- Email: [email protected]
The effect of different psychological profiles and timings of stress exposure on humoral immune response
Physiology & Behavior 76 (2002) 321 – 326
The effect of different psychological profiles and timings of stress exposure on humoral immune response O.C. Gasparottoa,*, Z.M. Igna´cioa, K. Lina, S. Gonc˛alvesb a
Department of Physiological Sciences, Biological Sciences Centre, Federal University of Santa Catarina, Trindade, Floriano´polis, SC 88040-900, Brazil b Department of Microbiology and Parasitology, Biological Sciences Centre, Federal University of Santa Catarina, Trindade, Floriano´polis, SC 88040-900, Brazil Received 16 January 2001; received in revised form 22 November 2001; accepted 27 March 2002
Abstract The aim of this study was to analyse the effects of different timings of stress exposure on humoral immune response in mice previously distinguished by their own psychological profile. The Swiss mice were submitted to two different protocols of social confrontation, based on the timing of stress exposure in relation to an immune challenge: animals socially confronted daily for 2 weeks (SINT) and another group confronted daily for 3 weeks (LINT). Control groups were individually housed in a different room. All groups were intraperitoneally injected with sheep red blood cells (SRBC). The SINT group was challenged on the 1st and 7th days of confrontation, whereas the LINT group was challenged on the 7th and 14th days. Two days prior to the period of social conflict, the animals were tested in the elevated plus-maze (PM). The SINT protocol caused a more depressed primary immune response in the submissive mice than that observed in the dominants. The LINT protocol induced a marked increase in the primary immune response, which was more evident in the dominant mice, whilst no changes were observed in the secondary immune response. In the control and dominant groups, the correlation analysis attributed a higher anti-SRBC titre to the more anxious animals; by contrast, higher anti-SRBC titres were found in the less anxious submissive mice. These studies show that different physiological and behavioural adaptations to environmental demands over time, as well as different psychological profiles, constitute important factors to a better understanding of neuro-immune interactions. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Psychosocial stress; Humoral immune response; Mice; Social status; Anxiety; Psychological profile
1. Introduction For a long time, researchers have tried to understand the symptoms associated with diverse kinds of stressful stimuli. The reaction to stress is dependent not only on the nature and intensity of the stressors but also on the individual’s psychological profile and means of coping with such stimuli [1,2]. Independent of individual characteristics, the activation of the pituitary –adrenal axis and the sympatho-adrenal system are markers of the stress reaction [3]. Such activations lead to broad functional and behavioural changes, which can modulate anxiety [4], depression [5,6], blood pressure [7] and gastric ulceration [8]. It has also been shown that social stress has a powerful impact on the immune system [9– 14]. To understand the neurobiology of psychosocial stress and their effects on social creatures, it
is important to consider all the inherent variables involved. The psychological profile was considered as a determinant factor for the animal’s ability to deal with the environment. The dominant and submissive status in the social conflicts may be favoured by its own psychological profile. Considering that the individual differences demand different ways of coping with the stressful stimuli, it might be expected that it should differently affect neuroimmunological interactions. Based on these assumptions, the aim of this work was to consider the psychological profile of animals that were further subjected to different timings of social stress exposure in relation to an immune challenge.
2. Materials and methods 2.1. Mice
* Corresponding author. Tel.: +55-48-3319352; fax: +55-48-3319672. E-mail address: [email protected] (O.C. Gasparotto).
The animals were bred in the Animal Facilities of the Federal University of Santa Catarina and were maintained
0031-9384/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 1 - 9 3 8 4 ( 0 2 ) 0 0 7 1 8 - 7
322
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
according to the Animal Care Guidelines from the National Institute of Health of the United States of America. Three-month-old male SWISS mice (n = 78) were individually housed in clear polypropylene cages (19 30 13 cm) and were maintained for an adaptation period of 30 days in a temperature-controlled room (24 C) under standard lighting (lights on from 05:30 to 17:30 h) with free access to food and water. The animals were handled once a week at the time of cage cleaning. For technical reasons, the data from nine animals could not be included in the statistical analysis. 2.2. Social confrontation Male SWISS mice were randomly separated into two groups and were identified as intruders (I) or residents (R). The residents previously isolated in a propylene cages were transferred to a wooden cage (19 30 13 cm) with a transparent front wall 3 days prior to the beginning of social interactions. The residents were always confronted with the same intruders, which were maintained individually in the same room in propylene cages. The agonistic interaction was performed using animals from different litters. The stress procedures took place when the intruder was transferred into the resident’s cage. To distinguish the resident from intruder animals during the confrontation, the backs of the residents were lightly painted with gentian-violet 2 days before testing. The social interaction, conducted under red lighting, started at 17:30, and the behaviour of the animals was analysed for 30 min. The dominant mice were identified by lateral threat postures, as well as offensive attacks. The subordinate mice were identified as animals that vocalised and displayed defensive upright or upside down postures, as well as freezing and flight when attacked by the opponent. The dominance was only considered established when one of the animals triggered 100% of offensive attacks (dominants) towards their opponents (submissives). 2.3. Immunisation The immunisation procedure was performed using sheep red blood cells (SRBC) washed in a phosphate-buffered saline (PBS; 0.01 M, pH 7.2), and resuspended in a volume containing 109 cells/ml. All the animals were intraperitoneally (ip) injected with 108 SRBC/0.1 ml at the beginning of the dark phase (18:00– 18:30 h). 2.4. Serum samples and haemaglutination assay Mice were anaesthetised with ether and 150 ml of blood was collected by retro-orbital puncture. In all serum samples, the presence of anti-SRBC antibodies was evaluated using the technique described earlier [15]. Samples of 25 ml of serum were added to a U-shaped 96-well-microtiter plate, and a twofold serial dilution was prepared, using PBS as the diluent. The same volume of a 2% suspension of SRBC in PBS
