- Email: [email protected]
Social buffering enhances extinction of conditioned fear responses in male rats
Physiology & Behavior 163 (2016) 123–128
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/phb
Social buffering enhances extinction of conditioned fear responses in male rats Kaori Mikami, Yasushi Kiyokawa ⁎, Yukari Takeuchi, Yuji Mori Laboratory of Veterinary Ethology, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
H I G H L I G H T S • • • •
Insufficient extinction training did not extinguish fear responses. Social buffering during insufficient extinction training suppressed fear responses. The effect of social buffering during extinction training was context specific. We concluded that social buffering enhanced extinction of fear responses.
a r t i c l e
i n f o
Article history: Received 18 March 2016 Received in revised form 3 May 2016 Accepted 4 May 2016 Available online 5 May 2016 Keywords: Social buffering Fear extinction Freezing Stress Amygdala
a b s t r a c t In social species, the phenomenon in which the presence of conspecific animals mitigates stress responses is called social buffering. We previously reported that social buffering in male rats ameliorated behavioral fear responses, as well as hypothalamic-pituitary-adrenal axis activation, elicited by an auditory conditioned stimulus (CS). However, after social buffering, it is not clear whether rats exhibit fear responses when they are reexposed to the same CS in the absence of another rat. In the present study, we addressed this issue using an experimental model of extinction. High stress levels during extinction training impaired extinction, suggesting that extinction is enhanced when stress levels during extinction training are low. Therefore, we hypothesized that rats that had received social buffering during extinction training would not show fear responses to a CS, even in the absence of another rat, because social buffering had enhanced the extinction of conditioned fear responses. To test this, we subjected male fear-conditioned rats to extinction training either alone or with a non-conditioned male rat. The subjects were then individually re-exposed to the CS in a recall test. When the subjects individually underwent extinction training, no responses were suppressed in the recall test. Conversely, when the subjects received social buffering during extinction training, freezing and Fos expression in the paraventricular nucleus of the hypothalamus and lateral amygdala were suppressed. Additionally, the effects of social buffering were absent when the recall test was conducted in a different context from the extinction training. The present results suggest that social buffering enhances extinction of conditioned fear responses. © 2016 Elsevier Inc. All rights reserved.
1. Introduction In social species, stress responses induced by exposure to distressing stimuli can be ameliorated when an animal is exposed along with a conspecific animal(s). For example, the presence of a conspecific animal has been found to suppress corticosterone or cortisol release in response to Abbreviations: BA, basal amygdala; CeL, lateral division of the central amygdala; CeM, medial division of the central amygdala; CS, conditioned stimulus; HPA, hypothalamicpituitary-adrenal; IL, infralimbic region of the prefrontal cortex; LA, lateral amygdala; pmOP, posteromedial olfactory peduncle; PVN, paraventricular nucleus of the hypothalamus. ⁎ Corresponding author. E-mail addresses: [email protected] (K. Mikami), [email protected] (Y. Kiyokawa), [email protected] (Y. Takeuchi), [email protected] (Y. Mori).
http://dx.doi.org/10.1016/j.physbeh.2016.05.001 0031-9384/© 2016 Elsevier Inc. All rights reserved.
a novel environment [1,2] or to predator-associated stimuli [3]. These phenomena are called social buffering [4]. In previous studies, we analyzed social buffering in male rats using fear conditioning. When fear-conditioned rats were exposed to an auditory conditioned stimulus (CS), they exhibited both behavioral fear responses and hypothalamic-pituitary-adrenal (HPA) axis activation. The presence of a non-conditioned unfamiliar male rat (associate) completely blocked these responses [5], suggesting that the fearconditioned rat received social buffering from the associate rat. This social buffering of conditioned fear responses occurred even if the dyad was separated by a wire-mesh partition or by double wire-mesh partitions separated by 5 cm [6]. We identified several additional characteristics of social buffering. For example, we found that the presence of guinea pigs [6] or some strains of rats [7] did not induce social buffering, that familiar associates were more effective for social buffering than
124
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
unfamiliar associates [8], that social buffering was similarly observed in female rats [9], and that associate-derived volatile olfactory signals can mediate social buffering [6,8,10]. In our investigations of the neural mechanisms underlying social buffering, we found that activation in the lateral amygdala (LA) in response to the CS was suppressed during social buffering [11]. We also found evidence that the posteromedial olfactory peduncle (pmOP) [12], most likely the posterior complex of the anterior olfactory nucleus within the pmOP [10], relays the olfactory signals responsible for social buffering from the main olfactory bulb to the amygdala. However, in all of our previous studies, rats were not reexposed to the CS after the test. Therefore, whether rats that have experienced social buffering show fear responses when they are re-exposed to the CS in the absence of another rat remains unclear. Extinction of conditioned fear responses appears to be a suitable experimental model for addressing this issue. In this model, fearconditioned animals undergo extinction training 1 day after conditioning, in which they are repeatedly exposed to the CS. If the extinction training is sufficient, the animals will exhibit suppressed fear responses when they are re-exposed to the CS in a recall test on the following day, i.e., extinction of conditioned fear responses [13,14]. Thus, one possibility for assessing the effects of social buffering on subsequent fear responses to the CS would be to subject fear-conditioned rats to social buffering during extinction training and then conduct a recall test in the absence of another rat. The stress levels of an animal during extinction training appear to affect the extinction of conditioned responses. For example, fearconditioned rats showed little or no extinction of conditioned fear responses when they were stressed via foot shocks [15] or placement on an elevated platform immediately prior to extinction training [16]. These findings suggest that extinction will be enhanced when animals undergoing extinction training have low levels of stress. Based on our findings that social buffering ameliorated stress caused by an auditory CS, we hypothesized that rats that experienced social buffering during extinction training would not show fear responses to the CS in a recall test. Indeed, we would expect that social buffering would lower the stress status during extinction training and thus enhance extinction of conditioned fear responses. To test this hypothesis, we subjected a group of rats to fear conditioning using an auditory CS. The rats underwent extinction training either alone or with an associate. The rats were then individually reexposed to the CS in a recall test. The effects of social buffering during extinction training were evaluated by examining freezing behavior and HPA axis activity, as reflected by Fos expression in the paraventricular nucleus of the hypothalamus (PVN), during the recall test. Because the amygdala plays an important role in conditioned fear responses [17,18], we examined Fos expression in the amygdala to assess the underlying neural mechanisms (Experiment 1). Next, we assessed whether the effects of social buffering during extinction training could be observed even if the recall test was conducted in a different context than that of extinction training (Experiment 2).
associates (rat placed with the subject during extinction training), which ensured unfamiliarity between the subjects and associates. All rats were individually housed and handled for 5 min per day for 3 days before the conditioning day. All behavioral procedures were performed between 09:00 and 16:00.
2.2. Procedure
2.1. Animals
2.2.1. Experiment 1 Fear conditioning was performed in an acrylic conditioning box with a metal grid floor (28 × 20 × 27 cm) under a white light (Context A). The rats received seven repetitions of a 3-s auditory CS (8 kHz, 70 dB) that terminated concurrently with a 0.5-s foot shock (0.55 mA). We presented the CS twice before and twice after the conditioning procedure to measure pre- and post-conditioning freezing. The inter-trial interval was randomly varied between 90 and 220 s. The subjects underwent extinction training either with (social situation) or without an associate (alone situation) 24 h after fear conditioning. Based on the post-conditioning freezing, the subjects were divided into extinction and no-extinction groups, which showed comparable freezing in each situation. The rats were placed in an acrylic extinction box with clean woodchip bedding (28 × 44 × 20.5 cm) under a dim red light (Context B). In the alone situation, the subjects in the extinction group (n = 9) received 24 CS presentations in the absence of other animals. In the social situation, the subjects in the extinction group (n = 9) underwent extinction training with an associate. We varied the inter-trial interval randomly between 60 and 120 s. For the rats in the no-extinction group (alone: n = 9, social: n = 9), we placed the subject and associate (in the social situation) in the extinction box for the same length of time as for the extinction group, without any CS presentation. Twenty-four hours after extinction training, we conducted the recall test with the individual subjects in both situations in the same context as for the extinction training (Context B). The subjects were exposed to the CS twice with an interval of 90 s. After the test, the subjects were returned to their home cages. Sixty minutes after the recall test [19], each subject was deeply anesthetized with sodium pentobarbital and intracardially perfused with saline, followed by 4% paraformaldehyde in 0.1 M phosphate buffer. The brain was removed and immersed overnight in 4% paraformaldehyde in 0.1 M phosphate buffer, and then placed in 30% sucrose/phosphate buffer for cryoprotection. We used the avidin-biotin-peroxidase immunohistochemistry method to detect Fos expression, as previously described [8,20]. Briefly, we collected six successive 30-μm sections containing the PVN (Bregma −1.80 mm) or LA, basal amygdala (BA), and lateral (CeL) and medial (CeM) division of the central amygdala (Bregma −2.76 mm). The sections were incubated with a primary antibody to c-Fos protein (1:8000; ABE457, Merck Millipore, Billerica, MA, USA) for 65 h and a biotinylated anti-rabbit secondary antibody (BA1000, Vector Laboratories, Burlingame, CA, USA) for 2 h. The sections were then processed using the ABC kit (Vector Laboratories). Staining was developed by incubating the tissue in a diaminobenzidine solution with nickel intensification.
