🧠 neuroscience

Stress-Enhanced Fear Learning (SEFL) is Associated with Enhanced Reactivation of Fear Engrams in Ventral but not Dorsal Dentate Gyrus

This study demonstrates that stress-enhanced fear learning (SEFL) potentiates fear memory by specifically increasing the reactivation of fear engram cells in the ventral (caudal) dentate gyrus, while leaving cell numbers and reactivation in the dorsal (rostral) dentate gyrus unaffected.

Original authors: Paredes, D., Drew, M. R.

Published 2026-03-13

📖 5 min read🧠 Deep dive

CC0 1.0

Original authors: Paredes, D., Drew, M. R.

Original paper dedicated to the public domain under CC0 1.0 (https://creativecommons.org/publicdomain/zero/1.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The Big Picture: Why Do Some Scars Stick?

Imagine your brain is a massive library. Every time you learn something new, a librarian writes a note on a card and files it away. This is how we form memories. Usually, if you have a bad experience, you remember it, but you can eventually calm down and move on.

However, for some people (and animals in this study), a traumatic event acts like a "super-glue." It makes their brain stick to fear memories so tightly that even a tiny, unrelated reminder later on causes a massive panic attack. This is a key feature of PTSD (Post-Traumatic Stress Disorder).

Scientists call this phenomenon Stress-Enhanced Fear Learning (SEFL). They wanted to know: Where in the brain does this "super-glue" happen, and how does it work?

The Experiment: The "Tagging" Game

To solve this mystery, the researchers used mice and a special genetic trick. Think of it like a high-tech highlighter pen.

The Setup: They used mice that could have their brain cells "tagged" with a glowing green light (called zsGreen) if those cells were active at a specific moment.
The Stress: First, some mice were put in a room and given a few mild, scary electric shocks (the "Stress" group). Others just sat in the room without shocks (the "Control" group).
The New Fear: Five days later, all the mice were put in a different room and given just one mild shock.
- The Twist: Immediately after this single shock, the researchers injected the mice with a chemical that "locked" the green tag onto the brain cells that were active during that one shock.
The Test: A few days later, the mice were put back in that second room. The researchers watched how much they froze (a sign of fear).
The Autopsy: Finally, they looked at the mice's brains under a microscope to see what the "green tags" looked like.

The Results: The "Two-Part" Brain

The researchers looked at two specific neighborhoods in the mouse's hippocampus (the brain's memory center). Let's call them The Library (Rostral/Dorsal) and The Emotion Center (Caudal/Ventral).

1. The Library (Rostral Dentate Gyrus)

What it does: This part is like a precise mapmaker. It remembers the exact details of a place (the shape of the walls, the smell of the floor).
What happened: The researchers checked the "green tags" here.
- Did the stressed mice have more tags? No.
- Did the stressed mice have more "active" cells when they remembered the fear? No.
The Takeaway: The stress didn't change how the brain stored the facts of the fear in this area. The map was the same for everyone.

2. The Emotion Center (Caudal Dentate Gyrus)

What it does: This part is like a volume knob for feelings. It connects memory to anxiety and emotional intensity. It doesn't care about the exact shape of the room; it cares about how scary the room feels.
What happened:
- The Tags: Just like in the Library, the number of green tags (the original memory) was the same for stressed and non-stressed mice.
- The Reactivation: This is where the magic happened. When the mice were tested later, the researchers looked to see how many of those green-tagged cells lit up again (turned on) during the fear test.
- The Shock: In the stressed mice, the "Emotion Center" cells lit up much brighter and more frequently than in the non-stressed mice.
The Takeaway: The stress didn't create more memory cells, but it made the existing memory cells super-sensitive. When the mouse thought about the fear, the "Emotion Center" screamed "DANGER!" much louder than usual.

The Analogy: The Volume Knob vs. The Recording

Imagine you have a recording of a scary sound.

The Non-Stressed Mouse: When it hears the sound, the volume is set to a normal level. It remembers the sound, but it's not overwhelming.
The Stressed Mouse: The stress didn't change the recording itself (the memory is the same). Instead, the stress turned the volume knob up to 11. Now, when the mouse hears the sound, the noise is deafening and terrifying, even though the recording hasn't changed.

Why Does This Matter?

The study found that stress specifically targets the Caudal (Ventral) part of the brain—the "Emotion Center"—to turn up the volume on fear. It leaves the "Library" (Rostral/Dorsal) alone.

This explains why people with PTSD often react so strongly to things that shouldn't be scary. Their brain isn't necessarily remembering the event differently; it's just reacting to the memory with way too much emotional intensity.

In short: Stress doesn't write a new, scarier story in your brain. It just turns the volume of the fear up so high that even a whisper sounds like a scream. And it does this specifically in the part of the brain that handles emotions, not the part that handles facts.

1. Problem Statement

Traumatic stress often leads to maladaptive fear responses, such as those seen in Post-Traumatic Stress Disorder (PTSD), characterized by heightened sensitivity and generalization of fear. The Stress-Enhanced Fear Learning (SEFL) paradigm models this by showing that prior exposure to severe stress (footshocks) sensitizes rodents to subsequent fear conditioning in a novel context.

While it is established that the hippocampus is required for SEFL, the specific neural mechanisms remain unclear. Specifically, it is unknown how prior stress alters the encoding and reactivation of fear memory engrams (sparse neural ensembles) within the Dentate Gyrus (DG) along its long axis. The DG exhibits functional gradients: the rostral (dorsal) DG is associated with fine-grained spatial cognition, while the caudal (ventral) DG is linked to emotional processing and anxiety. This study aims to determine if stress differentially affects the size and reactivation of fear ensembles in these distinct sub-regions.

