Aging and the Changing Landscape of Memory

Gathering evidence suggests that the brain’s ability to make new neurons in a region called the hippocampus—specifically in a sub-region called the dentate gyrus—is important for the acquisition of memories. This capacity for neurogenesis declines as we age, and the result is a decline in at least one kind of memory, studies show. But that’s not where the story ends. As neuroscientists develop a more complex understanding of how memory works and the impact of aging, they’re learning how exercise, an improved diet and staying mentally active may boost our ability to make new neurons where they’re needed to preserve and maintain memory.

Three neuroscientists talk about this emerging consensus and what they hope to learn in coming years about the changing landscape of memory as we age.

The participants:

• Fred "Rusty" Gage – Professor in the Laboratory of Genetics at the Salk Institute for Biological Studies and member of the executive committee of the Kavli Institute for Brain and Mind (KIBM) at the University of California, San Diego.
• Scott A. Small – Professor of Neurology at Columbia University, Director of the Alzheimer's Disease Research Center at Columbia, and member of the Kavli Institute for Brain Science (KIBS).
• Craig Stark – Professor, Department of Neurobiology and Behavior, and Director of the Center for the Neurobiology of Learning and Memory, University of California, Irvine

The following is an edited transcript.

THE KAVLI FOUNDATION: Rusty, you’ve called this an exciting time for thinking about how a decline in our ability to make new brain cells as we age is connected to age-related memory loss. Why is that?

FRED GAGE: It's an exciting time for several reasons. Along with gaining a better understanding of the different types of memory, we’re getting a better understanding of how we acquire, store and retrieve information. For example, with the hippocampus, which we agree is involved in the acquisition of new memories, the wiring diagram is being worked out. And we’re understanding better how the dentate gyrus receives input from a structure called the entorhinal cortex, which interfaces between the hippocampus and the neocortex, where higher functions like conscious thought and language occur. What this means is we’re able to talk about memory in terms of cells and connections among cells in ways that we really haven’t been able to before. We’re able to talk now about the birth of new neurons in terms of their influence on existing circuitry. And that gives us a lot more confidence about understanding something about the mechanisms involved in memory and age-related memory decline.

SCOTT SMALL: For me, it’s an especially exciting time because of the technologies we use to image the brain in living human beings. My perspective on this is also very much influenced and colored by my focus on dysfunction – on aging and disease. I’m interested in studying how different regions seem to be targeted by different processes or disorders. And, as Rusty said, it’s an exciting time because we’re building a more sophisticated and complex understanding of memory. There is really very good evidence that different regions do slightly different things, and that's important, as all of us here are trying to understand why that is and how we might intervene to ameliorate age-related memory decline.

CRAIG STARK: I would add one more idea. In our studies of memory and its decline as we age, researchers have tried to look at what's going on at the level of an individual neuron and relate that to a behavior or dysfunction. In many areas of neuroscience, it's incredibly difficult to make these leaps. However, when we study the hippocampus and the dentate gyrus in particular – brain structures that are so important for memory – there’s been a lot of progress in this area, so we can make those leaps. This is helping us move forward a lot faster. There's a lot of excitement, and it's a lot of fun to do.

TKF: When it comes to the brain, what is the general picture of aging and memory loss that’s emerging?

SMALL: We know that neurogenesis in the dentate gyrus declines with aging. Knowing that, the next question becomes: “What kind of memory is most linked to the dentate gyrus?” This is very much along the lines of what Craig was saying, that once you pinpoint an area of the hippocampus that's affected, you can zoom up and down the different levels of analysis to ask different questions. What’s causing this decline in neurogenesis, what kinds of symptoms does this decline lead to, and then how can we intervene? At the same time, it’s important to recognize that the hippocampus is part of a circuit in the brain, and so if one part of that circuit becomes affected other parts of that circuit, in theory, could be affected.