All experiments were approved by the Animal Care and Use Committee of the Faculty of Agriculture at The University of Tokyo, according to guidelines that were adapted from the Consensus Recommendations on Effective Institutional Animal Care and Use Committees by the Scientists Center for Animal Welfare. Experimentally naïve male Wister rats were purchased at 8 weeks of age from Charles River Laboratories Japan (Kanagawa, Japan). The rats were housed 2–4 per cage in a room with an ambient temperature of 24 ± 1 °C and a humidity of 45 ± 5%. The room had a 12-h light/12-h dark cycle (lights switched on at 08:00). Food and water were available ad libitum. In each cage, rats were assigned to be either subjects or
2.2.2. Experiment 2 Fear conditioning (Context A) and extinction training (Context B) were performed as described above with the exception that all subjects underwent extinction training in the social situation. The subjects then individually underwent the recall test in the same (Context B; same situation, extinction group: n = 8, no-extinction group: n = 8) or different context (Context C; novel situation, extinction group: n = 5, no-extinction group: n = 6) from that of extinction training. In context C, the recall test was conducted in a cylindrical acrylic box (28 × 28 × 25) with paper bedding under a white light.
2. Material and methods
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
2.3. Data analyses Data are expressed as mean ± standard error of the mean and significance was set at P b 0.05. An experimenter who was blinded to the experimental groups evaluated the behavioral responses. We measured the duration of freezing and the frequency of walking using Microsoft Excel-based Visual Basic software that recorded the duration and number of key presses on a computer keyboard. Freezing was defined as an immobile posture (except for movements associated with respiration) and was measured during the 20-s period after the onset of each CS [21]. We calculated the percentage of time spent in a freezing posture with respect to the 20-s period and calculated the average of every 2 CS. During extinction training and the recall test, we measured baseline freezing that occurred during the 20-s period 90 s before the onset of the first CS. Baseline freezing was also expressed as a percent with respect to the 20-s period. During the recall test, the walking that occurred during the 5-m period before the onset of the first CS served as baseline activity. Walking was defined as the number of steps taken with the hind paws. Freezing during fear conditioning and the recall test was analyzed via two-way repeated ANOVA. Freezing during extinction training was analyzed via one-way repeated ANOVA. Additional comparisons of the baseline activity between the situations were conducted via Student's t-test. For immunohistochemical analyses, sections of the PVN, LA, BA, CeL, and CeM were captured using a microscope equipped with a digital camera (DP30BW, Olympus, Tokyo, Japan). An experimenter who was blind to the experimental groups counted the number of Fosimmunoreactive cells within a 0.5-mm square. Quantification was conducted bilaterally in six sections using Scion image Beta 4.0.2 software. When the designated region was smaller than the boundaries of the 0.5mm square, only the cells in the region of interest were counted. Although some sections were lost because of technical problems, we were able to analyze between six and nine subjects in all groups. The Fos expression in each situation was analyzed via Student's t-test. 3. Results 3.1. Experiment 1 For rats in the alone situation (Fig. 1A), we confirmed that an establishment of fear conditioning was not affected by future extinction training (group × conditioning, F(1, 16) = 0.07, P = 0.79) (Fig. 1B). The percent of freezing was affected by fear conditioning (F(1, 16) = 57.3, P b 0.0001), but not by group (F(1, 16) = 0.07, P = 0.79). In extinction training, we confirmed that the percent of baseline freezing was low (2.0 ± 2.0). Additionally, the percent of freezing decreased when the number of presented CSs increased (F(11, 88) = 5.32, P b 0.0001) (Fig. 1B). In the recall test, extinction training did not extinguish fear responses (group × test, F(1, 16) = 0.95, P = 0.34). The percent of freezing was affected by the test (F(1, 16) = 39.4, P b 0.0001), but not by group (F(1, 16) = 0.06, P = 0.82) (Fig. 1B). After the recall test, we observed Fos expression in various brain regions, including the PVN and LA (Fig. 1C). Fos expression was not different between the extinction and no-extinction group in the PVN (t12 = − 0.28, P = 0.78), LA (t13 = 0.38, P = 0.71), BA (t13 = − 0.93, P = 0.37), CeL (t13 = 0.18, P = 0.86), and CeM (t13 = −0.28, P = 0.78) (Fig. 1D). In the social situation (Fig. 1E), we confirmed that an establishment fear conditioning was not affected by future extinction training (group × conditioning, F(1, 16) = 0.003, P = 0.96) (Fig. 1F). The percent of freezing was affected by fear conditioning (F(1, 16) = 40.6, P b 0.0001), but not by group (F(1, 16) = 0.24, P = 0.64). In extinction training, we confirmed that the percent of baseline freezing was low (0 ± 0) and was not affected by an increase in the number of presented CSs (F(11, 88) = 1.14, P = 0.34) (Fig. 1F). In the recall test, extinction training extinguished fear responses (group × test, F(1, 16) = 18.9, P = 0.0005).
125
The percent of freezing was affected by the test (F(1, 16) = 36.5, P b 0.0001) and by group (F(1, 16) = 15.1, P = 0.0013) (Fig. 1F). Additional analyses comparing the situations confirmed that the extinction group showed equivalent baseline activity between the situations (alone situation: 61.9 ± 10.2, social situation: 81.9 ± 8.3, t16 = 1.52, P = 0.15). After the recall test, we observed Fos expression in various brain regions, including the PVN and LA (Fig. 1G). The rats in the extinction group exhibited decreased Fos expression in the PVN (t16 = −2.46, P = 0.0258) and LA (t13 = −3.73, P = 0.0025) compared with those in the no-extinction group (Fig. 1H). Fos expression in the BA (t13 = − 0.51, P = 0.617), CeL (t13 = − 0.60, P = 0.56), and CeM (t13 = −1.34, P = 0.20) did not differ between the extinction and noextinction group (Fig. 1H). 3.2. Experiment 2 In the same situation (Fig. 2A), we confirmed that an establishment of fear conditioning was not affected by future extinction training (group × conditioning, F(1, 14) = 0.037, P = 0.85) (Fig. 2B). The percent of freezing was affected by fear conditioning (F(1, 14) = 39.3, P b 0.0001), but not by group (F(1, 14) = 0.002, P = 0.97). In extinction training, we confirmed that the percent of baseline freezing was low (0 ± 0) and was not affected by an increase in the number of presented CSs (F(11, 77) = 1.27, P = 0.26) (Fig. 2B). In the recall test, extinction training extinguished fear responses (group × test, F(1, 14) = 6.36, P = 0.024). The percent of freezing was affected by the test (F(1, 14) = 53.1, P b 0.0001) and tended to be affected by group (F(1, 14) = 4.58, P = 0.0505) (Fig. 2B). In the novel situation (Fig. 2C), we confirmed that an establishment of fear conditioning was not affected by future extinction training (group × conditioning, F(1, 9) = 0.19, P = 0.67) (Fig. 2D). The percent of freezing was affected by fear conditioning (F(1, 9) = 23.5, P = 0.0009), but not by group (F(1, 9) = 0.19, P = 0.67). In extinction training, we confirmed that the percent of baseline freezing was low (0 ± 0). Additionally, the percent of freezing was not affected by an increase in the number of presented CSs (F(11, 44) = 0.58, P = 0.84) (Fig. 2D). In the recall test, extinction training did not extinguish fear responses (group × test, F(1, 9) = 0.48, P = 0.50). The percent of freezing was affected by the test (F(1, 16) = 105, P b 0.0001), but not by group (F(1, 9) = 0.48, P = 0.50) (Fig. 2D). 4. Discussion In the present study, we observed no extinction of conditioned fear responses in the alone situation (Experiment 1). This suggests that extinction training was insufficient to extinguish fear responses in the recall test. However, in the social situation, we observed suppressed freezing in the recall test and decreased Fos expression in the PVN and LA. Additionally, the effects of social buffering were absent in the novel situation (Experiment 2). Based on these findings, we suggest that social buffering enhanced the extinction of conditioned fear responses in our subjects. In the present Fos analyses, the absolute values of expression seem not to reflect the intensity of fear responses. For example, rats in the no-extinction group in the social situation showed higher expression compared with both groups in the alone situation, although all three groups showed fear responses. Similarly, the extinction group in the social situation showed Fos expression that was comparable to that of both groups in the alone situation, even if the rats in the extinction group did not show fear responses. It is possible that social interaction during extinction training affected basal Fos expression in the recall test. In addition, we assessed each situation successively. Therefore, it seems more appropriate to interpret the data by comparing the noextinction and extinction groups within the situations, rather than between the situations. As a result, we believe that the present results
126
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
Fig. 1. Extinction of fear responses elicited by an auditory conditioned stimulus (CS) in the alone and social situation. (A) Schematic of the procedure in the alone situation. S, subject. (B) The percent of freezing elicited by the CS before (pre) and after (post) fear conditioning, during extinction training, and during the recall test. The responses are expressed as the average of every 2 CS [mean ± standard error of the mean (SEM)]. (C) Representative photomicrographs of the Fos-immunoreactive cells in the paraventricular nucleus (PVN) and lateral amygdala (LA). (D) The number of Fos-immunoreactive cells (mean + SEM) in the PVN, LA, basal amygdala (BA), and lateral (CeL) and medial (CeM) division of the central amygdala of the subject after the recall test. (E) Schematic of the procedure in the social situation. S, subject; A, associate. (F) The percent of freezing elicited by the CS pre and post fear conditioning during extinction training and the recall test. The responses are expressed as the average of every 2 CS. *P b 0.05 compared to the no-extinction group in the social situation, as assessed by a two-way repeated ANOVA. (G) Representative photomicrographs of the Fos-immunoreactive cells in the PVN and LA. (H) The number of Fosimmunoreactive cells in the PVN, LA, BA, CeL, and CeM of the subject after the recall test. *P b 0.05 compared to the no-extinction group in the social situation, as assessed by a Student's t-test.
