- Email: [email protected]
Social and structural housing conditions influence the development of a depressive-like phenotype in the learned helplessness paradigm in male mice
Behavioural Brain Research 164 (2005) 100–106
Research report
Social and structural housing conditions influence the development of a depressive-like phenotype in the learned helplessness paradigm in male mice Sabine Chourbaji a,∗ , Christiane Zacher a , Carles Sanchis-Segura b , Rainer Spanagel b , Peter Gass a a b
Department of Behavioural Biology, Central Institute of Mental Health Mannheim (ZI), University of Heidelberg, J5, D-68159 Mannheim, Germany Department of Psychopharmacology, Central Institute of Mental Health Mannheim (ZI), University of Heidelberg, J5, D-68159 Mannheim, Germany Received 25 April 2005; received in revised form 2 June 2005; accepted 2 June 2005 Available online 19 July 2005
Abstract Structural and social factors are known to play a crucial role in the pathogenesis of depression. Since animal models of depression are a major tool to gain insights into the mechanisms involved in the pathophysiology of this disease it is important not only to exploit but also to be aware of factors that may affect these models. As housing represents a fundamental external factor, which is controversially debated to affect the animals’ emotionality, this study aimed to investigate the impact of different social and structural housing conditions on the development of a depressive-like syndrome in the learned helplessness paradigm. Group housing in an impoverished environment led to an increased vulnerability in the learned helplessness paradigm. Groups that were housed enriched, however, were less helpless. Furthermore impoverished conditions did not increase the vulnerability in single housed animals. Regarding emotionality in the animals, basal anxiety was reduced and the exploration was enhanced by group housing and enriched environment. These results suggest that housing conditions significantly influence the outcome of learned helplessness studies. © 2005 Published by Elsevier B.V. Keywords: Housing; Enrichment; Depression; Learned helplessness; Mice
1. Introduction Working on behavioural animal models for disease states requires exact knowledge and control of “internal” and “external” factors that may influence the outcome of the experiments. Internal factors comprise the animals’ gender, age, hormone state, satiety and genetic background; external factors are housing and social conditions as well as many other factors. While the importance of internal factors has been widely recognized in recent years (e.g. Ban∗
Corresponding author. Tel.: +49 621 1703 2932; fax: +49 621 1703 2005. E-mail addresses: [email protected] (S. Chourbaji), [email protected] (C. Zacher), [email protected] (C. SanchisSegura), [email protected] (R. Spanagel), [email protected] (P. Gass). 0166-4328/$ – see front matter © 2005 Published by Elsevier B.V. doi:10.1016/j.bbr.2005.06.003
bury conference [8]), housing of experimental animals is frequently reported as “standardized” without a clear definition. The housing conditions for mice, which are considered as “standardized” in most laboratories are barren compared to any natural environment, and thereby imply behavioural and social (if kept single caged) deprivation. This results in a reduction of the natural behavioural repertoire [36] as well as in alterations of specific brain functions of the animals [23,37]. In contrast, it has been shown that structurally enriched or super-enriched home-cages significantly alter physiological parameters thought to be relevant for stressrelated disorders, such as corticosterone levels [9], hippocampal neurogenesis [13], the release of neurotrophins [11,28] or body weight [1]. Thus, the specific conditions of keeping the animals seem to be particularly relevant for animal models of stress-related disorders [4,32]. However, the effects
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
of housing conditions on the development of depressionlike behaviours in a generally recognized model, such as the learned helplessness paradigm have yet not been studied elaborately. The learned helplessness paradigm, a rodent model of depressive-like states with appropriate construct and face validity [18,26,30,31,33], may reflect the described housinginduced physiological changes at the behavioural level. For instance, exposure to an opponent, may act as a source of chronic moderate stress altering depression-relevant parameters [3,16,35] and if so, according to the current views of depression, housing of animals in groups might evoke an increase in coping deficits. Supporting this hypothesis, group housing, a common way of animal husbandry, in which the mice, being territorial animals, are permanently confronted to potential rivals, increased depressive-like behaviour such as immobility in the forced swim and in tail suspension test, suggesting that long-lasting group housing may constitute a stressful environment [17]. These evidences seem to be in agreement with the consideration of social interaction as an important factor regarding stress vulnerability and ability to cope with demanding situations in rats and human beings [5,7,22]. Therefore variation in animal husbandry may be employed as a source of stress itself and consequently constitutes an appropriate instrument for the construction of a natural stressor. The present study was designed to assess the influence of different housing conditions on 4 months old male C57Bl/6N mice with regard to anxiety-related behavioural features in the open field and dark–light box, as well as the development of depressive-like behaviour in the learned helplessness paradigm. For this purpose we analysed the effects of single versus group housing as well as the influence of impoverished versus enriched structural conditions in these models. The final goal of this study was (i) to elucidate the potential role of group housing as a model for chronic stress; and (ii) to determine, if any of the “standardized” keeping conditions hold subtle advantages or limitations with respect to the learned helplessness paradigm.
2. Materials and methods 2.1. Animals All behavioural tests were conducted in 4 months old male C57Bl/6N mice, which is the most frequently used background strain in genetic engineering. The animals were purchased from Charles River (Sulzfeld, Germany) at an age of 8 weeks and were then housed for 7 weeks under specific conditions in type II cages for single and type III cages for group rearing (four mice per cage). Four cohorts of mice were investigated: (1) single housed, impoverished conditions (n = 16), (2) single housed, enriched conditions (n = 16), (3) group housed, impoverished conditions (n = 16), (4) group housed, enriched conditions (n = 16). Enrichment consisted of
101
red, transparent plastic mouse igloos and tunnels (EMSICON Jung GmbH, Forstinning, Germany) that were cleaned weekly, when the cage was changed. Additionally, nesting material (tissue) was provided. Impoverished cages simply contained bedding material. All mice were kept in the same room, in a 12 h:12 h reversed dark–light cycle, lights on at 6.00 p.m. Water and food pellets were available ad libitum. Body weight was assessed once a week when the cages were cleaned. All experiments were approved by German animal welfare authorities. 2.2. Behavioural procedures After 7 weeks of housing under the aforementioned conditions, all mice were subjected to tests for locomotion, exploration, and anxiety, followed by the learned helplessness procedure. Following earlier recommendations for repetitive behavioural testing, animals were initially tested in the experiments ranked as less stressful [21,29]. Considering that the determination of the pain threshold by the hot plate test could have a direct influence on helpless behaviour, we assessed the effect of housing conditions on pain sensitivity in a separate group of mice reared under identical housing conditions. Test procedures were essentially performed as described earlier [6,12,25]. Between individual tests was a pause of at least 24 h. Prior to each test, mice were acclimatized to the experimental room for 30 min. All behavioural tests were conducted during the dark cycle, i.e. during the animals’ active phase. 2.3. Open field test Activity monitoring was performed in a square, white open field, measuring 50 cm × 50 cm and illuminated from above by 25 lx. Mice were placed individually into the arena and monitored for 15 min by a Video camera (Sony CCD IRIS). The resulting data were analysed using the image processing system EthoVision 2.3 (Noldus Information Technology, Wageningen, The Netherlands). For each sample, the system recorded position and the occurrence of defined events. Parameters assessed in the present study were total distance moved, velocity, and thigmotaxis (i.e. the percentage of time spent in a corridor with maximal distance of 10 cm to the walls). 2.4. Dark–light box test The dark–light box consisted of two plastic chambers, connected by a small tunnel. The dark chamber measured 20 cm × 15 cm and was covered by a lid. The light chamber, measuring 30 cm × 15 cm, was white and illuminated from above with an intensity of 600 lx. Mice were placed into the dark compartment, and the latency to the first exit, the number of exits and the total time spent in the light compartment were recorded for 5 min. 2.5. Learned helplessness The learned helplessness paradigm was conducted as described earlier for male mice [25]. Mice received 360 footshocks (0.150 mA) on two consecutive days in a transparent plexiglas shock chamber (18 cm × 18 cm × 30 cm), equipped with a stainless steel grid floor (Coulborn precision regulated animal shocker, Coulborn Instruments, D¨usseldorf, Germany). The footshocks applied were unpre-
102
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
dictable with varying shock-durations (1–3 s) and interval-episodes (1–15 s), amounting to a total session duration of approximately 52 min. Twenty-four hours after the second shock exposure, learned helplessness was assessed by testing two-way avoidance in a shuttle box (Coulborn Instruments, D¨usseldorf, Germany). Spontaneous initial shuttle activity from one compartment to the other was monitored during the first two minutes by red-light beams at the bottom of each of the two compartments. Thereafter, specific avoidance learning was tested over 30 trials. Each trial started with a light stimulus of 5 s, announcing a subsequent footshock (0.150 mA; maximal duration: 10 s). The duration of this procedure varied (total testing time averaged 20 min) according to each animal’s individual performance in terms of failures, i.e. not attempting to escape from the shock, and escape latencies, i.e. latency to escape after onset of the shock. An additional batch of mice, reared under the equal four conditions as described above (n = 8 per condition) was tested in the same two-way avoidance paradigm. However, these mice were not exposed to uncontrolled shocks before shuttle-box testing, thus providing a control for a possible effect of housing conditions per se on the performance in this test. These animals are referred as “nonshocked” mice. 2.6. Hotplate test To exclude altered pain sensitivity as a modulating factor for learned helplessness, an additional batch of mice (n = 8 per condition) was tested on the hotplate test (ATLab, Vendargues, France). Temperature was set at 53 ◦ C and a 45 s cut-off was introduced to prevent injury of mice. Latency to first reaction, i.e. licking hind paws or jumping, was assessed.
