Hormonal, reproductive, and behavioural predictors of fear extinction recall in female rats
Introduction
Fear extinction refers to the decline in conditioned fear that occurs when subjects are repeatedly presented with a fear conditioned stimulus (CS) in the absence of an aversive outcome (US, e.g., a noise previously paired with a shock) until their fear subsides. A subject's ability to recall extinction can be tested by measuring their fear responses to the CS at a later time point (extinction recall). Low levels of CS-elicited fear are indicative of good extinction recall, while high levels of CS-elicited fear are indicative of poor extinction recall. CS-elicited fear during extinction recall is influenced by a subject's ability to acquire, consolidate and/or retrieve a fear extinction memory. Fear extinction forms the procedural basis of exposure therapy, a highly efficacious and effective treatment for anxiety disorders (Bandelow et al., 2015; Foa and McLean, 2016). Recent evidence also indicates that fear extinction performance measured prior to exposure therapy may predict exposure therapy success (Ball et al., 2017; Craske et al., 2018; Forcadell et al., 2017; Lange et al., 2019; Waters and Pine, 2016). Understanding the factors that influence fear extinction is therefore worthwhile to identify what factors might influence the success of exposure therapy.
To date, the vast majority of animal studies examining fear extinction have exclusively used male subjects. Human studies also often fail to analyse results by sex (S. H. Li and Graham, 2017). This is problematic given that women are twice as likely to be diagnosed with an anxiety disorder compared to men. Moreover, anxiety disorders in women are associated with greater symptom severity, higher rates of comorbidity and a higher burden of disease (Bandelow and Michaelis, 2015; Bekker and van Mens-Verhulst, 2007; McLean et al., 2011). Understanding the factors that influence fear extinction in female subjects may better allow us to optimise exposure therapy outcomes in women. Naturally cycling female rats and women who undergo fear extinction during periods of high endogenous estradiol (the major form of estrogen) show good extinction recall relative to female rats and women that undergo fear extinction during periods of low endogenous estradiol (Glover et al., 2012; Graham and Daher, 2016; Graham and Scott, 2018a; Gruene et al., 2015; Li and Graham, 2016; Milad et al., 2009; Milad et al., 2010; Milligan-Saville and Graham, 2016; Pineles et al., 2016; Rey et al., 2014; Wegerer et al., 2014; White and Graham, 2016; Zeidan et al., 2011). Poor extinction recall is also observed in female rats that experience chronically low levels of estradiol as a result of ovariectomy (Chang et al., 2009; Parrish et al., 2019) or the administration of hormonal contraceptives (Graham and Milad, 2013; Parrish et al., 2019; White and Graham, 2016). Systemic administration of estradiol or selective estrogen receptor agonists pre- or post-extinction training improve extinction recall in rats extinguished during periods of low endogenous estradiol (Chang et al., 2009; Graham and Milad, 2013; Graham and Scott, 2018a, Graham and Scott, 2018b; Maeng et al., 2017; Milad et al., 2009; Zeidan et al., 2011), while systemic administration of an estrogen receptor antagonist pre-extinction training impairs extinction recall (Milad et al., 2009), suggesting that estradiol not only enhances, but is also necessary for fear extinction in female rats.
Notably, however, we have recently shown that the involvement of estradiol in fear extinction may be altered by reproductive experience. In contrast to nulliparous (virgin) rats, primiparous (one prior reproductive experience) rats extinguished during proestrus (i.e., high estradiol phase of estrous cycle) showed comparable levels of extinction recall as those extinguished during metestrus (i.e., low estradiol phase of estrous cycle) (Graham, 2018; Milligan-Saville and Graham, 2016). Notably, these effects persisted for up to three months post-weaning, suggesting that changes in the features of fear extinction following reproductive experience may be enduring. Primiparous rats also appear to differ from nulliparous rats in that they show extinction that appears to be resistant to relapse (Milligan-Saville and Graham, 2016), independent of the activation of N-methyl-d-aspartate (NMDA) receptors (Tang and Graham, 2019b), and resistant to the augmenting effects of pharmacological adjuncts, including d-cycloserine (DCS) and estradiol (Tang and Graham, 2019a). We have also recently demonstrated that female reproductive experience appears to influence the quality of extinction recall, with some (but not all) studies indicating that primiparous (one prior reproductive experience) rats show worse extinction recall relative to nulliparous (virgin) rats. For instance, in one study, we showed that an increased number of extinction trials was required for primiparous rats to exhibit levels of extinction recall that were comparable to that of nulliparous rats (Tang and Graham, 2019b). In another recent study, we found that reproductive experience may alter the relationship between within-session extinction and extinction recall. In this study, CS-elicited freezing during extinction recall was higher among nulliparous rats that showed high levels of freezing at the end of extinction training relative to those that showed low levels of freezing at the end of extinction training. No such difference in extinction recall was found among primiparous rats, suggesting a lack of a relationship between freezing at the end of extinction training and freezing during extinction recall in primiparous rats (Tang and Graham, 2019a).
