Key features of Alzheimer’s disease (AD), amyloid plaque formation and tau tangles, have been assumed to be the underlying cause of AD. However, it appears likely that these represent independent manifestations of a multi-factorial disease process with a common underlying defect in brain stem cell function.

The Wnt signaling pathway plays an important role in the brain and regulates hippocampal brain stem cell activity. The research funded by CART helped us demonstrate that pharmacologically manipulating Wnt signaling has the potential to ameliorate neurodegeneration and improve memory in aged adults.

They identified multiple astrocyte genes as being altered in aging, and have now found that some of these genes are also altered in expression at the earliest stages of AD. They are in the process of asking if blocking these gene expression changes in mouse models of AD is protective to synapses.

In a parallel approach the lab is asking if re-expressing signals that astrocytes use to regulate synapses in the young brain, is also protective of synapses in AD. Their findings provide insight into how certain genes could become dysregulated, contributing to synapse loss in AD.

When microglia are stimulated by such damage-released molecules, they secrete inflammatory signals called cytokines, which are small proteins through which microglia can affect neuronal functions. IL-1 and IL-18 are two of these cytokines produced by activated microglia and they can amplify the initial damage to neuronal cells and thereby worsen the neuronal loss that eventually gives rise to the dementia and impaired cognitive ability associated with Alzheimer’s disease.

IL-1 and IL-18 are made by a large intracellular complex of proteins called the NLRP3 inflammasome. We have proposed that the most effective way to halt the production of IL-1 and IL-18 by activated microglia, and thereby protect early Alzheimer’s disease patients from progressive neuronal loss, is to inhibit, or shut down, the NLRP3 inflammasome factory. We therefore investigated how the NLRP3 inflammasome factory is powered and identified an important role for the cellular “power stations” called mitochondria.

Mitochondria are organelles (sub-cellular organs) that produce energy for cells. Just like us, mitochondria have their own DNA. We found that the NLRP3 inflammasome and the production of mitochondrial DNA are stimulated at the same time. However, the conditions that stimulate the NLRP3 inflammasome also damage the newly produced mitochondrial DNA. To get rid of their bad DNA and repair the damage, the mitochondria spit out fragments of bad DNA to the cytosol, where they bind to the NLRP3 inflammasome and start the production of IL-1 and IL-18.

Although short-term production of IL-1 and IL-18 is important for tissue repair, their chronic production worsens neuronal damage and accelerates the progression of Alzheimer’s disease. Based on these findings, we suggest that a highly effective way to shut down the NLRP3 inflammasome factory is to prevent the generation of bad mitochondrial DNA. To this end, we identified a critical enzyme that helps produce mitochondrial DNA in microglia and have shown that stopping this enzyme, called CMPK2, prevents the generation of bad mitochondrial DNA, thereby blocking NLRP3 inflammasome activation and production of IL-1 and IL-18.

So, to put all of this in an easier to understand context, the NLRP3 inflammasome can be considered a factory, IL-1 and IL-18 are the products made by the factory and bad mitochondrial DNA and CMPK2 are the control boards that control how fast the factory spits out its products.

Since the initiation of our CART-sponsored project, we have confirmed that exposing microglia to inflammatory signals begins the production of CMPK2. Furthermore, we recently produced mice that are deficient in CMPK2 and are now expanding this line of mice so that we will be able to crossbreed them with a mouse model of Alzheimer’s disease. This will enable us to combine our discoveries and determine if shutting down CMPK2 will be useful for halting the progression of the neuronal degeneration and dementia that characterize Alzheimer’s disease.

In particular, we found activation of PI3K signaling by STX improves mitochondrial function, which generate cellular energy needed to keep nerve cells healthy. 

Using CART funds, we also showed that STX can be safely fed to mice for many months (with no side effects). More importantly, treatment with STX improved behavioral responses and memory in a mouse model of AD. The results of our CART project will provide the foundation for a more extensive analysis of STX, with the long-term goal of advancing STX to clinical trials.

We are testing the hypothesis that sleep enhancement can improve memory function and reverse the deficits associated with Alzheimer’s Disease, using – for the first time – gain-of-sleep experiments. We have previously validated a mouse model of SWS enhancement.

The CART grant allowed us to validate a mouse model of REM sleep enhancement, establish cognitive testing in our lab, and produce and validate two triple transgenic mouse models to enhance sleep in Alzheimer’s Disease mice.