2.7. Statistical analyses For statistical analyses, values of each housing condition were evaluated by two-way ANOVAs followed by Fisher’s (LSD) post hoc tests (XLstat program Version 7.5, Addinsoft) calculating the statistically significant effects of factors as well as interactions at p < 0.05. Correlations between variables were analysed by Pearson’s correlation reaching significant levels at p < 0.05. Comparisons of escape deficits in the learned helplessness paradigm between shocked animals and non-shocked controls were performed by ttests.
3. Results 3.1. Group housing affects exploratory behaviour but not locomotion In the open field test, the housing conditions did not affect basal locomotor activity of the animals, i.e. the total distance moved (Fig. 1A) or the locomotor velocity (Table 1). However, a single exposure to the open field test also confronts the animals with an approach-avoidance conflict, in which the exploration of the animals, measured by the time spent in the centre of the open field arena [10], was significantly influenced by social housing conditions [F(1,62) = 10.962; p < 0.002]. Post hoc tests demonstrated, that both group housed cohorts (impoverished and enriched) spent more time in the centre than the respective single housed groups (Fig. 1B). p-values of Fisher’s (LSD) post hoc tests are pre-
Fig. 1. Locomotor, exploratory and anxiety-like behaviour under different housing conditions. (A) In the open field test, no changes in locomotor activity occur under the different housing conditions applied. (B) Group housed mice spend significantly more time in the centre of the arena than single housed animals. (C and D) In the dark–light box test, both social as well as structural factors affect anxiety-like behaviour, but there is no interaction between these two factors (cf. Section 3). * p < 0.05, ** p < 0.01.
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
103
Table 1 Significant inter-group differences: Fisher’s (LSD) post hoc test p-values Single impoverished
Single enriched
Group impoverished
Group enriched
Openfield locomotion Openfield velocity Openfield centre time Dark–light box latency
n.s. n.s vs. group imp.: p = 0.042 n.s.
n.s. n.s vs. group enr.: p = 0.002 vs. group enr.: p = 0.008
n.s. n.s vs. single imp.: p = 0.042
Dark–light box exits Dark–light box time in lit compartment Learned helplessness failures
n.s. vs. group imp.: p = 0.044
n.s. n.s.
vs. group enr.: p = 0.031 vs. group enr.: p = 0.016 vs. single imp.: p = 0.044
n.s. n.s vs. single enr.: p = 0.002 vs. single enr.: p = 0.008 vs. group imp.: p = 0.031 vs. group imp.: p = 0.016 n.s.
vs. group imp.: p = 0.008 ↓
n.s.
vs. single imp.: p = 0.008 ↑
vs. group imp.: p = 0.014 ↓
n.s.
vs. group enr.: p = 0.005 ↑ vs. single imp.: p = 0.014 ↑
vs. group imp.: p = 0.005 ↓
Learned helplessness escape latency Hotplate pain sensitivity
n.s.
n.s.
vs. group enr.: p = 0.007 ↑ n.s.
vs. group imp.: p = 0.007 ↓ n.s.
Group comparisons: p-values indicate inter-group differences in all parameters assessed, according to a Fisher’ (LSD) post hoc comparison. n.s.: not significant; enr.: enriched; imp.: impoverished. (↑) enhanced helplessness; (↓) reduced helplessness.
sented in Table 1. There was no statistical effect for the interaction between the factors structural and social housing. 3.2. Group housing and enrichment evoke reduced anxiety-like behaviour Both housing factors, social [F(1,62) = 7.797; p < 0.007] and structural conditions [F(1,62) = 4.86; p < 0.031], significantly affected the latency to enter the aversive light compartment, while the interaction between factors did not show a significant effect (Fig. 1C). Post hoc tests (Table 1) demonstrated significantly reduced anxiety-like behaviour, i.e. latency to enter the lit compartment, in group enriched animals compared to single enriched (p < 0.008) or compared to group impoverished mice (p < 0.031). Single impoverished mice showed a statistical trend for increased anxiety-like behaviour when compared with group impoverished animals (p = 0.06, Fig. 1C). Moreover, ANOVA showed that structural rearing (impoverished versus enriched) affected the number of exits into the aversive compartment [F(1,62) = 6.005; p < 0.017]. Post hoc tests also revealed a significant difference between single impoverished and group impover-
ished rearing (p < 0.016, Fig. 1D). The time spent in the aversive light compartment was also dependent on social conditions [F(1,62) = 8.025; p < 0.006], with group housing resulting in decreased in anxiety levels, while structural housing conditions only caused a trend for reduced anxiety under enriched conditions [F(1,62) = 3.613; p < 0.062]. The interaction between the factors social and structural housing did not reach statistical significance for any of the three parameters investigated (i.e. latency, number of exits, time in lit compartment). All p-values of Fisher’s (LSD) post hoc test are presented in Table 1. 3.3. Housing conditions affect the coping behaviour in the learned helplessness paradigm Before analysing potential effects of housing on learned helplessness, we wanted to know, whether housing affects the two-way avoidance task, which is the “behavioural readout” for helplessness in this paradigm. Therefore, groups of non-shocked mice were compared by ANOVA for possible effects of the different housing conditions on the parameters escape failures and escape latencies, respectively. However,
Fig. 2. Group housed impoverished mice show increased helplessness. (A and B) In the learned helplessness task, no difference occurs between the different rearing conditions in unshocked animals, both with respect to escape failures and latencies. Following shock exposure, however, ANOVA shows a significant interaction of the two housing factors for both, failures and latencies. Post hoc analyses demonstrate that group impoverished mice exhibit significantly more helplessness behaviour than group enriched or single impoverished animals. * p < 0.05, ** p < 0.01.