was added to each well and was incubated for 2 –3 h at room temperature. The titre of the natural agglutinating activity of the serum sample was expressed as the reciprocal of the highest dilution showing a positive pattern of agglutination. 2.5. Timings of stress exposure and immune challenge 2.5.1. Short agonistic interaction (SINT) Mice (n = 15) were subjected to social confrontation for a period of 30 min/day for 2 weeks. This protocol was used in order to analyse the effects of short agonistic interaction on humoral immune response to SRBC, in which the animals were intraperitoneally inoculated with SRBC following the social interaction on the 1st and 7th day of experimentation. The blood samples were collected on the 7th day after each antigen inoculation, when the confrontation did not take place, in order to avoid stress overload. The sampling procedure was used in all groups. 2.5.2. Long agonistic interaction (LINT) Mice (n = 26) were subjected to social confrontation for a period of 30 min/day for 3 weeks. To analyse the effects of long agonistic interaction on humoral immune response to SRBC, the animals were intraperitoneally inoculated with SRBC on the 7th and 14th days of experimentation. The blood samples were obtained on the 14th and 21st days and as in the SINT protocol, the animals did not interact in these days. 2.5.3. Control group (CT) Mice were individually housed in polypropylene cages (n = 28), were kept in a separate room for 2 weeks and were identified as control. This group was inoculated with SRBC on the 1st and 7th days of experimentation. 2.6. Anxiety test To measure anxiety levels, each animal was individually tested in an elevated plus-maze (PM) adapted model [16]. The PM apparatus was made of opaque Plexiglas and consisted of two open arms (30 5 cm) and two enclosed arms (30 5 15 cm). The apparatus was mounted on a wooden base, which raised it to 40 cm above the floor. The test was performed in a room with diffuse red light 1 day before starting the agonistic interaction. A normal 5-min test duration was employed, with the maze cleaned between subjects. All the tests were recorded using a video camera. The anxiety levels were expressed as the percentage of time spent in open arms (%TO). The percentage of entries into the open arms of the maze is not shown since this gave similar results to %TO. 2.7. Data analysis Data were evaluated using nonparametric tests for independent multiple variables: Dunn test and Kruskal – Wallis test followed by a nonparametric test for two inde-
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
323
pendent variables (Mann – Whitney U test). The Spearman test was employed to correlate the immunological data with anxiety levels. The antibody titres P were expressed as geometric mean, GM = antilog[(1/n) (logY)], and the confidence interval (L1 and L2) was calculated as described bellow [17]: h pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffii L1 ¼ antilog MAðlogY Þ tð:05, nÞ ðS 2 ðlogY Þ=nÞ ,
3.2. Effects of social stress on humoral immune response to SRBC
h pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffii L2 ¼ antilog MAðlogY Þ þ tð:05, nÞ ðS 2 ðlogY Þ=nÞ ,
3.2.1. Primary immune response The multiple comparison analysis of the primary immune response to SRBC (Fig. 1) showed a significant difference between groups [H(4, n = 69) = 19.6776, P < .001]. The pair comparison analysis of the SINT group showed that submissive mice (SINT-S) presented a primary immune response significantly lower than that found in the control group (Mann – Whitney, P < .01). A higher primary immune response was observed in the LINT group when compared to the CT group, and it was statistically significant in the response of the dominant mice (Mann –Whitney, P < .05). In addition, the statistical analysis showed significant differences between LINT and SINT protocols, specifically by comparison between LINT-S SINT-S and LINT-D SINT-D. The Mann – Whitney and Dunn tests showed, respectively, P < .01 and P < .05 when the submissive or dominant mice from different protocols (timings of stress exposure) were compared.
where the formulas above use the following: MA(logY) = arithmetic mean of the log of the data; (.05, n) = value of the Student’s t test for an a of .05 and n values; and S2(logY) = variance of the log of the data.
3. Results 3.1. Behavioural analysis Our results showed that the residents became dominants in 65% of the dyads subjected to agonistic interaction. Generally, the dominance behaviour is already established in the first social conflict. However, alternations in the dominance may occur up to the 5th session of agonistic interaction. The alternating behaviour of dominance diminished in each social conflict session from 37% (1st to 2nd session) to 12% (4th to 5th session). After the 5th session, the dyads showed stability, with the same dominant mice triggering 100% of the offensive attacks.
These studies were performed to determine the consequences of social stress on the antibody response to SRBC, considering the time course between the SRBC injection and the social stress.
3.2.2. Secondary antibody response There is a tendency of SINT-D mice to show a higher secondary immune response when compared to the SINT-S group. However, the multiple comparison analysis of the secondary immune response to SRBC (Fig. 2) showed no significant difference between groups.