suggest the suppression of the PVN and LA in the extinction group in the social situation. In Experiment 1, we found that rats in the social situation exhibited reduced fear responses to a CS, as well as suppressed HPA axis and LA activation, in the recall test. We interpret these results to mean that social buffering enhances extinction of conditioned fear responses. Although we cannot deny the possibility that fear responses to the 3-s
CS were resistant to extinction, this seems unlikely because fear responses to a similarly short 2-s CS were successfully extinguished by extinction training [15]. Additionally, this interpretation is consistent with the established characteristics of extinction. For example, extinction training induces extinction of conditioned fear responses only in the context in which the subjects underwent training [22,23]. In addition, the extinction of conditioned fear responses has been found to be
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
127
Fig. 2. Extinction of fear responses elicited by the auditory conditioned stimulus (CS) in the same and novel situation. (A) Schematic of the procedure in the same situation. S, subject; A, associate. (B) The percent of freezing elicited by the CS before (pre) and after (post) fear conditioning, during extinction training, and during the recall test. The responses are expressed as the average of every 2 CS [mean ± standard error of the mean (SEM)]. *P b 0.05 compared to the no-extinction group in the same situation as assessed by a two-way repeated ANOVA. (C) Schematic of the procedure in the novel situation. S, subject; A, associate. (D) The percent of freezing elicited by the CS pre and post fear conditioning, during extinction training, and during the recall test. The responses are expressed as the average of every 2 CS (mean ± SEM).
accompanied by inhibition of the LA during the recall test [24,25]. Therefore, although the extinction training in the present study was insufficient, social buffering during training enhanced the effects of this insufficient extinction training, which induced extinction of conditioned fear responses in the recall test. As a result, the effects of social buffering were observed in a specific context and were accompanied by inhibition of the LA during the recall test. It is worthwhile to consider alternative explanations. First, it is possible that social buffering erased the fear memory of the rats in the extinction group in the social and same situations [26–29]. However, this explanation seems unlikely because the rats in the extinction group in the novel situation exhibited robust fear responses (Experiment 2). Second, it is possible that social buffering during extinction training established an association between the extinction box and stress amelioration, as observed during safety conditioning [30,31], which then suppressed fear responses in the recall test. However, this explanation is unlikely because the rats in the no-extinction group in the social and same situations exhibited robust fear responses, even when an associate accompanied them in the extinction box. Third, it is possible that the presence of an associate during extinction training established an association between the extinction box and hyperactivity. However, this explanation appears to be inappropriate, because the baseline activity in the extinction group was equivalent between the alone and social situations. In the present study, social buffering took place during extinction training while we observed Fos expression after the recall test. Therefore, the present Fos data do not provide information about how social buffering enhanced acquisition of extinction. The infralimbic region of the prefrontal cortex (IL) [32,33], BA [33,34], and ventral hippocampus [33] have been suggested to participate in the acquisition of extinction. As Fos expression in the IL is reduced after extinction training in rats with high stress responses [35] and impairment of extinction [36], it is possible that a low stress status during extinction training as a result of social buffering can activate the IL and enhance extinction. Similarly, it is possible that the associate-derived olfactory signals that are responsible for social buffering activate the BA and enhance extinction when the detected signals are transmitted from the pmOP to the BA via direct projections [12]. However, in our previous studies, social buffering was not necessarily accompanied by increased Fos expression in the IL [5,6] and BA [5,8,10], suggesting that these regions may not be involved. Therefore, further research is necessary to clarify the neural mechanisms underlying the enhancement of extinction by social buffering. In extinction training in this study, the presence of an associate induced social buffering of conditioned fear responses, as seen in our
previous research. However, in some studies, the presence of a conspecific did not affect fear responses to the auditory CS [37–39]. One reason for this discrepancy might be differences between species. Although social buffering can be induced by a mother, mate, or a conspecific with which the subject does not have a sexual relationship [4], this last type of social buffering has been scarcely observed in mice. Indeed, in mice, the presence of another mouse did not reduce freezing in response to an auditory CS [39]. Another possible reason might be differences in CS duration. Indeed, the presence of a conspecific rat did not appear to induce social buffering of conditioned fear responses to a 20-s CS [37,38]. The characteristics of fear responses elicited by a 3-s CS are different than those elicited by a 20-s CS [21]. Therefore, it is possible that the presence of another rat reduced freezing elicited by a 3-s CS, as tested in the present study, while a 20-s CS, tested in previous studies, had no effect. Because this study is the first to assess the effects of social buffering on the extinction of conditioned fear responses, our findings provoke many interesting questions. Given that co-housing with an associate additively induced another type of social buffering [5,40–42], it is possible that the extinction of conditioned fear responses is further enhanced when the subject is housed with an associate between the conditioning, extinction training, and recall test periods. Additionally, extinction of conditioned fear responses might be more strongly enhanced by familiar associates, because familiar associates have been found to be more effective than unfamiliar associate for social buffering of conditioned fear responses [8]. Further research is needed to address these questions. 5. Conclusions The results of the present study demonstrate that insufficient extinction training extinguished fear responses in a recall test when subjects received social buffering during extinction training. This phenomenon was context specific. These results suggest that social buffering enhances extinction of conditioned fear responses. Further analyses are necessary to fully understand the neural mechanisms underlying social buffering, and may contribute to the development of treatments for anxiety and fear responses. Acknowledgments This study was supported by the Japan Society for the Promotion of Science KAKENHI Grant Numbers 15H05635 and 15H05782.
