals kept on a reversed day–night cycle (unpublished observations). While this was not formally tested, there appeared to be few differences in the general pattern of behavioural responsivity of the control groups tested at different time points. Reduced escape-oriented stress behaviour has been reported in the dark versus light phase in a different class of novel challenge, the forced swim test [9]. Although no clear differences are found over a 24-h period in the consumption of a sweet solution between preshocked and control rats [17], there is evidence for differential consumption in light and dark phases after chronic stress [6], indicating that examination of altered hedonia or consumatory behaviour at different times of day in preshocked rats may merit further attention. In contrast to experiments 1 and 2, the increased immobility response in preshocked males in experiment 3 was much clearer in the noise on than in the noise off period, preshocked males showing more immobility in the noise on period and control males in the noise off period. This could relate to a combination of challenging the animals a week later and (olfactory) influences of females in combined housing before testing. The results of experiment 3 again demonstrate that differences in the expression of stress-induced behavioural (but not necessary hormonal or autonomic) sensitisation depend critically on the testing environment. When the noise test is conducted in a low-arousal home cage situation, expression of preshock-sensitised behaviour is negligible, and no interactions with sex are found [14]. In the present study, the same noise challenge conducted in an empty, unfamiliar cage induced clear differences in behavioural responsivity between
74
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
preshocked and control rats, which varied in intensity and direction with sex. It cannot simply be argued that females are more or less vulnerable to the effects of stress-sensitisation than males. They appear to show a
different pattern of sensitised behavioural responsivity to the same challenge, possibly pointing to differential alterations in the neuronal substrates involved [8]. In a similar vein, initial exposure to either of two different
Fig. 2. Duration of six mutually exclusive behavioural components during 5 min in an empty cage 8 days, and in an empty cage with electrified prod 9 days after a brief session of foot shocks (prs) or no shocks (con), in groups of rats tested either in the first half of the light phase (AM), second half of the light phase (PM1) or first half of the dark phase (PM2). Significant differences between prs and con: * PB 0.05; ** PB 0.01; *** P B0.001. locom, locomotion; rear, rearing; groom, grooming; immob, immobility; atten, attention; prod, prod-directed behaviour.
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
75
Fig. 3. Duration of five mutually exclusive behavioural components during 5 min after onset of noise (noise on) and 5 min after cessation of noise (noise off), 10 days after a brief session of foot shocks (prs) or no shocks (con), in groups of rats tested either in the first half of the light phase (AM), second half of the light phase (PM1) or first half of the dark phase (PM2). Significant differences between prs and con: * PB 0.05. locom, locomotion; rear, rearing; groom, grooming; immob, immobility; atten, attention.
sensitising stressful experiences can result in opposite patterns of behavioural sensitisation in a subsequent novel challenge that involve different endogenous neuropeptide mediators [15]. Clarifying the mechanisms
involved may help our understanding of different pathophysiological vulnerabilities in humans [5,7]. The robustness of the model of shock-induced sensitisation makes it a useful tool to that end.
76
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11] [12]
[13] Fig. 4. Duration of five mutually exclusive behavioural components during 5 min after onset of noise (noise on) and 5 min after cessation of noise (noise off), 14 days after a brief session of foot shocks (prs) or no shocks (con), in groups of male (m) and female (f) rats. For significant effects see the results section. locom, locomotion; rear, rearing; groom, grooming; immob, immobility; atten, attention.
[14]
Acknowledgements
[16]
[15]
T.J. van Laar was supported by the Janssen Research Foundation. Supported by NWO grant 902-22-184.