GAGE: I think we would all agree with Scott’s point: you don’t want to focus just on the dentate gyrus because it’s only part of a circuit. There's very important input from the cortex to the dentate gyrus, so as soon as you look at it you also want to look at the connections it has – the inputs and the outputs. You really would not want to think about it in isolation but rather as part of the circuit. This is a change in how neuroscientists are thinking about function – that you shouldn’t really assign one structure to a function. Instead, you want to think about that structure in the context of its connections.

TKF: Do we know why only a few structures of the brain, including the dentate gyrus, make new cells in adulthood and others don’t?

GAGE: In many mammalian species, neurogenesis is limited to certain structures. And I think there are a couple of ways to think about that. One is that it's very costly to have cells divide. It takes a lot of energy. The other is it’s pretty dangerous to have cells dividing in any organ without a fair number of controls over it, in terms of cancer issues. So you have these two sides of biology working to restrict cell division and maturation. That said, the fact that there is neurogenesis at all in the adult brain is rather remarkable.

TKF: Is this why some people are concerned about boosting neurogenesis in people where memory has declined – because there could be health repercussions?

GAGE: I've certainly heard that. As we understand the mechanism for how cells divide, it involves a molecular signaling pathway that we know is also activated during the development of tumors. So if you're using drugs to activate this certain state, there is some concern that it might cause an uncontrollable growth of tumors.

STARK: There's another reason why boosting neurogenesis can be a bad idea. Information processing in the brain is happening in a massive network, and there’s a delicate balance of passing signals around in this network. Now you toss in a bunch of random new neurons with all sorts of random connections, and perhaps they are very excitable neurons. You're throwing a bit of a wrench into the works. And so making new neurons in the brain may be a good thing because they may give you the flexibility to rapidly store something new. But there might be times when it's not such a good idea, and tossing some kind of new character into the mix is just going to upset the cart. Neuroscientists who study brain development in childhood know this. In fact, if the brain prunes back neurons and connections too slowly during childhood, that leads to all sorts of problems. So it's not always the case that more is better.

TKF: What are some of the debates among neuroscientists about the role that various parts of the brain play in the acquisition of memories – particularly the importance now given to the dentate gyrus?

SMALL: One of the more interesting debates that I’ve heard is whether the dentate gyrus is important at all for the acquisition of information. And this comes out of a whole different school of thought that the focus on the hippocampus is for allowing us to navigate through space and move around in our environment. But we now know, however, that the dentate gyrus is important for a particular kind of memory called pattern separation, which allows us to distinguish things that are similar, like faces, but also clearly distinct. Anytime I hear a person say, “Gee whiz, I'm having a hard time remembering names of people” – assuming they don't have Alzheimer's – it certainly is a declaration that something is up in the dentate gyrus.

TKF: Scott, Is there a way to tell if people experienced age-related memory decline in the past, when in general they died younger? Or do we see this decline in neurogenesis and memory only because humans are living longer today?

SMALL: There is a sort of linear decline, a slide from around 30 or 40 years old onward, so if most people in the past died in their 40's they would not have noticed it is as much as if they died in their 80's. Now, it does seem to be the case that all mammals that have a hippocampus and live into old age clearly have hippocampal dysfunction, and there’s growing evidence that the seat of this dysfunction is the dentate gyrus. For humans, we are living longer and more stressful lives, and in some respects more unhealthy lives. On top of that, I do really think there is something inherent in the aging process against which the dentate gyrus is sensitive.

GAGE: I would add that the things we do, how we behave, what we eat and what we do physically all can play a role in how well the brain functions. Our memories are formed in the brain, and the brain is an organ. Just like other organs that deteriorate with age, the brain is going to deteriorate with age in a global sense. But what we do as individuals can impact the rate of that decline. Maybe 100 years ago – and I don't know that this is true, Scott and Craig – there may have been just as much decline in memory because of the general ill health of people.

STARK: I agree Rusty, and it's an interesting point. I mean, is aging’s effect a kind of internal clock in the dentate gyrus that just kind of peters out with the passage of time, or are there external events that play a role? I think it's probably both. There are many things that happen as we age. We often become sedentary, have glucose intolerance, and we may well have high cholesterol and high blood pressure. All these things deserve to be on the list of potential causes of age-related dentate gyrus dysfunction. So, to the extent that this is at least partially true, it's also true that if we could improve our general health, our diet, and our exercise regimen, there are ways of ameliorating or preventing even profound age-related memory loss.

