Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 1.
Published in final edited form as: Addict Biol. 2018 Aug 14;24(5):958–968. doi: 10.1111/adb.12666

Social stress-potentiated methamphetamine seeking

Ashley M Blouin 1,2, Swathi Pisupati 1,2, Colton G Hoffer 1,2, Madalyn Hafenbreidel 1,2, Sarah E Jamieson 1,2, Gavin Rumbaugh 2, Courtney A Miller 1,2,#
PMCID: PMC6375809  NIHMSID: NIHMS982175  PMID: 30105771

Abstract

Human studies of substance use disorder (SUD) show that psychological stress and drug availability interact following rehabilitation, contributing to the high relapse potential. Social stressors trigger particularly strong motivation for drug, but how this affects neuronal function to increase relapse is unknown. Animal models, which allow for the dissection of neural mechanisms, primarily utilize physical stressors to trigger relapse. To recapitulate psychosocial post-rehabilitation challenges in animals, we developed a model of social stress-potentiated methamphetamine (METH) seeking. Rats receive a single social defeat session after completion of self-administration and extinction of lever pressing. While a reminder of the social defeat was insufficient to reinstate METH seeking on its own, rats that received a reminder of social defeat followed by a METH-priming injection displayed potentiated reinstatement over METH-priming alone. Examination of neuronal activation patterns of the METH-primed reinstatement session identified c-Fos-immunoreactivity (-ir) in the basolateral amygdala (BLA) as correlated with social defeat (SD) score, a measure of defeat latency. Rapidly defeated rats showed potentiated METH-primed reinstatement and elevated BLA c-Fos compared to controls. Conversely, rats that were undefeated during the social stress did not show potentiated METH-primed reinstatement or elevated BLA c-Fos. Interestingly, inactivation of the BLA with baclofen/muscimol (bac/musc) prior to the stress reminder and METH-priming generated a potentiation of METH seeking in the undefeated rats, suggesting the BLA may mediate resilience to the stressor. This model provides a tool for the further dissection of neural mechanisms mediating social stress-potentiated relapse and for the development of relapse-reducing therapeutics.

Keywords: social defeat, stress, methamphetamine, reinstatement

INTRODUCTION

Drug users identify stressors as frequent reasons for drug use. Social stressors, in particular, induce strong motivation to seek the drug, serving as potent relapse factors (Bradley et al., 1989; Marlatt and George, 1984; Sinha et al., 2000; Sinha et al., 2006; Wallace, 1989). Most existing animal models utilize physical stressors, such as footshock, that are variable in their ability to reinstate drug seeking. In addition, social factors have rarely been integrated into research models of drug abuse (Heilig et al., 2016). Therefore, development of animal models that utilize psychosocial stressors have the potential to more closely match the human experience, producing more relevant and possibly more consistent results. Furthermore, animal models that include the complex interaction between psychosocial factors and drug availability could be highly relevant to the human situation, as relapse is often multi-determined (Soderpalm et al., 2003; Wallace, 1989). For example, a person experiencing emotional distress or social pressure is more likely to relapse when drugs are available (Wallace, 1989). Furthermore, a negative emotional state can change the subjective and behavioral effects of the drug, influencing how much drug is consumed (Soderpalm et al., 2003). Specifically, Soderpalm et al. demonstrated that social stress combined with METH administration increased wanting for drug in humans and that this differed from the effect of social stress alone (Soderpalm et al., 2003). Recapitulating this general phenomenon in animals would enable study of the underlying neurobiology and development of therapeutics to prevent relapse triggered by psychosocial stress. Finally, existing studies on the effects of social stress on drug use have either failed to find effects or do not model relapse (Al-Hasani et al., 2013; Caldwell and Riccio, 2010; Covington and Miczek, 2001; Cruz et al., 2011; Zou et al., 2014). Most studies that report potentiation of drug seeking with social stress utilize a stressor introduced prior to or during training, likely modeling an effect on development of compulsive drug taking, rather than relapse. Modeling effects of social stress on reinstatement is important, as the neurobiological mechanisms mediating the effects of social stress on drug seeking are likely unique during periods of abstinence (Covington et al., 2011).

Therefore, we sought to develop an animal model of relapse triggered by the interaction of psychosocial stress and drug availability. Specifically, rats received a reminder of social defeat prior to METH-primed reinstatement, or social stress and METH-primed (SSM) reinstatement. Using this model, we investigated c-Fos-ir following SSM reinstatement and identified three brain regions with elevated c-Fos in stressed animals compared to controls. This included the BLA, which we determined is important for relapse triggered by SSM reinstatement. This model is useful for understanding the neurobiological mechanisms that mediate relapse and can facilitate the development of therapeutics targeting the chronic, relapsing nature of SUD.

METHODS

Animals

Adult male Sprague-Dawley rats (250-300g; Charles River Laboratories) and adult male Long Evans rats (500g; Envigo) were housed under a 12:12 light/dark cycle (lights off at 0800), with food and water ad libitum. Animals were handled for 3 days prior to the start of food training and experimenter blinding was performed in all cases.

Surgery

Sprague-Dawley rats were anesthetized with dexmedetomidine (0.25mg/kg, IM) and ketamine (0.5mg/kg, IM) and implanted bilaterally with 26 gauge guide cannulae (Plastics One) 2mm above the BLA (AP −2.9mm, ML ± 5mm; DV −6.7mm; (Paxinos and Watson, 2007)). Immediately after surgery, buprenex (0.05ml, SC) was administered. For intravenous self-administration (IVSA) experiments, rats were implanted with chronic indwelling jugular catheters 48hrs after food training. Catheters were flushed once daily for 5 days with 100 μl of cefazolin (10 mg/ml) in heparinized saline (70 U/ml) and 100 μl of low dose heparinized saline (10 U/ml) to ensure patency. Patency was confirmed with 100μl of propofol (10 mg/ml) prior to starting METH training.

