Deep within the brains of people with Alzheimer’s disease, three main problems exist: the buildup of amyloid plaques between neurons, the accumulation of tau tangles inside neurons, and long-lasting neuroinflammation that further promotes amyloid and tau pathology and drives neuronal damage.

By the time someone has symptoms of cognitive decline and receives a diagnosis, all three of these issues have already taken root in their brain — likely for decades. Currently approved therapies include monoclonal antibodies that only target the amyloid plaques and drugs that modulate acetylcholine and glutamate neurotransmitter levels in the brain. But no therapies are yet approved to target tau tangles or neuroinflammation.

Paul Seidler, a structural biologist at the University of Southern California, believes that these additional therapies will be required to fully treat or prevent the entirety of the disease mechanism. “My thinking is that as we continue to add medications to the arsenal to tackle neuroinflammation, to tackle tau tangles — that's where we're going to see the total resolution of progressing cognitive impairment,” he said.

Tau proteins can clump together into fibers that spread throughout the brain by acting like 'seeds,' causing more clumps to form in nearby neurons. Seidler’s lab is focused on identifying and developing small molecule drugs that can break up these tau aggregates by reducing the thermodynamic stability of the tau fibers until they can’t combine. “If you decompose the fiber enough, it can no longer seed,” he said.

DDN spoke with Seidler about his team’s work on structure-based small molecules, including both natural and synthetic drug compounds, and the future of curative treatments for Alzheimer’s disease.

Continue reading below...

WebinarsReliable fluid biomarker strategies for clinical neuroscience research

Explore how validated fluid biomarker assays advance clinical research for neurological diseases.

Read More

Why does your team focus on developing small molecule drugs?

We think small molecules have broader potential. Monoclonal antibodies have challenges in delivery, as patients need to be near an infusion center, at least initially. Plus, biologics tend to be rather expensive to bring to market. They're more complex to produce, preserve, and deliver.

We think that if we can develop a small molecule that is effective and safe, it could be used similar to how statins are used in patients with high cholesterol, where someone goes to the doctor and finds they have this biomarker that says, “You're becoming at risk,” and then they start taking a statin pill in the morning at home. This is how we see treatment for someone at risk of tau aggregation in the brain who could start taking a pill early on, and really, we're trying to move in the same direction for amyloid beta plaques as well — both with the use of small molecules that can be used safely at home in the decades prior to when cognitive impairment begins.

Your small molecule drugs under development are structure-based inhibitors. Why is that?

It’s very difficult to make a drug that has all the hallmarks that we're interested in for brain disease. We need specificity, meaning that it needs to bind to a three-dimensional target site that is fairly unique. Structural biology is really a direct route to visualize those sites. For example, a straightforward case would involve targeting an enzyme. We need the structure of the enzyme to understand where the substrate binds and then we can design analogs that will get stuck in the active site.

However, the problem with protein aggregation is that it’s not clear where the active site is, or where we should put the small molecule. There is no enzymatic activity that we understand.

While I was a postdoctoral fellow in David Eisenberg’s lab at the University of California, Los Angeles (UCLA), my work focused on visualizing the three-dimensional binding site on the tau tangle that we knew had a bioactive effect, which was to destabilize the fiber. That work was important because it gave us a bull's eye. It gave us the active site, so to speak, where we can put something to elicit an inhibitory or disaggregase effect. We used a molecule derived from green tea, epigallocatechin gallate (EGCG), which binds to all sorts of different enzymes in the proteome. So as a drug molecule, it's not good, but it’s great as a probe to indicate where to aim with better chemistry.

Once you understand where to put it, you can engineer the right hydrogen bond from the right places to drive that potency and affinity. Of course, all that needs to be balanced with physiochemical characteristics that are going to allow brain permeation and cell membrane permeation to get the drug where the target is. A lot of people have been trying to target extracellular tau to try to intercept it when it goes between cells, but 99 plus percent of the tau resides in the neuron. That's a huge amount of tau that we're missing. We really need to get drugs into the cells where the target predominantly resides.

What other natural compounds has your lab studied as tau inhibitors?

Spinning off from that EGCG work, we asked what other natural products have this disaggregase capability, or the ability to bind to fibers and inhibit and destabilize them. That led to some really great discoveries, in my opinion, about endogenous neurotransmitters in the brain that help buffer the protein aggregation environment. We've discovered a whole host of these in my lab now that are already in the brain as small molecules that can disaggregate fibers. We think these small molecules are involved in another layer of protein homeostasis that has been unrecognized prior to this.

