Neural plasticity, LTP and Memory

Question 1

What are the different types of neural plasticity?

Answer 1

Lecture 15, slide 20, 22-24, 49

Question 2

What is structural plasticity important in?

Answer 2

Lecture 15, slide 25

Question 3

Summarise some studies that show synaptic plasticity in humans?

Answer 3

Lecture 15, slide 26-48

Question 4

Descibe the tri-synaptic ciruit?

Answer 4

Lecture 15, slide 54

Question 5

What is the role of AMPA receptors in glutamate transmission? What are its features? What are the characteristics of its current?

Answer 5

Lecture 15, slide 58-60

Question 6

What is the obligatory subunit of NMDA receptors?

Answer 6

GluN1

Question 7

What are the properties of NMDA receptors? What are the characteristics of its current?

Answer 7

Lecture 15, slide 63-68

Question 8

What is co-localisation of glutamate receptors? What does it underlie?

Answer 8

Lecture 15, slide 69-70

Question 9

What is Hebb’s rule?

Answer 9

The rule states that: learning and memory are based on changes in synaptic strength (driven by synaptic plasticity) among neurons that are simultaneously active.

Question 10

What is long-term potentiation?

Answer 10

Lecture 15, slide 73

Question 11

What are the properties of LTP?

Answer 11

Lecture 15, slide 77-91
-LTP is rapidly induced and long-lasting. It is also cooperative/associative, meaning that strong inputs can strengthen weak inputs—if neurons in the synapse are activated coincidentally. If the inputs are not coincident then the strong input may induce LTP, while the weak input does not (input-specificity).

Question 12

What evidence is there that NMDARs are necessary for LTP induction?

Answer 12

Lecture 15, slide 95-101

Question 13

What changes, which support LTP expression, does calcium influx through NMDARs lead to?

Answer 13

Lecture 15, slide 102-103

Question 14

What are two theories for the role of hippocampal LTP in memory?

Answer 14

• LTP in the hippocampus (particularly the CA1 subfield) underlies/plays a role in encoding hippocampal-dependent forms of associative spatial learning and memory
• hippocampal LTP is not required to form spatial memories (spatial knowledge), but instead, it is important for disambiguating between overlapping and/or competing spatial memories during memory retrieval (spatial choice).

Question 15

What did early attempts to investigate the role of LTP in learning and memory find? How did Moser et al. show a flaw in these studies?

Answer 15

Lecture 16, slide 22-27
-Early attempts to study the role of hippocampal LTP in learning reported that an increased fEPSP could be observed after spatial learning, however, Moser and colleagues showed that most changes in fEPSP observed in these studies could be explained by temperature changes. They did so by recording fEPSPs and temperature in the dentate gyrus while the test animal explored a novel environment. The results illustrated that increased fEPSPs observed in explorative learning tasks correlated with a temperature rise in the brain and could be replicated through passive warming of the animal and running on a treadmill.

Question 16

What are some correlational studies that support the theory that hippocampal LTP underlies spatial learning and memory?

Answer 16

Lecture 16, slides 28-58

• A study by Whitlock and colleagues supports the aforementioned hypothesis of hippocampal LTP function. In the study, rats carried out an Inhibitory Avoidance task (IA, which is an associative learning paradigm), where the experimental group were trained with footshocks to avoid the dark chamber and stay in the illuminated chamber. Recording from all over the CA1, taken using a multi-electrode array, showed that there was no significant change in fEPSPs after a rat was IA trained compared to the baseline fEPSP level and control groups. However, some electrodes showed a substantial increase in fEPSPs, but this was only observed in individuals of the IA trained group and not controls. The increase was rapidly induced and persisted throughout the trial session, suggesting that hippocampal LTP was involved in the learning of the IA task. Unlike the earlier studies, the results observed here are unlikely to be due to temperature because the temperature would have increased fEPSPs uniformly, whereas most electrodes in the experiment recorded a decrease in fEPSP after IA training. The use of a multi-electrode array enabled the primary evidence to be obtained in this experiment—I.e. without recording at multiple points of the CA1, the fEPSP increases may not have been observed.
• In the same study, Whitlock also demonstrated that GluA1 and GluA2 levels (AMPAR subunits) increased in the postsynaptic membrane following IA training. AMPAR increase in postsynaptic membrane is required for the expression of LTP and so trafficking of AMPAR subunits is a molecular marker of LTP. Therefore, these particular experimental results show that LTP occurs after IA training, and suggest that it is important in learning and the formation of memories. In addition, Whitlock showed that these results were specific to the hippocampus (and thus hippocampal LTP and learning) and that the increase in AMPAR subunits was due to trafficking, rather than an increase in the generation of AMPAR subunits. The latter was illustrated through the use of a crude hippocampal homogenate which showed no overall change in AMPAR subunit levels.

Question 17

What are some causal studies that support the theory that hippocampal LTP underlies spatial learning and memory? What are some problems with these studies?

Answer 17

Lecture 16, slide 61-77

• Morris et al. pharmacologically intervened with NMDARs using the antagonists D-AP5 and L-AP5. The mice with the intervention had impaired performance on the spatial reference memory water maze test and did not have a quadrant preference during the transfer test. This evidence seems to support the hypothesis that hippocampal LTP (via NMDAR activation) plays a role in associative spatial learning and memory, however, the antagonists used also blocks extra-hippocampal structures, meaning that the results observed could be mediated by another brain area. Also, AP5 does induce some sensory-motor impairments which would inhibit animals in this task. Therefore, you cannot conclude that the results are the effect of impaired learning—they could be the effect of impaired performance.
• A different study by Tsien and colleagues attempted to obtain tissue-specific causal evidence for hippocampal LTP function in spatial learning by deleting the gene for GluN1 (the obligatory subunit of NMDARs) in the CA1. In the hidden water maze variant of the Morris water maze task, these KOs were impaired in learning, however, they were also impaired in the landmark task variant. In the landmark task, the platform is visible and even animals with no hippocampus can complete the test, suggesting that GluN1 deletion was not as tissue-specific as intended. This was determined to be the case by Rondi-Reig et al. who showed that there was a clear reduction of GluN1 levels in the cortex of KO mice.

Question 18

What is a causal study that supports the theory that hippocampal LTP is important for disambiguating between overlapping and/or competing spatial memories during memory retrieval (spatial choice)?

Answer 18

Lecture 16, slide 79-102

• GluN1 was knocked out post-natally from granule cells in the dentate gyrus and pyramidal cells in CA1. Although the KO was not perfect, it was mostly hippocampal specific (especially compared to the other studies). In this study, the KO animals had impaired spatial learning in a radial maze but not in a water maze. This suggests that hippocampal NMDARs and hippocampal LTP are not always required to encode associative spatial memories.
• To determine the role of hippocampal LTP, Bannerman and colleagues generated a novel water maze task with two beacons—one of which was a decoy and the other was situated on a platform. The logic behind this was that in the radial maze the cues are ambiguous because all 6 arms are identical and can to lead to reward, whereas in the standard water maze the animal is able to use the extra-maze spatial cues to learn to navigate to the hidden platform and the cues are unambiguous/different. The novel water maze task creates the same ambiguity that is experienced in the radial maze. In the new water maze experiment, KO mice made significantly more errors than the control mice, despite knowing exactly where the plastform was located according to the probe tests. These results suggest that hippocampal LTP is important for spatial choice (novel water maze task is impaired in KOs), but not for spatial knowledge (standard water maze task is not impaired in KOs).