128
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
References [1] M.B. Hennessy, D.S. Maken, F.C. Graves, Consequences of the presence of the mother or unfamiliar adult female on cortisol, ACTH, testosterone and behavioral responses of periadolescent guinea pigs during exposure to novelty, Psychoneuroendocrinology 25 (2000) 619–632. [2] M.L. Terranova, F. Cirulli, G. Laviola, Behavioral and hormonal effects of partner familiarity in periadolescent rat pairs upon novelty exposure, Psychoneuroendocrinology 24 (1999) 639–656. [3] J.L. Vogt, C.L. Coe, S. Levine, Behavioral and adrenocorticoid responsiveness of squirrel monkeys to a live snake: is flight necessarily stressful? Behav. Neural Biol. 32 (1981) 391–405. [4] Y. Kiyokawa, Social odors: alarm pheromones and social buffering, Curr. Top. Behav. Neurosci. (2015). [5] Y. Kiyokawa, Y. Takeuchi, Y. Mori, Two types of social buffering differentially mitigate conditioned fear responses, Eur. J. Neurosci. 26 (2007) 3606–3613. [6] Y. Kiyokawa, Y. Takeuchi, M. Nishihara, Y. Mori, Main olfactory system mediates social buffering of conditioned fear responses in male rats, Eur. J. Neurosci. 29 (2009) 777–785. [7] Nakamura, K., Ishii, A., Kiyokawa, Y., Takeuchi, Y., Mori, Y. The strain of an accompanying conspecific affects the efficacy of social buffering in male rats. Horm. Behav. (in press). [8] Y. Kiyokawa, A. Honda, Y. Takeuchi, Y. Mori, A familiar conspecific is more effective than an unfamiliar conspecific for social buffering of conditioned fear responses in male rats, Behav. Brain Res. 267 (2014) 189–193. [9] A. Ishii, Y. Kiyokawa, Y. Takeuchi, Y. Mori, Social buffering ameliorates conditioned fear responses in female rats, Horm. Behav. 81 (2016) 53–58. [10] Y. Takahashi, Y. Kiyokawa, Y. Kodama, S. Arata, Y. Takeuchi, Y. Mori, Olfactory signals mediate social buffering of conditioned fear responses in male rats, Behav. Brain Res. 240 (2013) 46–51. [11] F. Fuzzo, J. Matsumoto, Y. Kiyokawa, Y. Takeuchi, T. Ono, H. Nishijo, Social buffering suppresses fear-associated activation of the lateral amygdala in male rats: behavioral and neurophysiological evidence, Front. Neurosci. 9 (2015) 99. [12] Y. Kiyokawa, Y. Wakabayashi, Y. Takeuchi, Y. Mori, The neural pathway underlying social buffering of conditioned fear responses in male rats, Eur. J. Neurosci. 36 (2012) 3429–3437. [13] K.M. Myers, M. Davis, Behavioral and neural analysis of extinction, Neuron 36 (2002) 567–584. [14] G.J. Quirk, D. Mueller, Neural mechanisms of extinction learning and retrieval, Neuropsychopharmacology 33 (2008) 56–72. [15] S. Maren, C.H. Chang, Recent fear is resistant to extinction, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 18020–18025. [16] I. Akirav, M. Maroun, The role of the medial prefrontal cortex-amygdala circuit in stress effects on the extinction of fear, Neural Plast. 2007 (2007) 30873. [17] S. Maren, G.J. Quirk, Neuronal signalling of fear memory, Nat. Rev. Neurosci. 5 (2004) 844–852. [18] D. Pare, G.J. Quirk, J.E. Ledoux, New vistas on amygdala networks in conditioned fear, J. Neurophysiol. 92 (2004) 1–9. [19] K.J. Kovacs, c-Fos as a transcription factor: a stressful (re)view from a functional map, Neurochem. Int. 33 (1998) 287–297. [20] T. Kobayashi, Y. Kiyokawa, Y. Takeuchi, Y. Mori, Neural correlates underlying naloxone-induced amelioration of sexual behavior deterioration due to an alarm pheromone, Front. Neurosci. 9 (2015) 52. [21] Y. Kiyokawa, K. Mikami, Y. Mikamura, A. Ishii, Y. Takeuchi, Y. Mori, The 3-second auditory conditioned stimulus is a more effective stressor than the 20-second auditory conditioned stimulus in male rats, Neuroscience 299 (2015) 79–87.
[22] M.E. Bouton, Context and behavioral processes in extinction, Learn. Mem. 11 (2004) 485–494. [23] S. Maren, K.L. Phan, I. Liberzon, The contextual brain: implications for fear conditioning, extinction and psychopathology, Nat. Rev. Neurosci. 14 (2013) 417–428. [24] J.A. Hobin, K.A. Goosens, S. Maren, Context-dependent neuronal activity in the lateral amygdala represents fear memories after extinction, J. Neurosci. 23 (2003) 8410–8416. [25] E. Knapska, S. Maren, Reciprocal patterns of c-Fos expression in the medial prefrontal cortex and amygdala after extinction and renewal of conditioned fear, Learn. Mem. 16 (2009) 486–493. [26] M.H. Monfils, K.K. Cowansage, E. Klann, J.E. LeDoux, Extinction-reconsolidation boundaries: key to persistent attenuation of fear memories, Science 324 (2009) 951–955. [27] K.M. Myers, K.J. Ressler, M. Davis, Different mechanisms of fear extinction dependent on length of time since fear acquisition, Learn. Mem. 13 (2006) 216–223. [28] K. Nader, G.E. Schafe, J.E. Le Doux, Fear memories require protein synthesis in the amygdala for reconsolidation after retrieval, Nature 406 (2000) 722–726. [29] G.J. Quirk, D. Pare, R. Richardson, C. Herry, M.H. Monfils, D. Schiller, et al., Erasing fear memories with extinction training, J. Neurosci. 30 (2010) 14993–14997. [30] M. Mohammadi, J.R. Bergado-Acosta, M. Fendt, Relief learning is distinguished from safety learning by the requirement of the nucleus accumbens, Behav. Brain Res. 272 (2014) 40–45. [31] M.T. Rogan, K.S. Leon, D.L. Perez, E.R. Kandel, Distinct neural signatures for safety and danger in the amygdala and striatum of the mouse, Neuron 46 (2005) 309–320. [32] F.H. Do-Monte, G. Manzano-Nieves, K. Quinones-Laracuente, L. Ramos-Medina, G.J. Quirk, Revisiting the role of infralimbic cortex in fear extinction with optogenetics, J. Neurosci. 35 (2015) 3607–3615. [33] D. Sierra-Mercado, N. Padilla-Coreano, G.J. Quirk, Dissociable roles of prelimbic and infralimbic cortices, ventral hippocampus, and basolateral amygdala in the expression and extinction of conditioned fear, Neuropsychopharmacology 36 (2011) 529–538. [34] T. Amano, S. Duvarci, D. Popa, D. Pare, The fear circuit revisited: contributions of the basal amygdala nuclei to conditioned fear, J. Neurosci. 31 (2011) 15481–15489. [35] R. Landgraf, A. Wigger, F. Holsboer, I.D. Neumann, Hyper-reactive hypothalamopituitary-adrenocortical axis in rats bred for high anxiety-related behaviour, J. Neuroendocrinol. 11 (1999) 405–407. [36] P. Muigg, A. Hetzenauer, G. Hauer, M. Hauschild, S. Gaburro, E. Frank, et al., Impaired extinction of learned fear in rats selectively bred for high anxiety–evidence of altered neuronal processing in prefrontal-amygdala pathways, Eur. J. Neurosci. 28 (2008) 2299–2309. [37] A.K. Bruchey, C.E. Jones, M.H. Monfils, Fear conditioning by-proxy: social transmission of fear during memory retrieval, Behav. Brain Res. 214 (2010) 80–84. [38] C.E. Jones, P.D. Riha, A.C. Gore, M.H. Monfils, Social transmission of Pavlovian fear: fear-conditioning by-proxy in related female rats, Anim. Cogn. 17 (2014) 827–834. [39] A. Nowak, T. Werka, E. Knapska, Social modulation in extinction of aversive memories, Behav. Brain Res. 238 (2013) 200–205. [40] Y. Kiyokawa, A. Ishida, Y. Takeuchi, Y. Mori, Sustained housing-type social buffering following social housing in male rats, Physiol. Behav. 158 (2016) 85–89. [41] Y. Kiyokawa, Y. Kodama, Y. Takeuchi, Y. Mori, Physical interaction is not necessary for the induction of housing-type social buffering of conditioned hyperthermia in male rats, Behav. Brain Res. 256 (2013) 414–419. [42] Y. Kodama, Y. Kiyokawa, Y. Takeuchi, Y. Mori, Twelve hours is sufficient for social buffering of conditioned hyperthermia, Physiol. Behav. 102 (2011) 188–192.