To find these, we used machine learning rules that govern the potency of disaggregases and screened about 1,000 small molecules of related chemical structure to EGCG. We discovered that disaggregases don’t just inhibit fibers at their endpoint, they also stop prion-like seeding. However, there is a continuum. If a disaggregase isn’t potent enough, instead of breaking fibers down in an inhibitory way, it fragments them such that it then actually promotes aggregation in a catalytic way.

We found that the monoaminergic neurotransmitters like dopamine and noradrenaline all have this biphasic relationship in seeding behavior. We think this is really important because one of the first places that tau aggregates appear in the brain in Alzheimer's disease is in noradrenergic neurons of the locus coeruleus. These neurons are making loads of noradrenaline, and their concentrations start to diminish as a function of aging.

We believe the initial aggregation could be very much a function of these endogenous neurotransmitter concentrations in locus coeruleus cells, and as they deplete, potentially you start to get some enhancement of tau-seeding potential. And the disease process doesn’t stop there. Noradrenaline levels decrease to the extent that they not only no longer suppress aggregation, but instead, noradrenaline levels actually potentiate aggregation through its limited disaggregation activity. The biphasic relationship means falling noradrenaline levels aren’t just a little less potent as a disaggregase. The relationship flips and noradrenaline becomes an enabler of disease.

As a result, we’re now investigating the role of noradrenaline in the disease.

Was that discovery a surprise to you?

The realization that you can go from a protective concentration of a molecule to the opposite — a disease-potentiating concentration — was surprising to me. That’s been the most profound realization for me because that explains both a disease mechanism and also a therapeutic strategy, in which we could focus on preserving noradrenaline levels to stunt tau aggregation at an early stage in the disease process. The challenge is keeping the drug in the therapeutic range to ensure the right concentration. But just realizing that this is something that's happening in our brains as we age due to our neurotransmitter fluctuations is really important for understanding disease mechanism.

The realization that you can go from a protective concentration of a molecule to the opposite — a disease-potentiating concentration — was surprising to me.
- Paul Seidler, University of Southern California

What progress has your lab made in developing synthetic small molecules to inhibit tau aggregation?

Our classic approach was looking at typical central nervous system drug molecules that follow Lipinski rules, so they can pass cell membranes and the blood-brain barrier. We've gotten a couple of leads in the lab that we've developed, and we've got some private funding and foundational funding to keep moving these forward, and we’ve started testing a few. These compounds leverage the same binding pockets on tau that are engaged by known tau ligands: chemical probes and positron emission tomography (PET) agents. Their core structures resemble these well-characterized tau binders, but we’ve modified them to promote tau disaggregation.

The other pathway that we've been investigating recently is transporting more of the endogenous-like molecules, although synthetic ones, using biological transporters. After recognizing that monoaminergic neurotransmitters have this disaggregase potential, we realized that these precursors are transported by LAT1 (L-type amino acid transporter 1). Its whole job is to transport aromatic and phenolic molecules that have this potentially disaggregase-like activity. So, we started generating a whole series of synthetic molecules in our medicinal chemistry lab that resemble these monoaminergic neurotransmitters, but are now increased in potency and metabolic stability through synthetic engineering. These molecules do not need to follow Lipinski rules because they're transported by a different mechanism to achieve the same kind of outcome, but through a totally different chemical footprint.

When do you hope to move a candidate drug into human testing?

What I need to be able to move this work closer to humans is a greater investment in basic science to help us optimize the potency of these small molecules, but a lot of times this is not something that is very popular to fund. Often, people only want to fund preclinical testing, but we need to improve the molecules through basic research first. That's the thing that's going to catalyze the success in the end.

Do you think we will see potentially curative treatments for Alzheimer’s disease, and if so when?

If you look at the progress with amyloid beta, I believe people will look back 10 years from now and recognize, “That was the first hit on target for Alzheimer's disease.” They'll realize that when we bring a second drug to market to add on that targets neuroinflammation or maybe tau, these additive effects are where we're going to start to become more convinced as a society that we really are on the right path. We could see that the FDA-approved monoclonal antibody Leqembi plus a new tau or neuroinflammation drug allows us to reach 50 or 60 or 70 percent slowing in cognitive decline. Then when we add another drug and we have all three aspects targeted, it will add even more potency.

This happened with research and therapeutics for human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS). I see the progression of that success very much as a template for the Alzheimer's field. The first HIV drugs weren't a cure by any stretch of the imagination, but they were an improvement. And then it took two decades to add on treatments to the point where now people are living more or less chronically with the disease or in some cases, potentially indications of cures. That may have taken 20 years to get to that point, but within 10 years, there was enough dramatic progress to push ahead and keep refining the targets and the potency of drugs. I believe we're going to get to that same horizon, without a doubt, within 10 years or so for Alzheimer’s disease.

This interview has been condensed and edited for clarity.