Sunday, February 26, 2017

Week 5

For the majority of the time that I was reading the Herry et al paper, I was convinced the the two neuronal populations were just representative of two different associations being made with a stimulus (tone) or stimuli (tone and context) and shock, but Fig 6 (along with the distinct connectivity shown in Fig 5) dramatically changed my opinion by showing that BA activity is not necessary for the expression of either the fear or extinction memory, just the switching between the two.

The main question I’m still left with is why there were never any experiments where the animals are returned to the conditioned context and after extinction training. In the same way that they have “Post-FC” data in Fig 1, in which they show that the fear neurons fire in response to tone even in the new context (aka the extinction context), I would have expected a similar “Post-Ext” experiment in which they show that the “extinction” neurons still fire even in the conditioned context (after several trials because this change in context would still induce freezing due to the context-fear association if not the tone-fear association) because they have learned that the tone is no longer associated with shock. Of course, the initial return to the conditioned context would surely cause an increase in the firing of the fear neurons again because of the context-fear association, but Fig 1d shows that these fear neurons are not only responsible for the context associated fear but also the tone associated fear (the animals show freezing in response to tone even in the new condition before going through extinction training) so they should be able to show extinction behavior after some time in the conditioned context.

Just as the animals are able to switch quickly from non-fear to fear states (Fig 4), they should be able to switch from a fear to non-fear state on a relatively fast time scale. Even if it is not as fast as the transition to fear, it should take less than a 24 hour period (the time is takes for extinction consolidation) to switch and therefore prove that it is not a matter of relearning an association, rather it is switching to an already existing one as in the case of the switch to fear after fear renewal. Without this, it is still possible that the BA is necessary for the the acquisition of the extinction and fear memory, if not the retrieval or expression of them, and since this paper focuses on this idea of bidirectional switching between fear states rather than fear-learning or just the rapid switch to fear after extinction, this is an important distinction to make.

I suppose it is still possible that being in the condition context would be sufficient for fear response and that the animals would have to undergo additional extinction training to dissociate the room, in addition to the tone, from the shock but I would be surprised if this were the case since the tone and not the room is the only stimulus time-locked to the shock and therefore should be the main association with fear, and if that association is extinguished, then reduced freezing in the conditioned room after tone fear extinction should be a much faster process.

As for the second paper, it was cool to see a follow up on the circuitry described in the Herry et al paper that showed the upstream modulation of fear behavior by the PVINs in the dmPFC. I feel like I missed something or perhaps didn’t quite understand this paper well enough because I was confused as to how they knew that the PNS are necessary or responsible for the fear expression. Based on what I understand, the PVINs are important because they coordinate the firing of the PNs and it is that coordination that is necessary for the PNs to produce a fear response. Although they show that PVINs activation inhibited PNs and prevents CS+ induced activation and reduced freezing (and conversely disinhibited PNs to increase freezing when PVINs were inhibited), they do not show that the PNs are either necessary or sufficient for fear behavior.

The PNs are 1) responsive to CS tone, 2) project to the BLA which we know to be important for fear behavior, and 3) their synchronized firing is correlated to fear behavior, but I am unclear as to why PN regulation is definitely the cause of the fear behavior and not just a consequence of PVIN activity that is not necessary for fear. As far as I can tell, the PVIN activation and inhibition experiments are not limited to changes in PN activity. Perhaps the key is understanding the cross-correlogram analysis? Or do the PVINs only project to the perisomatic region of PNs? Maybe the PNs being necessary/sufficient for fear behavior is not as important as I am thinking. At the very least, I do agree that the data shows that PVIN inhibition is causally related to and necessary to drive fear expression.

No comments:

Post a Comment