104
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
no differences were observed (Fig. 2). These results indicated that housing conditions do not produce an effect on avoidance per se. Non-shocked and shocked groups subjected to the same specific housing conditions significantly differed in their shuttle-box performance, respectively (failures: single impoverished p < 0.034; single enriched p < 0.038; group impoverished p < 0.021; group enriched p < 0.067, escape latency: single impoverished p < 0.035; single enriched p < 0.45 × 10−7 ; group impoverished p < 0.15 × 10−5 ; group enriched p < 0.26 × 10−9 , Fig. 2). These results confirmed for all four housing conditions, that exposure to the uncontrollable shock protocol produced increased helplessness. In the learned helplessness paradigm, ANOVA of the shocked groups subjected to the different housing conditions did not show a main effect for enrichment nor for group housing, but a significant interaction of both factors (Fig. 2). This finding was true with regard to both major parameters of interest, escape failures [F(1,64) = 8.46; p < 0.005, Fig. 2A] and escape latencies [F(1,64) = 5.223; p < 0.026, Fig. 2B]. This effect was confirmed by post hoc analyses, which revealed that group housed impoverished mice show significantly increased helplessness with more failures and prolonged latencies, compared to single impoverished as well as to group enriched mice (Fig. 2, Table 1). Thus, when reared in an impoverished environment, group housed animals showed the poorest coping abilities. However, group housing per se as well as impoverished environment per se did not affect the extent of helpless behaviour (p-values of Fisher’s (LSD) post hoc tests are presented in Table 1). 3.4. Housing conditions do not affect pain sensitivity As analysed by a two-way ANOVA, the different housing conditions did not affect the pain sensitivity in the hotplate test (Table 1).
4. Discussion The present study examined the effects of different housing conditions for the development of learned helplessness, an animal model of stress-induced depression-like behaviour. In our learned helplessness paradigm, group housed mice kept in impoverished cages displayed significantly increased helplessness as indicated by an elevated number of escape failures and prolonged escape latencies compared to mice kept under respective other conditions investigated. Interestingly, this effect was neither present when group housing was combined with an enriched cage structure nor when impoverished animals were kept single caged. Consequently, both factors – impoverished environment and group housing – seem to be risk factors for increased helplessness but only if combined. The increased helplessness observed under impoverished group housing conditions could have resulted from alterations
of general home cage behaviours, which are important to consider with respect to the performance of the test, or from interference of stress with different stages of the complex paradigm, which are determined to assess helplessness. First, the elevated number of failures and increased escape latencies could have resulted from locomotor deficits produced by the specific housing conditions. However, as assessed in the open field (total distance moved, locomotor velocity), locomotion was not influenced by the various housing conditions (Fig. 1A). Second, housing could affect the pain threshold, thus changing the impact of the electroshocks used for the induction of helplessness. However, the pain threshold as measured by the hotplate test was not altered by housing conditions (Table 1). Third, shuttle-box performance, i.e. the behavioural read out in the helplessness paradigm following the uncontrollable footshock exposures, could have been influenced by the respective animal husbandry. However, non-shocked animals displayed similar avoidance behaviour irrespective of their housing conditions, indicating that the mode of housing did not affect two-way avoidance per se (Fig. 2). In conclusion, increased helplessness due to impoverished group housing conditions seems to be a true effect of altered stress-sensitivity, which might rely on the different housing constellations. The lack of an adequate ability to cope, when confronted to a stressful situation, represents one of the behavioural core symptoms of depression, which may be imitated in animal models for depression. This fact suggests the learned helplessness model as a suitable tool for the investigation of distinct aspects of a depressive state including a classification of the severity, since it allows the reflection of distinct levels of stress caused by different factors, which also affected emotionality in our study. Therefore, it could be that housing conditions may alter helplessness by changing anxiety-related parameters. Emotional behaviour, anxiety, and mood disorders like depression show partial overlapping in both humans and rodents [34]. However, although some differences among groups were found in anxiety-related measures, the pattern of these results was not coincident with those of the shuttle performance, and in fact no significant correlation between both sets of data was found, which is in line with previous findings in another stress model in which a direct association between a depressive-like state and anxiety is lacking [27]. The results of the present study reveal a specific effect of housing conditions on the development of helpless behaviour. Thus, group housing can be a source of stress in territorial animals such as mice [14] that may result in enhanced behavioural despair [17], in enriched conditions due to possible defence of the elements included in the enrichment. Moreover, in our study, group housed animals kept in an impoverished environment, in which there is a lack of possibilities to shelter, showed the highest scores in two indices of helplessness, the number of failures and the latency to escape in a two-way avoidance task. Vice versa, enrichment, which was reported to have positive effects in mice in terms
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
of behavioural and immunological parameters without negatively affecting standardization or variability of physiological parameters [19,20], can have a protective effect in the learned helplessness paradigm, as reflected by the improved performance in the group-enriched housed animals. This supports the conclusion of Marashi et al. [19,20], in which a positive effect of structural enrichment, also in male mice, is reported. These behavioural observations seem to be in accordance with previous studies that have shown that an enriched environment affects the release of stress hormones and improves some physiological parameters like learning and memory, which accompany depression-like states in both rodents and humans [9,11,13,28]. Consequently, there is no reason for a general deprivation of group-housed male mice from a structural enrichment of their housing cage. However, it should be noted, that, according to our findings, the often-criticized single housing of mice does not seem to have a negative effect on stress sensitivity, but performance of individually housed mice was not sensitive to structural enrichment. Supported by previous results showing an enhanced efficacy of imipramine treatment in the Porsolt and Tail-suspension test only in group housed mice [17] our study suggests, that impoverished group housing in male mice may represent a model of chronic stress, which is associated with a depressive-like state. In summary, the results of the present study show that, besides their known modulating effects on anxiety-like behavioural features [1,24], housing conditions represent an anxiety-independent critical factor for the development of learned helplessness and have to be considered in the experimental design. Furthermore, manipulation of the housing conditions can be a suitable procedure to study behavioural despair in mice, but it has to be regarded, that our results only refer to C57Bl/6N, which is one of the most common used strain in behavioural phenotyping in mutant mice, which may not be transferable to other strains. We consider that this procedure presents some advantages over other available alternatives. First, it allows a bi-directional manipulation of this behavioural phenomenon, which may be manipulated by social (i.e. group housing) as well as structural (i.e. enrichment) factors. Moreover, the utility of this type of manipulation as a model for chronic stress has been partially pharmacologically validated [17]. In addition, as observed in our non-shocked animals, the manipulation of housing conditions does not produce helpless behaviour per se, but rather results in a predisposition that is only expressed after a precipitant factor (i.e. uncontrollable shock exposure), then mimicking the diathesis-stress model currently accepted for affective disorders such as major depression [15]. On the other hand, our model deserves to be extended by the investigation of female mice, overcoming the obstacle of variances during the oestrous cycle, since depression is a disorder, which is not restricted to the male gender. Furthermore, our findings have implications when considering animal welfare and the attempts to standardize the rearing of rodents. Thus, individual housing, which is usually criticized to be stress-
105
ful [2] due to the isolation from (litter)mates, may also be an adequate type of housing depending on the species and strains as well as on the experimental needs.
Acknowledgements This work was supported by grants to PG (GA427/4-2, SFB636) and RS (SFB636) from the Deutsche Forschungsgemeinschaft. SC had a scholarship from the GK 791, University of Heidelberg.