They found that tau is full-length in the brain, but CSF tau is truncated at the middle of the protein. Neurons derived from human skin fibroblasts using induced pluripotent stem cell (iPSC) technologies were also analyzed. Tau is full-length inside the cells but when it is released in the media, it is truncated, recapitulating tau profiles in the brain and CSF, respectively. This supports that neurons in a dish can serve as a translational model for studying tau in the human central nervous system (Sato, Barthelemy et al., Neuron, 2018).

They further analyzed p-tau profiles in the individuals with AD in the CSF (Barthelemy et al., Front Aging Neurosci, 2019) and brain. In AD CSF and brain, p-tau at specific sites were increased. When brain samples were biochemically analyzed for local correlation between amyloid and p-tau, soluble p-tau was more associated with the amount of local amyloid compared to insoluble p-tau. This suggests a direct link between local amyloid and p-tau in AD brains (Horie, Barthelemy, Sato et al., submitted).

Furthermore, they found that specific peptides in the Microtubule Binding Region (MTBR) of tau that is closer to the end of the protein increase in AD (Horie, Barthelemy, Sato et al., under revision), potentially serving as another CSF biomarker of AD.

Second, they tested whether other CSF p-tau species can serve as more sensitive AD biomarkers compared to the most commonly used phosphorylation site at amino acid Threonine 181. They found that the ratio of phosphorylation at Threonine 217 (pT217) to T217 (phosphorylation occupancy at T217) in the middle of the protein associates with amyloid more strongly than the ratio of pT181 to T181 in late onset sporadic AD (Barthelemy, Sato et al., Alz Res Ther 2020). This led to another study in individuals with familial AD mutations in the Dominantly Inherited Alzheimer’s Network (DIAN) cohort at Washington University. Dr. Barthelemy and the team found that the ratio of pT217 to T217 increases almost 20 years before onset of AD symptoms, right at the time amyloid aggregation starts in the brain as detected by positron emission tomography (PET) imaging (Barthelemy, McDade et al., Nat Med 2020).

This computational project proceeded in three stages. The first was to create a computer model of the interactions between many of the key factors that are known to underlie AD. The second was to analyze an extensive database containing data on real AD patients, which also included lists of all the drugs they took. Fortunately, the patients in the database took commonly prescribed drugs in many different combinations, so the third stage was to see if the benefits to the patients of specific combinations of drugs could be correlated with predictions of the efficacies of those same drug combinations from the computer model.

Statistical analysis showed that the database benefits and the predicted model efficacies were significantly correlated! That meant that the computer model was picking up something real about the complex, multifactorial, pathological processes that underlie AD. The main payoff was that the agreement between the database and the model could be used to identify new drug combinations that might actually be effective in treating AD, and they involve commonly used drugs that elderly people already take.

The approach identified many potentially effective combinations, but combinations of certain drugs rose to the top. They included the NSAIDs ibuprofen (and similar drugs) and aspirin, along with drugs that are commonly used to treat high blood pressure. This CART funded project suggests that these combinations should be explored experimentally and/or clinically to see if they really can be effective weapons in the fight against AD.

AD is caused by progressive brain cell death over time. We believe that AD, as well as other neurodegenerative disorders, can be treated by addressing breakdown in the cells’ autophagic pathways. If the pathways can return to functioning properly, the cells may able to engage in their own housekeeping and free themselves of the damaging detritus.

My team is looking at ways to clear the autophagic pathways in these cells. Because of advanced technologies, we can use chemical libraries to screen for and identify existing drug molecules that promote the proper function of these cleaning pathways.

With this generous grant, we have identified such molecules. We have also synthesized the molecules because in doing so we can improve their drug properties.

As we move forward, we’re continuing to develop this class of molecules as drug candidates to further understand, and possibly treat, AD. Our progress has brought us closer to our goal of bringing relief to individuals and families living with this difficult and progressive neurological disease.

The ultimate goal of this research project is to search for therapeutic agents that can enhance the tau-degrading activity of microglia, thus reducing tau pathology and preventing cognitive decline in AD patients. During these funded years, we have made scientific discoveries about the mechanism how pathologic tau is internalized and degraded by microglia. We also identified the activation conditions as well as genetic factors that regulate this tau degradation process in microglia. Finally, we have collected data showing that brain microglia approach pathologic tau injected into the mouse brain (see figure to right).

We deeply appreciate the tremendous help provided by The CART fund to help us initiate and extend this project. The scientific observation and conclusion are significant for the AD basic research and also impact greatly in AD therapeutic study.