2. Methodology

The study utilized a combination of behavioral paradigms, transgenic mouse models, and immunohistochemistry.

Subjects: Adult male and female FosTRAP2/Ai6 mice. These mice express a Cre-ERT2 fusion protein under the Fos promoter and a ZsGreen reporter. This allows for the permanent tagging of neurons active during a specific time window with ZsGreen upon administration of 4-hydroxytamoxifen (4-OHT).
Experimental Design (SEFL Paradigm):
- Day 1 (Stress/Context A): Mice were exposed to either 10 scrambled footshocks (1 mA, 2s) in Context A (Stress group) or equivalent context exposure without shock (No-Stress group).
- Days 2–4: Extinction training in Context A (20 min sessions).
- Day 5 (Conditioning/Context B): Mice underwent 1-shock Contextual Fear Conditioning (CFC) in a novel Context B (0.75 mA shock). Control groups received context exposure without shock.
- Tagging: Immediately after the CFC session, mice received an intraperitoneal injection of 4-OHT (55 mg/kg) to permanently tag neurons active during fear acquisition with ZsGreen. Mice were then housed in the dark for 72 hours to prevent non-specific tagging.
- Day 8 (Recall): Mice were returned to Context B to test for fear recall (freezing behavior).
- Tissue Collection: Mice were perfused 90 minutes after the recall test to capture immediate early gene (c-Fos) expression associated with memory retrieval.
Histology & Analysis:
- Brains were sectioned to analyze Rostral DG (-1.22 to -1.94 mm from Bregma) and Caudal DG (-3.16 to -3.80 mm from Bregma).
- Immunohistochemistry was performed for ZsGreen (acquisition engram) and c-Fos (recall activation).
- Quantification involved counting:
  1. Total ZsGreen+ cells (Engram size).
  2. Total c-Fos+ cells (Recall activation).
  3. % Reactivation: The percentage of ZsGreen+ cells that also expressed c-Fos.

3. Key Contributions

Regional Specificity: The study provides the first evidence that stress-induced sensitization of fear learning is mediated by specific changes in the caudal (ventral) DG, while the rostral (dorsal) DG remains unaffected regarding engram reactivation.
Mechanism of Sensitization: It demonstrates that stress does not increase the number of neurons recruited during fear learning (engram size) but rather enhances the probability of those specific neurons being reactivated during memory retrieval.
Engram Specificity: The stress effect is specific to fear memories; stress did not enhance the reactivation of ensembles encoding neutral context exposure.

4. Results

Behavioral Findings:

Enhanced Freezing: Stressed mice exhibited significantly higher freezing levels during the recall test in Context B compared to non-stressed mice ( $p=0.03$ ), confirming the SEFL phenotype.
No Generalization: There was no difference in freezing between groups prior to the shock in Context B, indicating that the enhanced fear was not due to generalization between Context A and B.

Neural Findings (Rostral DG):

Engram Size: Stress had no effect on the number of ZsGreen+ cells (acquisition) in the rostral DG.
Reactivation: Stress did not significantly alter the percentage of ZsGreen+ cells reactivated (c-Fos+) during recall in the rostral DG.
Conclusion: The rostral DG fear ensemble is not modulated by prior stress in terms of size or reactivation efficiency.

Neural Findings (Caudal DG):

Engram Size: Stress did not change the total number of ZsGreen+ cells in the caudal DG.
Reactivation: There was a significant Stress $\times$ CFC interaction on the percentage of reactivated cells.
- Stressed mice that underwent fear conditioning showed a significantly higher percentage of reactivation of their fear engram cells compared to non-stressed conditioned mice ( $p=0.0198$ ).
- Stress did not enhance reactivation in mice that did not receive fear conditioning (neutral context).
Correlation: Interestingly, the magnitude of reactivation in the caudal DG did not significantly correlate with the level of freezing behavior, suggesting the mechanism may be part of a broader network process.

5. Significance and Discussion

Functional Dissociation: The results support the hypothesis that the hippocampal long axis serves distinct functions. The caudal (ventral) DG, which connects heavily with the amygdala and hypothalamus, appears to be the locus where stress "primes" fear circuits. This aligns with the ventral hippocampus's role in emotional valence and anxiety.
Mechanism of PTSD-like Symptoms: The study suggests that stress does not create more fear neurons but makes the existing fear memory trace hyper-reactive. This hyper-reactivation in the caudal DG may underlie the generalization of fear (a core symptom of PTSD), where neutral stimuli trigger fear responses because the fear engram is too easily reactivated.
Circuit Implications: The findings implicate the Basolateral Amygdala (BLA) as a potential driver, given its known role in SEFL and its strong projections to the ventral hippocampus. Stress may enhance BLA-ventral DG communication, lowering the threshold for fear engram reactivation.
Therapeutic Potential: By identifying the caudal DG as a critical node for stress-enhanced fear, the study suggests that therapeutic interventions targeting this specific sub-region or its inputs (e.g., from the BLA or prefrontal cortex) could potentially mitigate stress-induced fear sensitization without affecting general spatial memory (rostral DG).

In summary, this paper elucidates a precise neural mechanism for stress-induced fear sensitization: prior stress selectively enhances the reactivation of fear engrams in the caudal dentate gyrus, leaving the rostral dentate gyrus and the total size of the engram unaffected.