Food Training

Sprague-Dawley rats were first food trained in a 20hr overnight session to lever press on a fixed ratio (FR) 1 for 45mg grain pellets (Bioserv) in operant conditioning chambers (Coulbourn Instruments). The chambers were placed in sound attenuating isolation cubicles (Coulbourn Instruments) and were equipped with two retractable levers and a food dispenser positioned between the levers. Pressing the lever positioned to the right of the dispenser (active lever) resulted in delivery of a food pellet, whereas pressing the left (inactive) lever had no programmed consequences. Surgery was performed 48hrs after reaching a criteria of 100 lever presses during a single food training session.

METH Self-Administration

After surgery and recovery, Sprague-Dawley rats began METH self-administration training, which was conducted during 2hr sessions (between the hours of 1200 and 1600) for 14 days. Training, extinction and reinstatement were performed at the same time of day and in the same context, consisting of a solid blue Plexiglas floor and two house lights (one inside and one outside the operant chamber). Animals were weighed and received a 0.1ml IV infusion of high dose heparin (70 U/ml) prior to the start of each training session. Catheters were then connected to polyethylene 20 tubing encased in a steel spring leash (Plastic One) leading to a liquid swivel suspended above the chamber. The liquid swivel was connected to a fixed rate pump (Coulbourn Instruments). Animals were trained to press the active lever on an FR1 schedule for METH reinforcement (0.02mg in a 0.05ml infusion), followed by a 20s time-out period. Lever presses on the inactive lever were not reinforced. After each training session, animals received an infusion of 100 μl of cefazolin (10 mg/ml) in heparinized saline (70 U/ml) and 100 μl of low dose heparinized saline (10 U/ml) to ensure patency. Animals received 20g of food pellets after each behavioral session to maintain their weight.

Extinction

Extinction training consisted of 2hr sessions free of reinforcement. All animals underwent at least 5 days of extinction and continued until extinction criteria was met. Extinction criteria was set for individual animals at 2 consecutive days where active lever presses were less than 25% of their average lever presses over the final three days of training. Groups were matched such that there were no differences in training or extinction levels between animals to serve as control or stressed. Social defeat occurred 24 hrs after the last day of extinction.

Social Defeat

Social defeat stress was performed in a procedure room separate from the self-administration room under red lighting at the beginning of the rats’ dark cycle between the hours of 0900-1200. The Sprague-Dawley (intruder) was placed into the home cage (38cm length × 26cm width × 21cm height) of a Long Evans retired breeder (resident) until it was defeated (supine for >3 sec or until a maximum of 15 min of fighting occurred). The intruder was assigned an SD score of 3 if it was defeated in less than 5 min, 2 if it was defeated in 5-15 min, and 1 if it was not defeated. Following defeat, the intruder was placed into a cylindrical wire mesh cage (16cm length × 12cm diameter) and remained in the resident’s home cage for the remainder of the 30 min session. The cage served to protect the intruder from physical injury, but allowed olfactory and visual perception of the resident. Control animals were placed in a clean home cage for 30 min. Prior to the social stress experiments, all residents were screened three times. Only those showing aggression toward all three test intruders within 2min of placement into their cage were used.

Reinstatement

24hrs after social defeat, rats underwent social stress (SS) reinstatement where they were given a social stress reminder 30min prior to placement into the IVSA chambers. For the social stress reminder, the intruder was caged in the resident’s home cage for 5min with the resident present. Control animals were placed in a clean home cage for 5min, followed by placement into the IVSA chambers 30min later. 24hrs later, rats underwent SSM reinstatement, where they were again given a social stress reminder, followed by a METH-priming injection (1mg/kg) 30min later and placement into the IVSA chambers. Control animals were placed in a clean home cage for 5min, followed by a METH-priming injection 30min later and placement into the IVSA chambers. Reinstatement testing consisted of 2hr sessions, free of reinforcement.

BLA Inactivation

Animals received an intra-BLA infusion of bac/musc (0.03/0.3 nmol in 0.5 μl/side) or vehicle (saline) 15min prior to SSM reinstatement. Intra-BLA infusions were delivered at a rate of 0.25 ul/min over 2min. Infusers were left in place for 1min to allow for diffusion of the drug. Infusion sites are presented in Figure S1. While the BLA was targeted, it is possible infusions spread up the needle into the lateral amygdala. Groups were matched so that training and extinction levels were equivalent across all groups.

Immunostaining

Immediately following SSM reinstatement, rats were transcardially perfused with saline followed by 4% paraformaldehyde. Brains were postfixed overnight, placed in 30% sucrose for one week, and frozen and sliced using a microtome. 50μm sections (1/12 sections) were immunolabeled for c-Fos (rabbit anti-c-Fos, 1:5000; Santa Cruz) or double labeled for c-Fos (rabbit anti-c-Fos, 1:5000; Synaptic Systems) and GAD67 (mouse anti-GAD67, 1:1000; EMD Millipore). Secondary antibodies were biotinylated goat anti-rabbit combined with diaminobenzidine (Vector Laboratories) or, for colabeling, Alexa Flour 488 goat anti-rabbit (1:500; Invitrogen) and Alexa Flour 568 goat anti-mouse (1:200; Invitrogen). c-Fos+ or c-Fos/GAD67+ cells were counted with Image J by a trained observer blind to treatment condition. Percent colabeling was calculated as the number of c-Fos/GAD67 colabeled neurons per mm2 divided by the number of GAD67 or c-Fos labeled neurons per mm2 and multiplied by 100. For colabeling, two sections from the BLA at coordinates −2.4 and −3.0 were analyzed. For the c-Fos/DAB staining, two sections from the following brain regions were analyzed at the following bregma coordinates: PrL +3.6, +3.0; NAcc +1.8, +1.2; BLA −2.4, −3.0; piriform cortex Pir −3.0, −3.6 and PrH −3.6, −4.2. One section from the supramammillary nucleus SuM was analyzed at bregma coordinate −4.2.

Statistical Analysis

Spearman’s rank-order or Pearson’s correlations, unpaired t-tests and one or two-way ANOVA with Tukey’s or Fisher’s LSD post hoc tests were used to analyze all data. Statistical significance was set at p ≤ 0.05.