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/phb
Social buffering enhances extinction of conditioned fear responses in male rats Kaori Mikami, Yasushi Kiyokawa ⁎, Yukari Takeuchi, Yuji Mori Laboratory of Veterinary Ethology, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
H I G H L I G H T S • • • •
Insufficient extinction training did not extinguish fear responses. Social buffering during insufficient extinction training suppressed fear responses. The effect of social buffering during extinction training was context specific. We concluded that social buffering enhanced extinction of fear responses.
a r t i c l e
i n f o
Article history: Received 18 March 2016 Received in revised form 3 May 2016 Accepted 4 May 2016 Available online 5 May 2016 Keywords: Social buffering Fear extinction Freezing Stress Amygdala
a b s t r a c t In social species, the phenomenon in which the presence of conspecific animals mitigates stress responses is called social buffering. We previously reported that social buffering in male rats ameliorated behavioral fear responses, as well as hypothalamic-pituitary-adrenal axis activation, elicited by an auditory conditioned stimulus (CS). However, after social buffering, it is not clear whether rats exhibit fear responses when they are reexposed to the same CS in the absence of another rat. In the present study, we addressed this issue using an experimental model of extinction. High stress levels during extinction training impaired extinction, suggesting that extinction is enhanced when stress levels during extinction training are low. Therefore, we hypothesized that rats that had received social buffering during extinction training would not show fear responses to a CS, even in the absence of another rat, because social buffering had enhanced the extinction of conditioned fear responses. To test this, we subjected male fear-conditioned rats to extinction training either alone or with a non-conditioned male rat. The subjects were then individually re-exposed to the CS in a recall test. When the subjects individually underwent extinction training, no responses were suppressed in the recall test. Conversely, when the subjects received social buffering during extinction training, freezing and Fos expression in the paraventricular nucleus of the hypothalamus and lateral amygdala were suppressed. Additionally, the effects of social buffering were absent when the recall test was conducted in a different context from the extinction training. The present results suggest that social buffering enhances extinction of conditioned fear responses. © 2016 Elsevier Inc. All rights reserved.
1. Introduction In social species, stress responses induced by exposure to distressing stimuli can be ameliorated when an animal is exposed along with a conspecific animal(s). For example, the presence of a conspecific animal has been found to suppress corticosterone or cortisol release in response to Abbreviations: BA, basal amygdala; CeL, lateral division of the central amygdala; CeM, medial division of the central amygdala; CS, conditioned stimulus; HPA, hypothalamicpituitary-adrenal; IL, infralimbic region of the prefrontal cortex; LA, lateral amygdala; pmOP, posteromedial olfactory peduncle; PVN, paraventricular nucleus of the hypothalamus. ⁎ Corresponding author. E-mail addresses: [email protected] (K. Mikami), [email protected] (Y. Kiyokawa), [email protected] (Y. Takeuchi), [email protected] (Y. Mori).
http://dx.doi.org/10.1016/j.physbeh.2016.05.001 0031-9384/© 2016 Elsevier Inc. All rights reserved.
a novel environment [1,2] or to predator-associated stimuli [3]. These phenomena are called social buffering [4]. In previous studies, we analyzed social buffering in male rats using fear conditioning. When fear-conditioned rats were exposed to an auditory conditioned stimulus (CS), they exhibited both behavioral fear responses and hypothalamic-pituitary-adrenal (HPA) axis activation. The presence of a non-conditioned unfamiliar male rat (associate) completely blocked these responses [5], suggesting that the fearconditioned rat received social buffering from the associate rat. This social buffering of conditioned fear responses occurred even if the dyad was separated by a wire-mesh partition or by double wire-mesh partitions separated by 5 cm [6]. We identified several additional characteristics of social buffering. For example, we found that the presence of guinea pigs [6] or some strains of rats [7] did not induce social buffering, that familiar associates were more effective for social buffering than
124
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
unfamiliar associates [8], that social buffering was similarly observed in female rats [9], and that associate-derived volatile olfactory signals can mediate social buffering [6,8,10]. In our investigations of the neural mechanisms underlying social buffering, we found that activation in the lateral amygdala (LA) in response to the CS was suppressed during social buffering [11]. We also found evidence that the posteromedial olfactory peduncle (pmOP) [12], most likely the posterior complex of the anterior olfactory nucleus within the pmOP [10], relays the olfactory signals responsible for social buffering from the main olfactory bulb to the amygdala. However, in all of our previous studies, rats were not reexposed to the CS after the test. Therefore, whether rats that have experienced social buffering show fear responses when they are re-exposed to the CS in the absence of another rat remains unclear. Extinction of conditioned fear responses appears to be a suitable experimental model for addressing this issue. In this model, fearconditioned animals undergo extinction training 1 day after conditioning, in which they are repeatedly exposed to the CS. If the extinction training is sufficient, the animals will exhibit suppressed fear responses when they are re-exposed to the CS in a recall test on the following day, i.e., extinction of conditioned fear responses [13,14]. Thus, one possibility for assessing the effects of social buffering on subsequent fear responses to the CS would be to subject fear-conditioned rats to social buffering during extinction training and then conduct a recall test in the absence of another rat. The stress levels of an animal during extinction training appear to affect the extinction of conditioned responses. For example, fearconditioned rats showed little or no extinction of conditioned fear responses when they were stressed via foot shocks [15] or placement on an elevated platform immediately prior to extinction training [16]. These findings suggest that extinction will be enhanced when animals undergoing extinction training have low levels of stress. Based on our findings that social buffering ameliorated stress caused by an auditory CS, we hypothesized that rats that experienced social buffering during extinction training would not show fear responses to the CS in a recall test. Indeed, we would expect that social buffering would lower the stress status during extinction training and thus enhance extinction of conditioned fear responses. To test this hypothesis, we subjected a group of rats to fear conditioning using an auditory CS. The rats underwent extinction training either alone or with an associate. The rats were then individually reexposed to the CS in a recall test. The effects of social buffering during extinction training were evaluated by examining freezing behavior and HPA axis activity, as reflected by Fos expression in the paraventricular nucleus of the hypothalamus (PVN), during the recall test. Because the amygdala plays an important role in conditioned fear responses [17,18], we examined Fos expression in the amygdala to assess the underlying neural mechanisms (Experiment 1). Next, we assessed whether the effects of social buffering during extinction training could be observed even if the recall test was conducted in a different context than that of extinction training (Experiment 2).
associates (rat placed with the subject during extinction training), which ensured unfamiliarity between the subjects and associates. All rats were individually housed and handled for 5 min per day for 3 days before the conditioning day. All behavioral procedures were performed between 09:00 and 16:00.
2.2. Procedure
2.1. Animals
2.2.1. Experiment 1 Fear conditioning was performed in an acrylic conditioning box with a metal grid floor (28 × 20 × 27 cm) under a white light (Context A). The rats received seven repetitions of a 3-s auditory CS (8 kHz, 70 dB) that terminated concurrently with a 0.5-s foot shock (0.55 mA). We presented the CS twice before and twice after the conditioning procedure to measure pre- and post-conditioning freezing. The inter-trial interval was randomly varied between 90 and 220 s. The subjects underwent extinction training either with (social situation) or without an associate (alone situation) 24 h after fear conditioning. Based on the post-conditioning freezing, the subjects were divided into extinction and no-extinction groups, which showed comparable freezing in each situation. The rats were placed in an acrylic extinction box with clean woodchip bedding (28 × 44 × 20.5 cm) under a dim red light (Context B). In the alone situation, the subjects in the extinction group (n = 9) received 24 CS presentations in the absence of other animals. In the social situation, the subjects in the extinction group (n = 9) underwent extinction training with an associate. We varied the inter-trial interval randomly between 60 and 120 s. For the rats in the no-extinction group (alone: n = 9, social: n = 9), we placed the subject and associate (in the social situation) in the extinction box for the same length of time as for the extinction group, without any CS presentation. Twenty-four hours after extinction training, we conducted the recall test with the individual subjects in both situations in the same context as for the extinction training (Context B). The subjects were exposed to the CS twice with an interval of 90 s. After the test, the subjects were returned to their home cages. Sixty minutes after the recall test [19], each subject was deeply anesthetized with sodium pentobarbital and intracardially perfused with saline, followed by 4% paraformaldehyde in 0.1 M phosphate buffer. The brain was removed and immersed overnight in 4% paraformaldehyde in 0.1 M phosphate buffer, and then placed in 30% sucrose/phosphate buffer for cryoprotection. We used the avidin-biotin-peroxidase immunohistochemistry method to detect Fos expression, as previously described [8,20]. Briefly, we collected six successive 30-μm sections containing the PVN (Bregma −1.80 mm) or LA, basal amygdala (BA), and lateral (CeL) and medial (CeM) division of the central amygdala (Bregma −2.76 mm). The sections were incubated with a primary antibody to c-Fos protein (1:8000; ABE457, Merck Millipore, Billerica, MA, USA) for 65 h and a biotinylated anti-rabbit secondary antibody (BA1000, Vector Laboratories, Burlingame, CA, USA) for 2 h. The sections were then processed using the ABC kit (Vector Laboratories). Staining was developed by incubating the tissue in a diaminobenzidine solution with nickel intensification.
