Play Live Radio
Next Up:
0:00 0:00
Available On Air Stations

New Clues to Memory Glitch Behind 'Senior Moments'


This is SCIENCE FRIDAY. I'm Ira Flatow. Age-related memory loss is exactly what it sounds like. The older you get, the less reliable your recall. I'm sure you know what I'm talking about. But you might be surprised to learn that age-related memory loss can start in your late 30s, your early 40s, eventually manifesting itself as, where did I put my glasses or the car keys, and so on and so on.

My next guest has spent the last 50 years unraveling some of the greatest mysteries about memory and the brain picking up the Nobel Prize along the way. And now, he and his team have discovered a culprit for those bugs in our memory, a gene that's starts to, well, retire maybe even before you do. Their work appears this week in the Journal of Science Translational Medicine.

Eric Kandel is winner of the 2000 Nobel Prize in physiology or medicine. He's also a senior investigator at the Howard Hughes Medical Institute, university professor and (unintelligible) professor at Columbia University here in New York. Always a pleasure to welcome you back, Eric.

ERIC KANDEL: It's always a pleasure to be here, Ira. Turns out that in the last six or seven years, every time something pleasant has happened to me, I have the added privilege of having a little conversation with you.

FLATOW: Oh, that's so nice of you to say that. Thank you. We'll start a mutual admiration society. Are we interrupting your vacation, Eric?

KANDEL: Not seriously. It turns out that one of the places I like to swim is a pool right near the studio so this took me five minutes out of my way.

FLATOW: All right.

KANDEL: For you, I can easily do that, Ira.

FLATOW: Thank you. Thank you very — well, let's talk business now. This is really interesting. Tell us what you did in this experiment to figure out a cause for this decline in memory age.

KANDEL: So let me put this into a little bit of perspective for you. When I was a medical student in the 1950s, we practically never spoke about Alzheimer's disease. And why is that so? And that is because people didn't live long enough to have Alzheimer's disease. You've got to be in your late 60s or early 70s to get what is called spontaneous Alzheimer's disease.

There's a very rare genetic form that has an early onset, but the common variety of Alzheimer's disease occurs much later in life. But as people lived longer, you began to see a significant amount of Alzheimer's disease. But as people lived longer, living in their 70s, their 80s and their 90s, they also lived better and you began to see these cognitive moments more frequently that you talked about.

And it wasn't all clear whether these age-related cognitive impairments were a distinct entity in their own right like the stiffening of the muscles as you age, weakening of various kinds, or was it an early sign of Alzheimer's disease. So Scott Small, Elias Pavlopoulos and I wanted to get a better grasp of that problem.

There was indirect evidence to suggest that age-related memory loss is a separate entity. And that was based upon the fact that it has its origin, apparently, in a different region of the brain than Alzheimer's does. Alzheimer's begins in a part of the brain called the entorhinal cortex. Age-related memory loss begins in the dentate gyrus.

And on imaging experiments, you can see some slight abnormality in the dentate gyrus with aging. And so we thought we would do a detail comparison of these two regions of the brain. So we took autopsy material of people who died between the ages of 38 and 90. Didn't have any evidence of Alzheimer's disease. And we compared in these people the dentate gyrus and the entorhinal cortex.

And we did an Affymetrix chip analysis, which is a fancy way of saying we looked at lots of genes and we wanted to see changes in expression that were specific to the dentate gyrus did not occur in the entorhinal cortex, that is specific to the region presumed to be important for age-related memory loss.

And we came up with about 17, 18 candidates, and one was particularly impressive because it declined systematically as a function of age only in the dentate gyrus. So here we had an interesting candidate called RB-AB 48, interesting protein involved in his histone binding. And we asked the question, can we manipulate this in some way?

Can we do causal experiments? So we turned to mice and we asked ourselves the question, what happens when mice age? Do they show a decline in this protein? And if so, is it specific to the dentate gyrus? And so we looked at mice, and sure enough, we found that there's a systematic decline and elderly mice show a defect in RB-AB 48.