References [1] Augustsson H, van de Weerd HA, Kruitwagen CL, Baumans V. Effect of enrichment on variation and results in the light/dark test. Lab Anim 2003;37:328–40. [2] Bartolomucci A, Palanza P, Sacerdote P, Ceresini G, Chirieleison A, Panerai AE, et al. Individual housing induces altered immunoendocrine responses to psychological stress in male mice. Psychoneuroendocrinology 2003;28:540–58. [3] Blanchard DC, Sakai RR, McEwen B, Weiss SM, Blanchard RJ. Subordination stress: behavioural, brain, and neuroendocrine correlates. Behav Brain Res 1993;58:113–21. [4] Cabib S, Ventura R, Puglisi-Allegra S. Opposite imbalances between mesocortical and mesoaccumbens dopamine responses to stress by the same genotype depending on living conditions. Behav Brain Res 2002;129:179–85. [5] Calvo-Torrent A, Brain PF, Martinez M. Effect of predatory stress on sucrose intake and behaviour on the plus-maze in male mice. Physiol Behav 1999;67:189–96. [6] Chourbaji S, Hellweg R, Brandis D, Z¨orner B, Zacher C, Lang UE, et al. Mice with reduced BDNF expression show decreased choline acetyltransferase levels, but regular brain monoamine levels and unaltered emotional behaviour. Mol Brain Res 2004;121:28– 66. [7] Cobb S. Presidential Address-1976. Social support as a moderator of life stress. Psychosom Med 1976;38:300–14. [8] Conference B. Mutant mice and neuroscience: recommendations concerning genetic background. Proceedings of the Banbury conference on genetic background in mice. Neuron 1997;19:755–9. [9] de Jong IC, Prelle IT, van de Burgwal JA, Lambooij E, Korte SM, Blokhuis HJ, et al. Effects of environmental enrichment on behavioural responses to novelty, learning, and memory, and the circadian rhythm in cortisol in growing pigs. Physiol Behav 2000;68:571–8. [10] Falkenberg T, Ernfors P, Persson H, Lindefors N. Cortical transynaptic activation of tyrosine kinase receptor trkB messenger RNA expression in rat hippocampus. Neuroscience 1992;51:883–9. [11] Dulawa SC, Grandy DK, Low MJ, Paulus MP, Geyer MA. Dopamine D4 receptor-knock-out mice exhibit reduced exploration of novel stimuli. J Neurosci 1999;19:9550–6. [12] Fleischmann A, Hvalby O, Jensen V, Strekalova T, Zacher C, Layer LE, et al. Impaired long-term memory and NR2A-type NMDA receptor-dependent synaptic plasticity in mice lacking c-Fos in the CNS. J Neurosci 2003;23:9116–22. [13] Garcia A, Steiner B, Kronenberg G, Bick-Sander A, Kempermann G. Age-dependent expression of glucocorticoid- and mineralocorticoid receptors on neural precursor cell populations in the adult murine hippocampus. Aging Cell 2004;3:363–71. [14] Gariepy JL, Rodriguiz RM, Jones BC. Handling, genetic and housing effects on the mouse stress system, dopamine function, and behaviour. Pharmacol Biochem Behav 2002;73:7–17.
106
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
[15] Grazioli R, Terry DJ. The role of cognitive vulnerability and stress in the prediction of postpartum depressive symptomatology. Br J Clin Psychol 2000;39(Pt 4):329–47. [16] Hendrie CA, Neill JC. Exposure to the calls of predators of mice activates defensive mechanisms and inhibits consummatory behaviour in an inbred mouse strain. Neurosci Biobehav Rev 1991;15:479–82. [17] Karolewicz B, Paul IA. Group housing of mice increases immobility and antidepressant sensitivity in the forced swim and tail suspension tests. Eur J Pharmacol 2001;415:197–201. [18] Maier SF. Exposure to the stress environment prevents the temporal dissipation of behavioural depression/learned helplessness. Biol Psychiatry 2001;49. [19] Marashi V, Barnekow A, Ossendorf E, Sachser N. Effects of different forms of environmental enrichment on behavioral, endocrinological, and immunological parameters in male mice. Horm Behav 2003;43:281–92. [20] Marashi V, Barnekow A, Sachser N. Effects of environmental enrichment on males of a docile inbred strain of mice. Physiol Behav 2004;82:765–76. [21] McIlwain KL, Merriweather MY, Yuva-Paylor LA, Paylor R. The use of behavioural test batteries: effects of training history. Physiol Behav 2001;73:705–17. [22] Monleon S, D’Aquila P, Parra A, Simon VM, Brain PF, Willner P. Attenuation of sucrose consumption in mice by chronic mild stress and its restoration by imipramine. Psychopharmacology (Berlin) 1995;117:453–7. [23] O’Donoghue PN, Bunyan J, Buckwell AC, Gregory JA, Griffiths PH, Howard BR, et al. LASA recommendations on education and training for licence holders under the UK Animals (Scientific Procedures) Act 1986—FELASA categories B and C. Report of a committee accepted by LASA Council June 1992. Lab Anim 1993;27:189–205. [24] Palanza P. Animal models of anxiety and depression: how are females different? Neurosci Biobehav Rev 2001;25:219–33. [25] Reif A, Schmitt A, Fritzen S, Chourbaji S, Bartsch C, Urani A, et al. Differential effect of endothelial nitric oxide synthase (NOS-III) on
[26]
[27]
[28]
[29] [30]
[31] [32]
[33] [34] [35]
[36]
[37]
the regulation of adult neurogenesis and behaviour. Eur J Neurosci 2004;20:885–95. Seligman ME, Weiss J, Weinraub M, Schulman A. Coping behaviour: learned helplessness, physiological change and learned inactivity. Behav Res Ther 1980;18:459–512. Strekalova T, Spanagel R, Bartsch D, Henn FA, Gass P. Stressinduced anhedonia in mice is associated with deficits in forced swimming and exploration. Neuropsychopharmacology 2004;29:2007–17. Turner CA, Lewis MH. Environmental enrichment: effects on stereotyped behaviour and neurotrophin levels. Physiol Behav 2003;80:259–66. van Gaalen MM, Steckler T. Behavioural analysis of four mouse strains in an anxiety test battery. Behav Brain Res 2000;115:95–106. Vollmayr B, Bachteler D, Vengeliene V, Gass P, Spanagel R, Henn F. Rats with congenital learned helplessness respond less to sucrose but show no deficits in activity or learning. Behav Brain Res 2004;150:217–21. Vollmayr B, Henn FA. Learned helplessness in the rat: improvements in validity and reliability. Brain Res Brain Res Protoc 2001;8:1–7. Von Frijtag JC, Reijmers LG, Van der Harst JE, Leus IE, Van den Bos R, Spruijt BM. Defeat followed by individual housing results in long-term impaired reward- and cognition-related behaviours in rats. Behav Brain Res 2000;117:137–46. Weiss JM. Effects of coping responses on stress. J Comp Physiol Psychol 1968;65:251–60. Willner P. Animal models as simulations of depression. Trends Pharmacol Sci 1991;12:131–6. Willner P, Muscat R, Papp M. Chronic mild stress-induced anhedonia: a realistic animal model of depression. Neurosci Biobehav Rev 1992;16:525–34. Wolfer DP, Litvin O, Morf S, Nitsch RM, Lipp HP, Wurbel H. Laboratory animal welfare: cage enrichment and mouse behaviour. Nature 2004;432:821–2. Wurbel H. Ideal homes? Housing effects on rodent brain and behaviour. Trends Neurosci 2001;24:207–11.
