Research on human brains reveals proteins, structures important in cognition

Upstate Medical University in Syracuse, New York invites you to be "The Informed Patient" with the podcast that features experts from Central New York's only academic medical center. I'm your host, Amber Smith. Neuroscientists have identified molecules, synapses, and anatomy in the brain that give rise to cognitive function, but it's been a challenge to figure out how they all work together. Now a team of neuroscientists, including a computational neuroscientist from Upstate, have assembled an unprecedented set of data showing that the brain's connectivity is influenced by specific proteins and molecular structures. Dr. Christopher Gaiteri is here to explain. He's an associate professor of psychiatry and behavioral sciences at Upstate, and he's also an assistant professor in Rush University's Alzheimer's Disease Center. Welcome back to "The Informed Patient," Dr. Gaiteri.

[00:00:55] Christopher Gaiteri, PhD: Hey Amber, it's a pleasure to be here. I'm really excited to talk about this study.

[00:01:00] Host Amber Smith: Well, your work was featured in the journal, Nature Neuroscience, and I understand it could lay the groundwork for major advances in the treatment or prevention of brain disorders, including Alzheimer's, dementia, and mood disorders. Can you first talk about the 98 people who agreed to be part of the study?

[00:01:18] Christopher Gaiteri, PhD: Yes. Most of those individuals were from the Chicago area, and half of them were somewhat unusual in that they're from the Religious Orders Study, so that's a study of monks, priests, nuns, brothers and sisters, and all of them have chosen to provide annual health information to us. They come in, and they'll take cognitive tests. And upon death, all of them donate their brains to science. The other half of the individuals were also from the Chicago area, but they were older adults who were not in a religious order.

[00:01:54] Host Amber Smith: So men and women?

[00:01:56] Christopher Gaiteri, PhD: Yes. About two thirds were female, but both sexes were there.

[00:02:02] Host Amber Smith: So was it hard to find people willing to participate in a study like this where they would commit to giving their brain to science after they died?

[00:02:11] Christopher Gaiteri, PhD: It is hard, and we have people whose full-time job it is to find individuals who are willing to do this. It's a tremendous gift that these people are, are doing for society. And you might think that when we have -- you know, this subset of individuals in this study comes from a subset of thousands of individuals who've donated their brains -- and you might think that that sounds like a lot, maybe that's enough. But actually, when you're older, most brains have a mixture of several different pathologies in them. And so if you're looking for this exact combination of pathologies, by the time you really drill down to that, you might have two or three people. So that's really not enough to study.

So we're always . looking for more. And even Alzheimer's disease, there's sort of several different mechanisms that can be at work in your brain. So it's not really like there's just one form of the disease. So for all of those reasons, it's really important to obtain more brains from people who are willing to do this.

[00:03:05] Host Amber Smith: So the 98 people who were part of the study, were they all used in the study, or did you have to whittle down within that 98?

[00:03:14] Christopher Gaiteri, PhD: We whittled down to the 98 people. So those 98 people were very unique because they had provided neuroimaging while they were alive and performing all of these cognitive tests.

And then after death we obtained their brains and a whole bunch of different measurements in their brain, so we have the same brain information from it while alive, and then a whole bunch of molecular levels after death. And so that paired data in these 98 people, that's what makes them really special.

[00:03:44] Host Amber Smith: Well, I'd like to ask you about the various types of data that were collected by your colleagues at Rush University, University of Alabama, and Emory University. What is functional MRI neuroimaging, and what can it tell you?

[00:03:57] Christopher Gaiteri, PhD: So you get fMRI signals out of the brain, based on the amount of oxygen that's being used in a particular part of the brain. Basically, the scanner picks up a different signal when the blood, when hemoglobin is oxygenated versus deoxygenated. And, and you might imagine if part of your brain is working harder, it's going to be using more oxygen. And so that's, that's what fMRI is essentially picking up.

As you're performing some tasks, some mental task in a scanner, we can be tracking "OK, this part of the brain is using more oxygen, versus doing some other task." But actually the people in this study were not doing any task. They were just in the scanner. They could think about whatever they wanted to think about. They had to remain awake. We made sure they remained awake. But the reason that that's useful is because if you're giving someone a task to do, especially this sort of older population, there's going to be some people for whom this task is trivial. There's going to be some people for whom it's actually too hard.

So you kind of have these floor and ceiling effects. And so what we do is we just gather what's called resting state data. And it turns out that if you just observe the brain at rest, with someone lying in the scanner, thinking about whatever they want to think about, that actually provides really useful data that relates to a whole variety of different diseases.