To our knowledge, no studies have examined the relationship between within-session extinction (e.g., length of training, fear exhibited at the beginning of training, extent of fear reduction achieved) and extinction recall in females, despite this being previously examined reasonably extensively in males. For example, Graham and Richardson (2019) recently examined the relationship between early extinction (i.e., CS-elicited fear exhibited at the beginning of extinction training), late extinction (i.e., CS-elicited fear exhibited at the end of extinction training), and extinction recall in male rats administered either vehicle or fibroblast growth factor-2 (FGF2, a pharmacological enhancer of fear extinction) prior to extinction training. Using a hierarchical multiple regression analysis (MRA), they found that neither early extinction, nor late extinction predicted freezing during extinction recall among rats administered vehicle. However, a positive relationship between late extinction and extinction recall emerged for FGF2-treated rats. Moreover, a number of studies have demonstrated that CS-elicited freezing during extinction recall is lower among adult male rats that exhibit a faster rate of within-session extinction relative to rats that exhibit a slower rate of within-session extinction (Bush et al., 2007; King et al., 2017). It is important to understand the relationship between within-session extinction and extinction recall in females as it may point to potential ways in which we can enhance extinction among females (e.g., by increasing the efficiency or effectiveness of within-session extinction).
As we have now conducted multiple experiments on fear extinction in female rats in our laboratory, in this study we compiled our previously published results from 245 rats in order to achieve strong statistical power to analyse the various factors that predict fear extinction recall in female rats using a hierarchical MRA. Several key variables were included in this analysis: extinction length, reproductive status, estrous phase, early extinction (as defined above), late extinction (as defined above) and extinction rate (i.e., the first extinction trial in which the subject shows 0% CS-elicited freezing). Early extinction was included as a measure of the strength of fear conditioning, late extinction was included as a measure of within-session extinction (i.e., the degree to which CS-elicited fear decreases over the course of extinction training), and extinction rate was included in light of research showing that this variable may be predictive of extinction recall in adult male rats (Bush et al., 2007; King et al., 2017).
Section snippets
Subjects
Data from 245 Sprague-Dawley females rats that had undergone handling and context pre-exposure, conditioning, extinction training and extinction recall in nine previously published experiments (Graham, 2018; Milligan-Saville and Graham, 2016; Tang and Graham, 2019a, Tang and Graham, 2019b) were analysed in this study (see Table 1). Rats were obtained from the Animal Resources Centre, Perth, WA, Australia. Upon arrival, rats were housed in groups of 8–10 in plastic boxes (67 cm long × 30 cm wide
Results
Results of the hierarchical MRA are presented in Table 2. The first model was significant (p = .04), with estrous phase emerging as a significant predictor of extinction recall. Undergoing extinction during metestrus resulted in higher levels of CS-elicited freezing during extinction recall relative to undergoing extinction training during proestrus. The second model was also significant (p < .01), with late extinction, and extinction rate both emerging as significant predictors of extinction
Discussion
In the current study, a hierarchical MRA was performed on data collated from nine previously published experiments to compare the behavioural, hormonal and reproductive predictors of fear extinction in female rats. This analysis revealed that estrous cycle effects on fear extinction may be dependent on reproductive status. While estrous phase predicts extinction recall, and the relationship between early extinction and extinction recall in nulliparous females, it may not predict either of these
Funding and disclosure
This work was supported by grants from the Australian Research Council (DE140100243 and DP180101563) to BMG, and an Australian Postgraduate Award to ST. The authors have no conflicts of interest to declare.
References (51)
- et al.
Anxiety disorders: sex differences in prevalence, degree, and background, but gender-neutral treatment
Gend. Med.
(2007) - et al.
The effects of safety behaviors during exposure therapy for anxiety: critical analysis from an inhibitory learning perspective
Clin. Psychol. Rev.
(2016) - et al.
Optimizing inhibitory learning during exposure therapy
Behav. Res. Ther.
(2008) - et al.
Maximizing exposure therapy: an inhibitory learning approach
Behav. Res. Ther.
(2014) - et al.
Does fear extinction in the laboratory predict outcomes of exposure therapy? A treatment analog study
Int. J Psychophysiol.
(2017) - et al.
Estrogen levels are associated with extinction deficits in women with posttraumatic stress disorder
Biol. Psychiatry
(2012) Postnatal stress is associated with impaired fear conditioning and extinction, and heightened hippocampal fibroblast growth factor 2, in mother rats
Horm. Behav.
(2018)- et al.
Blockade of estrogen by hormonal contraceptives impairs fear extinction in female rats and women
Biol. Psychiatry
(2013) - et al.
Effects of systemic estradiol on fear extinction in female rats are dependent on interactions between dose, estrous phase, and endogenous estradiol levels
Horm. Behav.
(2018) - et al.
Estradiol-induced enhancement of fear extinction in female rats: the role of NMDA receptor activation
Prog. Neuro-Psychopharmacol. Biol. Psychiatry
(2018)