Research report
Social and structural housing conditions influence the development of a depressive-like phenotype in the learned helplessness paradigm in male mice Sabine Chourbaji a,∗ , Christiane Zacher a , Carles Sanchis-Segura b , Rainer Spanagel b , Peter Gass a a b
Department of Behavioural Biology, Central Institute of Mental Health Mannheim (ZI), University of Heidelberg, J5, D-68159 Mannheim, Germany Department of Psychopharmacology, Central Institute of Mental Health Mannheim (ZI), University of Heidelberg, J5, D-68159 Mannheim, Germany Received 25 April 2005; received in revised form 2 June 2005; accepted 2 June 2005 Available online 19 July 2005
Abstract Structural and social factors are known to play a crucial role in the pathogenesis of depression. Since animal models of depression are a major tool to gain insights into the mechanisms involved in the pathophysiology of this disease it is important not only to exploit but also to be aware of factors that may affect these models. As housing represents a fundamental external factor, which is controversially debated to affect the animals’ emotionality, this study aimed to investigate the impact of different social and structural housing conditions on the development of a depressive-like syndrome in the learned helplessness paradigm. Group housing in an impoverished environment led to an increased vulnerability in the learned helplessness paradigm. Groups that were housed enriched, however, were less helpless. Furthermore impoverished conditions did not increase the vulnerability in single housed animals. Regarding emotionality in the animals, basal anxiety was reduced and the exploration was enhanced by group housing and enriched environment. These results suggest that housing conditions significantly influence the outcome of learned helplessness studies. © 2005 Published by Elsevier B.V. Keywords: Housing; Enrichment; Depression; Learned helplessness; Mice
1. Introduction Working on behavioural animal models for disease states requires exact knowledge and control of “internal” and “external” factors that may influence the outcome of the experiments. Internal factors comprise the animals’ gender, age, hormone state, satiety and genetic background; external factors are housing and social conditions as well as many other factors. While the importance of internal factors has been widely recognized in recent years (e.g. Ban∗
Corresponding author. Tel.: +49 621 1703 2932; fax: +49 621 1703 2005. E-mail addresses: [email protected] (S. Chourbaji), [email protected] (C. Zacher), [email protected] (C. SanchisSegura), [email protected] (R. Spanagel), [email protected] (P. Gass). 0166-4328/$ – see front matter © 2005 Published by Elsevier B.V. doi:10.1016/j.bbr.2005.06.003
bury conference [8]), housing of experimental animals is frequently reported as “standardized” without a clear definition. The housing conditions for mice, which are considered as “standardized” in most laboratories are barren compared to any natural environment, and thereby imply behavioural and social (if kept single caged) deprivation. This results in a reduction of the natural behavioural repertoire [36] as well as in alterations of specific brain functions of the animals [23,37]. In contrast, it has been shown that structurally enriched or super-enriched home-cages significantly alter physiological parameters thought to be relevant for stressrelated disorders, such as corticosterone levels [9], hippocampal neurogenesis [13], the release of neurotrophins [11,28] or body weight [1]. Thus, the specific conditions of keeping the animals seem to be particularly relevant for animal models of stress-related disorders [4,32]. However, the effects
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
of housing conditions on the development of depressionlike behaviours in a generally recognized model, such as the learned helplessness paradigm have yet not been studied elaborately. The learned helplessness paradigm, a rodent model of depressive-like states with appropriate construct and face validity [18,26,30,31,33], may reflect the described housinginduced physiological changes at the behavioural level. For instance, exposure to an opponent, may act as a source of chronic moderate stress altering depression-relevant parameters [3,16,35] and if so, according to the current views of depression, housing of animals in groups might evoke an increase in coping deficits. Supporting this hypothesis, group housing, a common way of animal husbandry, in which the mice, being territorial animals, are permanently confronted to potential rivals, increased depressive-like behaviour such as immobility in the forced swim and in tail suspension test, suggesting that long-lasting group housing may constitute a stressful environment [17]. These evidences seem to be in agreement with the consideration of social interaction as an important factor regarding stress vulnerability and ability to cope with demanding situations in rats and human beings [5,7,22]. Therefore variation in animal husbandry may be employed as a source of stress itself and consequently constitutes an appropriate instrument for the construction of a natural stressor. The present study was designed to assess the influence of different housing conditions on 4 months old male C57Bl/6N mice with regard to anxiety-related behavioural features in the open field and dark–light box, as well as the development of depressive-like behaviour in the learned helplessness paradigm. For this purpose we analysed the effects of single versus group housing as well as the influence of impoverished versus enriched structural conditions in these models. The final goal of this study was (i) to elucidate the potential role of group housing as a model for chronic stress; and (ii) to determine, if any of the “standardized” keeping conditions hold subtle advantages or limitations with respect to the learned helplessness paradigm.
2. Materials and methods 2.1. Animals All behavioural tests were conducted in 4 months old male C57Bl/6N mice, which is the most frequently used background strain in genetic engineering. The animals were purchased from Charles River (Sulzfeld, Germany) at an age of 8 weeks and were then housed for 7 weeks under specific conditions in type II cages for single and type III cages for group rearing (four mice per cage). Four cohorts of mice were investigated: (1) single housed, impoverished conditions (n = 16), (2) single housed, enriched conditions (n = 16), (3) group housed, impoverished conditions (n = 16), (4) group housed, enriched conditions (n = 16). Enrichment consisted of
101
red, transparent plastic mouse igloos and tunnels (EMSICON Jung GmbH, Forstinning, Germany) that were cleaned weekly, when the cage was changed. Additionally, nesting material (tissue) was provided. Impoverished cages simply contained bedding material. All mice were kept in the same room, in a 12 h:12 h reversed dark–light cycle, lights on at 6.00 p.m. Water and food pellets were available ad libitum. Body weight was assessed once a week when the cages were cleaned. All experiments were approved by German animal welfare authorities. 2.2. Behavioural procedures After 7 weeks of housing under the aforementioned conditions, all mice were subjected to tests for locomotion, exploration, and anxiety, followed by the learned helplessness procedure. Following earlier recommendations for repetitive behavioural testing, animals were initially tested in the experiments ranked as less stressful [21,29]. Considering that the determination of the pain threshold by the hot plate test could have a direct influence on helpless behaviour, we assessed the effect of housing conditions on pain sensitivity in a separate group of mice reared under identical housing conditions. Test procedures were essentially performed as described earlier [6,12,25]. Between individual tests was a pause of at least 24 h. Prior to each test, mice were acclimatized to the experimental room for 30 min. All behavioural tests were conducted during the dark cycle, i.e. during the animals’ active phase. 2.3. Open field test Activity monitoring was performed in a square, white open field, measuring 50 cm × 50 cm and illuminated from above by 25 lx. Mice were placed individually into the arena and monitored for 15 min by a Video camera (Sony CCD IRIS). The resulting data were analysed using the image processing system EthoVision 2.3 (Noldus Information Technology, Wageningen, The Netherlands). For each sample, the system recorded position and the occurrence of defined events. Parameters assessed in the present study were total distance moved, velocity, and thigmotaxis (i.e. the percentage of time spent in a corridor with maximal distance of 10 cm to the walls). 2.4. Dark–light box test The dark–light box consisted of two plastic chambers, connected by a small tunnel. The dark chamber measured 20 cm × 15 cm and was covered by a lid. The light chamber, measuring 30 cm × 15 cm, was white and illuminated from above with an intensity of 600 lx. Mice were placed into the dark compartment, and the latency to the first exit, the number of exits and the total time spent in the light compartment were recorded for 5 min. 2.5. Learned helplessness The learned helplessness paradigm was conducted as described earlier for male mice [25]. Mice received 360 footshocks (0.150 mA) on two consecutive days in a transparent plexiglas shock chamber (18 cm × 18 cm × 30 cm), equipped with a stainless steel grid floor (Coulborn precision regulated animal shocker, Coulborn Instruments, D¨usseldorf, Germany). The footshocks applied were unpre-
102
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
dictable with varying shock-durations (1–3 s) and interval-episodes (1–15 s), amounting to a total session duration of approximately 52 min. Twenty-four hours after the second shock exposure, learned helplessness was assessed by testing two-way avoidance in a shuttle box (Coulborn Instruments, D¨usseldorf, Germany). Spontaneous initial shuttle activity from one compartment to the other was monitored during the first two minutes by red-light beams at the bottom of each of the two compartments. Thereafter, specific avoidance learning was tested over 30 trials. Each trial started with a light stimulus of 5 s, announcing a subsequent footshock (0.150 mA; maximal duration: 10 s). The duration of this procedure varied (total testing time averaged 20 min) according to each animal’s individual performance in terms of failures, i.e. not attempting to escape from the shock, and escape latencies, i.e. latency to escape after onset of the shock. An additional batch of mice, reared under the equal four conditions as described above (n = 8 per condition) was tested in the same two-way avoidance paradigm. However, these mice were not exposed to uncontrolled shocks before shuttle-box testing, thus providing a control for a possible effect of housing conditions per se on the performance in this test. These animals are referred as “nonshocked” mice. 2.6. Hotplate test To exclude altered pain sensitivity as a modulating factor for learned helplessness, an additional batch of mice (n = 8 per condition) was tested on the hotplate test (ATLab, Vendargues, France). Temperature was set at 53 ◦ C and a 45 s cut-off was introduced to prevent injury of mice. Latency to first reaction, i.e. licking hind paws or jumping, was assessed.
