Learning to resist Alzheimer’s disease:Novel molecular mechanisms that protect neurons against amyloid-beta

During the early stages of Alzheimer’s disease (AD) an excess of amyloid-beta (Abeta) impairs memory formation by corrupting synaptic function. However, not all people are equally affected by the presence of Abeta in the brain. In particular an active, learning brain is capable of building a ‘cognitive reserve’ that protects against Abeta-mediated memory deficits. A biological explanation for this protective cognitive reserve is still lacking. We are taking a unique and original approach by studying the molecular mechanisms that occur in a learning brain to provide protection against the detrimental effects of Abeta. This program builds on two mechanisms that we recently discovered and that show that neurons can defend themselves against Abeta-induced synapse dysfunction and loss. First, we discovered that the early growth response transcription factors (Egr1-4) are up-regulated in the brains of AD-patients during early non-symptomatic stages of AD, and are down-regulated in late symptomatic stages. Experiments in mouse brain slices show that expression of Egr1 alone is sufficient to neutralize Abeta-driven synapse loss. We propose to delineate this molecular pathway that protects synapses against excess Abeta upon Egr expression. We will perform a detailed analysis of the changes in the molecular composition of the synapse that occur after Egr1-4 induction. These studies aim to elucidate the Egr-driven molecular signaling that protects synapses against Abeta. Second, we discovered that neurons become insensitive to Abeta by changing the AMPA-receptor (AMPAR) subunit usage. Excitatory neurons mainly express two types of AMPARs: those that contain subunits GluA1 and GluA2 and those that contain GluA3 and GluA2. To establish whether Abeta selectively targets synapses with a particular AMPAR subunit-composition, we used mice that lack either GluA1 or GluA3. We found that mice are more sensitive to Abeta-driven memory deficits when they lack the AMPA-receptor subunit GluA1. Reversely, GluA3-deficient mice were fully resistant to AD-related synapse and memory impairment. Our data may explain how learning, which leads to an increase of GluA1 in synapses, can make neuron less susceptible to Abeta. By using proteomics technologies, we aim to identify the AMPAR-associated factors that render synapses resistant or susceptible to Abeta. This research program aims to fully decipher these key plasticity pathways, which are induced by learning behavior and are critically involved in protecting neurons against Abeta. We hypothesize that in an active, learning brain neurons become less susceptible to the detrimental effects of Abeta due to elevated expression levels of Egr1-4. We propose that this protection is achieved by an increase of GluA1-containing AMPARs in synapses. To test this hypothesis, we will apply advanced techniques in electrophysiology, live-imaging and proteomics of amyloidosis-based AD mouse-models. Importantly, our findings will be verified in human AD-brain samples and brain samples from people who, despite their old age, were not affected by dementia. We take a very promising and original approach in AD-research by studying the neuronal mechanisms that underlie the cognitive reserve for AD. The results of our studies will have important implications for the design of novel pharmacological, gene therapy-based, and behavioral therapeutic strategies that effectively protect against AD.