We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.


Interview: Can Sensory Stimulation Fight Alzheimer’s Disease?

Interview: Can Sensory Stimulation Fight Alzheimer’s Disease? content piece image
Listen with
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 6 minutes

At SfN 2019, we interviewed Li-Huei Tsai, director of the Picower Institute for Learning and Memory in the Department of Brain and Cognitive Sciences at the Massachusetts Institute of Technology, on the topic of her fascinating research, which explores the effects of modulating brain rhythms in Alzheimer's disease. Tsai presented her work in a plenary lecture at the conference ahead of the findings being explored in a clinical trial. You can read our reporting of her research here. In this interview, we take a deeper look at the field of non-invasive stimulation and hear why it has such a substantial therapeutic promise.

Ruairi Mackenzie (RM): Could you summarize the research you presented in your plenary?

Li-Huei Tsai (LT):
We’ve been working on Alzheimer’s disease for many years and in this lecture, I particularly focused on impairment of neural network activity that we think is very relevant to Alzheimer’s disease.

Historically, people who work on disease tend to focus on molecules or pathology or genes but the idea we had is that all these molecular and cellular events converge at a systems level to avert functionality of the brain. We found that a particular network activity known as gamma oscillations impair early in the course of the disease and we think that this contributes to the progression of the disease. We used various means and eventually this very non-invasive sensory stimulation to enhance gamma oscillations. We found that this can reduce hallmark pathological features of Alzheimer’s including amyloid plaques, [tau] tangles, neuronal loss, synaptic loss and even improved cognition.  This whole approach is extremely non-invasive, simply by using light and sound exposure. I think all of the observations that I presented today provide some very promising avenues for potential studies in humans.

RM: Where are these rhythms produced from? You mentioned GABA interneurons in your lecture, are these the only source?

System neuroscientists are still debating between two models. Either the interneurons by themselves are sufficient to generate a rhythm or the pyramidal cells in the neuronal network are necessary. In our studies, it is very clear if we silence these GABAergic interneurons, the rhythm is no longer produced, so regardless of whether they originally generated them, they are absolutely essential for these rhythms.

RM: What is the function of these rhythms in healthy brains?

So in healthy brains, as I alluded to in my lecture, enhanced gamma rhythms are particularly associated with higher order brain functions such as when the animals are effectively engaged in some sort of task, like working memory, where they’re trying to figure out where they are and where to go and they are processing sensory information, that’s when gamma rhythms are very active.

RM: One thing I had noticed in the data you presented is that when you stimulated mice with random rhythms rather than steady ones, there was a worsening of cognitive functions in animals instead of the improvement caused by the steady gamma rhythms. Why is that?

Behaviorally we haven’t really explored the other frequencies extensively but when you mention the random rhythms… So in the initial optogenetics experiment, we found out that when we drive these interneurons with random rhythms, like random frequency, but average 40Hz (that was our initial control) then what we did notice was that this increased amyloid levels, this random rhythm. But that was optogenetics; after we switched to sensory stimulation, random light flicker no longer produced such a large increase in amyloid. I think that’s the difference between directly driving these neurons with random rhythms, after which they become very random, and with sensory stimulation which I think perhaps involves some sort of filtering of the gamma rhythms so that it doesn’t produce such random oscillations in the brain.

RM: Are these random rhythms something humans would encounter in our environment?

I think there are instances when you will see random rhythm like if you go to a nightclub or whatever, the flashes of light are low frequency and sometimes can be random. And so, a lot of people asked me if you go to nightclubs do you get gamma rhythms? The frequency is very, very slow, so it’s probably the opposite of gamma oscillations.

RM: Are these GABAergic interneurons affected at a particular stage of Alzheimer’s disease?

I think now there is a lot of effort in the field to look at each subtype of cells very closely to look at not just pyramidal neurons and how the interneurons are affected. I would say a lot of this work is still ongoing, for instance one thing that people do, and my laboratory is very active in doing it as well, is to do the so-called single-cell RNA sequencing to really capture the profile of every single cell in the brain and to see how it changes. Our group and others are also very actively looking into which classes of interneurons are particularly affected by Alzheimer’s disease and we look not only in mouse models, but post-mortem human brains as well.