[17]
References
[18]
[1] Breslau N, Kessler RC, Chilcoat HD, Schultz LR, Davis GC, Andreski P. Trauma and posttraumatic stress disorder in the community. Arch Gen Psychiatry 1998;50:626 –32. [2] Bruijnzeel AW, Stam R, Croiset G, Wiegant VM. Long-term sensitization of cardiovascular responses after a single stressful experience. Physiol Behav 2001;73:81 –6. [3] Bruijnzeel AW, Stam R, Wiegant, VM. Effect of a benzodiazepine receptor agonist and CRH receptor antagonists on long-
[19]
[20]
term foot shocks-induced increase in defensive withdrawal behavior. Psychopharmacology, in press. Charney DS, Deutch AY, Krystal JH, Southwick SM, Davis M. Psychobiologic mechanisms of posttraumatic stress disorder. Arch Gen Psychiatry 1993;50:294 – 305. Curle CE, Williams C. Post-traumatic stress reactions in children: gender differences in the incidence of trauma reactions at 2 years and examinations of factors infuencing adjustment. Br J Clin Psychol 1996;35:297 – 309. D’Aquila PS, Newton J, Willner P. Diurnal variation in the effect of chronic mild stress on sucrose intake and preference. Physiol Behav 1997;62:421 – 6. Davidson JRT, Hughes D, Blazer DG, George LK. Post-traumatic stress disorder in the community: an epidemiological study. Psychol Med 1991;21:713 – 21. Farabollini F, Fluck E, Albonetti ME, File SE. Sex differences in benzodiazepine binding in the frontal cortex and amygdala of the rat 24 hours after restraint stress. Neurosci Lett 1996;218:177 – 80. Kelliher P, Connor TJ, Harkin A, Sanchez C, Kelly JP, Leonard BE. Varying responses to the rat forced swim test under diurnal and nocturnal conditions. Physiol Behav 2000;69:531 – 9. Lemieux AM, Coe CL. Abuse-related posttraumatic stress disorder: evidence for chronic neuroendocrine activation in women. Psychosom Med 1995;57:105 – 15. Levine S, Madden JI. Psychological and behavioral effects of prior aversive stimulation. Physiol Behav 1973;10:467 –71. Murison R, Overmier JB. Comparison of different animal models of stress reveals a non-monotonic effect. Stress 1998;2:227 – 30. Stam R, Bruijnzeel AW, Wiegant VM. Long-lasting stress sensitisation. Eur J Pharmacol 2000;405:217 – 24. Stam R, Croiset G, Bruijnzeel AW, Visser TJ, Akkermans LMA, Wiegant VM. Sex differences in long-term stress-induced colonic, behavioural and hormonal disturbances. Life Sci 1999;65:2837 – 49. van den Berg CL, Lamberts RR, Wolterink G, Wiegant VM, van Ree JM. Emotional and footshock stimuli induce differential long-lasting behavioural effects in rats. Brain Res 1998;799:6 –15. van Dijken HH, de Goeij DCE, Sutanto W, Mos J, de Kloet ER, Tilders FJH. Short inescapable stress produces long-lasting changes in the brain-pituitary-adrenal axis of adult male rats. Neuroendocrinology 1993;58:57 – 64. van Dijken HH, Mos J, van der Heyden JAM, Tilders FJH. Characterization of stress-induced long-term behavioural changes in rats: evidence in favor of anxiety. Physiol Behav 1992;52:945 – 51. van Dijken HH, van der Heyden JAM, Mos J, Tilders FJH. Inescapable footshocks induce progressive and long-lasting behavioral changes in male rats. Physiol Behav 1992;51:787 – 94. Yehuda R, McFarlane AC. Conflict between current knowledge about posttraumatic stress disorder and its original conceptual basis. Am J Psychiatry 1995;152:1705 – 13. Yehuda R, Teicher MH, Trestman RL, Levengood RA, Siever LJ. Cortisol regulation in posttraumatic stress disorder and major depression: a chronobiological analysis. Biol Psychiatry 1996;40:79 – 88.
Research report
Variability factors in the expression of stress-induced behavioural sensitisation Rianne Stam a,*, Teus-Jan van Laar a, Louis M.A. Akkermans b, Victor M. Wiegant a a
Di6isions of Pharmacology and Anatomy, Rudolf Magnus Institute for Neurosciences, Uni6ersity Medical Center Utrecht, P.O. Box 85060, 3508 AB Utrecht, The Netherlands b Di6ision of Surgery, Rudolf Magnus Institute for Neurosciences, Uni6ersity Medical Center Utrecht, P.O. Box 85060, 3508 AB Utrecht, The Netherlands Received 4 July 2001; received in revised form 13 September 2001; accepted 13 September 2001
Abstract Altered behavioural and physiological responsivity following a short session of foot shocks in the rat has proven to be a stable and clinically relevant model of stress-induced sensitisation. However, a number of key factors influencing effect size or direction have not previously been reported. Rats underwent a single, 15-min session of foot shocks and were exposed to a variety of novel stressful challenges 1 or 2 weeks later. Sensitised behavioural responses (increased immobility) in preshocked rats remained present over 3 days of repeated exposure to noise stress. In mild novel challenges (open field, empty cage), behavioural sensitisation and defecation was most clearly expressed at the beginning of the dark phase (evening). Higher-arousal challenges (prod, noise) caused increased behavioural inhibition in preshocked rats at all three time points (morning, afternoon, evening). Female preshocked rats showed a different pattern of behavioural and defecation sensitisation than preshocked males. The robustness of the model makes it suitable for further investigations into the mechanisms and vulnerability factors involved in the long-term consequences of stress. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Stress; Sensitisation; Rat; Behaviour
1. Introduction Increasing attention in recent years has focused on long-lasting sensitisation-like phenomena following intense but relatively short-lasting stressors (for a recent review, see Ref. [13]). The neurobiological principles involved differ from those in adaptive and dysregulated responsivity after chronic stress, and are thought to be relevant to post-traumatic psychiatric disturbances in humans and their physiological correlates [19]. One of the more widely used and well-characterised animal models of stress-induced sensitisation uses a short-lasting exposure to a grid cage in which electric currents are repeatedly, and usually unpredictably, applied [12]. * Corresponding author. Tel.: + 31-30-2538830; fax: +31-302539032. E-mail address: [email protected] (R. Stam).