Fig. 1. Geometric mean of anti-SRBC titres in the primary humoral immune response. Mann – Whitney (m) and Dunn tests (d) were used. * ( P < .05) or ** ( P < .01) indicate the statistical significance between experimental groups [SINT-S (n = 6), SINT-D (n = 9) and LINT-S (n = 13), LINT-D (n = 13)] CT group (n = 28). # ( P < .05) or ## ( P < .01) indicate the statistical significance between LINT-S SINT-S and LINT-D SINT-D groups.
324
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
Fig. 2. Geometric mean of the antibody titres in the secondary humoral immune response. A significant difference ( P < .05) between submissive and dominant mice of the same group is marked with @. The number of animals analysed is indicated in Fig. 1.
3.3. Exploratory activity in the elevated PM The purpose of this experiment was to analyse the anxiety levels in the animals subjected to the aversive PM stimulus in light of the individual animal’s characteristics preceding social stress. The animals that later became dominants during the agonistic interactions tended to spend less time in the open arms than SINT-S and CT mice (Table 1), although the differences among groups were not statistically significant.
correlation ( .54) proved to be significant ( P < .001), meaning that the more anxious controls (spending less time in the open arms of the PM) exhibited the higher anti-SRBC titres. In the SINT-S group, the positive correlation index (.97) also proved to be significant ( P < .001), meaning at this time that subordinates with lower anxiety levels yielded higher anti-SRBC titres. In the SINT-D group, the negative correlation index ( .69) was not significant.
4. Discussion 3.4. Correlation analysis between humoral immune response and anxiety levels The anxiety measured in the PM test showed to be correlated with the secondary humoral immune response in some groups (Table 1). In the CT group, the negative
Table 1 Correlation index (CI) between the humoral immune response (HIR) to SRBC and the %TO in the elevated PM test in mice submitted or not submitted to social interaction Group
n
%TO (mean ± S.E.M.)
HIR
CI
CT
28
22.6 ± 4.8
SINT-S
6
22.3 ± 7.4
SINT-D
9
9.9 ± 3.9
LINT-S
13
19.7 ± 5.8
LINT-D
13
13.5 ± 6.4
H1 H2 H1 H2 H1 H2 H1 H2 H1 H2
.33 .54** .87 .97** .17 .69 .17 .03 .34 .40
H1: primary immune response. H2: secondary immune response. ** P < .01 (Spearman test).
The present study shows that the most evident changes occurred in the SINT group, where a significant depression of the primary immune response was observed mainly in the subordinates when compared to the controls. Although the secondary immune response in the SINT group did not differ from the control group, the subordinate mice still presented a humoral immune response that was lower than dominants. By contrast, the primary immune response in the LINT group tended to show an increase that was not seen in the secondary immune response when compared to the control group. These data are in agreement with those observed earlier [18,19] in which the temporal relation between the immune challenge and the stress presentation are analysed. Exposure to an electric shock 72 h after SRBC inoculation induces a decrease in the anti-SRBC titres and in the plaque-forming cells (PFC) [18]. In contrast, SRBC injections immediately after shock or preceded by 13 daily short sessions induce an increase in the anti-SRBC titres [19]. Thus, all these data point out to the relevance of the temporal dynamic of social stress exposure in relation to the immune challenge induction. However, we cannot discard the effects caused during the process of hierarchical establishment in the dyads over time. As observed initially, the
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
animals were fighting to establish dominance, and at this time, it was even possible to find alternations of dominance in some dyads. Therefore, the instability generated a stress condition that affected both mice, which may explain the reason for immune depression that is more evident in the subordinates of the SINT mice. In the second week of interaction, a hierarchical stability was already established. At this moment, it is necessary to consider the different effects that the social interaction exerts in animals of different social status. Since the subordinates continued under an uncontrollable situation, the still strong activation of the HPA axis impaired the humoral immune response of these mice as a result of high adrenal corticoid levels [20]. On the other hand, the dominant position mitigated the effect of stress on the HPA axis over time. This adaptation over time may be the reason for the predominant increase of the anti-SRBC titres in the second week of interaction that is more evident in the dominant mice. Our results showed that the residents became dominants in 65% of the dyads subjected to agonistic interaction. Nevertheless, contradictory results were obtained using the same model of agonistic interaction in previous experiments, in which the intruders preponderated as dominants (data not shown). The attempts to link differences in performance in behavioural tests to social status or psychological profile have led to controversies. After an agonistic interaction, dominant mice showed higher levels of anxiety in the elevated PM when compared to submissive mice [21]. Conversely, no difference was found when dominants and subordinates were subjected to several behavioural tests, including the elevated PM test, which were applied before a period of interaction [22]. In our experiments, the PM test preceding the social interaction in mice, which were later classified as dominant or subordinate, showed no statistically significant difference. In spite of that, a consistent correlation analysis was found between anxiety and the humoral immune response. A negative correlation was found in control and SINT-D mice, suggesting that in a nonstress condition or under a controllable stress condition, the anxiety may exert a stimulatory effect on the humoral immune response. The submissive mice showed a positive correlation, meaning that in a defeated condition, the anxiety measured in the PM contribute to a humoral immune depression. On the other hand, when using the LINT protocol, the individual’s psychological profile showed a lower and not statistically significant correlation analysis of anxiety and humoral immune response. Altogether, these data allow us to conclude that the time of stress exposure and the psychological profile are important modulators of the humoral immune response. The effect of the time of stress exposure was evident after 1 week of social interaction, with a significant depression on the primary humoral immune response. It was also evident that in the second week of interaction, with the hierarchy well
325
established, the most evident effect of the social interaction was an increase of the immune response mainly in the dominants. We postulate that the strong and uncontrollable stress faced by the animals at the beginning of the interactions contribute to the failure of the immune response. On the other hand, after establishing dominance, mice with different hierarchical positions adapt differently over time. The controllable stress faced by the dominant mice favours the peripheral lymphoid organs to increase the anti-SRBC production, although in the submissive mice that are still under uncontrollable condition, the humoral immune response is impaired. These results suggest that the PM test may be useful as an instrument to predict the consequences of stress on immunological functions. Nevertheless, this method by itself is insufficient without considering the coping ability to the environmental demands and the temporal relation between stress and the immune challenge.
