Engrams and memory

Question 1

Engram

Answer 1

refers to the enduring offline physical and or chemical changes that were elicited by learning and underlie the newly formed memory associations
- neural memory traces are what engrams are ⇒ latent modification in the irritable substance produced by a stimulus
- Long term record in brain tissue of some prior event
- Connection between engrams and memory

Question 2

Engram cells

Answer 2

populations of cells that constitute critical cellular component of a given engram
- These cells may or may not also be critical components of engrams supporting other memories

Question 3

what are engram cells activated by? (3)

Answer 3

• A learning experience
• Physically or chemically modified by the learning experience
• Reactivated by subsequent presentation of the stimuli present at the learning experience (or some portion thereof) resulting in memory retrieval ⇒ aka relevant stimuli

Question 4

what does reactivation of engram cells produce?

Answer 4

a behavioral change relevant to the memory

Question 5

when are engrams active?

Answer 5

they activate during encoding and then there is a consolidation period where they are not active but changes are happening in the neurons storing the memory
- later on when animals are exposed to relevant stimuli the engram ensembles become reactivated and can undergo another reconsolidation phase with additional changes in terms of the strength of connections
- can be repeatedly activated during memory retrieval

Question 6

what are brain regions implicated in memory consolidation? (4)

Answer 6

• Areas of prefrontal cortex marked in red
• Amygdala and basolateral amygdala
• Hippocampus
• Entorhinal cortex

Question 7

T/F rodent cortexes have accepted homologous areas in the human brain?

Answer 7

False
- rodent cortex don’t always have accepted homologous areas in the human

Question 8

where was the damage in patient H.M that caused anterograde amnesia and temporally graded retrograde amnesia?

Answer 8

damage to amygdala, parts of hippocampus and associated cortex
- He couldn’t form new declarative memories after damage and had better memories for past events
- There was damage to the amygdala in the coronal slice and more anterior areas of the hippocampus ⇒ less damage to more posterior hippocampal areas

Question 9

Cued fear conditioning

Answer 9

there is an auditory cue that predicts a foot shock at the end of the sound ⇒ when they repeatedly experience the sound and foot shock association they will freeze when they hear the sound even if there is no shock presented

Question 10

Contextual conditioning

Answer 10

foot shock is associated with the box itself and not a particular cue to look at how animals learn the environment predicts likelihood of foot shock

Question 11

Basolateral amygdala (BLA)

Answer 11

brain region necessary for both learning and expression of conditioned fear

Question 12

what happened with lesions to the amygdala via NMDA infusions to kill amygdala neurons?

Answer 12

Prior to fear conditioning animals have very low freezing behavior but animals with SH should have more of a fear freezing response with more shock trials because the BLA is not affected; Animals with lesions to the BLA have minimal increases in freezing behavior across trials
- You need an intact BLA to learn this freezing but they also found if they lesion the BLA after training they can disrupt the conditioned responses

Question 13

Tet-tag system

Answer 13

tagging of activated neurons is achieved by two trans genes in a tet-tag mice and when the animals have doxycycline neural activation cannot induce expression of the transgene called tauLacZ
- When neurons are activated it allows expression of the transgene under the control of a Fos promoter which is activated in recently activated neurons
- animals are raised on a diet that has doxycycline which prevents tTA that forms in response to activation of the fos promoter ⇒ it prevents tTA from acting on the TetO promoter to drive tauLacZ and none of the cells express LacZ even if they are activated
- after the period of time where there is no doxycycline and we allow LacZ expression, the mice go back onto the doxycycline diet to close the window so that neurons that are activated in response to new stimuli are no longer going to be allowed to express LacZ

Question 14

Fos

Answer 14

drives the activation of tetracycline controlled trans activator protein (tTA) which in the absence of doxycycline can act to drive expression of LacZ ⇒ with doxycycline present the tTA can no longer drive tauLacZ expression

Question 15

Activation of early genes like ZIF or Fos is going to result in what?

Answer 15

lacZ expression in activated neurons from fear conditioning

Question 16

what happens when animals go back on doxy?

Answer 16

The animals go back onto doxycycline to close the LacZ window and have a retrieval test where they are exposed to the fear cue and results in activation of early gene but these neurons will not produce LacZ

Question 17

if there are cells active with LacZ and ZIF what does that mean?

Answer 17

suggests that those neurons might be part of an engram or neural ensemble in encoding the fear

Question 18

what did reactivation of neurons by fear conditioning memory retrieval show?

Answer 18

the % of cells with LacZ had an increase in neurons for fear conditioning groups but is still a relatively small % of overall cells in the amygdala and there is an increase in overlap between Lac and ZIF that had fear conditioning and retrieval which suggests the neurons could underlie a fear memory trace
- lower percentage in the no retrieval group that showed % LacZ +ZIF

Question 19

what did time for fear conditioning extinction show for mice that had previously been fear conditioned?

Answer 19

The lateral amygdala (slightly above BLA) had no relationship between % cells with LacZ + ZIF overlapping with freezing but positive correlation between overlap and how much they froze to the tone suggesting there could be some dissociation between which cells in different parts of the amygdala encode context related to fear and discrete cues related to the fear
- can see even in animals with strongest freezing after extinction there is still relatively low % of neurons that actually have overlap ⇒ small portion join the engrams or memory related ensembles

Question 20

how does CREB determines which neurons are allocated to a specific memory ensemble?