And the nice thing is, it doesn't have these sort of floor and ceiling effects that you might get if you were actually asking someone to do something specific while in the scanner. So that's sort of where the fMRI signal comes from. And again, it's just someone at rest in the scanner thinking about whatever they want to think about. That's what we're measuring.

[00:05:33] Host Amber Smith: And then you end up matching that up somehow with the brain biopsy, which can only be done postmortem?

[00:05:40] Christopher Gaiteri, PhD: Mm-hmm . Absolutely. And so that's a process of essentially following these people bi-annually with scans. So some of these people have come in several different times to be scanned, providing cognitive tests.

Once they pass away, we then quickly obtain tissue samples from their brain. Usually it's less than 12 hours after death, we have the brains. And your brain is actually, it's quite beautiful after death. The body really takes care of your brain and buffers it more than anything else in your body. It's your body's trying to take care of your brain.

So there's an opportunity to get really accurate data from it even after death. And so again, to your initial question, though, these are the same people who did the scans and then also provided all of these molecular measurements after death.

[00:06:26] Host Amber Smith: Now, there was a term in your paper I haven't heard before. What is a dendritic spine? And I'm not, maybe I'm not even saying that correctly.

[00:06:35] Christopher Gaiteri, PhD: No, you nailed it. That's what it is. It's a dendritic spine. So, two words, dendritic and spine. Let's talk about the dendrite first. So when you information flow in your brain, it's gathered from the dendrites. These are these little extensions off of neurons. And, they'll gather information from other cells and kind of feed it up into the cell body where it's integrated and then sent out to other cells where it's picked up by their dendrites. This is essentially how information flows among the cells of the brain.

And so on the dendrite, there's actually these little extensions. If you sort of zoom in, you can see them. There's thousands of these on any given dendrite. And these are the spines. These are the spines of the dendrites, or dendritic spines. And they come in all different shapes. It sort of looks like a forest of mushrooms. There's thin ones. There's ones that are actually called mushroom shaped. And these are the points of contact between two cells. And so that's like a very important part in the brain because the brain's all about communication. And so we're studying the exact point where two cells meet when we study these dendritic spines.

Actually you can... These dendritic spines, they move. It's really cool. Over the course of hours to days, they can sort of get bigger or smaller or shrink up or form stronger or weaker contacts between different cells. And that's really how you actually instantiate your memory of the world is by these little tiny extensions between cells, the dendritic spines.

[00:08:08] Host Amber Smith: So are the dendritic spines involved in the brain protein data that you were collecting?

[00:08:15] Christopher Gaiteri, PhD: Yes. In particular we found that a combination of protein levels and measuring the shapes of these spines was able to predict how statistically connected two different brain regions were. So that's, it's kind of cool because they're very different types of data.

So the dendritic spine data, our colleague Jeremy Herskowitz at Alabama uses microscopes and then actually maps out the three dimensional structure of individual spines across tens of thousands of these things. So we have that data.

And then you have the abundance of protein, which is just a level more, sort of, more or less. And those two things together, we're able to predict how synchronized two different areas of the brain involved in Alzheimer's disease were. So really, if you think about it, that's three very different types of data that are actually shown to be in sync. You've got the correlations between two distant parts of the brain. You've got the dendritic spines, which is a structure, a cellular structure, and then you've got the protein levels that are coming from inside of a cell. All of those are sort of shown to be in sync by this study.

[00:09:20] Host Amber Smith: This is Upstate's "The Informed Patient" podcast. I'm your host, Amber Smith. I'm talking with computational biologist, Dr. Christopher Gaiteri, about some innovative research into brain connectivity.

Can you describe what connectivity is and why it's important?

[00:09:35] Christopher Gaiteri, PhD: Yes. There's a couple different types of connectivity that mean different things in terms of the brain. So the most straightforward one is structural connectivity, which is the kind of thing that we're actually measuring with dendridic spines. We're measuring essentially what is the physical strength of how two things are coupled together.

On the scale of the whole brain, sort of zooming out to the level of the whole brain, you can measure how many fibers, how many sort of cellular structures are connecting to different brain regions. So that's very much a concrete kind of measure of structural connectivity.

FMRI, you can also measure statistical connectivity, which is basically just two brain regions over time. Their activities going up and down. Is it going up and down in sync? That's functional connectivity. And so you don't necessarily know why they're in sync. There could be some other third brain region that maybe is synchronizing them together, but you can see that there's a statistical correlation.

So that's the other main type of connectivity that neuroscientists will talk about is functional connectivity, which is more of a statistical construct versus structural connectivity. There are relationships between the two of them, but those are sort of the two main types that we talk about.