Acknowledgments The authors would like to thank Fundac˛a˜o Coordenac˛a˜o de Aperfeic˛oamento de Pessoal de Nı´vel Superior (CAPES) for the financial support and Dr. Marcus Vinicius C. Baldo for the statistical data analysis contribution.
References [1] Koolhaas JM, Korte SM, De Boer SF, Van Der Vergt BJ, Van Reenen CG, Hopster H, De Jong IC, Ruis MAW, Blokhuis HJ. Coping styles in animals: current status in behavior and stress-physiology. Neurosci Biobehav Rev 1999;23:925 – 35. [2] Stefanski V. Social stress in loser rats: opposite immunological effects in submissive and subdominant males. Physiol Behav 1998;63: 605 – 13. [3] Selye H. The stress of life. New York: McGraw-Hill, 1956. [4] Heinrichs C, Menzaghi F, Pich EM, Britton KT, Koob GF. The role of CRF in behavioral aspects of stress. Ann NY Acad Sci 1995;771: 92 – 104. [5] Leonard BE, Song C. Stress and the immune system in the etiology of anxiety and depression. Pharmacol Biochem Behav 1996;54(1): 299 – 303. [6] Song C, Earley B, Leonard BE. Behavioral, neurochemical, and immunological responses to CRF administration. Is CRF a mediator of stress? Ann NY Acad Sci 1995;771:55 – 72. [7] Henry JP, Liu YY, Nadra WE, Qian CG, Mormede P, Lemaire V, Ely D, Hendley ED. Psychosocial stress can induce chronic hypertension in normotensive strains of rats. Hypertension 1993;21:714 – 23. [8] Overmier JB, Murison R. Anxiety and helplessness in the face of stress predisposes, precipitates, and sustains gastric ulceration. Behav Brain Res 2000;110:161 – 74. [9] Ader R, Felten D, Cohen N. Interactions between the brain and the immune system. Annu Rev Pharmacol Toxicol 1990;30:561 – 602. [10] Anismam H, Zalcman S, Zacharco RM. The impact of stressors on immune and central neurotransmitter activity: bidirectional communication. Rev Neurosci 1993;4:147 – 80. [11] Dhabhar FS, Miller AH, McEwen BS, Spencer RL. Stress-induced changes in blood leukocyte distribution. Role of adrenal steroid hormones. J Immunol 1996;157(4):1638 – 44.
326
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
[12] Dantzer R. Stress and immunity: what have we learned from psychoneuroimmunology? Acta Physiol Scand 1997;161:43 – 6. [13] Hardy CA, Quay J, Livnat S, Ader R. Altered T-lymphocyte response following aggressive encounters in mice. Physiol Behav 1990;47: 1245 – 51. [14] Bohus B, Koolhaas JM, Heijnen CJ, Boer O. Immunological responses to social stress: dependence on social environment and coping abilities. Neuropsychobiology 1993;28:95 – 9. [15] Hudson L, Hay FC. Antibody interaction with antigen. Practical immunology. 4th ed. Oxford: Blackwell, 1991. pp. 252 – 4. [16] Lister RG. The use of a Plus-Maze to measure anxiety in the mouse. Psychopharmacology 1990;92:180 – 5. [17] Sokal RR, Rohlf J. Assumptions of analysis of variance. Biometry: the principles and practice of statistics in biological research. 3rd ed. New York: Freeman, 1995. pp. 392 – 450. [18] Zalcman S, Minkiewicz-Janda A, Richte M, Anisman H. Critical
[19]
[20]
[21]
[22]
periods associated with stressor effects on antibody titers and on the plaque forming cell response to sheep red blood cells. Brain Behav Immun 1988;2:254 – 66. Zalcman S, Anisman H. Acute and chronic stressor effects on the antibody response to sheep red blood cells. Pharmacol Biochem Behav 1993;4:445 – 52. Raab A, Dantzer R, Michaud B, Mormed P, Taghzouti K, Simon H, Le Moal M. Behavioral, physiological and immunological consequences of social status and aggression in chronically coexisting resident – intruder dyads of male rats. Physiol Behav 1986;36:223 – 8. Ferrari PF, Palanza P, Rodgers RJ. Interindividual variability in Swiss male mice: relationship between social factors, aggression, and anxiety. Physiol Behav 1998;63(5):821 – 7. Hilakivi-Clarke LA, Lister RG. Are there preexisting behavioral characteristics that predict the dominant status of male NIH Swiss mice (Mus musculus)? J Comp Psychol 1992;106:184 – 9.