RESULTS

Rats underwent the IVSA procedure shown in Figure 1A. One day after the completion of extinction (Figure 1B), rats underwent a 30 minute social defeat session that produced equal numbers of rats assigned an SD score of 1 (no defeat), 2 (defeat in 5-15 min) or 3 (defeat in < 5 min; Figure 1C). These groups, along with control animals that were placed in a clean cage instead of receiving social stress, did not differ in training (One-way ANOVA: F3,32 = 0.7614, p = 0.5241 active lever; F3,32 = 1.958, p = 0.1401 infusions; Table S1) or extinction levels (One-way ANOVA: F3,32 = 0.4394, p = 0.7264 active lever; Table S1 and Figure 1D), which occurred prior to introduction of the stressor. One day after the social defeat session, rats underwent SS reinstatement, where they were given a 5 minute social stress reminder, or 5 minutes in a clean cage in the case of control animals, 30 minutes before placement in the IVSA chambers. This was insufficient to reinstate lever pressing, as it did not differ from the number of lever presses at the end of extinction (Figure 1D; Two-way ANOVA: group F3,32 = 0.3755; p = 0.7712; session F1,32 = 0.5814; p = 0.4513; interaction F3,32 = 0.2521, p = 0.8592). Further, there were no differences among controls and SD score 1-3 animals (Figure 1D and S2A). The next day, rats underwent SSM reinstatement, where they were given a social stress reminder (or clean cage for controls) prior to a METH-priming injection and placement into the IVSA chambers. Active lever presses during SSM reinstatement were higher than during the last day of extinction for controls and rats with SD score 3 (Figure 1D-E; Two-way ANOVA: group F3,32 = 4.649; p= 0.0083; session F1,32 = 42.16; p < 0.0001; interaction F3,32 = 5.047; p = 0.0056; Sidak’s multiple comparisons test: p<0.05 controls; p<0.0001 SD 3; p>0.05 SD 1-2). Rats with an SD score of 3 had higher total active lever presses (ALPs) when compared to rats with an SD score of 1 and controls during the 2 hour reinstatement session (Figure 1E; One-way ANOVA: F3,32 = 4.902, p = 0.0065; Tukey’s post hoc test: controls vs SD 3: p<0.01 and SD 3 vs SD 1: p<0.01) and over the course of the 2 hours (Figure 1F; Two-way ANOVA: group F3,32 = 4.902, p = 0.0065; Tukey’s post hoc test: controls vs SD 3: p<0.01 and SD 3 vs SD 1: p<0.01). Consistent with an effect of time to defeat, SD 2 rats displayed an intermediate phenotype, particularly pronounced during the first 30 minutes (Figure 1F). Together, this indicates a connection between rapid defeat during social stress and response to a METH-priming injection that produced a marked potentiation of drug seeking.

Figure 1. Social stress potentiates METH-primed reinstatement with interindividual variability.

Figure 1

Open in a new tab

A) Self-administration and social stress procedure; SS reinstatement = social stress reminder + reinstatement test, SSM reinstatement = social stress reminder + METH-primed reinstatement test. B) Training and extinction data from stressed (n = 19) and control (n = 17) animals combined. C) Number of animals with an SD score of 1 (no defeat), 2 (defeat in 5-15 min) or 3 (defeat < 5 min). D) Active and inactive lever presses for the 2 hr SS reinstatement session and the last day of extinction (2 hr session). E) Active and inactive lever presses for SSM reinstatement during the entire 2 hrs and F) over the course of 2 hrs shown in 30 min bins. ** p<0.01 for SD 3 vs control and SD 1. Error bars represent ±SEM.

To begin to determine the neural circuitry governing SSM reinstatement potentiation, the rats from Figure 1 were perfused immediately following SSM reinstatement and c-Fos was quantified in several brain regions (Figure 2A-B). Overall, stressed (SD 1-3) rats had more c-Fos-positive (+) cells in the prelimbic cortex (PrL; p = 0.0181), nucleus accumbens core (NAcc; p = 0.0486) and BLA (p = 0.0351) compared to controls. In contrast, no differences were found between stressed rats and controls in the piriform cortex (Pir; p = 0.9563), perirhinal cortex (PrH; p = 0.8622) or supramammillary nucleus (SuM; p =0.9399). Because increased activation could merely reflect the elevated motor activity accompanying lever pressing, the number of c-Fos+ cells were compared to ALPs and SD score (Figure 2C-D and Table 1). The BLA was the only region in which increased c-Fos+ cells correlated with increased SD score (Figure 2C left; Spearman’s correlation: rs = 0.52, p = 0.02). There was no such correlation with ALPs (Figure 2C right; Pearson’s correlation: r = 0.14, p = 0.56), suggesting that the neural activation in this region was influenced by the effects of social stress on METH seeking, rather than the motor activity accompanying lever pressing. Interestingly, the increased mean for BLA activation was entirely driven by the defeated rats (SD 2-3); undefeated rats (SD 1) showed no change in c-Fos compared to controls (Figure S2B; One-way ANOVA: F2,33=9.846, p<0.001; Tukey’s post hoc test: defeat vs control p<0.01, defeat vs undefeated p<0.01, control vs undefeated p>0.05). In stressed animals, c-Fos-ir in the PrL and NAcc did not correlate with SD score or ALPs, although the PrL showed a trend to correlate with SD score (Table 1; Figure 2D). c-Fos+ cells in both the PrL and NAcc trended toward a correlation with ALPs in control animals (Figure S2C-D; Table 1; Pearson’s correlation: r = 0.56, p = 0.10 for PrL and r = 0.57, p = 0.09 for NAcc), suggesting these regions may mediate nonstress-associated lever pressing. Because we quantified c-Fos+ cells in the first 2 of 3 cohorts of rats (10 stressed rats and 10 control rats) and found that activation in the NAcc and PrL showed a trend to correlate with ALPs in controls (p ≤ 0.1), we focused on the BLA with cohort 3. Therefore, the results presented in Table 1 and Figure 2, represent 19 stressed and 17 control rats for BLA, and 10 stressed and 10 control rats for the other regions.

Figure 2. Social stress-potentiated METH seeking is associated with increased c-Fos in the BLA.