[00:10:51] Host Amber Smith: Does functional or structural connectivity change as we age, like, naturally? Do you expect to see changes in that?

[00:10:58] Christopher Gaiteri, PhD: There's a ton of changes in connectivity as you develop. So if we're talking about before birth, right after birth, there's massive changes in connectivity.

Actually, you tend to lose a lot of synapses early on. It's not so much a matter of like gaining them as sort of pruning out ones that the brain doesn't need. But all through adolescence, there's massive changes in connectivity. A lot of them are related to global integration. Instead of different parts of your brain sort of being isolated, you're kind of developing this global brain network. And that continues up into the 20s. And then it sort of stabilizes until later in life.

Exactly how late depends on the person. But then you start to also see some neurodegenerative changes, which are generally not beneficial, whereas the connectivity changes in development as you're growing up are beneficial. So, there's definitely changes in connectivity. It's not necessarily, it depends on the individual.

The late life changes in your brain connectivity are generally not as big as the ones that you see growing up. But there are some really severe cases though, where it might almost be on par with that.

[00:12:14] Host Amber Smith: So does connectivity help you measure a person's cognition?

[00:12:19] Christopher Gaiteri, PhD: We think that, actually, cognition in a lot of cases is essentially generated by connectivity, either the structural connectivity or some of these statistical connectivities that I talked about before. So that's, that's really our perspective is that essentially cognition is given rise to by the way that your brain and your cells are connected. So, yeah, I think there's a very close connection between those two things.

[00:12:46] Host Amber Smith: What's important to know about synchrony?

[00:12:49] Christopher Gaiteri, PhD: Synchrony between two brain regions relates to how much information you can actually transfer between them. And synchrony's also important for building up connections. So a lot of times in the brain if you see processes that are synchronous, they tend to become stronger and stronger, more strongly connected to each other. So those are sort of two main things to keep in mind about synchrony.

[00:13:12] Host Amber Smith: Now let's talk about proteins. How many proteins influence the brain's connectivity?

[00:13:19] Christopher Gaiteri, PhD: So in this study we found about 300 different proteins were related to the strength of functional connectivity between two brain regions. And, you might say, OK is that a lot more than we thought? Is that less than we thought? How does that change?

And the thing is, until this study, there hasn't actually been a really definitive human study of how many proteins are involved in the connectivity of the brain. There's definitely been excellent attempts to measure this, but what we're doing here is we're actually saying in real living humans as opposed to, for instance, dead mice, this is how many proteins that we can actually observe being related to changes in functional connectivity.

And that's really important because in a lot of clinical studies from whatever disease you'll see that, OK, in our patient cohort, people with ADHD or whatever the particular condition of interest is, we notice that there's altered connectivity between these two brain regions.

That's great. That's useful. But, if we're trying to develop drugs for whatever this condition is, it's very difficult to connect those changes in connectivity to molecular levels. And molecular levels are what you're going to develop the drug for, right? So in this study that we did, that's part of why it's, I think, quite important is because we don't only show variation in functional connectivity, but we're actually relating that variation to molecular levels, which is a much more actionable thing than this sort of, just the beautiful connectivity of the brain is one thing, but we're trying to tie it to something that's more actionable, which is molecular levels. So I think that broadly across diseases, that ability is, going to be useful.

[00:15:02] Host Amber Smith: Which molecular structures are involved with the proteins?

[00:15:06] Christopher Gaiteri, PhD: In particular, we found that the dendritic spines were very helpful in improving our ability to predict changes in functional connectivity, in addition to the protein levels. So essentially, you could count up how many of each particular class of dendritic spine shape was in a particular brain, and that would actually help you to predict how connected different brain regions were with each other. So that's the thing, the dendritic spines, that improved our predictions.

[00:15:34] Host Amber Smith: So how could the results of this study be used to help with Alzheimer's or dementia or mood disorders?

[00:15:43] Christopher Gaiteri, PhD: Well, the two brain regions where we gathered the protein data in this study were picked because they're related to resilience, to Alzheimer's disease. Resilience is this idea that really comes from examining brains where we found that there are older people whose brains are loaded with Alzheimer's proteins, but cognitively they were fine, until death. So in a way, that's kind of what you want to be, right? If you have to have some Alzheimer's proteins, you don't want to be affected by it. Like, you want to be resilient.