Rats exposed to a single session of repeated foot shocks show sensitised behavioural responsivity in a variety of novel stressful challenges that develops over time [17,18]. The general pattern of behavioural sensitisation is stable after 7 days, though individual behavioural components may still show small increases after that [18]. Altered behavioural responsivity can be accompanied by neuroendocrine [16] and autonomic [2,14] hyperresponsivity, although their relative expression depends on the testing conditions used. It has been assumed that the expression of sensitisation depends to a high degree on the novelty of the subsequent challenge [18], but it is unclear whether this is a general phenomenon or test-dependent. While there is evidence for circadian variability in the expression of post-traumatic changes in humans [20], information on optimal time-points for the expression of shock-induced sensiti-
0166-4328/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 3 2 8 ( 0 1 ) 0 0 3 8 7 - 4
70
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
sation in rats is not yet available. Likewise, few studies have addressed sex-differences in the development and expression of stress-sensitisation, although there is clear evidence for gender-differences in vulnerability for stress-related psychiatric and physiological abnormalities [1,10]. The present study sought to address these gaps in our knowledge by systematically investigating the contribution of these three factors (repeated testing, circadian and sex variability) on the expression of foot shock-induced behavioural sensitisation.
2. Materials and methods
2.1. Animals and sensitisation procedure Experimental procedures were approved by the Commission on Laboratory Animal Experiments of the Faculty of Medicine, Utrecht University. Ten week old male and female Wistar rats (U:WU, Central Animal Laboratory, Utrecht University) were housed individually in macrolon cages measuring 20× 30 × 15 cm (l× w × h) and received standard laboratory chow and tap water ad libitum. Ambient temperature was kept at 21 °C and lighting regulated (on at 07:00, off at 19:00 h). All rats were briefly handled on 4 successive days in week 1. In week 2, between 15:00 and 17:30 h, half the rats were placed in a metal grid cage (32× 30 × 38 cm l ×w × h) and underwent a 15-min session of scrambled electric foot shocks (10×6 s, 0.5 mA, with randomised intervals of 20– 210 s; preshocked), the other half serving as non-shocked controls.