Figure 2

Open in a new tab

A) c-Fos expression following social stress-potentiated METH-primed reinstatement in the PrL, BLA, NAcc, Pir, PrH and SuM. N = 10 stressed (SD 1-3) and n = 10 control for PrL, Nacc, Pir, PrH, SuM; n = 19 stressed and n = 17 control for BLA; *p<0.05, Holm-Sidak. B) Examples of immunostained tissue showing increased c-Fos expression in the PrL, NAcc and BLA but not SuM of stressed (SD 3 shown) rats. C) Correlations between c-Fos expression and SD score (left) and c-Fos expression and active lever presses during SSM reinstatement (right) in the BLA (n = 19) D) Correlations between c-Fos expression and SD score (left) and c-Fos expression and active lever presses during SSM reinstatement (right) in the NAcc (n = 10). Scale bar = 100μm. Error bars represent ±SEM.

Table 1.

Correlation of c-Fos expression with social defeat score and active lever presses during SSM reinstatement.

SD Score ALP
Stressed rs p r p
PrL 0.59 0.09 −0.05 0.89
BLA 0.52 0.02 0.14 0.56
NAcc 0.16 0.68 −0.14 0.70
Pir 0.16 0.68 −0.19 0.60
PrH −0.16 0.68 −0.24 0.50
SuM −0.14 0.73 −0.07 0.85
Control rs p r p
PrL N/A N/A 0.56 0.10
BLA N/A N/A 0.24 0.36
NAcc N/A N/A 0.57 0.09
Pir N/A N/A 0.22 0.54
PrH N/A N/A 0.31 0.39
SuM N/A N/A −0.07 0.84
Open in a new tab

BLA c-Fos expression in stressed rats correlates with SD score but not with active lever presses while PrL and NAcc c-Fos expression in control rats tends to correlate with active lever presses during the METH-primed reinstatement session. n = 10 stressed and n = 10 control for PrL, Nacc, Pir, PrH, SuM; n = 19 stressed and n = 17 control for BLA.

Because the c-Fos results suggested a link between the BLA and the impact of social stress on METH-primed reinstatement, the IVSA procedure was repeated with BLA inactivation by co-infusion of bac/musc prior to the SSM reinstatement session. Similar to the data in Figure 1E, ALPs during SSM reinstatement were higher than during the last day of extinction for defeated rats (SD 2-3) receiving intra-BLA vehicle infusions, but not for undefeated rats (SD 1) receiving intra-BLA vehicle infusions (Figure S3A; Two-way ANOVA: session F1,27 = 16.80; p = 0.0003; Sidak’s multiple comparisons test: p<0.05 defeated; p>0.05 undefeated). Also similar to Figure 1E, defeated rats receiving intra-BLA vehicle infusions had higher ALPs compared to undefeated, vehicle-infused rats (Figure 3A; Two-way ANOVA: interaction p = 0.0223; Fisher LSD: defeat-veh vs no defeat-veh p = 0.0217 and Figure S3B). There was a trend toward a bac/musc-induced decrease in reinstatement among the defeated rats (Figure 3A; Fisher LSD: defeat-veh vs defeat-bac/musc p = 0.1415 and Figure S3C) and these rats did not reinstate compared to their last day of extinction (Figure S3A; p>0.05). However, implanting rats with cannulae biased the distribution toward “fighting back” during social defeat, as evidenced by the lower percentage of defeated rats in the bac/musc study (32.3%; 21 undefeated vs 10 defeated) versus the study in Figure 1, which did not involve cannulations (63.2%; 7 undefeated vs 12 defeated). Thus, statistical significance could not be reached with the defeated group. Conclusive results could, however, be obtained from the well-powered no defeat group, in which intra-BLA bac/musc produced the unexpected effect of enhanced SSM reinstatement (Figure 3A; Fisher LSD: no defeat-veh vs no defeat-bac/musc p = 0.0477), producing reinstatement compared to their extinction levels (Figure S3A; p<0.05). This effect was particularly pronounced during the first hour of reinstatement testing (915% increase over vehicle; Figure 3B; Two-way ANOVA: group F1,19 = 7.341, p = 0.0139; time F3,57 = 3.846, p = 0.0141; interaction F3,57 = 7.723, p = 0.0002; Tukey: no defeat-veh vs no defeat-bac/musc p < 0.05 for the first 30 min and p < 0.001 for the second 30 min).

Figure 3. Inactivation of the BLA potentiates METH seeking in undefeated rats.

Figure 3

Open in a new tab

A) Active lever presses during the 2 hr METH-primed reinstatement session (veh SD 1 n = 11, SD 2-3 n = 5; bac/musc SD 1 n = 10, SD 2-3 n = 5). B) Data for no defeat (SD 1) rats shown over the course of the 2 hr METH-primed reinstatement session in 30 min bins. C) Effect of bac/musc on BLA c-Fos expression and D) CeA c-Fos expression. E) Examples of immunostained tissue from the BLA and CeA of rats that had veh or bac/musc infusion into the BLA. *p<0.05; ***p<0.001; Scale bar = 100μm. Error bars represent ±SEM.

To confirm that the BLA was inactivated by the bac/musc infusion, c-Fos-ir was assessed in the BLA and found to be reduced compared to vehicle infused rats (Figure 3C and 3E; Two-way ANOVA: drug F1,27 = 21.09 p < 0.0001). Interestingly, there was a large increase in c-Fos in the central amygdala (CeA) of bac/musc rats (Figure 3D-E; Two-way ANOVA: drug F1,27 = 12.02 p=0.0018). From this, we hypothesized that this CeA activation may contribute to SSM reinstatement of drug seeking. Therefore, we retrospectively analyzed the CeA from SSM reinstatement rats in which the BLA was not manipulated (Figures 1–2). There was a trend for defeated rats to have higher CeA c-Fos when compared to nondefeated rats (Figure S3D; unpaired t-test: no defeat (SD 1) vs defeat (SD 2-3) p = 0.29), suggesting there may be a critical interplay between the BLA and CeA, with social stress susceptibility serving as an important co-variate.

