Skip to main content

Understanding mechanisms regulating endoplasmic reticulum dynamics in hippocampal synaptic plasticity

Final Report Summary - SYNAPTIC ER (Understanding mechanisms regulating endoplasmic reticulum dynamics in hippocampal synaptic plasticity)

In the hippocampus, neurons are the principal cells that mediate and store information as we learn and later recall the information as memories. How learning and memories are formed remains an important question as deterioration in cognitive function is a common feature of many neurological conditions, such as dementia. It is anticipated that findings from this area of research will ultimately lead to treatments that improve the quality of life, especially when memory loss and deteriorating cognitive functions are a common problem faced by an aging population.

Past efforts in the examination of neuronal function have revealed that neurons communicate with each other using patterns of electrical activity at synapses. The synapse is made up of two elements, a pre-synaptic and a post-synaptic compartment. Changes in both pre- and post-synaptic compartments are often related to the strength of electrical transmission between neurons. For example, strong electrical transmission strengthens the synaptic connections and maintains increases in the pre- and post-synaptic compartment size while weak transmission weakens and reduces pre- and post-synaptic compartment size. Most importantly, these functional and structural correlations are thought to underlie learning and memory processes.

Dendritic spines are the post-synaptic compartments concentrated with glutamate receptors. There are four types of glutamate receptors in neurons, Alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR), N-methyl-D-aspartate receptor (NMDAR), metabotropic glutamate receptor (mGluR) and kainate receptor (KAR). Of these, it is AMPARs that drive the synaptic transmission while NMDARs dictate the direction of changes in synaptic activity, termed synaptic plasticity. In the field, it is well recognized that there are two forms of synaptic plasticity, long-term potentiation (LTP) is the result of a strengthening and long-term depression (LTD) for the weakening of synaptic transmission. Both forms of plasticity are thought to be the major contributors to the cellular basis for memory. Both NMDARs and mGluRs are involved in modulating the synaptic response and downstream signaling pathways in LTP and LTD.

Like many eukaryotic cells, neurons have an endoplasmic reticulum (ER). The ER is a heterogeneous organelle comprising of branching tubules and flattened sacs that extends throughout the neuronal structures. The ER has multiple functions including intracellular signaling, protein synthesis and cell regulation. In hippocampal neurons, only subsets of dendritic spines contain ER. These ER-containing spines are highly dynamic. Whilst research has thus far focused on the contribution of glutamate receptors and its associated protein complexes for controlling the neuronal functions, the role of ER in dendritic spines has not been studied in detail. In this connection, this proposal aimed to test the hypothesis that the sporadic distribution of ER in dendritic spines can contribute to synaptic plasticity by elucidating whether there are any common mechanisms regulating the glutamate receptors and ER dynamics in spines.

AMPAR-mediated synaptic transmission is the most commonly studied response in LTP and LTD. This receptor, like the NMDA receptor, is a multimeric assembly of different subunits. The incorporation of one particular subunit, GluA2, is important to preserve the AMPAR function at synapses and this occurs at the ER. Considering this intimate relationship between the AMPAR GluA2 subunits and ER, we aimed to elucidate whether mechanisms that redistribute the AMPAR GluA2 subunits during LTD also impacted the spine ER dynamics. We chose to induce LTD with group I mGluR selective agonist DHPG, a method widely adopted in the field. Previous work has shown that DHPG expelled ER from hippocampal dendritic spines (Ng and Toresson, 2011). We therefore reasoned that if the DHPG-LTD induction expels ER from spines, it then follows that synaptic AMPAR will be removed, resulting in fewer AMPAR in particular the GluA2 subunits at synapses. We expressed two fluorescent markers in dispersed hippocampal neuronal cultures, a superecliptic pHluorin (SEP)-GluA2 marker to follow the internalization of AMPAR at synapses and a red fluorescent marker to visualize the ER. This live-cell imaging method has allowed us to simultaneously observe two modes of dynamic events in real-time, one at the level of AMPAR GluA2 trafficking and the other at the level of ER content in spines. However, there was no positive correlation between SEP-GluA2 signal and ER content in spines following DHPG application. It could be that the SEP-GluA2 reporter is not sensitive towards subtle changes in the GluA2 signal at synapses.

The lack of correlation between AMPAR internalization and ER content was a surprise, given previous results. We thus turned our attention to the NMDA receptor, which acts as a coincident detector and modulator for LTP and LTD. We hypothesized that this receptor would act as a common denominator for both AMPAR trafficking and spine ER dynamics. Here, we chose to express two fluorescent markers in dispersed neuronal cultures, a cytosolic green fluorescent marker to label the dendritic spine structures and a red fluorescent marker to visualize the ER. We found that some spines appear or disappear while others alter in size and shape under basal conditions. Within spines, many showed an increase or decrease in the ER content. By combining the imaging experiments with molecular and pharmacological manipulations and by performing a detailed analysis on ER-containing and ER-lacking spines, we have shown that NMDA receptor activity is responsible for coordinating the changes in spine size and the associated ER dynamics. The activation of NMDARs promoted a rapid ER expansion in growing spines. We have further linked the striatal-enriched protein tyrosine phosphatase (STEP), an endogenous negative regulator of NMDAR function, to the regulation of spine ER dynamics in neurons. We showed that STEP inhibits the NMDAR-mediated ER expansion in growing spines. In this sense, we have provided a novel mechanistic pathway on how NMDARs, STEP and ER regulate spine growth in neurons. These findings have been recently published (Ng et. al. 2014) and constitute an important milestone towards defining the roles and mechanisms regulating spine ER dynamics in synaptic physiology. The reviewers for this published manuscript have commented, “this is an important study, in particular, the results on spine ER dynamics are novel and interesting”.

We are currently characterising the effects of DHPG-LTD on spine ER in dispersed neuronal cultures. Thus far, we have found that, on average, the induction of DHPG-LTD inhibits ER growth in spines with marginal effects on dendritic spine size. In addition, we have extended this experimental approach to a more physiologically relevant model - organotypic hippocampal slices. We are currently continuing these experiments in the laboratory on whether and how DHPG-LTD and synaptic mGluR-LTD induction redistribute the ER from spines and its relevance to synaptic physiology.

We anticipate that findings from this project will be of interest to academic researchers in the area of neurobiology/neuroscience and potentially, the industry as novel molecular targets for therapeutic drug developments.