And so in studying the brain that we found that the structure of these two brain regions we studied in this study were involved in resilience to Alzheimer's disease. So this study's kind of intrinsically couched in terms of Alzheimer's disease. So we're doing more work currently to tie it specifically to Alzheimer's disease. And in particular, what we're trying to do is, it's fairly easy to obtain a brain scan. It's non-invasive, right? But for studying Alzheimer's disease, developing drugs, you would like to know what are the actual molecular levels in this person's brain? Because given those molecular levels, I might think that this clinical trial makes more sense for this person, and this other clinical trial makes more sense for another person.

So I think we can actually do that, and that's one thing that we're working on now with this dataset is, given the brain scan, what are the molecular levels in this person's brain? We can predict that fairly accurately. And that's the sort of thing that could be used for sending people to one clinical trial or another.

[00:17:15] Host Amber Smith: Now, how typical of the general population do you think the religious people in your study are likely to be? Should we consider it a representative sample?

[00:17:28] Christopher Gaiteri, PhD: That's an excellent question. There's similarities and differences.So for instance, having a purpose in life is associated with living longer, being healthier in various regards, having slower cognitive design, and individuals in these religious orders tend to score higher in terms of their purpose in life. So, in that sense, they are not entirely representative of the population. They tend to be more educated, and education is another thing that's been shown to be associated with more resilience to the effect of Alzheimer's disease. You get it later than people who have fewer years of education.

So yeah, there are definitely differences between these people. For the purposes of this study, we tried to statistically control for all of those. And the other half of the individuals, were just sort of drawn from the general population, albeit the population that was willing to donate their brains to science. So again, there may be some special things about them too.

[00:18:32] Host Amber Smith: Well, can you tell us about the data set, because I understand you're making it available to other scientists to use.

[00:18:39] Christopher Gaiteri, PhD: Yes, yes. It's a huge data set. It is freely available. And many other aspects of it, too, are freely available to any researcher. And these individuals who've all donated their brains to science, their data is used in several hundred papers, actually. So they're having a huge impact.

[00:18:59] Host Amber Smith: Is it a spreadsheet, or are there biological samples that you share? How does it exist?

[00:19:07] Christopher Gaiteri, PhD: It's so big it's going to break Excel. It's not a spreadsheet, but it's definitely addressable with any sort of programming language. And there are also biological samples. So most of the brains are still available. We don't use up all of them. So if you want to run a study, and you're interested in a particular brain region, and you need it on hundreds of people, that's totally available. So it's a huge resource.

[00:19:34] Host Amber Smith: So you think this is already inspiring scientists. Would it inspire them to collect similar data from other areas of the country, maybe, or from other populations?

[00:19:45] Christopher Gaiteri, PhD: That would be great. And we do have, there are additional studies,in particular trying to get data from African American and Hispanic populations as well, because those are sort of underrepresented in the data that we have currently.

And I do hope it inspires other people to collect such data. If you're trying to get the exact data we're talking about here today, though, which is the, the fMRI plus the brainomics, that really unusual combination of data, that would probably take another decade to acquire. If you wanted it faster, we're talking about a hundred million dollars. So I would definitely make use of the brains that are already available for this.

There is one similar project -- you're talking about inspiration -- there's a similar or a related project at least called the Living Brain Project. In that one people with typically really severe Parkinson's disease who are getting brain surgery allow scientists to take a small sample of their brains sort of en route to where they're going anyway, surgically. So in that case you do have brain data coupled to people who are actually still alive.

So that's, I guess, another sort of similar study. And that's the only study in the whole world that I'm aware of, that has any sort of similar data. Although in that case, they're mainly Parkinson's disease because almost no healthy person would ever volunteer for this. So it's sort of difficult to have the comparison data. Whereas in the dataset that we've gathered, there's both healthy and diseased individuals. So you, you can make those comparisons.

[00:21:19] Host Amber Smith: It really is a unique set of data.

[00:21:22] Christopher Gaiteri, PhD: Yeah, I was so excited. I remember the moment I heard about this data several years ago. I thought, this is awesome. I need to do something with this. So, yeah, it's really unique.

[00:21:33] Host Amber Smith: You're listening to Upstate's "The Informed Patient" podcast. I'm your host, Amber Smith. I'm talking with computational biologist, Dr. Christopher Gaiteri, about some innovative research into brain connectivity.

So what sorts of research do you anticipate other scientists may want to do using this data set?

[00:21:56] Christopher Gaiteri, PhD: Well, one thing that we're doing, so I anticipate other people would want to do it, is looking at how brain dynamics relate to molecular levels. And you might think that, how is that different from what we did before? Actually as we're talking now, there's a certain set of brain regions that will be active in each of our brains.