To further determine the relationship between reinstatement and SD score, we combined rats that received intra-BLA vehicle infusions (Figure 3) with rats that were not cannulated (Figure 1), since lever pressing during SSM reinstatement was equivalent between these groups (Two-way ANOVA: treatment F1,27 = 0.01663, p = 0.8984). During SSM reinstatement, ALPs positively correlated with SD score (Figure 4A; Spearman’s correlation: rs= 0.52; p = 0.0015), such that the more rapid the social defeat, the greater the reinstatement. BLA inactivation with bac/musc disrupted this correlation (Figure 4B; Spearman’s correlation: rs = −0.34, p = 0.22), indicating that the BLA is important for the effects of social defeat on drug seeking during SSM reinstatement. While bac/musc abolished the correlation between SD score and ALPs, it produced only a slight trend in the opposite direction for rapidly defeated rats to display lower reinstatement.

Figure 4. BLA inactivation disrupts the correlation between social defeat score and drug seeking.

Figure 4

Open in a new tab

Correlation between SD score and active lever presses during METH-primed reinstatement in rats that received A) an intra-BLA veh infusion or no infusion (n = 34) or B) BLA inactivation with bac/musc (n = 15).

Because BLA inactivation increased METH seeking in undefeated rats, but tended to decrease it in defeated rats, we hypothesized that different populations of BLA neurons may be activated during SSM reinstatement in the two groups of rats, as has been found in the medial amygdala for social and asocial behaviors (Hong et al., 2014). First, we examined interneuron activation in defeated and undefeated rats following SSM reinstatement, using double immunolabeling for c-Fos and GAD67, because interneurons can have broad effects on local circuitry. The total number of c-Fos+/GAD67+ colabeled cells tended to be higher in the defeated (SD 2-3) group compared to the undefeated (SD 1) group, but this did not reach significance (Figure S4A; unpaired t-test: no defeat vs defeat p = 0.075). The total number of GAD67+ neurons did not differ across groups (Figure S4B; unpaired t-test: no defeat vs defeat p = 0.40) and was low, consistent with reports that interneurons make up approximately 15% of BLA neurons (McDonald, 1992). When the number of colocalized cells was calculated as a percentage of the total number of GAD67+ cells, the percent of colabeled cells was higher in the defeated group (Figure 5A; unpaired t-test: no defeat vs defeat p = 0.048) and positively correlated with SD score (Figure 5B; Spearman’s correlation: rs = 0.45, p = 0.05), but not ALPs during SSM reinstatement (Figure S4C; Pearson’s correlation: r = 0.09, p = 0.73), suggesting that activated GAD67+ neurons of the BLA region were related to the effects of social stress on METH seeking, rather than the motor activity accompanying lever pressing. However, when the number of colocalized cells was calculated as a percentage of the total number of c-Fos+ cells (Figure S4D; unpaired t-test: no defeat vs defeat p = 0.0002;), percent colabeling did not differ between defeated and undefeated rats (Figure 5C; unpaired t-test: p = 0.33) and was not correlated with SD score (Figure 5D; Spearman’s correlation: rs = 0.30, p = 0.21) or ALPs (Figure S4E; Pearson’s correlation: r = 0.00, p = 0.99), suggesting that the increase in colabeled cells in the defeated rats was due to the overall increase in c-Fos+ cells and that the proportion of activated interneurons did not differ between the two groups (Figure S4F).

Figure 5. c-Fos expression in GAD67+ neurons of the BLA is higher in defeated rats compared to undefeated rats due to an overall increase in c-Fos in defeated rats.

Figure 5

Open in a new tab

A) Defeated rats (n = 12) have a significantly higher percentage of cells expressing c-Fos compared to undefeated rats (n = 7) when c-Fos+GAD67+ cells are expressed as a percentage of GAD67+ neurons B) and this % colabeling correlates with SD score in the BLA (n = 19). C) The difference is not present when c-Fos+GAD67+ neurons are expressed as a percentage of c-Fos+ neurons D) and this % colabeling does not correlate with SD score. *p<0.05; Error bars represent ±SEM.

DISCUSSION

We found that social defeat plus a social stress reminder just before testing was insufficient to induce reinstatement without a METH-priming injection and may actually suppress seeking. This result is consistent with previous findings showing that acute exposure to social defeat did not reinstate alcohol self-administration (Funk et al., 2005), although exposure to a peppermint odor previously paired with social defeat was sufficient to reinstate alcohol and cocaine seeking (Funk et al., 2005; Manvich et al., 2015). Social stress increases alcohol consumption in rats when it occurs 2, but not 4 hours prior to self-administration (Caldwell and Riccio, 2010; van Erp and Miczek, 2001), suggesting the timing of stress is important. We found that when social stress was combined with a METH-priming injection, it dramatically potentiated reinstatement. Interestingly, this potentiation was greatest at 1-2 hours following the stress reminder. We also found that a dichotomy emerged that was linked to the rate at which individual rats were defeated. While social stress significantly potentiated METH-primed reinstatement in rapidly defeated rats, it tended to suppress it in rats that were not defeated. Because all resident rats were selected to ensure that they were highly aggressive, this variability was not dependent on the resident, but instead depended on the intruder rats’ stress coping style. Rapidly defeated rats displayed passive coping by quickly submitting, whereas undefeated rats displayed active coping by aggressively fighting the resident. Our study suggests that active coping may suppress METH reinstatement, as passive, but not active coping resulted in significantly higher lever pressing during SSM reinstatement compared to extinction. Consistent with our study, passive coping with social defeat has been associated with vulnerability to a range of behavioral and physiological disturbances, as well as to increased neuronal activation in rats (Walker et al., 2009; Wood et al., 2010). This aspect of our model also makes it highly relevant to humans, as passive coping strategies, such as avoidance behavior, are positively associated with SUD and relapse vulnerability and active coping strategies, such as problem solving and altering the source of stress, can decrease drug use (Aldao et al., 2010; Kronenberg et al., 2015; Sinha, 2001; Wagner et al., 1999).