2.2. No6el stressful challenges 2.2.1. Experiment 1: effects of repeated testing Seven days after foot shocks, between 15:00 and 17:00 h in accordance with earlier studies on behaviour and colonic motility, male preshocked or control rats (n= 8 per group) were taken to an adjacent room and placed in an empty perspex cage with stainless steel floor (25×25×35 cm, l ×w × h) where they were exposed to a 85 dB broad spectrum noise. After 5 min the noise was switched off and the rats observed for another 5 min. Behaviour was recorded on video and any fecal boli produced were counted and the cage thoroughly cleaned with detergent and rinsed with water. The noise test was repeated on day 8 and 9 after foot shocks following identical procedures. Light intensity in the test cage was 270 lux. 2.2.2. Experiment 2: effects of time of day Male preshocked and control rats (n =24 per group) were each divided in three subgroups of n= 8. On the
four successive challenge days, one control and one preshocked group was always challenged between 09:00 and 11:00 h (light; morning), one control and preshocked group between 15:00 and 17:00 h (light; afternoon), and one control and preshocked group between 21:00 and 23:00 h (dark; evening). At the latter time point, animals were observed in red light. Seven days after foot shocks, rats were picked up from their home cages, taken to an adjacent room and placed in the center of a black polyester circular arena (‘open field’, diameter 130 cm, height of wall 30 cm). Traveled distance during 5 min was recorded automatically via a digitised image using Ethovision tracking software (Noldus Information Systems, Wageningen, The Netherlands). Eight days after foot shocks, rats were picked up from their home cages, taken to an adjacent room and placed in an empty perspex cage with stainless steel floor (25× 25× 35 cm, l × w× h) for 5 min. Nine days after foot shocks, rats were picked up from their home cages, taken to an adjacent room and placed in the empty perspex cage where an electrified prod was inserted that delivered a 2 mA current on touch and removed after 5 min. Ten days after foot shocks, rats were picked up from their home cages, taken to an adjacent room and placed in the empty perspex cage where they were exposed to a 85 dB broad spectrum noise. After 5 min the noise was switched off and the rats observed for another 5 min. On each challenge day, behaviour was recorded on video and any fecal boli produced were counted and the cage thoroughly cleaned with detergent and rinsed with water. Light intensity in morning and afternoon periods was 280 lux (open field) or 400 lux (empty cage, prod and noise tests). Light intensity in the evening was 6–8 lux.
2.2.3. Experiment 3: effects of sex Noise exposure was part of a larger pilot study into the subsequent effects of colonic distention, and experimental conditions therefore varied slightly from experiment 1 and 2. The control group was exposed to the shock cage without receiving any shocks, and testing was conducted 14 days later at a slightly later time in the afternoon. Male and female rats were subdivided in control and preshocked groups of n= 10 each. Fourteen days after foot shocks, between 16:45 and 17:45 h, rats were picked up from their home cages, taken to an adjacent room and placed in a perspex cage with stainless steel floor (25× 25×35 cm, l ×w× h) where they were exposed to a 85 dB broad spectrum noise. After 5 min the noise was switched off and the rats observed for another 5 min. Behaviour was recorded on video. Male and female rats were tested on alternate days so as not to confound behavioural reactivity by olfactory
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
cues. Light intensity at the level of the test cage was 270 lux.
2.3. Beha6iour analysis and statistics Behaviour analysis of taped sessions was performed blindly by an experienced observer. For empty cage and noise exposure, mean duration of the following six mutually exclusive components was calculated: locomotion (movement of whole body), rearing (upright posture), grooming, immobility and attention (sitting while head and whiskers are moving). For the electrified prod test, in addition to locomotion, rearing, grooming, immobility and attention as defined above, the set of mutually exclusive behavioural components included prod-directed behaviour (‘prod’, a combination of approaching, sniffing or otherwise exploring the prod). Effects of repeated testing and preshock treatment on behaviour in experiment 1 were assessed with multivariate analysis of variance for repeated measures. Differences in open field travelled distance and duration of behavioural components between preshocked and control groups in experiment 2 were evaluated with unpaired Student’s t-tests. Effects of preshock treatment and sex on behaviour in the noise test in experiment 2 were tested with two-factor univariate analysis of variance. Defecation differences (number of boli) were analysed with the non-parametric Mann– Whitney Utest (experiment 1 and 2) or Kruskal– Wallis analysis of variance (experiment 3). SPSS for WINDOWS 9.0 was used for all statistical analyses and significance levels set at P= 0.05.
71
3.2. Experiment 2: effects of time of day Open field traveled distance on day 7 showed trends toward a decrease in preshocked rats on all three time points, but the decreased activity was significant only in the evening group. This appeared mainly due to greater traveled distance in control rats in the dark period (Table 1). A significant increase in defecation, too, was found only in the evening period (Table 1). Effects of preshock treatment on behaviour in the empty cage on day 8 were clearest in the evening, with significantly decreased locomotion, rearing and attention and increased immobility (Fig. 2). Rats tested in the morning also showed less locomotion and more immobility, but no behavioural differences were found in the afternoon. In contrast, in the prod test on day 9, a very clear increase in immobility was seen in preshocked rats at all three time points, with a reduction in more active behaviours (locomotion, rearing, grooming, attention; Fig. 2). During the noise on period on day 10, preshocked rats showed increased immobility in the morning period and decreased locomotion in the evening (Fig. 3). During the following noise off period, immobility in preshocked rats was greater than that in controls in the afternoon and evening, and attention decreased in the afternoon. In the morning period, preshocked rats only showed decreased grooming behaviour (Fig. 3). No significant differences in defecation were found in empty cage, prod or noise tests (results not shown).