And they show this specifically in the dentate gyrus. And now we could do systematic experiments. We could take a young mouse and we could knock out the gene for RB-AB 48 so we could reduce the level of the protein dramatically and the young mouse, three months old, showed age-related memory loss.


KANDEL: We could now do the opposite. We could take an old mouse, increase the level of RB-AB 48, and reverse the memory deficit.

FLATOW: So the old mouse, his memory returned like it was a young mouse?

KANDEL: As good as new, yes. It's quite remarkable.

FLATOW: It is. I'm just sitting here, flabbergasted, listening to you talk about it.

KANDEL: I just...

FLATOW: Were you flabbergasted, too?

KANDEL: Yes. I want to put in a cautionary comment because one gene does it means that gene is important, but it doesn't mean it's the only gene. It's possible it acts in concert with other genes that could also do it, but it's clearly an interesting candidate. One of the things that's of further interest is its function is quite fascinating.

As you pointed out, I'm older than I appear. I've been around a long time and I've been interested in memory for a long time. And one of my earlier interests in molecular biology of memory lead me to define the switch that converts short term to long term memory. And that's a gene called CREB, cyclic AMP response element binding protein.

And that switch turns on long term memory and leads to the growth of new synaptic connections between nerve cells, which is how long term memory manifests itself. You grow new anatomic ignitions. So Ira, if you or I remember anything about this program tomorrow, it's because, as a result of this conversation, each of us will have a slightly different brain than we had before.

And CREB is an important contributor of that. Now, it turns out CREB does not act by itself. It acts with two partners. One is called the CREB binding protein. And the CREB binding protein, if it's defective, leads to mental retardation. The other protein is RB-AB 48.

FLATOW: That's what we're talking about.

KANDEL: Exactly. So it fits into this cascade of gene regulation that is known to be important for normal memory. So now it opens up a number of avenues to explore. What leads to the decrease in RB-AB 48? Is there an enzyme that chews it up? Is there something in the diet? What happens if you went on a vegan diet? Would that keep the protein up?

Is this what is working these miracles for President Clinton? One could see. What does exercise do to this protein? We now have a target to look at chemical interventions, trying to develop drugs or use preexisting drugs that might elevate the level of the protein, but also other kinds of interventions. We could also look at the other candidates.

One candidate is quite interesting because it's been shown by George Martin to be involved in premature aging so that certainly is involved in cognitive disturbances so it would be interesting to see. Does it work through a different pathway? Does it work through the same pathway? How does it work?

So there are lots of - like any interesting problem, once you get an initial insight into it, it leads to five new questions.

FLATOW: Right, right. If you know the gene and you know the protein that is made by the gene, people are going to ask me, can I run out to the health food store and buy that protein made by the gene and take it and increase my memory?

KANDEL: There's no - first of all, no, they don't sell it, number one. Number two, most proteins, if you just eat them, they're digested very rapidly. They wouldn't get to the brain satisfactorily. So you would have to design in a very special way to get into the brain intact. But there might be easier ways to do it, to use substances that get into the brain very easily that are not proteins, smaller molecules that might elevate the level of the protein. So there are a number of pharmaceutical approaches that one could try.

FLATOW: And so far, there might be a lot of different things, you're saying, that trigger the protein to do its thing. You're saying maybe there is food, maybe there's exercise, maybe there's...

KANDEL: Absolutely.

FLATOW: ...maybe there's laughter like you do. Maybe that's one of your secrets of longevity.

KANDEL: Not a bad idea to enjoy life. It's more likely to increase longevity than being depressed. So I completely agree with you. Moreover, I think there's an important other point here, Ira, and that is I think although the suggestive evidence has been getting stronger, these are the strongest experiments so far to show that age-related memory loss is distinct from Alzheimer's disease. So every time you forget where you place your keys, it doesn't at all mean you're on the way to Alzheimer's.

FLATOW: Mm-hmm. You know, I know why...