All experiments were approved by the Animal Care and Use Committee of the Faculty of Agriculture at The University of Tokyo, according to guidelines that were adapted from the Consensus Recommendations on Effective Institutional Animal Care and Use Committees by the Scientists Center for Animal Welfare. Experimentally naïve male Wister rats were purchased at 8 weeks of age from Charles River Laboratories Japan (Kanagawa, Japan). The rats were housed 2–4 per cage in a room with an ambient temperature of 24 ± 1 °C and a humidity of 45 ± 5%. The room had a 12-h light/12-h dark cycle (lights switched on at 08:00). Food and water were available ad libitum. In each cage, rats were assigned to be either subjects or
2.2.2. Experiment 2 Fear conditioning (Context A) and extinction training (Context B) were performed as described above with the exception that all subjects underwent extinction training in the social situation. The subjects then individually underwent the recall test in the same (Context B; same situation, extinction group: n = 8, no-extinction group: n = 8) or different context (Context C; novel situation, extinction group: n = 5, no-extinction group: n = 6) from that of extinction training. In context C, the recall test was conducted in a cylindrical acrylic box (28 × 28 × 25) with paper bedding under a white light.
2. Material and methods
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
2.3. Data analyses Data are expressed as mean ± standard error of the mean and significance was set at P b 0.05. An experimenter who was blinded to the experimental groups evaluated the behavioral responses. We measured the duration of freezing and the frequency of walking using Microsoft Excel-based Visual Basic software that recorded the duration and number of key presses on a computer keyboard. Freezing was defined as an immobile posture (except for movements associated with respiration) and was measured during the 20-s period after the onset of each CS [21]. We calculated the percentage of time spent in a freezing posture with respect to the 20-s period and calculated the average of every 2 CS. During extinction training and the recall test, we measured baseline freezing that occurred during the 20-s period 90 s before the onset of the first CS. Baseline freezing was also expressed as a percent with respect to the 20-s period. During the recall test, the walking that occurred during the 5-m period before the onset of the first CS served as baseline activity. Walking was defined as the number of steps taken with the hind paws. Freezing during fear conditioning and the recall test was analyzed via two-way repeated ANOVA. Freezing during extinction training was analyzed via one-way repeated ANOVA. Additional comparisons of the baseline activity between the situations were conducted via Student's t-test. For immunohistochemical analyses, sections of the PVN, LA, BA, CeL, and CeM were captured using a microscope equipped with a digital camera (DP30BW, Olympus, Tokyo, Japan). An experimenter who was blind to the experimental groups counted the number of Fosimmunoreactive cells within a 0.5-mm square. Quantification was conducted bilaterally in six sections using Scion image Beta 4.0.2 software. When the designated region was smaller than the boundaries of the 0.5mm square, only the cells in the region of interest were counted. Although some sections were lost because of technical problems, we were able to analyze between six and nine subjects in all groups. The Fos expression in each situation was analyzed via Student's t-test. 3. Results 3.1. Experiment 1 For rats in the alone situation (Fig. 1A), we confirmed that an establishment of fear conditioning was not affected by future extinction training (group × conditioning, F(1, 16) = 0.07, P = 0.79) (Fig. 1B). The percent of freezing was affected by fear conditioning (F(1, 16) = 57.3, P b 0.0001), but not by group (F(1, 16) = 0.07, P = 0.79). In extinction training, we confirmed that the percent of baseline freezing was low (2.0 ± 2.0). Additionally, the percent of freezing decreased when the number of presented CSs increased (F(11, 88) = 5.32, P b 0.0001) (Fig. 1B). In the recall test, extinction training did not extinguish fear responses (group × test, F(1, 16) = 0.95, P = 0.34). The percent of freezing was affected by the test (F(1, 16) = 39.4, P b 0.0001), but not by group (F(1, 16) = 0.06, P = 0.82) (Fig. 1B). After the recall test, we observed Fos expression in various brain regions, including the PVN and LA (Fig. 1C). Fos expression was not different between the extinction and no-extinction group in the PVN (t12 = − 0.28, P = 0.78), LA (t13 = 0.38, P = 0.71), BA (t13 = − 0.93, P = 0.37), CeL (t13 = 0.18, P = 0.86), and CeM (t13 = −0.28, P = 0.78) (Fig. 1D). In the social situation (Fig. 1E), we confirmed that an establishment fear conditioning was not affected by future extinction training (group × conditioning, F(1, 16) = 0.003, P = 0.96) (Fig. 1F). The percent of freezing was affected by fear conditioning (F(1, 16) = 40.6, P b 0.0001), but not by group (F(1, 16) = 0.24, P = 0.64). In extinction training, we confirmed that the percent of baseline freezing was low (0 ± 0) and was not affected by an increase in the number of presented CSs (F(11, 88) = 1.14, P = 0.34) (Fig. 1F). In the recall test, extinction training extinguished fear responses (group × test, F(1, 16) = 18.9, P = 0.0005).
125
The percent of freezing was affected by the test (F(1, 16) = 36.5, P b 0.0001) and by group (F(1, 16) = 15.1, P = 0.0013) (Fig. 1F). Additional analyses comparing the situations confirmed that the extinction group showed equivalent baseline activity between the situations (alone situation: 61.9 ± 10.2, social situation: 81.9 ± 8.3, t16 = 1.52, P = 0.15). After the recall test, we observed Fos expression in various brain regions, including the PVN and LA (Fig. 1G). The rats in the extinction group exhibited decreased Fos expression in the PVN (t16 = −2.46, P = 0.0258) and LA (t13 = −3.73, P = 0.0025) compared with those in the no-extinction group (Fig. 1H). Fos expression in the BA (t13 = − 0.51, P = 0.617), CeL (t13 = − 0.60, P = 0.56), and CeM (t13 = −1.34, P = 0.20) did not differ between the extinction and noextinction group (Fig. 1H). 3.2. Experiment 2 In the same situation (Fig. 2A), we confirmed that an establishment of fear conditioning was not affected by future extinction training (group × conditioning, F(1, 14) = 0.037, P = 0.85) (Fig. 2B). The percent of freezing was affected by fear conditioning (F(1, 14) = 39.3, P b 0.0001), but not by group (F(1, 14) = 0.002, P = 0.97). In extinction training, we confirmed that the percent of baseline freezing was low (0 ± 0) and was not affected by an increase in the number of presented CSs (F(11, 77) = 1.27, P = 0.26) (Fig. 2B). In the recall test, extinction training extinguished fear responses (group × test, F(1, 14) = 6.36, P = 0.024). The percent of freezing was affected by the test (F(1, 14) = 53.1, P b 0.0001) and tended to be affected by group (F(1, 14) = 4.58, P = 0.0505) (Fig. 2B). In the novel situation (Fig. 2C), we confirmed that an establishment of fear conditioning was not affected by future extinction training (group × conditioning, F(1, 9) = 0.19, P = 0.67) (Fig. 2D). The percent of freezing was affected by fear conditioning (F(1, 9) = 23.5, P = 0.0009), but not by group (F(1, 9) = 0.19, P = 0.67). In extinction training, we confirmed that the percent of baseline freezing was low (0 ± 0). Additionally, the percent of freezing was not affected by an increase in the number of presented CSs (F(11, 44) = 0.58, P = 0.84) (Fig. 2D). In the recall test, extinction training did not extinguish fear responses (group × test, F(1, 9) = 0.48, P = 0.50). The percent of freezing was affected by the test (F(1, 16) = 105, P b 0.0001), but not by group (F(1, 9) = 0.48, P = 0.50) (Fig. 2D). 4. Discussion In the present study, we observed no extinction of conditioned fear responses in the alone situation (Experiment 1). This suggests that extinction training was insufficient to extinguish fear responses in the recall test. However, in the social situation, we observed suppressed freezing in the recall test and decreased Fos expression in the PVN and LA. Additionally, the effects of social buffering were absent in the novel situation (Experiment 2). Based on these findings, we suggest that social buffering enhanced the extinction of conditioned fear responses in our subjects. In the present Fos analyses, the absolute values of expression seem not to reflect the intensity of fear responses. For example, rats in the no-extinction group in the social situation showed higher expression compared with both groups in the alone situation, although all three groups showed fear responses. Similarly, the extinction group in the social situation showed Fos expression that was comparable to that of both groups in the alone situation, even if the rats in the extinction group did not show fear responses. It is possible that social interaction during extinction training affected basal Fos expression in the recall test. In addition, we assessed each situation successively. Therefore, it seems more appropriate to interpret the data by comparing the noextinction and extinction groups within the situations, rather than between the situations. As a result, we believe that the present results
126
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
Fig. 1. Extinction of fear responses elicited by an auditory conditioned stimulus (CS) in the alone and social situation. (A) Schematic of the procedure in the alone situation. S, subject. (B) The percent of freezing elicited by the CS before (pre) and after (post) fear conditioning, during extinction training, and during the recall test. The responses are expressed as the average of every 2 CS [mean ± standard error of the mean (SEM)]. (C) Representative photomicrographs of the Fos-immunoreactive cells in the paraventricular nucleus (PVN) and lateral amygdala (LA). (D) The number of Fos-immunoreactive cells (mean + SEM) in the PVN, LA, basal amygdala (BA), and lateral (CeL) and medial (CeM) division of the central amygdala of the subject after the recall test. (E) Schematic of the procedure in the social situation. S, subject; A, associate. (F) The percent of freezing elicited by the CS pre and post fear conditioning during extinction training and the recall test. The responses are expressed as the average of every 2 CS. *P b 0.05 compared to the no-extinction group in the social situation, as assessed by a two-way repeated ANOVA. (G) Representative photomicrographs of the Fos-immunoreactive cells in the PVN and LA. (H) The number of Fosimmunoreactive cells in the PVN, LA, BA, CeL, and CeM of the subject after the recall test. *P b 0.05 compared to the no-extinction group in the social situation, as assessed by a Student's t-test.
