The effects of diazepam on fear extinction in nulliparous and primiparous female rats

Benzodiazepines undermine the success of exposure therapy in humans with anxiety disorders, and impair the long-term memory of fear extinction (the laboratory basis of exposure therapy) in rodents. However, most rodent studies on fear extinction and benzodiazepines have been conducted in male rodents. In female rodents, the estrous cycle influences the consolidation of fear extinction memories and sensitivity to benzodiazepines. In addition, pregnancy leads to long-term changes in the neurobiological, hormonal, and behavioural features of fear extinction, as well as the responsivity to benzodiazepines. Therefore, the present experiments examined the impact of benzodiazepines on fear extinction in female rats with and without reproductive experience. Age-matched nulliparous (no reproductive experience) and primiparous (one prior reproductive experience; tested one-month post-weaning) rats received fear conditioning to a discrete cue. The next day, rats were administered the benzodiazepine diazepam (2 mg/kg, s.c), or vehicle, prior to or immediately after extinction training. Rats were then tested the next day, drug free, for extinction retention. Similar to previous findings in males, diazepam impaired extinction retention in both nulliparous and primiparous rats when administered either pre-or post-extinction training. These findings may have potential clinical implications as they suggest that benzodiaze-pine use in conjunction with exposure therapy may undermine long-term treatment success in women with and without reproductive experience, although this remains to be tested in human populations. Moreover, these findings are theoretically important when considered in light of previous studies showing dissociable mechanisms of fear extinction in females pre-versus post-pregnancy.


Introduction
Benzodiazepines (positive allosteric modulators of GABA A receptors) are widely prescribed to treat anxiety disorders, with 20 % of patients reporting usage over a 12-month period (Agarwal and Landon, 2019).However, benzodiazepines undermine exposure therapy, the first-line psychological treatment for anxiety and stress-related disorders, as demonstrated in patients with post-traumatic stress disorder and specific phobias (Rothbaum et al., 2014;Westra et al., 2002).The basis of exposure therapy is fear extinction, the learned inhibition of conditioned fear.In rodents, benzodiazepines impair long-term extinction retention when administered before extinction training (Bouton et al., 1990;Hart et al., 2009;Hart et al., 2014), likely driven by benzodiazepine-mediated suppression of activity in the amygdala, which prevents inhibitory learning (Hart et al., 2009).Benzodiazepines also impair extinction memory consolidation when administered immediately after extinction training (Bustos et al., 2009;Franzen et al., 2019).
The majority of rodent studies on benzodiazepines and fear extinction have been conducted in males, and the few studies examining females have used virgins (Bouton et al., 1990;Franzen et al., 2019).Pregnancy leads to long-term changes in neurobiological, hormonal, and behavioural features of fear extinction (Kershaw et al., 2023;Milligan-Saville and Graham, 2016;Pestana et al., 2021;Tang andGraham, 2019a, 2019b), and the sensitivity to benzodiazepines (Pestana and Graham, 2023).Of relevance to the present study, we recently showed that inhibiting the basolateral amygdala via an infusion of a GABA A agonist impaired fear extinction in nulliparous (no reproductive experience) female rats but not primiparous (one reproductive experience) rats (Kershaw et al., 2023).As such, it is possible that benzodiazepines may impair fear extinction in nulliparous, but not primiparous females, as the latter do not appear to require the amygdala to extinguish fear.Identifying the impact of benzodiazepines on fear extinction in females with different reproductive histories may therefore be clinically and theoretically informative.Accordingly, the present study assessed the impact of pre-and post-extinction training administration of diazepam (a benzodiazepine) on extinction retention in nulliparous and primiparous female rats.

Subjects
Experimentally naïve naturally cycling 6-month-old nulliparous (N = 48) and primiparous (N = 50) female Sprague-Dawley rats were obtained from the Animal Resources Centre (ARC), Australia.Breeding and weaning of offspring was conducted by ARC prior to their arrival at UNSW, as previously described (Pestana et al., 2021).The breeding procedure for primiparous rats was controlled between drug conditions.Testing occurred one-month post-weaning, when estrous cycle had recommenced (Milligan-Saville and Graham, 2016).Rats were housed in groups of 5-8 in plastic boxes (67x30x22cm) with corncob bedding, in a 20-22 • C colony room maintained on a 12 h light-dark cycle (lights on 0700 h).Food and water were available ad libitum.All procedures were approved by the Animal Care and Ethics Committee at UNSW.