3.3. Experiment 3: effects of sex 3. Results
3.1. Experiment 1: effects of repeated testing Rats generally showed somewhat more active behaviour during the 5-min noise on period and more immobility during the 5-min noise off period (Fig. 1). In the noise on period, there was a small but significant decrease in the duration of locomotion over the 3 testing days (F(2, 28)= 7.64, P =0.002) as well as a small increase in grooming (F(2, 28)=4.33, P =0.023). Overall immobility was increased in the preshocked group compared to controls (F(1, 14)=5.73, P= 0.031). In the noise off period, immobility decreased over the 3 testing days (F(2, 28)= 5.25, P =0.012). The days × preshocked interaction did not reach significance (P=0.086), showing that the increased immobility response due to the previous foot shock experience (F(1, 14)=79.4, P B 0.001) did not habituate after repeated testing. No significant effects of test day or preshock treatment were found on defecation (results not shown).
During the noise on period, a significant preshocked× sex interaction indicated that the reduction in locomotion (F(1, 36)=4.25, P=0.047) and increase in immobility (F(1, 36)= 4.54, P= 0.040) caused by the preshock experience were more pronounced in males than in females (Fig. 4). Preshocks reduced the duration of rearing (F(1, 36)= 9.72, P=0.004) and grooming (F(1, 36)= 9.77, P= 0.003) in both males and females. During the noise off period, a significant preshocked× sex interaction appeared to result from an increased locomotion (F(1, 36)= 12.0, P=0.001) and rearing (F(1, 36)=8.80, P=0.005) response in preshocked females compared to the other three experimental groups (Fig. 4). Compared to their respective controls, the immobility response increased in preshocked males but decreased in preshocked females (F(1, 36)= 13.0, P= 0.001). A significant group effect was found for noise-induced defecation ( 2 = 10.1, P= 0.018). Post-hoc analysis indicated that the number of boli was higher only in preshocked compared to control females (mean n= 3 and 7, respectively; critical difference: 13.79).
72
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
4. Discussion Three experiments were conducted to clarify variability factors in the effectiveness of stress-induced sensitisation, using a model foot shock experience. Sensitised
behavioural responsivity remained present during 3 days of repeated testing with a noise challenge. In a relatively mild or less complex novel challenge (open field, empty cage), behavioural sensitisation was expressed most clearly in the evening (dark), while sensiti-
Fig. 1. Duration of five mutually exclusive behavioural components during 5 min after onset of noise (noise on) and 5 min after cessation of noise (noise off), in sequential testing 7, 8 and 9 days after a brief session of foot shocks (prs) or no shocks (con). For significant effects see the results section. locom, locomotion; rear, rearing; groom, grooming; immob, immobility; atten, attention.
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
73
Table 1 Open field traveled distance and defecation, with S.E.M., during 5 min testing, at three different time points, 7 days after foot shocks (prs) or no shocks (con) AM
Con Prs
PM1
PM2
Distance (cm)
Defecation (number)
Distance (cm)
Defecation (number)
Distance (cm)
Defecation (number)
2064 9154 1581 9232
5 4
19359244 16349166
5 6
2314 9158 1692 9 135*
2 6**
* Significantly different from control, P = 0.01. ** PB0.001.