When correlations between c-Fos-ir and SD score were examined, we found that increased activation in the BLA was associated with rapid defeat. This increased activation was not due simply to the increased lever pressing during the reinstatement session, as it did not correlate with active lever pressing during SSM reinstatement. Instead, the data suggest that the BLA is involved in the effects of social stress on drug-primed reinstatement, perhaps through a mediation of resilience. Consistent with this hypothesis, we found that BLA inactivation enhanced METH seeking in the undefeated group and resulted in increased lever pressing during SSM reinstatement compared to extinction. BLA inactivation produced a trend toward decreased METH seeking in the defeated group and these animals did not show increased lever pressing during SSM reinstatement compared to extinction. As described in the Results section, cannulation shifted the response of rats toward the “undefeated phenotype”, leaving few defeated rats to work with. Cannulation is a mild stressor that may alter the rats’ coping strategy to a subsequent stressor, and in this case, resulted in more resilience to a social stressor. An interesting possibility is that the few rats that did display the “defeated phenotype” may represent particularly strong examples of passive coping. The effect of BLA inactivation in this group, to reduce drug seeking during SSM reinstatement, would suggest the BLA regulates coping strategy.

While quantifying c-Fos in the BLA to verify BLA inactivation with bac/musc, we noticed a marked increase in activation of the CeA of animals receiving bac/musc infusions. Since the CeA is composed predominantly of GABAergic neurons (McDonald, 1991; McDonald and Augustine, 1993; Pare and Smith, 1993), one possibility is that BLA infusion of bac/musc spilled over into the CeA, disinhibiting these cells and causing generalized activation. However, this is unlikely because we used the same dose and volume of bac/musc previously shown to inactivate the BLA without affecting the CeA (Li et al., 2015; McFarland and Kalivas, 2001). Additionally, activation of the CeA was equivalent in defeated and nondefeated animals, yet the behavioral effects of the bac/musc injection were opposite, suggesting a nonspecific mechanism such as global CeA activation is unlikely. Instead, the increased drug seeking was specific only to animals that used an active coping strategy and received bac/musc, which caused them to behave like rats that submitted quickly. This interesting effect suggests the presence of a BLA to CeA pathway that modulates coping strategy. Furthermore, evidence indicates that the CeA is involved in stress-induced reinstatement of drug seeking (Mantsch et al., 2016), the response to social defeat stress (Jasnow et al., 2004), and in regulation of passive versus active coping (Gozzi et al., 2010; Phelps and LeDoux, 2005). Walker et al. found that rats that were “low fighters” during social defeat stress had significantly higher c-Fos in the CeA following the stress when compared to “high fighter” rats, suggesting that this region is directly linked to coping with social stress (Walker et al., 2009). In addition, stimulation of the BLA’s projection to the lateral division of the central amygdala (CeL; accompanied by inhibition of the medial central amygdala [CeM]) decreases the passive coping response of unconditioned freezing (Ciocchi et al., 2010). Thus, it is possible that inactivation of the BLA in the undefeated rats during SSM reinstatement produced a similar bias toward active coping (reducing drug seeking) through its projection to the CeL and concomitant inhibition of the CeM. Upon BLA inactivation with bac/musc, the shift in undefeated rats from active to passive coping (increased drug seeking) may have been supported by CeM disinhibition. Future molecular and tracing studies will determine the identity of BLA and CeA cells in defeated and undefeated rats, with a particular focus on subdivisions of the CeA. Furthermore, cell-specific functional knockdown and optogenetic studies will address whether the CeA activation that accompanied local inactivation of the BLA by GABA agonism mediated the change in SSM reinstatement responses in both defeated and undefeated rats. These studies would also help to rule out a nonspecific bac/musc infusion spillover effect.

Because bac/musc infusion into the BLA had seemingly opposite effects in defeated and undefeated rats, we hypothesized that different populations of BLA neurons are recruited during SSM reinstatement based on coping strategy during the social stress. These populations could differ in a variety of ways, including their inputs, outputs, local connectivity and/or cell type. As a first step, we addressed the possibility that the opposing populations differed in cell type. For example, one could contain a larger proportion of activated interneurons, as has been found for social behavior in the MeA (Hong et al., 2014). We tested this by double labeling for c-Fos and GAD67, but found that only a minority of activated neurons in both defeated and undefeated rats were interneurons. The BLA of the defeated animals contained more activated interneurons, but only because it contained more c-Fos+ neurons overall. This finding is consistent with existing evidence that neuronal ensembles linked to drug-associated stimuli are composed of mostly excitatory cells and may differ only in their molecular markers and/or connections with other brain regions (Bossert et al., 2011; Cai et al., 2016; Carelli et al., 2003; Herry et al., 2008; Kim et al., 2016; Pfarr et al., 2015).

While a great deal is yet to be done to determine the specific neural circuit details, the current study presents a viable protocol for modeling the impact of social stress on drug-primed reinstatement. Our model uses the ethological-based stressor of social defeat in rats and demonstrates parallel findings to human studies of the effects of social stress on drug seeking. In particular, Soderpalm et al. found that drug craving in human subjects is potentiated following both social stress and METH administration, but not after social stress alone (Soderpalm et al., 2003), an effect we have recapitulated here in rats. Importantly, we have also identified individual variability, a core characteristic of humans, as a mediator of the response. Indeed, rats that employ the active stress coping strategy of fighting the aggressor rat for the full social stress period are less susceptible to reinstatement of drug seeking by a reminder of the stressor and a low dose drug priming injection. It will be of interest to identify the drugs of abuse that this general discovery extends to. We hope that this model is adopted by others in the field to begin untangling the molecular and circuit level mediators of social stress during (forced) abstinence on drug relapse, with an eye toward therapeutic discovery.

Supplementary Material

Supp info

Acknowledgments

The authors would like to thank Thomas Vaissiere for technical assistance. This work was funded by grants from the National Institute on Drug Abuse (R03DA033499 and R01DA034116 to CAM and K01DA040737 to AMB), and the Brain and Behavior Foundation (NARSAD Young Investigator Award to AMB).

Footnotes

CONFLICT OF INTEREST

The authors have no conflicts of interest to report.