Answer 20

CREB transcriptional regulator by protein kinase A which turns on several genes important for long-lasting changes in PKA activity and structure of synapses for long-term learning
- CREB is generally important for learning in the LA but we don’t know which neuron roles it plays

Question 21

what does CREB deficiency do?

Answer 21

CREB deficiency reduces learned freezing so CREB is important for learning

Question 22

what happens when neurons have enhanced CREB?

Answer 22

neurons with overexpression of CREB prior to training were more likely to be activated during the fear memory test
- enhanced CREB neurons are more likely to become part of the neural ensemble activated by shocked-paired tones

Question 23

how was enhanced CREB engineered?

Answer 23

they infused a vector for the overexpression of CREB in a subset of neurons and trained animals in fear conditioning (with high intensity shock to get lots of fear conditioning)
- Looked at % of neurons expressing Arc which is another early expression gene marking recent neuron activity => 20% of neurons expressed Arc after fear conditioning in response to the test
- Neurons expressing CREB were more likely to express Arc later than neurons without the CREB vector

Question 24

for the control vector that expressed GFP but not CREB what did that show?

Answer 24

similar number of neurons were activated (so not necessarily CREB activating more neurons) but these were no more likely to become part of Arc expressing engram
- You can bias which cells become part of a potential engram via overexpression of CREB

Question 25

how does CREB overexpression influence hippocampal neurons? What about when CREB is inhibited?

Answer 25

Even though a similar number of hippocampal neurons are inhibited only the ones overexpressing CREB are important for the engram => in animals that overexpress CREB there is a big decrease in freezing when the neurons that express CREB are inhibited via the dread receptor - suggests you can prime hippocampal neurons via overexpression of CREB to become part of the memory trace similar to what happens in the amygdala - Suggests engrams in the BLA and hippocampus have similar properties

Question 26

how did they do the CREB overexpression experiment in the hippocampus?

Answer 26

some animals received a virus that is just for the expression of hM4Di which is a designer receptor activated by a designer drug (dred) ⇒ these respond to drugs like CNO to inhibit or excite the neurons they are expressed in - For this experiment cells that expressed the virus are inhibited in response to the CNO infusion - they also had a group where a virus would overexpress CREB and express the designed receptor in the same neurons ⇒ green indicates which neurons were expressing the hM4Di receptor and red shows CREB - they gave this prior to any training and allowed enough time for overexpression of CREB and then animals underwent fear conditioning - on the test day the animals either got CNO to inhibit the cells that expressed the dread receptor and overexpressed CREB or they had a control injection

Question 27

how did researchers look at ensembles via optogenetics used to drive appetitive and aversive behavior in BLA and hippocampus?

Answer 27

they used a transgenic mouse that expresses tTa under the control of a cFos promoter ⇒ when dox is absent tTa can drive expression of channel rhodopsin or mCherry under TRE (tetracycline response element) - if dox is not on board neurons that are activated can produce fluorescent protein or Chr2 which is an excitatory opsin to give us access to neurons to activate them when we want - there were animals that went under fear vs reward conditioning in the absence of dox and they checked that similar numbers of neurons in each area were able to express ChR2 and mCherry after each of the manipulations

Question 28

can neurons drive place avoidance?

Answer 28

Activation of neurons previously activated during fear conditioning is aversive ⇒ true for the hippocampus and amygdala

Question 29

how was the place avoidance experiment done?

Answer 29

They looked at which side the animal naturally preferred and then activated neurons when on their baseline preferred side to see if they could drive them away - During the first part of the experiment animals are still on doxycycline so neural activation won’t result in light sensitive protein ⇒ light activation day 1 won’t do much to activate the neurons - then the animals go off doxycycline and have fear conditioning with foot shock to allow activation of cells in response to fear conditioning to express whatever is under the TRE promoter in response to tTA - Some animals have a virus for channel rhodopsin and some got a virus in the BLA as well as some controls - some animals didn’t get foot shock on day 3 and they can express channel rhodopsin but it shouldn’t be in response to fear conditioning but is related to whichever neurons are activated by exposure to the chamber - the animals go back on dox to prevent new neurons from expressing the virus and they were tested again to see if they would avoid activation

Question 30

can optogenetics affect place preference?

Answer 30

yes, day 1 none of them express channel rhodopsin so there is minimal interaction with non preferred side - the they tag a reward engram of neurons activated by time with a female mouse (experimental mice are male) - the mice go back on the dox to close the tagging window - for hippocampus and BLA there is an increase in the zone when paired with the reward engram

Question 31

Can the valence associated with each ensemble for the DG and BLA be switched?

Answer 31

hippocampal ensembles can be associated with aversive or appetitive stimuli to drive behavior whereas the BLA ensembles are more limited in their ability to drive fear or reward related behaviors - amygdala may play different roles in memory formation or that these amygdala neurons may have more prepotent roles in driving appetitive and aversive behaviors - It seems like the hippocampal ensemble has flipped by pairing the neuron activation with a reward but not the amygdala and same with opposite condition from reward to aversion