Abstract
Fear and anxiety are adaptive responses to environmental threats, with distinct neural circuits involved in their regulation. When dysregulated, they contribute to disorders such as generalised anxiety disorder (GAD) and post-traumatic stress disorder (PTSD). The infralimbic cortex (IL) plays a central role in the extinction of conditioned fear responses, a process critical for fear behaviour regulation. The bed nucleus of the stria terminalis (BNST) has been recently suggested to be implicated in sustained threat responses, but its role within fear circuits during extinction remains unclear. This study investigated IL-BNST connectivity and extinction-related engram activity in rats using viral tracing and engram-tagging approaches. Freezing behaviour and ultrasonic vocalisations (USVs) were measured as complementary behavioural markers of conditioned freezing response across two cohorts.A pilot study in fluorescent reporter mice was first conducted to validate a dual-viral engram tagging technique. This confirmed that the technique labels activity-dependent neurons, though at lower levels than a genetic approach.
In the tracing cohort, a retrograde virus was injected into the IL following cued-aversive fear conditioning to identify BNST subregions projecting to it. Retrograde expression from the BNST was confirmed locally at the injection site (IL), but no labelled neurons were observed at the BNST subregions, suggesting that BNST-to-IL projections may be sparse or absent. Behaviourally, rats acquired conditioned freezing, but freezing levels were low relative to previous studies. During extinction, rats maintained the freezing levels across trials, and no significant differences were observed between the extinction-expected (15CS) and no extinction-expected (1CS) subgroups. Similarly, spontaneous recovery testing showed no group differences. Moreover, USVs analysis provided additional insight into the conditioned response, however, only a minority of rats emitted 22 kHz alarm calls, and no direct relationship was observed between USV production and freezing behaviour during conditioning nor extinction.
In the tagging cohort, a dual-viral cFos-dependent technique was used to label extinction-related engrams within the IL. More engram labelling was observed in the injected IL hemisphere than in control hemisphere. However, the number of tagged neurons did not directly correlate with levels of freezing during extinction. This suggests that while extinction-related IL ensembles can be identified using this approach, the limited expression of extinction behaviour under this behavioural protocol may have hindered the ability to detect a clear relationship between behavioural output and engram identification.
Taken together, the experiments aimed to characterise how aversive experiences are encoded and extinguished in the brain by mapping IL-BNST connectivity while optimising a dual-viral tagging technique to identify engrams recruited during extinction. Retrograde tracing revealed no direct projections from the BNST to the IL, suggesting that prefrontal regulation during the extinction of sustained fear may be mediated indirectly, possibly via amygdala-BNST pathways, for example. Complementing these anatomical findings, dual-viral tagging arose as a potential technique for identifying engrams in rats, providing a basis for future studies linking ensemble activity with behaviour. Although, behavioural outcomes were limited by sample size or technical issues, these findings advance our understanding of the neural circuits underlying fear extinction and may aid the development of therapeutic interventions for anxiety disorders and PTSD.
| Date of Award | 28 Jan 2026 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Paul Banks (Supervisor) & Emma N Cahill (Supervisor) |
Cite this
- Standard