KANDEL: You're much more likely...

FLATOW: Yeah, I know why you lose your keys. I know the answer to why you can't find your keys and that's because you never look at the spot where you leave them. You're holding the grocery...

KANDEL: Good point.

FLATOW: If you're holding a bag of groceries with the keys in it, you just drop it on the table, your eyes never see that spot where you left them. So of course, you never know where they are.

KANDEL: One solution to that, if I may advise you, Ira - you probably don't need any advice on these matters - is to have specific spots where you keep specific items that are important to you.

FLATOW: Yeah, the key bowl.

KANDEL: Always go to the drawer and put it there.

FLATOW: Yeah, yeah, there's always a place but you don't see it. So this is - so your lab is as excited as we are in hearing about this discovery?

KANDEL: We're very thrilled with this. Also, wonderful collaboration between Scott Small and myself. Scott Small, many years ago when he was an undergraduate in NYU, was a student in my lab. He was quite extraordinary in those days, so just from a social point of view it's been a very pleasant set of collaborations.

FLATOW: Would you know why the gene just stops working at a certain age?

KANDEL: No. That's a fascinating question we want to look at. Is there some active degradative process going on or is it just being deprived of a nutrient or some other source of stimulation that should be there?

FLATOW: All right. Eric, can you stay with us a little bit longer?

KANDEL: Absolutely.

FLATOW: All right. Eric Kandel, winner of the 2000 Nobel Prize in physiology or medicine. Our number, 1-800-989-8255 if you'd like to talk with Dr. Kandel. He's also senior investigator at Howard Hughes Medical Institute, university professor and (unintelligible) professor at Columbia in New York. We're going to take a break. You can also Tweet us. Yeah, you want to Tweet a question in? We'd love to have it at scifri S-C-I-F-R-I. We'll be right back after this break. Stay with us.


FLATOW: This is SCIENCE FRIDAY. I'm Ira Flatow. We're talking this hour about aging and the brain with Nobel Prize-winning neurobiologist Dr. Eric Kandel. We have an extra special bonus for you. We have a Q and A with Dr. Kandel on our website at And if you've just joined us, we were talking about the discovery of, well, a gene that influences learning and memory (unintelligible) whether you - how well, at least in mice...laboratory mice. Dr. Kandel, can you give us a one-sentence summary of what that gene does?

KANDEL: That gene is important in retaining memory. So if that gene is not functioning well, if it produces less protein, you have less capability for storing memories. It weakens your memory a great deal. If you boost the protein up, it strengthens memory, it restores memory.

FLATOW: And you were able, in laboratory mice, to give that protein to the mice and...

KANDEL: old mice and restore their memory, yes.

FLATOW: Wow. Well, how do you test a mouse? Running a maze, something like that?

KANDEL: You have a maze. We use a number of tasks but one of them is novel object recognitions. So if you put a young mouse into a small enclosure with two identical objects, it spends an equal amount of time with each. If you now take one of the objects away and give it a novel object, it likes novelties so it'll spend more time with the novel object than with the familiar one. An old mouse forgets very soon what's novel and what's familiar and spends an equal amount of time to each.

So this is a very good test to use because it involves the dentate gyrus which is the area that is thought to be involved in age-related memory loss. So we're honing in with a task that is quite important for this particular deficit of memory.

FLATOW: And you make the good point of making sure we understand that we're not talking about Alzheimer's here. This is different...

KANDEL: No, no. Actually, these experiments show quite clearly that age-related memory loss, which many people have, is distinct from Alzheimer's disease.

FLATOW: Yeah. Do we really know what causes Alzheimer's disease? It seems like we hear sort of confusing and conflicting reports about the plaques, the knots. You know, maybe they're involved, maybe they're not. Maybe they're the result of the disease. Are we still - are we quite clear on what the real cause is?