Estrous cycle
Vaginal swabs were conducted daily (0900 h-1100 h) to determine estrous cycle phase (Graham and Daher, 2016).All rats received fear conditioning during diestrus, extinction training during proestrus, and extinction retention during estrus.No rats were excluded due to irregular cycling.

Drug administration
Rats received a subcutaneous injection (1.0 ml/kg of body weight) of diazepam (Troy Laboratories Australia) in vehicle (40 % propylene glycol, 10 % ethanol, 40 % water), or vehicle.A dose of 2.0 mg/kg body weight was chosen based on our pilot study in which this dose produced acute anxiolytic effects on the elevated plus maze, without causing sedative effects.

Apparatus
Two sets of experimental chambers served as distinct contexts for conditioning (Context A) and extinction procedures (Context B) as described in Graham and Daher (2016).The CS was a 62 dB 10 s white noise delivered through a speaker on the wall of each chamber, and the US was a 0.4 mA, 1.0 s footshock delivered through the floor.A computer running Med Associates Med-PC IV controlled the CS and US presentations.

Handling and pre-exposure
Rats were handled for 5 min for three consecutive days.After handling on the final two days, rats were individually placed in Context A for 10 min.

Fear conditioning
Rats were placed in Context A and after a 2 min adaptation period the CS was presented and co-terminated with the US.Rats received three CS-US pairings (intertrial interval 85-135 s, with an average of 110 s).

Extinction training
Rats underwent extinction training 24 h after conditioning in Context B. After a 2 min adaptation period, rats received 30 x CS presentations (intertrial interval 10s).Diazepam was injected 30 min prior to extinction training in Experiment 1, and immediately after extinction training in Experiment 2.

Extinction retention
Rats were returned to Context B 24 h after extinction training.Following a 1-min adaptation period, rats received 15 x CS presentations (intertrial interval 10s).

Scoring
Conditioned fear was measured using freezing, defined as the absence of movement except that related to respiration (Fanselow, 1980).Rats were manually scored as "freezing" or "not freezing" every three seconds using a time-sampling procedure.A percentage of observed freezing was calculated for each rat to determine the proportion of total observations spent freezing.20 % of the data was crossscored, and the inter-observer reliability was very high (r = 0.99; p < .001).

Statistical analysis
The data for extinction training and extinction retention were analysed in blocks, with each an average of five CS trials.Two-way ANOVAs with the between-subjects factors of reproductive status (nulliparous or primiparous) and drug (diazepam or vehicle) were used to assess group differences in pre-CS freezing prior to fear conditioning, extinction training, and extinction retention.Two-way ANOVAs with repeated measures were used to assess group differences in CS-elicited freezing during fear conditioning, extinction training, and extinction retention, with the Greenhouse-Geisser correction used if the assumption of sphericity was violated (as indicated by Mauchly's Test of Sphericity).Homogeneity of variance was confirmed via Levene's Statistic.Normality was confirmed via Shapiro-Wilk test and inspection of Q-Q plots.Effect sizes were determined using η2 for overall ANOVA tests (i.e., small effect = 0.01, medium effect = 0.06, large effect = 0.14), and Cohen's d was used for pairwise comparisons (small effect = 0.2, medium effect = 0.5, large effect = 0.8).Data for the last trial of conditioning for one primiparous-diazepam rat was missing in Experiment 1.