And then when you go to your next task and you do something different, you go get coffee, there's going to be a different set of brain regions that'll be co-activated. And actually, even as we're talking, second to second, there's different sets of brain regions that become active, more or less active. And it's been shown that your brain's ability to shift between those different types of networks is actually really related to cognitive functions. So when I say brain dynamics, that's what I'm talking about, the ability to shift between these different networks of co-activated brain regions. So we're sort of taking that perspective on this data as well.

I want to see ... There's something about, for instance, how rapidly someone can flip between different brain networks that relates to their protein level. So if I had to guess what someone's going to do next, that's what we're doing next. So that would be my guess.

[00:23:04] Host Amber Smith: Some of the other Alzheimer's research I've read about, they talk about tau and tangles, and I don't hear you talking about any of that. Is that not thought to have a role in Alzheimer's, or does it play a different role that relates to this in some way?

[00:23:22] Christopher Gaiteri, PhD: So about a third of the individuals in this study have Alzheimer's disease, and that's clinically confirmed Alzheimer's disease. So the pathologists have gone into their brains, examined slices, and they see those proteins, tau and amyloid proteins that you're talking about. So they definitely have it.

Another third of the people have those proteins, but actually were cognitively fine. So those are the resilient people. So we're talking about two thirds of the people in this study having those proteins. So those proteins are definitely evident in the brains that we're studying.

There wasn't exactly room to incorporate that into this study as well, but that's something that we're also working on currently is incorporating those proteins.

[00:24:03] Host Amber Smith: I was going to ask if there's additional research you and your team may do with this data set, but that's maybe one thing.

[00:24:10] Christopher Gaiteri, PhD: Yes, that is one thing. And another thing that we're doing that's maybe slightly less mainstream, but becoming more mainstream, is we want to tie these effects that we're seeing to a particular type of cell, like a really specific type of cell. Not just a neuron or oligodendrocyte or a microglia -- those are broad classes of cells -- but like really drill down within those. What type of cell might be responsible for this?

And the reason for caring about that is because if you can do that, then you can essentially focus in on that cell. You can grow it in a dish and do experiments on it. And it's much more, more rapid. And so we can essentially count up how many cells of each type are in the brain of each of these people because their brains are still around. We still have access to their brains. We can do this. It's just sort of held up by the fact that it's fairly expensive. It's about $4,000 per sample to do this.

And you want to do it in a couple hundred people in a couple brain regions. So you're well into the millions of dollars at that point. But, these people who've shared their brains and been neuro imaged are so rare. I think they could really be a Rosetta Stone that's going to be useful for a variety of diseases. So I think it's really important to continue investing and, and measuring data in these exact set of brains.

[00:25:28] Host Amber Smith: Is there a theory that Alzheimer's starts with one cell?

[00:25:31] Christopher Gaiteri, PhD: That is a very interesting hypothesis. You could name it after yourself if you want to. It's always up for grabs. There is a question about, like, what is the first cell that's going to sort of flip into an Alzheimer's state, and is that cell going to then influence other cells that it's connected to and induce those to flip into an Alzheimer's state? Or is it, for some individuals, more of a issue of genetic destiny, and there's these internal molecular aging clocks, and at some point the clock rolls over to a certain number, and it flips a switch and a whole bunch of cells begin to have issues?

Or is it that in perhaps a buildup of different exposures, environmental exposures begins to initiate this in a broad variety of cells? Or maybe there's something about the activity of a particular cell type. It just gets utilized by a way, sort of a high wear item in the brain, and it essentially becomes the first target of Alzheimer's disease.

So, if you want to join me in Alzheimer's research, we can try to figure this out. It's possible that it starts somewhere, and we're very interested in those earliest transitions towards Alzheimer's disease. So I like the way you think about this, but I think that's kind of an open question.

[00:26:53] Host Amber Smith: Well, it's clearly very fascinating research, and I really appreciate you making time for this interview, Dr. Gaiteri.

[00:26:59] Christopher Gaiteri, PhD: Oh, it's been such a pleasure to speak with you. Thank you.

[00:27:02] Host Amber Smith: My guest has been Dr. Christopher Gaiteri, a computational biologist and associate professor of psychiatry and behavioral sciences at Upstate. "The Informed Patient" is a podcast covering health, science and medicine, brought to you by Upstate Medical University in Syracuse, New York, and produced by Jim Howe with sound engineering by Bill Broeckel and graphic design by Dan Cameron. Find our archive of previous episodes at upstate.edu/informed. If you enjoyed this episode, please invite a friend to listen. You can also rate and review "The Informed Patient" podcast on Spotify, Apple podcasts, YouTube, or wherever you tune in. This is your host, Amber Smith, thanking you for listening.