The effect of different psychological profiles and timings of stress exposure on humoral immune response O.C. Gasparottoa,*, Z.M. Igna´cioa, K. Lina, S. Gonc˛alvesb a
Department of Physiological Sciences, Biological Sciences Centre, Federal University of Santa Catarina, Trindade, Floriano´polis, SC 88040-900, Brazil b Department of Microbiology and Parasitology, Biological Sciences Centre, Federal University of Santa Catarina, Trindade, Floriano´polis, SC 88040-900, Brazil Received 16 January 2001; received in revised form 22 November 2001; accepted 27 March 2002
Abstract The aim of this study was to analyse the effects of different timings of stress exposure on humoral immune response in mice previously distinguished by their own psychological profile. The Swiss mice were submitted to two different protocols of social confrontation, based on the timing of stress exposure in relation to an immune challenge: animals socially confronted daily for 2 weeks (SINT) and another group confronted daily for 3 weeks (LINT). Control groups were individually housed in a different room. All groups were intraperitoneally injected with sheep red blood cells (SRBC). The SINT group was challenged on the 1st and 7th days of confrontation, whereas the LINT group was challenged on the 7th and 14th days. Two days prior to the period of social conflict, the animals were tested in the elevated plus-maze (PM). The SINT protocol caused a more depressed primary immune response in the submissive mice than that observed in the dominants. The LINT protocol induced a marked increase in the primary immune response, which was more evident in the dominant mice, whilst no changes were observed in the secondary immune response. In the control and dominant groups, the correlation analysis attributed a higher anti-SRBC titre to the more anxious animals; by contrast, higher anti-SRBC titres were found in the less anxious submissive mice. These studies show that different physiological and behavioural adaptations to environmental demands over time, as well as different psychological profiles, constitute important factors to a better understanding of neuro-immune interactions. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Psychosocial stress; Humoral immune response; Mice; Social status; Anxiety; Psychological profile
1. Introduction For a long time, researchers have tried to understand the symptoms associated with diverse kinds of stressful stimuli. The reaction to stress is dependent not only on the nature and intensity of the stressors but also on the individual’s psychological profile and means of coping with such stimuli [1,2]. Independent of individual characteristics, the activation of the pituitary –adrenal axis and the sympatho-adrenal system are markers of the stress reaction [3]. Such activations lead to broad functional and behavioural changes, which can modulate anxiety [4], depression [5,6], blood pressure [7] and gastric ulceration [8]. It has also been shown that social stress has a powerful impact on the immune system [9– 14]. To understand the neurobiology of psychosocial stress and their effects on social creatures, it
is important to consider all the inherent variables involved. The psychological profile was considered as a determinant factor for the animal’s ability to deal with the environment. The dominant and submissive status in the social conflicts may be favoured by its own psychological profile. Considering that the individual differences demand different ways of coping with the stressful stimuli, it might be expected that it should differently affect neuroimmunological interactions. Based on these assumptions, the aim of this work was to consider the psychological profile of animals that were further subjected to different timings of social stress exposure in relation to an immune challenge.
2. Materials and methods 2.1. Mice
* Corresponding author. Tel.: +55-48-3319352; fax: +55-48-3319672. E-mail address: [email protected] (O.C. Gasparotto).
The animals were bred in the Animal Facilities of the Federal University of Santa Catarina and were maintained
0031-9384/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 1 - 9 3 8 4 ( 0 2 ) 0 0 7 1 8 - 7
322
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
according to the Animal Care Guidelines from the National Institute of Health of the United States of America. Three-month-old male SWISS mice (n = 78) were individually housed in clear polypropylene cages (19 30 13 cm) and were maintained for an adaptation period of 30 days in a temperature-controlled room (24 C) under standard lighting (lights on from 05:30 to 17:30 h) with free access to food and water. The animals were handled once a week at the time of cage cleaning. For technical reasons, the data from nine animals could not be included in the statistical analysis. 2.2. Social confrontation Male SWISS mice were randomly separated into two groups and were identified as intruders (I) or residents (R). The residents previously isolated in a propylene cages were transferred to a wooden cage (19 30 13 cm) with a transparent front wall 3 days prior to the beginning of social interactions. The residents were always confronted with the same intruders, which were maintained individually in the same room in propylene cages. The agonistic interaction was performed using animals from different litters. The stress procedures took place when the intruder was transferred into the resident’s cage. To distinguish the resident from intruder animals during the confrontation, the backs of the residents were lightly painted with gentian-violet 2 days before testing. The social interaction, conducted under red lighting, started at 17:30, and the behaviour of the animals was analysed for 30 min. The dominant mice were identified by lateral threat postures, as well as offensive attacks. The subordinate mice were identified as animals that vocalised and displayed defensive upright or upside down postures, as well as freezing and flight when attacked by the opponent. The dominance was only considered established when one of the animals triggered 100% of offensive attacks (dominants) towards their opponents (submissives). 2.3. Immunisation The immunisation procedure was performed using sheep red blood cells (SRBC) washed in a phosphate-buffered saline (PBS; 0.01 M, pH 7.2), and resuspended in a volume containing 109 cells/ml. All the animals were intraperitoneally (ip) injected with 108 SRBC/0.1 ml at the beginning of the dark phase (18:00– 18:30 h). 2.4. Serum samples and haemaglutination assay Mice were anaesthetised with ether and 150 ml of blood was collected by retro-orbital puncture. In all serum samples, the presence of anti-SRBC antibodies was evaluated using the technique described earlier [15]. Samples of 25 ml of serum were added to a U-shaped 96-well-microtiter plate, and a twofold serial dilution was prepared, using PBS as the diluent. The same volume of a 2% suspension of SRBC in PBS