Discussion
This study demonstrated that diazepam administered before or immediately after extinction training resulted in higher freezing the next day during a drug-free extinction retention test in nulliparous and primiparous rats.Different mechanisms may have produced impairments in extinction retention dependent on the timing of administration.For instance, pre-extinction training administration of diazepam may have caused state-dependent extinction learning, such that diazepam-treated rats exhibited a return of fear during extinction retention because their internal drug-free state was different to the diazepam-induced state during extinction learning (Bouton et al., 1990).Alternatively, the anxiolytic effects of pre-extinction diazepam may have attenuated prediction error (discrepancy between predicted and actual threat) required  for extinction (Hart et al., 2009;Hart et al., 2014).By contrast, postextinction training administration of diazepam likely impaired extinction retention by disrupting extinction consolidation, an effect that is independent of state-dependent learning or prediction error.To assess for state-dependent learning, future studies should administer diazepam both pre-extinction training and pre-extinction retention to determine whether this abolishes the impairment in extinction retention caused by pre-extinction diazepam alone (Bouton et al., 1990).Moreover, future studies should test multiple doses of diazepam on fear extinction given that lower doses of diazepam elicit divergent anxiolytic effects in nulliparous and primiparous rats (Pestana and Graham, 2023).
Taken together, these findings provide additional support that benzodiazepines may undermine exposure therapy success in both males and females (irrespective of reproductive status) via disruptions to longterm extinction memories.Although current guidelines stipulate benzodiazepines should be used as second-or third-line treatments only when indicated (National Institute for Health and Care Excellence, 2014), they do not discuss the potential detrimental ramifications of combining evidence-based psychotherapy with benzodiazepines.It may be beneficial to incorporate such information in the guidelines so that patients do not invest time and resources into two treatments, one of which may undermine the other.
In addition to these clinical implications, our finding that diazepam had equivalent impact on fear extinction in nulliparous and primiparous rats has theoretical relevance when considered through the lens of our past research, which has documented striking differences in the neurobiological features of fear extinction in nulliparous and primiparous females, despite reproductive experience producing a minimal overall effect (Tang and Graham, 2020).For instance, unlike nulliparous rats, fear extinction in primiparous rats is relapse resistant (Milligan-Saville and Graham, 2016), independent of NMDA receptors (Tang andGraham, 2019a, 2019b) or hormone modulation (Milligan-Saville and Graham, 2016;Pestana et al., 2021), and does not require activity in the basolateral amygdala (Kershaw et al., 2023).These findings have led us to hypothesise that extinction in primiparous rats may involve erasure of the original CS-US conditioning memory rather than new learning of a CS-no US extinction memory (Kershaw et al., 2023).Indeed, we predicted that diazepam may not impair extinction in primiparous rats because of the apparent lack of involvement of the amygdala in primiparous fear extinction; although diazepam also targets other brain regions (e.g., hippocampus, Heldt and Ressler, 2006) that could be involved in primiparous fear extinction.The current findings in primiparous rats, which ran counter to our predictions, are therefore quite illuminating.First, that pre-extinction training diazepam impaired extinction retention reinforces the notion that extinction in primiparous rats is an active process, as indicated by our previous finding that primiparous rats are not simply forgetting the CS-US association (Kershaw et al., 2023).Second, that post-extinction training diazepam impaired extinction retention suggests that primiparous rats undergo a consolidation period following fear extinction acquisition.The contents of what is being consolidated is currently unclear.E.g., the lack of relapse or involvement of NMDA receptors in primiparous fear extinction suggests it is unlikely that new inhibitory learning takes place, yet whether the alternative possibility, erasure, requires consolidation is unknown.A third account is that primiparous rats may engage in updating the original CS-US conditioning memory, a process that requires reconsolidation, and which may depend on different mechanisms to new learning (Ferrara et al., 2023).Clearly, our understanding of the mechanisms of fear extinction in primiparous rats is only just beginning, although we are making headway.Given that anxiety and stress-related disorders are twice as common in women compared to men (COVID-19 Mental Disorders Collaborators, 2021), and that 85 % of women become mothers before the age of 44 (Martinez et al., 2018), further investigation of fear extinction in nulliparous and primiparous rats is warranted.

Fig. 1 .
Fig. 1.In Experiment 1, nulliparous (vehicle; n = 10, diazepam; n = 9) and primiparous (vehicle; n = 10, diazepam; n = 10) rats underwent fear conditioning on day 1, were administered diazepam or saline 30 min prior to fear extinction on day 2, and were tested for extinction retention on day 3. A) Mean (± SEM) pre-CS and CSelicited freezing during fear conditioning.B) Mean (± SEM) pre-CS and CS-elicited freezing during fear extinction.The data are presented as 6 blocks of trials, each representing an average of five CS trials.C) Mean (± SEM) pre-CS and CS-elicited freezing during extinction retention.The data are presented as 3 blocks of trials, each representing an average of five CS trials.D) Mean (± SEM) and individual CS-elicited freezing during extinction retention.* = significant difference, p < .05).

Fig. 2 .
Fig. 2. In Experiment 2, nulliparous (vehicle; n = 13, diazepam; n = 13) and primiparous rats (vehicle; n = 15, diazepam; n = 15) underwent fear conditioning on day 1, fear extinction immediately followed by administration of diazepam or saline on day 2, and were tested for extinction retention on day 3. A) Mean (± SEM) pre-CS and CS-elicited freezing during fear conditioning.B) Mean (± SEM) pre-CS and CS-elicited freezing during fear extinction.The data are presented as 6 blocks of trials, each representing an average of five CS trials.C) Mean (± SEM) pre-CS and CS-elicited freezing during extinction retention.The data are presented as 3 blocks of trials, each representing an average of five CS trials.D) Mean (± SEM) and individual CS-elicited freezing during extinction retention.* = significant difference, p < .05).