suggest the suppression of the PVN and LA in the extinction group in the social situation. In Experiment 1, we found that rats in the social situation exhibited reduced fear responses to a CS, as well as suppressed HPA axis and LA activation, in the recall test. We interpret these results to mean that social buffering enhances extinction of conditioned fear responses. Although we cannot deny the possibility that fear responses to the 3-s
CS were resistant to extinction, this seems unlikely because fear responses to a similarly short 2-s CS were successfully extinguished by extinction training [15]. Additionally, this interpretation is consistent with the established characteristics of extinction. For example, extinction training induces extinction of conditioned fear responses only in the context in which the subjects underwent training [22,23]. In addition, the extinction of conditioned fear responses has been found to be
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
127
Fig. 2. Extinction of fear responses elicited by the auditory conditioned stimulus (CS) in the same and novel situation. (A) Schematic of the procedure in the same situation. S, subject; A, associate. (B) The percent of freezing elicited by the CS before (pre) and after (post) fear conditioning, during extinction training, and during the recall test. The responses are expressed as the average of every 2 CS [mean ± standard error of the mean (SEM)]. *P b 0.05 compared to the no-extinction group in the same situation as assessed by a two-way repeated ANOVA. (C) Schematic of the procedure in the novel situation. S, subject; A, associate. (D) The percent of freezing elicited by the CS pre and post fear conditioning, during extinction training, and during the recall test. The responses are expressed as the average of every 2 CS (mean ± SEM).
accompanied by inhibition of the LA during the recall test [24,25]. Therefore, although the extinction training in the present study was insufficient, social buffering during training enhanced the effects of this insufficient extinction training, which induced extinction of conditioned fear responses in the recall test. As a result, the effects of social buffering were observed in a specific context and were accompanied by inhibition of the LA during the recall test. It is worthwhile to consider alternative explanations. First, it is possible that social buffering erased the fear memory of the rats in the extinction group in the social and same situations [26–29]. However, this explanation seems unlikely because the rats in the extinction group in the novel situation exhibited robust fear responses (Experiment 2). Second, it is possible that social buffering during extinction training established an association between the extinction box and stress amelioration, as observed during safety conditioning [30,31], which then suppressed fear responses in the recall test. However, this explanation is unlikely because the rats in the no-extinction group in the social and same situations exhibited robust fear responses, even when an associate accompanied them in the extinction box. Third, it is possible that the presence of an associate during extinction training established an association between the extinction box and hyperactivity. However, this explanation appears to be inappropriate, because the baseline activity in the extinction group was equivalent between the alone and social situations. In the present study, social buffering took place during extinction training while we observed Fos expression after the recall test. Therefore, the present Fos data do not provide information about how social buffering enhanced acquisition of extinction. The infralimbic region of the prefrontal cortex (IL) [32,33], BA [33,34], and ventral hippocampus [33] have been suggested to participate in the acquisition of extinction. As Fos expression in the IL is reduced after extinction training in rats with high stress responses [35] and impairment of extinction [36], it is possible that a low stress status during extinction training as a result of social buffering can activate the IL and enhance extinction. Similarly, it is possible that the associate-derived olfactory signals that are responsible for social buffering activate the BA and enhance extinction when the detected signals are transmitted from the pmOP to the BA via direct projections [12]. However, in our previous studies, social buffering was not necessarily accompanied by increased Fos expression in the IL [5,6] and BA [5,8,10], suggesting that these regions may not be involved. Therefore, further research is necessary to clarify the neural mechanisms underlying the enhancement of extinction by social buffering. In extinction training in this study, the presence of an associate induced social buffering of conditioned fear responses, as seen in our
previous research. However, in some studies, the presence of a conspecific did not affect fear responses to the auditory CS [37–39]. One reason for this discrepancy might be differences between species. Although social buffering can be induced by a mother, mate, or a conspecific with which the subject does not have a sexual relationship [4], this last type of social buffering has been scarcely observed in mice. Indeed, in mice, the presence of another mouse did not reduce freezing in response to an auditory CS [39]. Another possible reason might be differences in CS duration. Indeed, the presence of a conspecific rat did not appear to induce social buffering of conditioned fear responses to a 20-s CS [37,38]. The characteristics of fear responses elicited by a 3-s CS are different than those elicited by a 20-s CS [21]. Therefore, it is possible that the presence of another rat reduced freezing elicited by a 3-s CS, as tested in the present study, while a 20-s CS, tested in previous studies, had no effect. Because this study is the first to assess the effects of social buffering on the extinction of conditioned fear responses, our findings provoke many interesting questions. Given that co-housing with an associate additively induced another type of social buffering [5,40–42], it is possible that the extinction of conditioned fear responses is further enhanced when the subject is housed with an associate between the conditioning, extinction training, and recall test periods. Additionally, extinction of conditioned fear responses might be more strongly enhanced by familiar associates, because familiar associates have been found to be more effective than unfamiliar associate for social buffering of conditioned fear responses [8]. Further research is needed to address these questions. 5. Conclusions The results of the present study demonstrate that insufficient extinction training extinguished fear responses in a recall test when subjects received social buffering during extinction training. This phenomenon was context specific. These results suggest that social buffering enhances extinction of conditioned fear responses. Further analyses are necessary to fully understand the neural mechanisms underlying social buffering, and may contribute to the development of treatments for anxiety and fear responses. Acknowledgments This study was supported by the Japan Society for the Promotion of Science KAKENHI Grant Numbers 15H05635 and 15H05782.