was added to each well and was incubated for 2 –3 h at room temperature. The titre of the natural agglutinating activity of the serum sample was expressed as the reciprocal of the highest dilution showing a positive pattern of agglutination. 2.5. Timings of stress exposure and immune challenge 2.5.1. Short agonistic interaction (SINT) Mice (n = 15) were subjected to social confrontation for a period of 30 min/day for 2 weeks. This protocol was used in order to analyse the effects of short agonistic interaction on humoral immune response to SRBC, in which the animals were intraperitoneally inoculated with SRBC following the social interaction on the 1st and 7th day of experimentation. The blood samples were collected on the 7th day after each antigen inoculation, when the confrontation did not take place, in order to avoid stress overload. The sampling procedure was used in all groups. 2.5.2. Long agonistic interaction (LINT) Mice (n = 26) were subjected to social confrontation for a period of 30 min/day for 3 weeks. To analyse the effects of long agonistic interaction on humoral immune response to SRBC, the animals were intraperitoneally inoculated with SRBC on the 7th and 14th days of experimentation. The blood samples were obtained on the 14th and 21st days and as in the SINT protocol, the animals did not interact in these days. 2.5.3. Control group (CT) Mice were individually housed in polypropylene cages (n = 28), were kept in a separate room for 2 weeks and were identified as control. This group was inoculated with SRBC on the 1st and 7th days of experimentation. 2.6. Anxiety test To measure anxiety levels, each animal was individually tested in an elevated plus-maze (PM) adapted model [16]. The PM apparatus was made of opaque Plexiglas and consisted of two open arms (30 5 cm) and two enclosed arms (30 5 15 cm). The apparatus was mounted on a wooden base, which raised it to 40 cm above the floor. The test was performed in a room with diffuse red light 1 day before starting the agonistic interaction. A normal 5-min test duration was employed, with the maze cleaned between subjects. All the tests were recorded using a video camera. The anxiety levels were expressed as the percentage of time spent in open arms (%TO). The percentage of entries into the open arms of the maze is not shown since this gave similar results to %TO. 2.7. Data analysis Data were evaluated using nonparametric tests for independent multiple variables: Dunn test and Kruskal – Wallis test followed by a nonparametric test for two inde-
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
323
pendent variables (Mann – Whitney U test). The Spearman test was employed to correlate the immunological data with anxiety levels. The antibody titres P were expressed as geometric mean, GM = antilog[(1/n) (logY)], and the confidence interval (L1 and L2) was calculated as described bellow [17]: h pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffii L1 ¼ antilog MAðlogY Þ tð:05, nÞ ðS 2 ðlogY Þ=nÞ ,
3.2. Effects of social stress on humoral immune response to SRBC
h pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffii L2 ¼ antilog MAðlogY Þ þ tð:05, nÞ ðS 2 ðlogY Þ=nÞ ,
3.2.1. Primary immune response The multiple comparison analysis of the primary immune response to SRBC (Fig. 1) showed a significant difference between groups [H(4, n = 69) = 19.6776, P < .001]. The pair comparison analysis of the SINT group showed that submissive mice (SINT-S) presented a primary immune response significantly lower than that found in the control group (Mann – Whitney, P < .01). A higher primary immune response was observed in the LINT group when compared to the CT group, and it was statistically significant in the response of the dominant mice (Mann –Whitney, P < .05). In addition, the statistical analysis showed significant differences between LINT and SINT protocols, specifically by comparison between LINT-S SINT-S and LINT-D SINT-D. The Mann – Whitney and Dunn tests showed, respectively, P < .01 and P < .05 when the submissive or dominant mice from different protocols (timings of stress exposure) were compared.
where the formulas above use the following: MA(logY) = arithmetic mean of the log of the data; (.05, n) = value of the Student’s t test for an a of .05 and n values; and S2(logY) = variance of the log of the data.
3. Results 3.1. Behavioural analysis Our results showed that the residents became dominants in 65% of the dyads subjected to agonistic interaction. Generally, the dominance behaviour is already established in the first social conflict. However, alternations in the dominance may occur up to the 5th session of agonistic interaction. The alternating behaviour of dominance diminished in each social conflict session from 37% (1st to 2nd session) to 12% (4th to 5th session). After the 5th session, the dyads showed stability, with the same dominant mice triggering 100% of the offensive attacks.
These studies were performed to determine the consequences of social stress on the antibody response to SRBC, considering the time course between the SRBC injection and the social stress.
3.2.2. Secondary antibody response There is a tendency of SINT-D mice to show a higher secondary immune response when compared to the SINT-S group. However, the multiple comparison analysis of the secondary immune response to SRBC (Fig. 2) showed no significant difference between groups.