Those experiments are ongoing, but we also have a collaboration with the Allen Brain Institute, mainly using mouse models to test if we silence the different classes of interneurons, how that affects results. We are doing quite a few different things. There are at least three classes of interneurons, probably within each class there are subtypes as well. They are known as the parvalbumin fast-spiking cells (PV cells), somatostatin positive cells (SST cells), which enervate into the PV cells to control them and then the surplus are the so-called vasoactive intestinal polypeptide-expressing cells (VIP cells).

RM: Do gamma rhythms come from one of these particular subtypes?

So, PV cells, the parvalbumin cells, we definitely showed originally 10 years ago, that optogenetically driving these particular cells is sufficient to produce gamma. I think PV cells definitely are the workhorses producing gamma but questions remain about these other two types; what kind of role do they play to regulate the PV cells? Do they play a role in these beneficial functions that we observe with gamma oscillations?

RM: You mentioned in your talk that the question still remains as to why targeting these gamma rhythms would lead to these effects. What do we know about this?

On the one hand, we are discovering all the different cell types that are responding to gamma oscillations – I talk about microglia, I talk about neurons, I talk about vasculatures, that they respond to gamma rhythms and they produce all these fascinating changes in their behavior – and it seems that all the changes converge to reduce pathology and enhance cognition. The fundamental question which we still don’t understand is “How do these different cells sense gamma oscillations?”, and so that is eluding us and we’re working very hard on this to make a connection. I say that this is the missing link right now, in our investigation.

RM: Have you been able to look at how gamma rhythms are propagated through the brain and which areas are stimulated in which order?

We don’t have that data yet. I think right now there’s a lot of new technology that can help us look into that. For instance, recently I just heard that people are building this new type of two-photon microscopy. Right now, two-photon technology is restricted to look at particular focal brain areas but now there’s a new microscopy called meso microscopy where you can look at the whole brain. You can imagine if we can have such a set up then we can use a calcium indicator to label all the neurons in the brain, and once we shine the gamma light, we can trace in real time the order of activation, throughout the different brain regions. This is the kind of thing that we hope we’ll be able to do.

RM: You’re starting a clinical trial and have done tests with young healthy volunteers to check that the stimulation has no negative side effects. Did you do any cognitive tests on your healthy volunteers?

No, no we didn’t do that. We really did it to see safety and compliance and also, honestly, in aged, wild-type mice, we can see improvement of cognitive function, but in the young mice we actually don’t see it. As young mice they’re already very, very good and with GENUS we really didn’t see a further improvement. That’s why in humans, we really didn’t look at the behavior.

RM: A broader question then about this non-invasive stimulation. There have been promising studies using transcranial magnetic stimulation (TMS), another type of non-invasive stimulation. Do you feel that the wider field is beginning to accept that we don’t have to be invasive in the brain to have positive effects?

I think it’s absolutely the way to go because it’s so safe. I think it doesn’t cost that much compared to all these drugs developed in the various pharmaceutical industries. I think that non-invasive sensory stimulation is particularly attractive because, say with TMS this patient is probably going to have to go to the hospital to receive stimulation but with light, all you need is a simple light panel and a speaker, everybody can do it at home. It’s very accessible, scalable, very non-invasive and one distinct advantage of sensory stimulation as I showed in my presentation is the property of propagation to large brain regions. Usually with brain stimulation, like deep brain stimulation, we stimulate at a very focal point; the effect stays at that very focal point. I think depending on the purpose if you want the effect to spread, you want sensory stimulation.

RM: Why do you think that these techniques haven’t been explored as a way for therapy before?

We just published our first paper in 2016 and I will say the field has already moved very fast but human testing always takes time. For the FDA to say this thing is really effective or not, they usually want to see it after you test in hundreds, if not thousands of patients, so it takes time.

Li-Huei Tsai was speaking to Ruairi J Mackenzie, Science Writer for Technology Networks. Interview has been edited for length and clarity.