AUTHOR CONTRIBUTION

AB, GR and CM were responsible for the study concept and design. AB, SP, CH, MH and SJ performed the experiments. AB and CM analyzed data and wrote the manuscript. MH provided helpful comments on the manuscript. All authors critically reviewed content and approved the final version for publication.

References

  1. Al-Hasani R, McCall JG, Bruchas MR. Exposure to chronic mild stress prevents kappa opioid-mediated reinstatement of cocaine and nicotine place preference. Front Pharmacol. 2013;4:96. doi: 10.3389/fphar.2013.00096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aldao A, Nolen-Hoeksema S, Schweizer S. Emotion-regulation strategies across psychopathology: A meta-analytic review. Clin Psychol Rev. 2010;30:217–237. doi: 10.1016/j.cpr.2009.11.004. [DOI] [PubMed] [Google Scholar]
  3. Bossert JM, Stern AL, Theberge FR, Cifani C, Koya E, Hope BT, Shaham Y. Ventral medial prefrontal cortex neuronal ensembles mediate context-induced relapse to heroin. Nat Neurosci. 2011;14:420–422. doi: 10.1038/nn.2758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bradley BP, Phillips G, Green L, Gossop M. Circumstances surrounding the initial lapse to opiate use following detoxification. Br J Psychiatry. 1989;154:354–359. doi: 10.1192/bjp.154.3.354. [DOI] [PubMed] [Google Scholar]
  5. Cai DJ, Aharoni D, Shuman T, Shobe J, Biane J, Song W, Wei B, Veshkini M, La-Vu M, Lou J, Flores SE, Kim I, Sano Y, Zhou M, Baumgaertel K, Lavi A, Kamata M, Tuszynski M, Mayford M, Golshani P, Silva AJ. A shared neural ensemble links distinct contextual memories encoded close in time. Nature. 2016;534:115–118. doi: 10.1038/nature17955. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Caldwell EE, Riccio DC. Alcohol self-administration in rats: Modulation by temporal parameters related to repeated mild social defeat stress. Alcohol. 2010;44:265–274. doi: 10.1016/j.alcohol.2010.02.012. [DOI] [PubMed] [Google Scholar]
  7. Carelli RM, Williams JG, Hollander JA. Basolateral amygdala neurons encode cocaine self-administration and cocaine-associated cues. J Neurosci. 2003;23:8204–8211. doi: 10.1523/JNEUROSCI.23-23-08204.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Ciocchi S, Herry C, Grenier F, Wolff SB, Letzkus JJ, Vlachos I, Ehrlich I, Sprengel R, Deisseroth K, Stadler MB, Muller C, Luthi A. Encoding of conditioned fear in central amygdala inhibitory circuits. Nature. 2010;468:277–282. doi: 10.1038/nature09559. [DOI] [PubMed] [Google Scholar]
  9. Covington HE, 3rd, Maze I, Sun H, Bomze HM, DeMaio KD, Wu EY, Dietz DM, Lobo MK, Ghose S, Mouzon E, Neve RL, Tamminga CA, Nestler EJ. A role for repressive histone methylation in cocaine-induced vulnerability to stress. Neuron. 2011;71:656–670. doi: 10.1016/j.neuron.2011.06.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Covington HE, 3rd, Miczek KA. Repeated social-defeat stress, cocaine or morphine. Effects on behavioral sensitization and intravenous cocaine self-administration "binges". Psychopharmacology (Berl) 2001;158:388–398. doi: 10.1007/s002130100858. [DOI] [PubMed] [Google Scholar]
  11. Cruz FC, Quadros IM, Hogenelst K, Planeta CS, Miczek KA. Social defeat stress in rats: escalation of cocaine and "speedball" binge self-administration, but not heroin. Psychopharmacology (Berl) 2011;215:165–175. doi: 10.1007/s00213-010-2139-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Funk D, Harding S, Juzytsch W, Le AD. Effects of unconditioned and conditioned social defeat on alcohol self-administration and reinstatement of alcohol seeking in rats. Psychopharmacology (Berl) 2005;183:341–349. doi: 10.1007/s00213-005-0194-1. [DOI] [PubMed] [Google Scholar]
  13. Gozzi A, Jain A, Giovannelli A, Bertollini C, Crestan V, Schwarz AJ, Tsetsenis T, Ragozzino D, Gross CT, Bifone A. A neural switch for active and passive fear. Neuron. 2010;67:656–666. doi: 10.1016/j.neuron.2010.07.008. [DOI] [PubMed] [Google Scholar]
  14. Heilig M, Epstein DH, Nader MA, Shaham Y. Time to connect: bringing social context into addiction neuroscience. Nat Rev Neurosci. 2016;17:592–599. doi: 10.1038/nrn.2016.67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Herry C, Ciocchi S, Senn V, Demmou L, Muller C, Luthi A. Switching on and off fear by distinct neuronal circuits. Nature. 2008;454:600–606. doi: 10.1038/nature07166. [DOI] [PubMed] [Google Scholar]
  16. Hong W, Kim DW, Anderson DJ. Antagonistic control of social versus repetitive self-grooming behaviors by separable amygdala neuronal subsets. Cell. 2014;158:1348–1361. doi: 10.1016/j.cell.2014.07.049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jasnow AM, Davis M, Huhman KL. Involvement of central amygdalar and bed nucleus of the stria terminalis corticotropin-releasing factor in behavioral responses to social defeat. Behav Neurosci. 2004;118:1052–1061. doi: 10.1037/0735-7044.118.5.1052. [DOI] [PubMed] [Google Scholar]