Fig. 1. Geometric mean of anti-SRBC titres in the primary humoral immune response. Mann – Whitney (m) and Dunn tests (d) were used. * ( P < .05) or ** ( P < .01) indicate the statistical significance between experimental groups [SINT-S (n = 6), SINT-D (n = 9) and LINT-S (n = 13), LINT-D (n = 13)] CT group (n = 28). # ( P < .05) or ## ( P < .01) indicate the statistical significance between LINT-S SINT-S and LINT-D SINT-D groups.
324
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
Fig. 2. Geometric mean of the antibody titres in the secondary humoral immune response. A significant difference ( P < .05) between submissive and dominant mice of the same group is marked with @. The number of animals analysed is indicated in Fig. 1.
3.3. Exploratory activity in the elevated PM The purpose of this experiment was to analyse the anxiety levels in the animals subjected to the aversive PM stimulus in light of the individual animal’s characteristics preceding social stress. The animals that later became dominants during the agonistic interactions tended to spend less time in the open arms than SINT-S and CT mice (Table 1), although the differences among groups were not statistically significant.
correlation ( .54) proved to be significant ( P < .001), meaning that the more anxious controls (spending less time in the open arms of the PM) exhibited the higher anti-SRBC titres. In the SINT-S group, the positive correlation index (.97) also proved to be significant ( P < .001), meaning at this time that subordinates with lower anxiety levels yielded higher anti-SRBC titres. In the SINT-D group, the negative correlation index ( .69) was not significant.
4. Discussion 3.4. Correlation analysis between humoral immune response and anxiety levels The anxiety measured in the PM test showed to be correlated with the secondary humoral immune response in some groups (Table 1). In the CT group, the negative
Table 1 Correlation index (CI) between the humoral immune response (HIR) to SRBC and the %TO in the elevated PM test in mice submitted or not submitted to social interaction Group
n
%TO (mean ± S.E.M.)
HIR
CI
CT
28
22.6 ± 4.8
SINT-S
6
22.3 ± 7.4
SINT-D
9
9.9 ± 3.9
LINT-S
13
19.7 ± 5.8
LINT-D
13
13.5 ± 6.4
H1 H2 H1 H2 H1 H2 H1 H2 H1 H2
.33 .54** .87 .97** .17 .69 .17 .03 .34 .40
H1: primary immune response. H2: secondary immune response. ** P < .01 (Spearman test).
The present study shows that the most evident changes occurred in the SINT group, where a significant depression of the primary immune response was observed mainly in the subordinates when compared to the controls. Although the secondary immune response in the SINT group did not differ from the control group, the subordinate mice still presented a humoral immune response that was lower than dominants. By contrast, the primary immune response in the LINT group tended to show an increase that was not seen in the secondary immune response when compared to the control group. These data are in agreement with those observed earlier [18,19] in which the temporal relation between the immune challenge and the stress presentation are analysed. Exposure to an electric shock 72 h after SRBC inoculation induces a decrease in the anti-SRBC titres and in the plaque-forming cells (PFC) [18]. In contrast, SRBC injections immediately after shock or preceded by 13 daily short sessions induce an increase in the anti-SRBC titres [19]. Thus, all these data point out to the relevance of the temporal dynamic of social stress exposure in relation to the immune challenge induction. However, we cannot discard the effects caused during the process of hierarchical establishment in the dyads over time. As observed initially, the
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
animals were fighting to establish dominance, and at this time, it was even possible to find alternations of dominance in some dyads. Therefore, the instability generated a stress condition that affected both mice, which may explain the reason for immune depression that is more evident in the subordinates of the SINT mice. In the second week of interaction, a hierarchical stability was already established. At this moment, it is necessary to consider the different effects that the social interaction exerts in animals of different social status. Since the subordinates continued under an uncontrollable situation, the still strong activation of the HPA axis impaired the humoral immune response of these mice as a result of high adrenal corticoid levels [20]. On the other hand, the dominant position mitigated the effect of stress on the HPA axis over time. This adaptation over time may be the reason for the predominant increase of the anti-SRBC titres in the second week of interaction that is more evident in the dominant mice. Our results showed that the residents became dominants in 65% of the dyads subjected to agonistic interaction. Nevertheless, contradictory results were obtained using the same model of agonistic interaction in previous experiments, in which the intruders preponderated as dominants (data not shown). The attempts to link differences in performance in behavioural tests to social status or psychological profile have led to controversies. After an agonistic interaction, dominant mice showed higher levels of anxiety in the elevated PM when compared to submissive mice [21]. Conversely, no difference was found when dominants and subordinates were subjected to several behavioural tests, including the elevated PM test, which were applied before a period of interaction [22]. In our experiments, the PM test preceding the social interaction in mice, which were later classified as dominant or subordinate, showed no statistically significant difference. In spite of that, a consistent correlation analysis was found between anxiety and the humoral immune response. A negative correlation was found in control and SINT-D mice, suggesting that in a nonstress condition or under a controllable stress condition, the anxiety may exert a stimulatory effect on the humoral immune response. The submissive mice showed a positive correlation, meaning that in a defeated condition, the anxiety measured in the PM contribute to a humoral immune depression. On the other hand, when using the LINT protocol, the individual’s psychological profile showed a lower and not statistically significant correlation analysis of anxiety and humoral immune response. Altogether, these data allow us to conclude that the time of stress exposure and the psychological profile are important modulators of the humoral immune response. The effect of the time of stress exposure was evident after 1 week of social interaction, with a significant depression on the primary humoral immune response. It was also evident that in the second week of interaction, with the hierarchy well
325
established, the most evident effect of the social interaction was an increase of the immune response mainly in the dominants. We postulate that the strong and uncontrollable stress faced by the animals at the beginning of the interactions contribute to the failure of the immune response. On the other hand, after establishing dominance, mice with different hierarchical positions adapt differently over time. The controllable stress faced by the dominant mice favours the peripheral lymphoid organs to increase the anti-SRBC production, although in the submissive mice that are still under uncontrollable condition, the humoral immune response is impaired. These results suggest that the PM test may be useful as an instrument to predict the consequences of stress on immunological functions. Nevertheless, this method by itself is insufficient without considering the coping ability to the environmental demands and the temporal relation between stress and the immune challenge.
