Liu et al: Hippocampal engram and fear memory recall

Question 1

Neural substrates of memory

Answer 1

•Engram: coined by Richard Semon
•Prior to this paper it was known that:
• Memories are correlated with a set of sparse neuronal activations (over
large timescales)
• These neurons (the memory trace) are also necessary for fear memory
• i.e. if you inhibit their activity, you reduce expression of a certain
behavior
•However, the sufficiency of a memory trace had not yet been
proven

Question 2

The hippocampus as a locus for fear

memory

Answer 2

•The Dentate gyrus (DG) is the input region of the
hippocampus
•Exposure to a context elicits sparse (2-4% of granule cells), specific DG activation
•DG encodes context (according to authors)
•Good target for manipulating and testing the engram

Question 3

Activating an engram

Answer 3

•Goal: Want to ‘tag’ a set of cells
that are activated with ChR2
• c-fos: immediate early gene (IEG) that is expressed during neuronal activity (promoter)
• tTA: drives expression of genes downstream of TRE when doxycycline is not present
• ChR2-EYFP: enhanced yellow fluorescent protein

•This is called an ‘off-Dox’ system
•We get activity-dependent labeling of neurons within a specific timeframe with ChR2
. In the absence of Dox, training-induced neuronal activity selectively labels active
c-Fos-expressing DG neurons with ChR2–EYFP, which can then be reactivated by light stimulation during testing.

Question 4

Behavioural timeline

Answer 4

•Label ‘engram’ of Context B, then reactivate in Context A

Context A = light stimulation and doxycycline
Context B = fear conditioning; no light and no doxycycline
Reactivate context a = testing.

Question 5

Experiment one - checking the activity tagging system

Answer 5

Mostly works.
- On-dox: no expression of ChR2-EYFP
- Off-dox, in home cage: ChR2-EYFP expression
• Off-dox, following fear conditioning : more ChR2-EYFP expression
- Off-dox, following context exposure sans fear: similar ChR2-EYFP expression
- 5 days post-FC: slight decrease in ChR2+ cells, but still elevated
- After 1 month, expression of ChR2-EYFP decays
- Seizures (lots of activity) lead to ‘complete’ labelling
- Interestingly, almost no inhibitory interneurons were labelled
- Sparse DG neurons in FC mice can be driven by light pulses
- When dox is removed, cells are labeled in an activity-dependent manner
• Some of these cells can be driven by LED stimulation

Question 6

Experiment 2 - stimulating the putative engram may recall fear memory

Answer 6

• Stimulating cells tagged during FC in a neutral context leads to freezing
  •Freezing = measure of fear
• If no shock is given during exposure to context B, then no freezing is induced by stimulation of context A tagged neurons
  •Stimulation of non ChR2, solely EYFP-injected mice does not induce freezing
• Limiting the tagging to one day prior to fear conditioning leads to stronger levels of freezing
  •Stimulating both hippocampal lobes increases freezing even more

Question 7

Experiment 3: how context specific is the putative engram

Answer 7

• Label context C with EYFP, and stain for c-Fos to identify cells active in context B
• Distinct contexts lead to distinct activity traces that overlap at chance frequency
• Activating a neutral context engram (C) does not elicit freezing behavior, even after fear conditioning in another context (B).

Question 8

Conclusions

Answer 8

Together, our findings indicate that
activating a sparse but specific ensemble ofhippocampal neurons that
contribute to a memory engram is sufficient for the recall of that
memory. Moreover, our experimental approach offers a general
method of mapping cellular populations bearing memory engrams.