KANDEL: Let me put it this way. I'm a strong believer of the fact that the beta amyloid is an important component of this. And the reason I say that is, I mentioned very briefly earlier, there's an early onset form that is very rare. That is, genetically determined; it's an inheritable form. And if you look at the mutations that led to the early onset Alzheimer's disease, every one of them involves one or another step in beta amyloid synthesis or breakdown.

So clearly if you screw up beta amyloid, you can produce Alzheimer's Disease in people.

FLATOW: Right.

KANDEL: With the other candidates, you cannot simply do it with a molecule that is incriminated. It may act as an ancillary factor but by itself the mutation in tau doesn't give you Alzheimer's disease. This is not to say the tau is not very important. It may be important in propagating the disorder from one cell to another. But as a causal mechanism the evidence is strongest for beta amyloid abnormalities.

FLATOW: With the discovery of this new gene, could it be working also in other parts of the nervous system that are (unintelligible)?

KANDEL: Absolutely. And we don't know what it does. Very likely it's involved in non-memory functions as well. And that's why it'd be interesting to see what is so special about the dentate gyrus that it goes down here and that interferes with memory. And this is a universal problem. In many diseases, we have a protein that is present in many, many cells of the brain. But only in a certain region does it cause havoc when it's disordered. Why is that so?

So looking at the dentate gyrus and seeing what is happening through this protein could provide very general information beyond informing us about the protein per se.

FLATOW: All right. So what are the next steps? Now that you know this happens, what do you want to know next? How do you find it out?

KANDEL: OK. So we're interested in exploring several aspects of it. One, the question you come back to, what causes the protein to go down in elderly people? Is this an active process? Is there some degradative enzyme that is chewing up RbAP48 at a very rapid clip in elderly people while it doesn't do it in young? Or is it that some nutrient, some metabolite is lacking that is required for synthesizing it? That is, is it a degradative process or is it the lack of a synthetic process? That's number one.

Number two, why the dentate gyrus? You know, why is that process active there and not elsewhere? We also want to see, can we get, you know, a chemical approach? Can we get drugs or other kinds of interventions, exercise, diet that enhance the - now that we have a marker we can look to see how it affects it. And finally, as I mentioned to you earlier, we don't want to deceive ourselves into thinking because we found one powerful actor, this is the only actor on the stage.

There may be other candidates among the ones that we've selected that also have important roles, either as co-conspirators or acting through an independent mechanism. So to look at some of the other candidates will be interesting.

FLATOW: One question about the mice experiments. When you sort of rejuvenated the memories of these older mice to become memories like younger mice, did they remember things they forgot or did they - was it just easy for them to create new memories?

KANDEL: You know, we really did not look to see whether or not tasks that they had learned, you know, months before got stronger. We primarily looked at tasks that we'd been looking at in all the other experiments, like object recognition, whether they held onto information they had learned within the last 24, 48 hours. But it's a very good question, Ira. Have you ever thought of getting into the lab and out of radio?

FLATOW: Well, if you saw me in college in a chemistry lab, you'd know I'm right where I belong right now.

KANDEL: Well, how you perform in a college chemistry lab is no predictor of ultimate outcome.

FLATOW: Well, I appreciate that coming from you. Well, let's talk a bit more in general about the brain. I mean, you've been studying it for years. And let me ask you about your Aplysia. You've been given a Nobel Prize for your work in that. Tell our listeners, who haven't followed your whole career, what was so interesting about that sea slug and why you chose that.

KANDEL: I chose the sea slug for, I think, good reasons although some of my friends were skeptical at the beginning. I started to work on the hippocampus. So 1956, '57 when I got going in neurobiology. Brenda Milner had just shown that the hippocampus is absolutely essential for memory storage in people. If you have a lesion in the hippocampus in both sides, you have short term memory, but you can convert that short term memory into long term memory.

And Alden Spencer, a colleague of mine, and I at the NIH, explored the cellular properties. I was - had learned in medical school how to record from single cells. I studied the nerve cells of hippocampus. And he - Alden and I found many fascinating things about it. And our seniors were very excited about it but we didn't learn a darn thing about memory, because to look at memory you can't just study the nerve cell. You have to see how information gets into that region, how it's maintained over time.