  18. Kim J, Pignatelli M, Xu S, Itohara S, Tonegawa S. Antagonistic negative and positive neurons of the basolateral amygdala. Nat Neurosci. 2016;19:1636–1646. doi: 10.1038/nn.4414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kronenberg LM, Goossens PJ, van Busschbach J, van Achterberg T, van den Brink W. Coping styles in substance use disorder (SUD) patients with and without co-occurring attention deficit/hyperactivity disorder (ADHD) or autism spectrum disorder (ASD) BMC Psychiatry. 2015;15:159. doi: 10.1186/s12888-015-0530-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Li X, Zeric T, Kambhampati S, Bossert JM, Shaham Y. The central amygdala nucleus is critical for incubation of methamphetamine craving. Neuropsychopharmacology. 2015;40:1297–1306. doi: 10.1038/npp.2014.320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Mantsch JR, Baker DA, Funk D, Le AD, Shaham Y. Stress-Induced Reinstatement of Drug Seeking: 20 Years of Progress. Neuropsychopharmacology. 2016;41:335–356. doi: 10.1038/npp.2015.142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Manvich DF, Stowe TA, Godfrey JR, Weinshenker D. A Method for Psychosocial Stress-Induced Reinstatement of Cocaine Seeking in Rats. Biol Psychiatry. 2015 doi: 10.1016/j.biopsych.2015.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Marlatt GA, George WH. Relapse prevention: introduction and overview of the model. Br J Addict. 1984;79:261–273. doi: 10.1111/j.1360-0443.1984.tb00274.x. [DOI] [PubMed] [Google Scholar]
  24. McDonald AJ. Topographical organization of amygdaloid projections to the caudatoputamen, nucleus accumbens, and related striatal-like areas of the rat brain. Neuroscience. 1991;44:15–33. doi: 10.1016/0306-4522(91)90248-m. [DOI] [PubMed] [Google Scholar]
  25. McDonald AJ. Projection neurons of the basolateral amygdala: a correlative Golgi and retrograde tract tracing study. Brain Res Bull. 1992;28:179–185. doi: 10.1016/0361-9230(92)90177-y. [DOI] [PubMed] [Google Scholar]
  26. McDonald AJ, Augustine JR. Localization of GABA-like immunoreactivity in the monkey amygdala. Neuroscience. 1993;52:281–294. doi: 10.1016/0306-4522(93)90156-a. [DOI] [PubMed] [Google Scholar]
  27. McFarland K, Kalivas PW. The circuitry mediating cocaine-induced reinstatement of drug-seeking behavior. J Neurosci. 2001;21:8655–8663. doi: 10.1523/JNEUROSCI.21-21-08655.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Pare D, Smith Y. Distribution of GABA immunoreactivity in the amygdaloid complex of the cat. Neuroscience. 1993;57:1061–1076. doi: 10.1016/0306-4522(93)90049-l. [DOI] [PubMed] [Google Scholar]
  29. Paxinos G, Watson C. The rat brain in Stereotaxic Coordinates. 4. 2007. [DOI] [PubMed] [Google Scholar]
  30. Pfarr S, Meinhardt MW, Klee ML, Hansson AC, Vengeliene V, Schonig K, Bartsch D, Hope BT, Spanagel R, Sommer WH. Losing Control: Excessive Alcohol Seeking after Selective Inactivation of Cue-Responsive Neurons in the Infralimbic Cortex. J Neurosci. 2015;35:10750–10761. doi: 10.1523/JNEUROSCI.0684-15.2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Phelps EA, LeDoux JE. Contributions of the amygdala to emotion processing: from animal models to human behavior. Neuron. 2005;48:175–187. doi: 10.1016/j.neuron.2005.09.025. [DOI] [PubMed] [Google Scholar]
  32. Sinha R. How does stress increase risk of drug abuse and relapse? Psychopharmacology (Berl) 2001;158:343–359. doi: 10.1007/s002130100917. [DOI] [PubMed] [Google Scholar]
  33. Sinha R, Fuse T, Aubin LR, O’Malley SS. Psychological stress, drug-related cues and cocaine craving. Psychopharmacology (Berl) 2000;152:140–148. doi: 10.1007/s002130000499. [DOI] [PubMed] [Google Scholar]
  34. Sinha R, Garcia M, Paliwal P, Kreek MJ, Rounsaville BJ. Stress-induced cocaine craving and hypothalamic-pituitary-adrenal responses are predictive of cocaine relapse outcomes. Arch Gen Psychiatry. 2006;63:324–331. doi: 10.1001/archpsyc.63.3.324. [DOI] [PubMed] [Google Scholar]
  35. Soderpalm A, Nikolayev L, de Wit H. Effects of stress on responses to methamphetamine in humans. Psychopharmacology (Berl) 2003;170:188–199. doi: 10.1007/s00213-003-1536-5. [DOI] [PubMed] [Google Scholar]
  36. van Erp AM, Miczek KA. Persistent suppression of ethanol self-administration by brief social stress in rats and increased startle response as index of withdrawal. Physiol Behav. 2001;73:301–311. doi: 10.1016/s0031-9384(01)00458-9. [DOI] [PubMed] [Google Scholar]
  37. Wagner EF, Myers MG, McIninch JL. Stress-coping and temptation-coping as predictors of adolescent substance use. Addict Behav. 1999;24:769–779. doi: 10.1016/s0306-4603(99)00058-1. [DOI] [PubMed] [Google Scholar]
  38. Walker FR, Masters LM, Dielenberg RA, Day TA. Coping with defeat: acute glucocorticoid and forebrain responses to social defeat vary with defeat episode behaviour. Neuroscience. 2009;162:244–253. doi: 10.1016/j.neuroscience.2009.04.041. [DOI] [PubMed] [Google Scholar]
  39. Wallace BC. Psychological and environmental determinants of relapse in crack cocaine smokers. J Subst Abuse Treat. 1989;6:95–106. doi: 10.1016/0740-5472(89)90036-6. [DOI] [PubMed] [Google Scholar]
  40. Wood SK, Walker HE, Valentino RJ, Bhatnagar S. Individual differences in reactivity to social stress predict susceptibility and resilience to a depressive phenotype: role of corticotropin-releasing factor. Endocrinology. 2010;151:1795–1805. doi: 10.1210/en.2009-1026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zou S, Funk D, Shram MJ, Le AD. Effects of stressors on the reinforcing efficacy of nicotine in adolescent and adult rats. Psychopharmacology (Berl) 2014;231:1601–1614. doi: 10.1007/s00213-013-3314-3. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp info
NIHMS982175-supplement-Supp_info.docx (397.3KB, docx)

RESOURCES