Acknowledgments The authors would like to thank Fundac˛a˜o Coordenac˛a˜o de Aperfeic˛oamento de Pessoal de Nı´vel Superior (CAPES) for the financial support and Dr. Marcus Vinicius C. Baldo for the statistical data analysis contribution.
References [1] Koolhaas JM, Korte SM, De Boer SF, Van Der Vergt BJ, Van Reenen CG, Hopster H, De Jong IC, Ruis MAW, Blokhuis HJ. Coping styles in animals: current status in behavior and stress-physiology. Neurosci Biobehav Rev 1999;23:925 – 35. [2] Stefanski V. Social stress in loser rats: opposite immunological effects in submissive and subdominant males. Physiol Behav 1998;63: 605 – 13. [3] Selye H. The stress of life. New York: McGraw-Hill, 1956. [4] Heinrichs C, Menzaghi F, Pich EM, Britton KT, Koob GF. The role of CRF in behavioral aspects of stress. Ann NY Acad Sci 1995;771: 92 – 104. [5] Leonard BE, Song C. Stress and the immune system in the etiology of anxiety and depression. Pharmacol Biochem Behav 1996;54(1): 299 – 303. [6] Song C, Earley B, Leonard BE. Behavioral, neurochemical, and immunological responses to CRF administration. Is CRF a mediator of stress? Ann NY Acad Sci 1995;771:55 – 72. [7] Henry JP, Liu YY, Nadra WE, Qian CG, Mormede P, Lemaire V, Ely D, Hendley ED. Psychosocial stress can induce chronic hypertension in normotensive strains of rats. Hypertension 1993;21:714 – 23. [8] Overmier JB, Murison R. Anxiety and helplessness in the face of stress predisposes, precipitates, and sustains gastric ulceration. Behav Brain Res 2000;110:161 – 74. [9] Ader R, Felten D, Cohen N. Interactions between the brain and the immune system. Annu Rev Pharmacol Toxicol 1990;30:561 – 602. [10] Anismam H, Zalcman S, Zacharco RM. The impact of stressors on immune and central neurotransmitter activity: bidirectional communication. Rev Neurosci 1993;4:147 – 80. [11] Dhabhar FS, Miller AH, McEwen BS, Spencer RL. Stress-induced changes in blood leukocyte distribution. Role of adrenal steroid hormones. J Immunol 1996;157(4):1638 – 44.
326
O.C. Gasparotto et al. / Physiology & Behavior 76 (2002) 321–326
[12] Dantzer R. Stress and immunity: what have we learned from psychoneuroimmunology? Acta Physiol Scand 1997;161:43 – 6. [13] Hardy CA, Quay J, Livnat S, Ader R. Altered T-lymphocyte response following aggressive encounters in mice. Physiol Behav 1990;47: 1245 – 51. [14] Bohus B, Koolhaas JM, Heijnen CJ, Boer O. Immunological responses to social stress: dependence on social environment and coping abilities. Neuropsychobiology 1993;28:95 – 9. [15] Hudson L, Hay FC. Antibody interaction with antigen. Practical immunology. 4th ed. Oxford: Blackwell, 1991. pp. 252 – 4. [16] Lister RG. The use of a Plus-Maze to measure anxiety in the mouse. Psychopharmacology 1990;92:180 – 5. [17] Sokal RR, Rohlf J. Assumptions of analysis of variance. Biometry: the principles and practice of statistics in biological research. 3rd ed. New York: Freeman, 1995. pp. 392 – 450. [18] Zalcman S, Minkiewicz-Janda A, Richte M, Anisman H. Critical
[19]
[20]
[21]
[22]
periods associated with stressor effects on antibody titers and on the plaque forming cell response to sheep red blood cells. Brain Behav Immun 1988;2:254 – 66. Zalcman S, Anisman H. Acute and chronic stressor effects on the antibody response to sheep red blood cells. Pharmacol Biochem Behav 1993;4:445 – 52. Raab A, Dantzer R, Michaud B, Mormed P, Taghzouti K, Simon H, Le Moal M. Behavioral, physiological and immunological consequences of social status and aggression in chronically coexisting resident – intruder dyads of male rats. Physiol Behav 1986;36:223 – 8. Ferrari PF, Palanza P, Rodgers RJ. Interindividual variability in Swiss male mice: relationship between social factors, aggression, and anxiety. Physiol Behav 1998;63(5):821 – 7. Hilakivi-Clarke LA, Lister RG. Are there preexisting behavioral characteristics that predict the dominant status of male NIH Swiss mice (Mus musculus)? J Comp Psychol 1992;106:184 – 9.