And when we tried to see what was the sensory information coming into the hippocampus, we couldn't figure it out. It turns out its space, it's very complicated. So I decided that even though this is a humanistic problem, memory, we need to take a reductionist approach. A reductionist approach means I want to take the problem that interests me and find the simplest possible example and try to understand it deeply.

So began to look for a simple animal that could learn and remember that had a very simple nervous system and large cells so you could be an electrode in it, take a lunch break, come back and still be in the same cell. And Aplysia was ideal for that. And I was able to find that Aplysia shows memory for different kinds of learning, habituation, sensitization, classical conditioning, oprin conditioning, short term memory and long term memory for each.

So now we were in a position, looking at a very simple behavior, to see what are the cellular mechanisms underlying it. And we could define, for the first time, number one, that what changes with learning is the strength of how nerve cells talk to one another. That's called synaptic transmission. The way one cell talks to another is strengthened in certain forms of learning, weakened in other forms of learning. That's number one.

Number two, the difference between short term memory and long term memory is the growth of synaptic connections. With learning processes that increase the strength of communication you get a growth of new synapses. With forms of learning like habituation, when you learn to ignore a stimulus that's trivial you actually lose synaptic connections.

So this is a critical finding, that anatomical changes underlie lone-term memory. And then we asked what are the molecules that underlie this, and this led us to CREB.


KANDEL: And if you interfered with CREB, you had short-term memory, but no long-term memory. And this brought us back to the mouse. By this time, you could use genetically modified mice. And when I had learned a lot (technical difficulties) other people's work about what the hippocampus does, when I had very good tasks for exploring it, and we began to test some of these ideas.

And one of the things that has intrigued me most recently - both in Aplysia, which we continue to work with, and in the mouse - is that since synapses change, one had known there is machinery for synthesizing proteins at the synapse. So if you take your two arms and stretch them out a little bit, make one - the right hand makes contact with a group of cells, the left hand makes contact with another group of cells.

Each one of those fingers has machine (technical difficulties) proteins, so you can change synaptic connections at one branch, but not another. We began to explore - this is Cal 6C, when he was in my laboratory - what controls that. And we found a fascinating protein called CPB that regulates local protein synthesis, and it had unusual properties. It was a self-propagating protein. It could exist in two states as a single protein, and as an aggregate.

Only the aggregate did the job. The single protein did nothing. But that aggregate, once it formed, could perpetuate itself. Now, this kind of mechanism had been discovered before. Stanley Prusiner had found it in mad cow's disease, Jakob-Creutzfeldt disease. People have gone on to find that mechanism in a number of different contexts, and invariably, once the protein goes into its aggregated form, it kills the cell.

FLATOW: Mm-hmm.

KANDEL: This is a disease-producing mechanism, and we found the first example of a functional case in which this happens, but it's beneficial for the cell. So this had been lots of fun.

FLATOW: Yeah. I can see that. Fifty years down in three minutes here. Is it - do you think - well, you've talked about the work you're doing now in memory and this protein. What's the ultimate question? I mean, it seems like you're getting to - all ready to solve this one. What do you want to look at next? What's the ultimate question you want to answer?

KANDEL: How - there's so many questions. You know, we've made very nice progress. I've been very fortunate. This is the beginning of the problem. Memory is so complicated. How do we create false memories? How do we recall memories? You know, how do I remember my first love experience for the rest of my life? I mean, there are dozens of questions.

Also, memory is only one of the many problems.

FLATOW: Mm-hmm. I'm Ira...

KANDEL: Consciousness.


KANDEL: A phenomenal problem that we're just beginning because of Stanislas Dehaene and a number of people working in France making progress in understanding. So...

FLATOW: You think it can be understood?

KANDEL: Yes. You know, biologists are delusional optimists.