sation was seen on all time points during more intense or complex novel challenges (prod, noise). Female rats showed a different pattern of expression of sensitisation than males. In general, sensitisation of defecation responses was much less consistently expressed than behavioural sensitisation. The results of repeated testing (experiment 1) show that the degree of novelty of the challenge may not be a key factor in the expression of shock-induced behavioural sensitisation. The decreased locomotion and increased grooming over the 3 testing days could indicate some habituation in overall behaviour, but the sensitised immobility response in prehocked rats remained present. Persistent sensitisation after repeated testing has been demonstrated previously, but using a more severe shock protocol [11]. In our hands, both reduced locomotion and increased immobility in controls, and increased locomotion and reduced immobility in preshocked rats after repeated testing was much less pronounced than those reported over a 1-week test interval, where these trends negated differences between preshocked rats and controls [18]. Though inter-trial intervals may play a role in the stability of sensitisation, we have found behavioural sensitisation in another challenge, the light– dark box, to remain stable in three successive challenges with 4-week intervals [3]. Thus, sensitised responsivity to a challenge different from the original foot shock experience may be a more stable phenomenon than previously thought and depend mainly on the perceived aversiveness of the stimulus. The results of experiment 2 show that sensitisation also remains clearly present when different challenges are given at 1-day intervals. Recent findings indicate that a similar principle may hold for the expression of autonomic sensitisation [2]. Interestingly, failure of extinction of generalised hyper arousal and stimulus responsivity is one of the key attributes of post-traumatic changes in humans [4]. While foot shock-induced sensitisation of responsivity in the noise test has been reported for both early light and early dark phases of the circadian cycle [18], differences in expression have not been examined within a single experiment. Our results (experiment 2) show
that the differential expression of behavioural sensitisation at different time points depends on the quality or intensity of the novel stimulus. In the prod test, clear effects of preshock treatment on activity are seen on all three time points; in the noise test more limited effects are seen mainly in afternoon and evening; in the open field and empty cage, differences are clearest in the evening and smaller or absent in the morning and afternoon periods. We have subsequently shown that locomotor differences in open field and empty cage in the early dark phase remain clear-cut in animals kept on a reversed day–night cycle (unpublished observations). While this was not formally tested, there appeared to be few differences in the general pattern of behavioural responsivity of the control groups tested at different time points. Reduced escape-oriented stress behaviour has been reported in the dark versus light phase in a different class of novel challenge, the forced swim test [9]. Although no clear differences are found over a 24-h period in the consumption of a sweet solution between preshocked and control rats [17], there is evidence for differential consumption in light and dark phases after chronic stress [6], indicating that examination of altered hedonia or consumatory behaviour at different times of day in preshocked rats may merit further attention. In contrast to experiments 1 and 2, the increased immobility response in preshocked males in experiment 3 was much clearer in the noise on than in the noise off period, preshocked males showing more immobility in the noise on period and control males in the noise off period. This could relate to a combination of challenging the animals a week later and (olfactory) influences of females in combined housing before testing. The results of experiment 3 again demonstrate that differences in the expression of stress-induced behavioural (but not necessary hormonal or autonomic) sensitisation depend critically on the testing environment. When the noise test is conducted in a low-arousal home cage situation, expression of preshock-sensitised behaviour is negligible, and no interactions with sex are found [14]. In the present study, the same noise challenge conducted in an empty, unfamiliar cage induced clear differences in behavioural responsivity between
74
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
preshocked and control rats, which varied in intensity and direction with sex. It cannot simply be argued that females are more or less vulnerable to the effects of stress-sensitisation than males. They appear to show a
different pattern of sensitised behavioural responsivity to the same challenge, possibly pointing to differential alterations in the neuronal substrates involved [8]. In a similar vein, initial exposure to either of two different
Fig. 2. Duration of six mutually exclusive behavioural components during 5 min in an empty cage 8 days, and in an empty cage with electrified prod 9 days after a brief session of foot shocks (prs) or no shocks (con), in groups of rats tested either in the first half of the light phase (AM), second half of the light phase (PM1) or first half of the dark phase (PM2). Significant differences between prs and con: * PB 0.05; ** PB 0.01; *** P B0.001. locom, locomotion; rear, rearing; groom, grooming; immob, immobility; atten, attention; prod, prod-directed behaviour.
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
75
Fig. 3. Duration of five mutually exclusive behavioural components during 5 min after onset of noise (noise on) and 5 min after cessation of noise (noise off), 10 days after a brief session of foot shocks (prs) or no shocks (con), in groups of rats tested either in the first half of the light phase (AM), second half of the light phase (PM1) or first half of the dark phase (PM2). Significant differences between prs and con: * PB 0.05. locom, locomotion; rear, rearing; groom, grooming; immob, immobility; atten, attention.
sensitising stressful experiences can result in opposite patterns of behavioural sensitisation in a subsequent novel challenge that involve different endogenous neuropeptide mediators [15]. Clarifying the mechanisms
involved may help our understanding of different pathophysiological vulnerabilities in humans [5,7]. The robustness of the model of shock-induced sensitisation makes it a useful tool to that end.
76
R. Stam et al. / Beha6ioural Brain Research 132 (2002) 69–76
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11] [12]
[13] Fig. 4. Duration of five mutually exclusive behavioural components during 5 min after onset of noise (noise on) and 5 min after cessation of noise (noise off), 14 days after a brief session of foot shocks (prs) or no shocks (con), in groups of male (m) and female (f) rats. For significant effects see the results section. locom, locomotion; rear, rearing; groom, grooming; immob, immobility; atten, attention.
