Induction of activity synchronization among primed hippocampal neurons out of random dynamics is key for trace memory formation and retrieval

Yuxin Zhou, Liyan Qiu, HaiYing Wang, Xuanmao Chen

TL;DR

Researchers have discovered that memory formation in the hippocampus isn't just about "turning on" neurons; it’s about a specific subset of "Primed Neurons" shifting from chaotic, random firing to a highly synchronized rhythm. Using deep-brain imaging in freely behaving mice, the study shows that this transition from noise to harmony is the physiological hallmark of both creating and retrieving a memory.

What Problem Does This Paper Address?

The "engram"—the physical trace of a memory in the brain—has been a holy grail of neuroscience for decades. While we know that specific ensembles of neurons encode memories, we haven't fully understood the dynamics of how they get recruited.

Most previous studies focused on "contextual" memory (remembering a place), but this paper tackles "trace" memory. Trace memory is more complex: it requires an animal to associate two events (like a tone and a shock) that are separated by a gap of silence. This requires the brain to maintain a "trace" of the first stimulus after it has ended. The researchers wanted to know: Do these neurons start from a silent state? How do they change as the animal learns? And what actually happens when the animal "remembers"?

Methods

The team employed a sophisticated setup combining trace fear conditioning with in vivo calcium imaging.

  • The Tech: They used Fiber-Optic Confocal (FOC) fluorescence endomicroscopy. Unlike traditional bulky microscopes, this uses a tiny 300-nm micro-probe that allows mice to move, jump, and explore freely while the researchers record from deep-brain regions (the hippocampal CA1) for up to seven hours.
  • The Indicator: They used GCaMP6m, a genetically encoded calcium indicator that glows when neurons fire, providing a proxy for electrical activity.
  • The Paradigm: Mice heard a tone, followed by a 30-second "trace" interval of silence, followed by a mild foot-shock. They repeated this over several cycles to see how neuronal behavior evolved as the mouse began to "freeze" (a sign of fear/memory).

Datasets

This was an experimental study involving 15 mice (9 male, 6 female). The researchers recorded activity from 50 to 200 neurons per animal, creating a longitudinal dataset of neuronal firing patterns across habituation, training, sleep, and recall phases. No external public datasets were used; the findings are based on these primary physiological recordings.

Key Insights and Ideas

  • The 10% Rule (Lognormal Distribution): Brain activity isn't a bell curve. The researchers found a "right-skewed lognormal" distribution. About 70% of neurons are "Silent," doing almost nothing even during training. Only a small elite—roughly 10%, termed "Primed Neurons"—are highly active and sensitive enough to become the basis of a memory.
  • From Chaos to Sync: Before training, Primed Neurons fire randomly. As the mouse learns the association, these neurons begin to fire in lockstep. This synchronization is the "aha!" moment of memory formation.
  • The "On-Phase" Calcium Surge: Synchronization isn't just a pattern; it’s a power-up. When neurons fire together in a "bursting" pattern, it triggers a 3-fold increase in calcium entry into the cells. This massive calcium influx is likely the signal that triggers gene expression to "solidify" the memory (consolidation).
  • Recall is Partial Sync: When the mouse hears the tone hours later and remembers the danger, the same Primed Neurons resynchronize. If the synchronization fails, the mouse doesn't freeze—even if the neurons are active. This suggests that synchrony, not just activation, is the key to retrieval.

Key Results

  • Recruitment: After training, the percentage of neurons responding to the tone jumped from a negligible 0.6% to 7.4%.
  • Behavioral Correlation: The emergence of activity synchronization in Primed Neurons perfectly coincided with the appearance of freezing behaviors in the mice.
  • Stability: Primed Neurons remained "primed" (highly active) for at least 5 hours across different contexts and even through short naps, showing they are a stable population ready for memory tasks.
  • Signal-to-Noise Ratio (SNR): The SNR of the Primed Neuron population increased significantly during training, indicating a more reliable transmission of the "memory signal" within the network.

Limitations and Future Work

While the FOC system is great for deep-brain access in moving animals, it has limited spatial (3 µm) and temporal (11.7 - 40 Hz) resolution. It cannot resolve individual action potentials (spikes), only the broader calcium waves they cause.

The authors also noted that while they can track the same neurons for several hours, they couldn't maintain the same field of view across multiple days. Future work will likely focus on how these "Primed Neurons" are selected in the first place—why are some cells born to be superstars while others stay silent? Additionally, developing better noise-reduction algorithms for this type of endoscopic imaging remains a technical challenge.

Summary updated

Abstract

Memory is thought to be encoded by sparsely distributed neuronal ensembles in memory-related regions. However, it is unclear how memory-eligible neurons react during learning to encode trace fear memory and how they retrieve a memory. We implemented a fiber-optic confocal fluorescence endomicroscope to directly visualize calcium dynamics of hippocampal CA1 neurons in freely behaving mice subjected to trace fear conditioning. Here we report that the overall activity levels of CA1 neurons showed a right-skewed lognormal distribution, with a small portion of highly active neurons (termed Primed Neurons) filling the long-tail. Repetitive training induced Primed Neurons to shift from random activity to well-tuned synchronization. The emergence of activity synchronization coincided with the appearance of mouse freezing behaviors. In recall, a partial synchronization among the same subset of Primed Neurons was induced from random dynamics, which also coincided with mouse freezing behaviors. Additionally, training-induced synchronization facilitated robust calcium entry into Primed Neurons. In contrast, most CA1 neurons did not respond to tone and foot shock throughout the training and recall cycles. In conclusion, Primed Neurons are preferably recruited to encode trace fear memory and induction of activity synchronization among Primed Neurons out of random dynamics is critical for trace memory formation and retrieval.
BibTeX
@article{zhou2020induction,
  author = {Yuxin Zhou and Liyan Qiu and HaiYing Wang and Xuanmao Chen},
  title = {{Induction of activity synchronization among primed hippocampal neurons out of random dynamics is key for trace memory formation and retrieval}},
  journal = {The FASEB Journal},
  year = {2020},
  doi = {10.1096/fj.201902274r},
}
Figure