2.7. Statistical analyses For statistical analyses, values of each housing condition were evaluated by two-way ANOVAs followed by Fisher’s (LSD) post hoc tests (XLstat program Version 7.5, Addinsoft) calculating the statistically significant effects of factors as well as interactions at p < 0.05. Correlations between variables were analysed by Pearson’s correlation reaching significant levels at p < 0.05. Comparisons of escape deficits in the learned helplessness paradigm between shocked animals and non-shocked controls were performed by ttests.
3. Results 3.1. Group housing affects exploratory behaviour but not locomotion In the open field test, the housing conditions did not affect basal locomotor activity of the animals, i.e. the total distance moved (Fig. 1A) or the locomotor velocity (Table 1). However, a single exposure to the open field test also confronts the animals with an approach-avoidance conflict, in which the exploration of the animals, measured by the time spent in the centre of the open field arena [10], was significantly influenced by social housing conditions [F(1,62) = 10.962; p < 0.002]. Post hoc tests demonstrated, that both group housed cohorts (impoverished and enriched) spent more time in the centre than the respective single housed groups (Fig. 1B). p-values of Fisher’s (LSD) post hoc tests are pre-
Fig. 1. Locomotor, exploratory and anxiety-like behaviour under different housing conditions. (A) In the open field test, no changes in locomotor activity occur under the different housing conditions applied. (B) Group housed mice spend significantly more time in the centre of the arena than single housed animals. (C and D) In the dark–light box test, both social as well as structural factors affect anxiety-like behaviour, but there is no interaction between these two factors (cf. Section 3). * p < 0.05, ** p < 0.01.
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
103
Table 1 Significant inter-group differences: Fisher’s (LSD) post hoc test p-values Single impoverished
Single enriched
Group impoverished
Group enriched
Openfield locomotion Openfield velocity Openfield centre time Dark–light box latency
n.s. n.s vs. group imp.: p = 0.042 n.s.
n.s. n.s vs. group enr.: p = 0.002 vs. group enr.: p = 0.008
n.s. n.s vs. single imp.: p = 0.042
Dark–light box exits Dark–light box time in lit compartment Learned helplessness failures
n.s. vs. group imp.: p = 0.044
n.s. n.s.
vs. group enr.: p = 0.031 vs. group enr.: p = 0.016 vs. single imp.: p = 0.044
n.s. n.s vs. single enr.: p = 0.002 vs. single enr.: p = 0.008 vs. group imp.: p = 0.031 vs. group imp.: p = 0.016 n.s.
vs. group imp.: p = 0.008 ↓
n.s.
vs. single imp.: p = 0.008 ↑
vs. group imp.: p = 0.014 ↓
n.s.
vs. group enr.: p = 0.005 ↑ vs. single imp.: p = 0.014 ↑
vs. group imp.: p = 0.005 ↓
Learned helplessness escape latency Hotplate pain sensitivity
n.s.
n.s.
vs. group enr.: p = 0.007 ↑ n.s.
vs. group imp.: p = 0.007 ↓ n.s.
Group comparisons: p-values indicate inter-group differences in all parameters assessed, according to a Fisher’ (LSD) post hoc comparison. n.s.: not significant; enr.: enriched; imp.: impoverished. (↑) enhanced helplessness; (↓) reduced helplessness.
sented in Table 1. There was no statistical effect for the interaction between the factors structural and social housing. 3.2. Group housing and enrichment evoke reduced anxiety-like behaviour Both housing factors, social [F(1,62) = 7.797; p < 0.007] and structural conditions [F(1,62) = 4.86; p < 0.031], significantly affected the latency to enter the aversive light compartment, while the interaction between factors did not show a significant effect (Fig. 1C). Post hoc tests (Table 1) demonstrated significantly reduced anxiety-like behaviour, i.e. latency to enter the lit compartment, in group enriched animals compared to single enriched (p < 0.008) or compared to group impoverished mice (p < 0.031). Single impoverished mice showed a statistical trend for increased anxiety-like behaviour when compared with group impoverished animals (p = 0.06, Fig. 1C). Moreover, ANOVA showed that structural rearing (impoverished versus enriched) affected the number of exits into the aversive compartment [F(1,62) = 6.005; p < 0.017]. Post hoc tests also revealed a significant difference between single impoverished and group impover-
ished rearing (p < 0.016, Fig. 1D). The time spent in the aversive light compartment was also dependent on social conditions [F(1,62) = 8.025; p < 0.006], with group housing resulting in decreased in anxiety levels, while structural housing conditions only caused a trend for reduced anxiety under enriched conditions [F(1,62) = 3.613; p < 0.062]. The interaction between the factors social and structural housing did not reach statistical significance for any of the three parameters investigated (i.e. latency, number of exits, time in lit compartment). All p-values of Fisher’s (LSD) post hoc test are presented in Table 1. 3.3. Housing conditions affect the coping behaviour in the learned helplessness paradigm Before analysing potential effects of housing on learned helplessness, we wanted to know, whether housing affects the two-way avoidance task, which is the “behavioural readout” for helplessness in this paradigm. Therefore, groups of non-shocked mice were compared by ANOVA for possible effects of the different housing conditions on the parameters escape failures and escape latencies, respectively. However,
Fig. 2. Group housed impoverished mice show increased helplessness. (A and B) In the learned helplessness task, no difference occurs between the different rearing conditions in unshocked animals, both with respect to escape failures and latencies. Following shock exposure, however, ANOVA shows a significant interaction of the two housing factors for both, failures and latencies. Post hoc analyses demonstrate that group impoverished mice exhibit significantly more helplessness behaviour than group enriched or single impoverished animals. * p < 0.05, ** p < 0.01.