128
K. Mikami et al. / Physiology & Behavior 163 (2016) 123–128
References [1] M.B. Hennessy, D.S. Maken, F.C. Graves, Consequences of the presence of the mother or unfamiliar adult female on cortisol, ACTH, testosterone and behavioral responses of periadolescent guinea pigs during exposure to novelty, Psychoneuroendocrinology 25 (2000) 619–632. [2] M.L. Terranova, F. Cirulli, G. Laviola, Behavioral and hormonal effects of partner familiarity in periadolescent rat pairs upon novelty exposure, Psychoneuroendocrinology 24 (1999) 639–656. [3] J.L. Vogt, C.L. Coe, S. Levine, Behavioral and adrenocorticoid responsiveness of squirrel monkeys to a live snake: is flight necessarily stressful? Behav. Neural Biol. 32 (1981) 391–405. [4] Y. Kiyokawa, Social odors: alarm pheromones and social buffering, Curr. Top. Behav. Neurosci. (2015). [5] Y. Kiyokawa, Y. Takeuchi, Y. Mori, Two types of social buffering differentially mitigate conditioned fear responses, Eur. J. Neurosci. 26 (2007) 3606–3613. [6] Y. Kiyokawa, Y. Takeuchi, M. Nishihara, Y. Mori, Main olfactory system mediates social buffering of conditioned fear responses in male rats, Eur. J. Neurosci. 29 (2009) 777–785. [7] Nakamura, K., Ishii, A., Kiyokawa, Y., Takeuchi, Y., Mori, Y. The strain of an accompanying conspecific affects the efficacy of social buffering in male rats. Horm. Behav. (in press). [8] Y. Kiyokawa, A. Honda, Y. Takeuchi, Y. Mori, A familiar conspecific is more effective than an unfamiliar conspecific for social buffering of conditioned fear responses in male rats, Behav. Brain Res. 267 (2014) 189–193. [9] A. Ishii, Y. Kiyokawa, Y. Takeuchi, Y. Mori, Social buffering ameliorates conditioned fear responses in female rats, Horm. Behav. 81 (2016) 53–58. [10] Y. Takahashi, Y. Kiyokawa, Y. Kodama, S. Arata, Y. Takeuchi, Y. Mori, Olfactory signals mediate social buffering of conditioned fear responses in male rats, Behav. Brain Res. 240 (2013) 46–51. [11] F. Fuzzo, J. Matsumoto, Y. Kiyokawa, Y. Takeuchi, T. Ono, H. Nishijo, Social buffering suppresses fear-associated activation of the lateral amygdala in male rats: behavioral and neurophysiological evidence, Front. Neurosci. 9 (2015) 99. [12] Y. Kiyokawa, Y. Wakabayashi, Y. Takeuchi, Y. Mori, The neural pathway underlying social buffering of conditioned fear responses in male rats, Eur. J. Neurosci. 36 (2012) 3429–3437. [13] K.M. Myers, M. Davis, Behavioral and neural analysis of extinction, Neuron 36 (2002) 567–584. [14] G.J. Quirk, D. Mueller, Neural mechanisms of extinction learning and retrieval, Neuropsychopharmacology 33 (2008) 56–72. [15] S. Maren, C.H. Chang, Recent fear is resistant to extinction, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 18020–18025. [16] I. Akirav, M. Maroun, The role of the medial prefrontal cortex-amygdala circuit in stress effects on the extinction of fear, Neural Plast. 2007 (2007) 30873. [17] S. Maren, G.J. Quirk, Neuronal signalling of fear memory, Nat. Rev. Neurosci. 5 (2004) 844–852. [18] D. Pare, G.J. Quirk, J.E. Ledoux, New vistas on amygdala networks in conditioned fear, J. Neurophysiol. 92 (2004) 1–9. [19] K.J. Kovacs, c-Fos as a transcription factor: a stressful (re)view from a functional map, Neurochem. Int. 33 (1998) 287–297. [20] T. Kobayashi, Y. Kiyokawa, Y. Takeuchi, Y. Mori, Neural correlates underlying naloxone-induced amelioration of sexual behavior deterioration due to an alarm pheromone, Front. Neurosci. 9 (2015) 52. [21] Y. Kiyokawa, K. Mikami, Y. Mikamura, A. Ishii, Y. Takeuchi, Y. Mori, The 3-second auditory conditioned stimulus is a more effective stressor than the 20-second auditory conditioned stimulus in male rats, Neuroscience 299 (2015) 79–87.
[22] M.E. Bouton, Context and behavioral processes in extinction, Learn. Mem. 11 (2004) 485–494. [23] S. Maren, K.L. Phan, I. Liberzon, The contextual brain: implications for fear conditioning, extinction and psychopathology, Nat. Rev. Neurosci. 14 (2013) 417–428. [24] J.A. Hobin, K.A. Goosens, S. Maren, Context-dependent neuronal activity in the lateral amygdala represents fear memories after extinction, J. Neurosci. 23 (2003) 8410–8416. [25] E. Knapska, S. Maren, Reciprocal patterns of c-Fos expression in the medial prefrontal cortex and amygdala after extinction and renewal of conditioned fear, Learn. Mem. 16 (2009) 486–493. [26] M.H. Monfils, K.K. Cowansage, E. Klann, J.E. LeDoux, Extinction-reconsolidation boundaries: key to persistent attenuation of fear memories, Science 324 (2009) 951–955. [27] K.M. Myers, K.J. Ressler, M. Davis, Different mechanisms of fear extinction dependent on length of time since fear acquisition, Learn. Mem. 13 (2006) 216–223. [28] K. Nader, G.E. Schafe, J.E. Le Doux, Fear memories require protein synthesis in the amygdala for reconsolidation after retrieval, Nature 406 (2000) 722–726. [29] G.J. Quirk, D. Pare, R. Richardson, C. Herry, M.H. Monfils, D. Schiller, et al., Erasing fear memories with extinction training, J. Neurosci. 30 (2010) 14993–14997. [30] M. Mohammadi, J.R. Bergado-Acosta, M. Fendt, Relief learning is distinguished from safety learning by the requirement of the nucleus accumbens, Behav. Brain Res. 272 (2014) 40–45. [31] M.T. Rogan, K.S. Leon, D.L. Perez, E.R. Kandel, Distinct neural signatures for safety and danger in the amygdala and striatum of the mouse, Neuron 46 (2005) 309–320. [32] F.H. Do-Monte, G. Manzano-Nieves, K. Quinones-Laracuente, L. Ramos-Medina, G.J. Quirk, Revisiting the role of infralimbic cortex in fear extinction with optogenetics, J. Neurosci. 35 (2015) 3607–3615. [33] D. Sierra-Mercado, N. Padilla-Coreano, G.J. Quirk, Dissociable roles of prelimbic and infralimbic cortices, ventral hippocampus, and basolateral amygdala in the expression and extinction of conditioned fear, Neuropsychopharmacology 36 (2011) 529–538. [34] T. Amano, S. Duvarci, D. Popa, D. Pare, The fear circuit revisited: contributions of the basal amygdala nuclei to conditioned fear, J. Neurosci. 31 (2011) 15481–15489. [35] R. Landgraf, A. Wigger, F. Holsboer, I.D. Neumann, Hyper-reactive hypothalamopituitary-adrenocortical axis in rats bred for high anxiety-related behaviour, J. Neuroendocrinol. 11 (1999) 405–407. [36] P. Muigg, A. Hetzenauer, G. Hauer, M. Hauschild, S. Gaburro, E. Frank, et al., Impaired extinction of learned fear in rats selectively bred for high anxiety–evidence of altered neuronal processing in prefrontal-amygdala pathways, Eur. J. Neurosci. 28 (2008) 2299–2309. [37] A.K. Bruchey, C.E. Jones, M.H. Monfils, Fear conditioning by-proxy: social transmission of fear during memory retrieval, Behav. Brain Res. 214 (2010) 80–84. [38] C.E. Jones, P.D. Riha, A.C. Gore, M.H. Monfils, Social transmission of Pavlovian fear: fear-conditioning by-proxy in related female rats, Anim. Cogn. 17 (2014) 827–834. [39] A. Nowak, T. Werka, E. Knapska, Social modulation in extinction of aversive memories, Behav. Brain Res. 238 (2013) 200–205. [40] Y. Kiyokawa, A. Ishida, Y. Takeuchi, Y. Mori, Sustained housing-type social buffering following social housing in male rats, Physiol. Behav. 158 (2016) 85–89. [41] Y. Kiyokawa, Y. Kodama, Y. Takeuchi, Y. Mori, Physical interaction is not necessary for the induction of housing-type social buffering of conditioned hyperthermia in male rats, Behav. Brain Res. 256 (2013) 414–419. [42] Y. Kodama, Y. Kiyokawa, Y. Takeuchi, Y. Mori, Twelve hours is sufficient for social buffering of conditioned hyperthermia, Physiol. Behav. 102 (2011) 188–192.