KANDEL: Until they hit a stone wall, they have no reason to believe that they can't overcome it. Because, periodically, you're told, you know, you can't do this and you can't do that. And you find that, you know, with new methodologies, you can do things you didn't think you could do before. And now, with the Obama initiative, I mean, brain science is recruiting extremely good people, developing fantastic methodologies that weren't at all available when I started out.

And all we need, to be honest with you, is good funding from the government. The government funding, by and large, has been disappointing. There have been serious cutbacks in funding. Howard Hughes is a magnificent organization. It supports several hundred scientists throughout the country and several abroad at a very satisfactory level.

But the national government, in part because of financial difficulties, has been cutting back on research. If there was unlimited resources coming along in size - I don't mean unlimited a realistic sense, but...


KANDEL: ...necessary to keep the workforce, which is quite effective and large and outgoing, I think we'd make progress. But this is not going to be solved in five years or 10 years. The brain is, you know...


KANDEL: ...a deep, deep problem. It'll take decades. We'll make progress all along. But a complete understanding of the human brain? Fifty or 100 years. So we've got to come back here 50 to 100 years...

FLATOW: All right. I'll meet you.

KANDEL: ...and review where we stand.

FLATOW: I'll meet you. I'm Ira Flatow. This is SCIENCE FRIDAY, from NPR. Writing in my calendar on this date 50 years from now...


FLATOW: meet Eric Kandel. How do you - you must take your - you must love your work so much that it keeps you young, Eric. I mean...

KANDEL: Well, I think this is true, in general.

FLATOW: must be so fascinating.

KANDEL: I think one of the most - you know, beyond choosing a partner, one of the most important decisions one has to make in life is the career you're going to have. And my career is my second passion. I like doing science more than most other things. I mean, I enjoy going to the opera. I love going to museums. You interviewed me about a book of art and science.

FLATOW: Right. Right.

KANDEL: I have a number of interests. But I love doing science. A tiny little idea in my head, putting two things together, almost obvious, but things I hadn't thought of before, maybe other people, gives me enormous pleasure.

FLATOW: Do ideas pop into your head all the time?

KANDEL: Not all the time.

FLATOW: I mean...

KANDEL: Certainly not good ideas. But, yes, when I drive, when I take a shower...

FLATOW: Right.

KANDEL: ...when I go for a walk. Yes.

FLATOW: Mm-hmm. And so it's a question of sort of doing triage on your ideas, about which ones you think will pan out? Or which you have the money to try?

KANDEL: Well, I mean, one of the wonderful things in science is, number one, you're surrounded by good colleagues, and they tell you what's bull and what's realistic. And also, you know, science can falsify. That's one of its powers. It's empirical. So if it's not too terribly time consuming, you run the experiment and you see.

FLATOW: Mm-hmm.

KANDEL: But deciding which experiments you want to run is an extremely important decision, because usually, it does take a commitment of time and of people.

FLATOW: Mm-hmm. Well, it's good to see that you're taking some time out, vacation time, to take a break.

KANDEL: I take a break and I enjoy it, but I must tell you, every day, I do a little bit of work, just to stay fit.



FLATOW: Yeah. Well, it's like running in place, just to, you know...

KANDEL: That's right.

FLATOW: keep up where you are. Thank you so much. It's always been a - it's always our pleasure to talk to you, Dr. Kandel.

KANDEL: It's very nice talking to you, Ira.

FLATOW: And please keep us up to date on how this is going. You know, where...

KANDEL: I certainly will.

FLATOW: We always want you to come back.

KANDEL: I certainly will.

FLATOW: And have a great weekend and have a good holiday.

KANDEL: You, too.

FLATOW: Happy New Year to you.

KANDEL: Happy New Year to you, Ira, and thank you for having me on this broadcast.

FLATOW: Always, Eric. Thanks a lot. We're going to take a short break, and when we come back, we're going to switch gears and talk about a space telescope called WISE, and how it's got a new mission to search for an asteroid, possibly, with our name on it. So stay with us. We'll be right back after this break. Transcript provided by NPR, Copyright NPR.