104
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
no differences were observed (Fig. 2). These results indicated that housing conditions do not produce an effect on avoidance per se. Non-shocked and shocked groups subjected to the same specific housing conditions significantly differed in their shuttle-box performance, respectively (failures: single impoverished p < 0.034; single enriched p < 0.038; group impoverished p < 0.021; group enriched p < 0.067, escape latency: single impoverished p < 0.035; single enriched p < 0.45 × 10−7 ; group impoverished p < 0.15 × 10−5 ; group enriched p < 0.26 × 10−9 , Fig. 2). These results confirmed for all four housing conditions, that exposure to the uncontrollable shock protocol produced increased helplessness. In the learned helplessness paradigm, ANOVA of the shocked groups subjected to the different housing conditions did not show a main effect for enrichment nor for group housing, but a significant interaction of both factors (Fig. 2). This finding was true with regard to both major parameters of interest, escape failures [F(1,64) = 8.46; p < 0.005, Fig. 2A] and escape latencies [F(1,64) = 5.223; p < 0.026, Fig. 2B]. This effect was confirmed by post hoc analyses, which revealed that group housed impoverished mice show significantly increased helplessness with more failures and prolonged latencies, compared to single impoverished as well as to group enriched mice (Fig. 2, Table 1). Thus, when reared in an impoverished environment, group housed animals showed the poorest coping abilities. However, group housing per se as well as impoverished environment per se did not affect the extent of helpless behaviour (p-values of Fisher’s (LSD) post hoc tests are presented in Table 1). 3.4. Housing conditions do not affect pain sensitivity As analysed by a two-way ANOVA, the different housing conditions did not affect the pain sensitivity in the hotplate test (Table 1).
4. Discussion The present study examined the effects of different housing conditions for the development of learned helplessness, an animal model of stress-induced depression-like behaviour. In our learned helplessness paradigm, group housed mice kept in impoverished cages displayed significantly increased helplessness as indicated by an elevated number of escape failures and prolonged escape latencies compared to mice kept under respective other conditions investigated. Interestingly, this effect was neither present when group housing was combined with an enriched cage structure nor when impoverished animals were kept single caged. Consequently, both factors – impoverished environment and group housing – seem to be risk factors for increased helplessness but only if combined. The increased helplessness observed under impoverished group housing conditions could have resulted from alterations
of general home cage behaviours, which are important to consider with respect to the performance of the test, or from interference of stress with different stages of the complex paradigm, which are determined to assess helplessness. First, the elevated number of failures and increased escape latencies could have resulted from locomotor deficits produced by the specific housing conditions. However, as assessed in the open field (total distance moved, locomotor velocity), locomotion was not influenced by the various housing conditions (Fig. 1A). Second, housing could affect the pain threshold, thus changing the impact of the electroshocks used for the induction of helplessness. However, the pain threshold as measured by the hotplate test was not altered by housing conditions (Table 1). Third, shuttle-box performance, i.e. the behavioural read out in the helplessness paradigm following the uncontrollable footshock exposures, could have been influenced by the respective animal husbandry. However, non-shocked animals displayed similar avoidance behaviour irrespective of their housing conditions, indicating that the mode of housing did not affect two-way avoidance per se (Fig. 2). In conclusion, increased helplessness due to impoverished group housing conditions seems to be a true effect of altered stress-sensitivity, which might rely on the different housing constellations. The lack of an adequate ability to cope, when confronted to a stressful situation, represents one of the behavioural core symptoms of depression, which may be imitated in animal models for depression. This fact suggests the learned helplessness model as a suitable tool for the investigation of distinct aspects of a depressive state including a classification of the severity, since it allows the reflection of distinct levels of stress caused by different factors, which also affected emotionality in our study. Therefore, it could be that housing conditions may alter helplessness by changing anxiety-related parameters. Emotional behaviour, anxiety, and mood disorders like depression show partial overlapping in both humans and rodents [34]. However, although some differences among groups were found in anxiety-related measures, the pattern of these results was not coincident with those of the shuttle performance, and in fact no significant correlation between both sets of data was found, which is in line with previous findings in another stress model in which a direct association between a depressive-like state and anxiety is lacking [27]. The results of the present study reveal a specific effect of housing conditions on the development of helpless behaviour. Thus, group housing can be a source of stress in territorial animals such as mice [14] that may result in enhanced behavioural despair [17], in enriched conditions due to possible defence of the elements included in the enrichment. Moreover, in our study, group housed animals kept in an impoverished environment, in which there is a lack of possibilities to shelter, showed the highest scores in two indices of helplessness, the number of failures and the latency to escape in a two-way avoidance task. Vice versa, enrichment, which was reported to have positive effects in mice in terms
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
of behavioural and immunological parameters without negatively affecting standardization or variability of physiological parameters [19,20], can have a protective effect in the learned helplessness paradigm, as reflected by the improved performance in the group-enriched housed animals. This supports the conclusion of Marashi et al. [19,20], in which a positive effect of structural enrichment, also in male mice, is reported. These behavioural observations seem to be in accordance with previous studies that have shown that an enriched environment affects the release of stress hormones and improves some physiological parameters like learning and memory, which accompany depression-like states in both rodents and humans [9,11,13,28]. Consequently, there is no reason for a general deprivation of group-housed male mice from a structural enrichment of their housing cage. However, it should be noted, that, according to our findings, the often-criticized single housing of mice does not seem to have a negative effect on stress sensitivity, but performance of individually housed mice was not sensitive to structural enrichment. Supported by previous results showing an enhanced efficacy of imipramine treatment in the Porsolt and Tail-suspension test only in group housed mice [17] our study suggests, that impoverished group housing in male mice may represent a model of chronic stress, which is associated with a depressive-like state. In summary, the results of the present study show that, besides their known modulating effects on anxiety-like behavioural features [1,24], housing conditions represent an anxiety-independent critical factor for the development of learned helplessness and have to be considered in the experimental design. Furthermore, manipulation of the housing conditions can be a suitable procedure to study behavioural despair in mice, but it has to be regarded, that our results only refer to C57Bl/6N, which is one of the most common used strain in behavioural phenotyping in mutant mice, which may not be transferable to other strains. We consider that this procedure presents some advantages over other available alternatives. First, it allows a bi-directional manipulation of this behavioural phenomenon, which may be manipulated by social (i.e. group housing) as well as structural (i.e. enrichment) factors. Moreover, the utility of this type of manipulation as a model for chronic stress has been partially pharmacologically validated [17]. In addition, as observed in our non-shocked animals, the manipulation of housing conditions does not produce helpless behaviour per se, but rather results in a predisposition that is only expressed after a precipitant factor (i.e. uncontrollable shock exposure), then mimicking the diathesis-stress model currently accepted for affective disorders such as major depression [15]. On the other hand, our model deserves to be extended by the investigation of female mice, overcoming the obstacle of variances during the oestrous cycle, since depression is a disorder, which is not restricted to the male gender. Furthermore, our findings have implications when considering animal welfare and the attempts to standardize the rearing of rodents. Thus, individual housing, which is usually criticized to be stress-
105
ful [2] due to the isolation from (litter)mates, may also be an adequate type of housing depending on the species and strains as well as on the experimental needs.
Acknowledgements This work was supported by grants to PG (GA427/4-2, SFB636) and RS (SFB636) from the Deutsche Forschungsgemeinschaft. SC had a scholarship from the GK 791, University of Heidelberg.