[14]
Acknowledgements
[16]
[15]
T.J. van Laar was supported by the Janssen Research Foundation. Supported by NWO grant 902-22-184.
[17]
References
[18]
[1] Breslau N, Kessler RC, Chilcoat HD, Schultz LR, Davis GC, Andreski P. Trauma and posttraumatic stress disorder in the community. Arch Gen Psychiatry 1998;50:626 –32. [2] Bruijnzeel AW, Stam R, Croiset G, Wiegant VM. Long-term sensitization of cardiovascular responses after a single stressful experience. Physiol Behav 2001;73:81 –6. [3] Bruijnzeel AW, Stam R, Wiegant, VM. Effect of a benzodiazepine receptor agonist and CRH receptor antagonists on long-
[19]
[20]
term foot shocks-induced increase in defensive withdrawal behavior. Psychopharmacology, in press. Charney DS, Deutch AY, Krystal JH, Southwick SM, Davis M. Psychobiologic mechanisms of posttraumatic stress disorder. Arch Gen Psychiatry 1993;50:294 – 305. Curle CE, Williams C. Post-traumatic stress reactions in children: gender differences in the incidence of trauma reactions at 2 years and examinations of factors infuencing adjustment. Br J Clin Psychol 1996;35:297 – 309. D’Aquila PS, Newton J, Willner P. Diurnal variation in the effect of chronic mild stress on sucrose intake and preference. Physiol Behav 1997;62:421 – 6. Davidson JRT, Hughes D, Blazer DG, George LK. Post-traumatic stress disorder in the community: an epidemiological study. Psychol Med 1991;21:713 – 21. Farabollini F, Fluck E, Albonetti ME, File SE. Sex differences in benzodiazepine binding in the frontal cortex and amygdala of the rat 24 hours after restraint stress. Neurosci Lett 1996;218:177 – 80. Kelliher P, Connor TJ, Harkin A, Sanchez C, Kelly JP, Leonard BE. Varying responses to the rat forced swim test under diurnal and nocturnal conditions. Physiol Behav 2000;69:531 – 9. Lemieux AM, Coe CL. Abuse-related posttraumatic stress disorder: evidence for chronic neuroendocrine activation in women. Psychosom Med 1995;57:105 – 15. Levine S, Madden JI. Psychological and behavioral effects of prior aversive stimulation. Physiol Behav 1973;10:467 –71. Murison R, Overmier JB. Comparison of different animal models of stress reveals a non-monotonic effect. Stress 1998;2:227 – 30. Stam R, Bruijnzeel AW, Wiegant VM. Long-lasting stress sensitisation. Eur J Pharmacol 2000;405:217 – 24. Stam R, Croiset G, Bruijnzeel AW, Visser TJ, Akkermans LMA, Wiegant VM. Sex differences in long-term stress-induced colonic, behavioural and hormonal disturbances. Life Sci 1999;65:2837 – 49. van den Berg CL, Lamberts RR, Wolterink G, Wiegant VM, van Ree JM. Emotional and footshock stimuli induce differential long-lasting behavioural effects in rats. Brain Res 1998;799:6 –15. van Dijken HH, de Goeij DCE, Sutanto W, Mos J, de Kloet ER, Tilders FJH. Short inescapable stress produces long-lasting changes in the brain-pituitary-adrenal axis of adult male rats. Neuroendocrinology 1993;58:57 – 64. van Dijken HH, Mos J, van der Heyden JAM, Tilders FJH. Characterization of stress-induced long-term behavioural changes in rats: evidence in favor of anxiety. Physiol Behav 1992;52:945 – 51. van Dijken HH, van der Heyden JAM, Mos J, Tilders FJH. Inescapable footshocks induce progressive and long-lasting behavioral changes in male rats. Physiol Behav 1992;51:787 – 94. Yehuda R, McFarlane AC. Conflict between current knowledge about posttraumatic stress disorder and its original conceptual basis. Am J Psychiatry 1995;152:1705 – 13. Yehuda R, Teicher MH, Trestman RL, Levengood RA, Siever LJ. Cortisol regulation in posttraumatic stress disorder and major depression: a chronobiological analysis. Biol Psychiatry 1996;40:79 – 88.