References [1] Augustsson H, van de Weerd HA, Kruitwagen CL, Baumans V. Effect of enrichment on variation and results in the light/dark test. Lab Anim 2003;37:328–40. [2] Bartolomucci A, Palanza P, Sacerdote P, Ceresini G, Chirieleison A, Panerai AE, et al. Individual housing induces altered immunoendocrine responses to psychological stress in male mice. Psychoneuroendocrinology 2003;28:540–58. [3] Blanchard DC, Sakai RR, McEwen B, Weiss SM, Blanchard RJ. Subordination stress: behavioural, brain, and neuroendocrine correlates. Behav Brain Res 1993;58:113–21. [4] Cabib S, Ventura R, Puglisi-Allegra S. Opposite imbalances between mesocortical and mesoaccumbens dopamine responses to stress by the same genotype depending on living conditions. Behav Brain Res 2002;129:179–85. [5] Calvo-Torrent A, Brain PF, Martinez M. Effect of predatory stress on sucrose intake and behaviour on the plus-maze in male mice. Physiol Behav 1999;67:189–96. [6] Chourbaji S, Hellweg R, Brandis D, Z¨orner B, Zacher C, Lang UE, et al. Mice with reduced BDNF expression show decreased choline acetyltransferase levels, but regular brain monoamine levels and unaltered emotional behaviour. Mol Brain Res 2004;121:28– 66. [7] Cobb S. Presidential Address-1976. Social support as a moderator of life stress. Psychosom Med 1976;38:300–14. [8] Conference B. Mutant mice and neuroscience: recommendations concerning genetic background. Proceedings of the Banbury conference on genetic background in mice. Neuron 1997;19:755–9. [9] de Jong IC, Prelle IT, van de Burgwal JA, Lambooij E, Korte SM, Blokhuis HJ, et al. Effects of environmental enrichment on behavioural responses to novelty, learning, and memory, and the circadian rhythm in cortisol in growing pigs. Physiol Behav 2000;68:571–8. [10] Falkenberg T, Ernfors P, Persson H, Lindefors N. Cortical transynaptic activation of tyrosine kinase receptor trkB messenger RNA expression in rat hippocampus. Neuroscience 1992;51:883–9. [11] Dulawa SC, Grandy DK, Low MJ, Paulus MP, Geyer MA. Dopamine D4 receptor-knock-out mice exhibit reduced exploration of novel stimuli. J Neurosci 1999;19:9550–6. [12] Fleischmann A, Hvalby O, Jensen V, Strekalova T, Zacher C, Layer LE, et al. Impaired long-term memory and NR2A-type NMDA receptor-dependent synaptic plasticity in mice lacking c-Fos in the CNS. J Neurosci 2003;23:9116–22. [13] Garcia A, Steiner B, Kronenberg G, Bick-Sander A, Kempermann G. Age-dependent expression of glucocorticoid- and mineralocorticoid receptors on neural precursor cell populations in the adult murine hippocampus. Aging Cell 2004;3:363–71. [14] Gariepy JL, Rodriguiz RM, Jones BC. Handling, genetic and housing effects on the mouse stress system, dopamine function, and behaviour. Pharmacol Biochem Behav 2002;73:7–17.
106
S. Chourbaji et al. / Behavioural Brain Research 164 (2005) 100–106
[15] Grazioli R, Terry DJ. The role of cognitive vulnerability and stress in the prediction of postpartum depressive symptomatology. Br J Clin Psychol 2000;39(Pt 4):329–47. [16] Hendrie CA, Neill JC. Exposure to the calls of predators of mice activates defensive mechanisms and inhibits consummatory behaviour in an inbred mouse strain. Neurosci Biobehav Rev 1991;15:479–82. [17] Karolewicz B, Paul IA. Group housing of mice increases immobility and antidepressant sensitivity in the forced swim and tail suspension tests. Eur J Pharmacol 2001;415:197–201. [18] Maier SF. Exposure to the stress environment prevents the temporal dissipation of behavioural depression/learned helplessness. Biol Psychiatry 2001;49. [19] Marashi V, Barnekow A, Ossendorf E, Sachser N. Effects of different forms of environmental enrichment on behavioral, endocrinological, and immunological parameters in male mice. Horm Behav 2003;43:281–92. [20] Marashi V, Barnekow A, Sachser N. Effects of environmental enrichment on males of a docile inbred strain of mice. Physiol Behav 2004;82:765–76. [21] McIlwain KL, Merriweather MY, Yuva-Paylor LA, Paylor R. The use of behavioural test batteries: effects of training history. Physiol Behav 2001;73:705–17. [22] Monleon S, D’Aquila P, Parra A, Simon VM, Brain PF, Willner P. Attenuation of sucrose consumption in mice by chronic mild stress and its restoration by imipramine. Psychopharmacology (Berlin) 1995;117:453–7. [23] O’Donoghue PN, Bunyan J, Buckwell AC, Gregory JA, Griffiths PH, Howard BR, et al. LASA recommendations on education and training for licence holders under the UK Animals (Scientific Procedures) Act 1986—FELASA categories B and C. Report of a committee accepted by LASA Council June 1992. Lab Anim 1993;27:189–205. [24] Palanza P. Animal models of anxiety and depression: how are females different? Neurosci Biobehav Rev 2001;25:219–33. [25] Reif A, Schmitt A, Fritzen S, Chourbaji S, Bartsch C, Urani A, et al. Differential effect of endothelial nitric oxide synthase (NOS-III) on
[26]
[27]
[28]
[29] [30]
[31] [32]
[33] [34] [35]
[36]
[37]
the regulation of adult neurogenesis and behaviour. Eur J Neurosci 2004;20:885–95. Seligman ME, Weiss J, Weinraub M, Schulman A. Coping behaviour: learned helplessness, physiological change and learned inactivity. Behav Res Ther 1980;18:459–512. Strekalova T, Spanagel R, Bartsch D, Henn FA, Gass P. Stressinduced anhedonia in mice is associated with deficits in forced swimming and exploration. Neuropsychopharmacology 2004;29:2007–17. Turner CA, Lewis MH. Environmental enrichment: effects on stereotyped behaviour and neurotrophin levels. Physiol Behav 2003;80:259–66. van Gaalen MM, Steckler T. Behavioural analysis of four mouse strains in an anxiety test battery. Behav Brain Res 2000;115:95–106. Vollmayr B, Bachteler D, Vengeliene V, Gass P, Spanagel R, Henn F. Rats with congenital learned helplessness respond less to sucrose but show no deficits in activity or learning. Behav Brain Res 2004;150:217–21. Vollmayr B, Henn FA. Learned helplessness in the rat: improvements in validity and reliability. Brain Res Brain Res Protoc 2001;8:1–7. Von Frijtag JC, Reijmers LG, Van der Harst JE, Leus IE, Van den Bos R, Spruijt BM. Defeat followed by individual housing results in long-term impaired reward- and cognition-related behaviours in rats. Behav Brain Res 2000;117:137–46. Weiss JM. Effects of coping responses on stress. J Comp Physiol Psychol 1968;65:251–60. Willner P. Animal models as simulations of depression. Trends Pharmacol Sci 1991;12:131–6. Willner P, Muscat R, Papp M. Chronic mild stress-induced anhedonia: a realistic animal model of depression. Neurosci Biobehav Rev 1992;16:525–34. Wolfer DP, Litvin O, Morf S, Nitsch RM, Lipp HP, Wurbel H. Laboratory animal welfare: cage enrichment and mouse behaviour. Nature 2004;432:821–2. Wurbel H. Ideal homes? Housing effects on rodent brain and behaviour. Trends Neurosci 2001;24:207–11.