TKF: So parts of the brain important for memory, including the dentate gyrus, may have prematurely aged in people long ago, as opposed to people today who have better nutrition and exercise habits, as well as more access to quality health care.

GAGE: Exactly. But there are other factors at play. There are individual genetic differences among people that are important. We all know that in the past there were people who lived very lengthy and very productive lives, as there are today. Part of the reason may well be that these people were born with a better set of DNA sequences that made them less vulnerable to the same challenges that other people would not respond well to.

SMALL: There is some really interesting work from twin studies, and it suggests that the ability to remember is much more intimately related to your genetic makeup in later life than earlier in life. So this does support what Rusty is saying. The mechanisms that cause age-related memory loss or decline probably have, at least in part, a genetic component.

TKF: So do we have some control, perhaps, over age-related memory decline?

GAGE: Most of the work that's been done in a rigorous manner on this topic has been done in experimental animals, so that's what I'll have to refer to. There hasn't been that much work done in humans, although it's beginning now and it's a very exciting time for that. Previous studies show that exercise increases blood flow, and increasing blood flow is related to elevated levels of certain proteins in the blood, such as IGF–1, that move more readily into the brain. And we know this particular protein plays an important role in cell proliferation and cell health in the hippocampus. So physical exercise appears to increase cognitive function in a well-controlled setting. Separate from that, when animals are exposed to a variety of complex environments over an extended period of time, there appear to be positive benefits for certain aspects of memory.

As far as diet, we want to understand what elements in the food we eat affect the biology of the hippocampus. The more we can understand about the cellular and molecular mechanisms of a correctly-functioning or optimally functioning circuit, the more we can think about regulating our intake of foods to optimize this. It's an area of study that’s in its infancy. But I do think that genetics, diet, consumption or lack of consumption, physical exercise, and blood circulation - plus this concept more loosely of complexity and enrichment - are all important considerations in memory function as we age.

TKF: How is this emerging picture of the interplay between neurogenesis and memory shaping your individual current research?

STARK: For me, the establishment of adult neurogenesis and its role in memory has opened up a whole new avenue to look at age-related cognitive decline. It’s a player in how memory operates and how memory changes with age. I’d like to figure out ways in which neurogenesis in humans can be manipulated. I’d also like to be able to measure levels of neurogenesis in the dentate gyrus. And, I want to learn what functions – memory included – are correlated with neurogenesis as well as what changes during aging are not correlated with neurogenesis.

GAGE: It's a very exciting time with all the tools that are currently available. When I go to meetings and talk to my ex-students and colleagues in the field, there’s just a plethora of new information that's helping us understand the molecular process of the birth of new cells in the dentate gyrus. We’re learning new things about how cells divide and migrate, how neurons initially send dendrites in the right direction, how axons make connections among neurons. And now we’re particularly benefitting from computer science. Pattern separation, this type of memory we talked about that occurs in the dentate gyrus, has a computational basis to it. There is an attempt now to merge the biology from a bottom-up manner with the top-down computational sciences. The goal is to understand cellular and molecular mechanisms in the context of a computer algorithm, if you will, that will help us better understand the memory concepts that we've been talking about. That merger between these different disciplines is very exciting. I'm sort of playing that role specializing in the bottom-up approach to things. And we all meet in the middle. It's fun.

SMALL: Listening to both Craig and Rusty, I notice how much I have evolved. If Craig is a sort of cognitive neuroscientist trying to understand memory, and Rusty defines himself as a basic neuroscientist, I would be the translational clinical neuroscientist. That's really emerged as my main question: now that we understand the anatomy of hippocampal memory, where it happens and some of the reasons why it happens, can we translate that into ways to prevent or ameliorate age-related memory decline?

Although drugs would be acceptable, I am particularly interested in exploring non-pharmacological roles – things like modifying behavior, diet and genes. At this point, I am very much wearing my clinical hat. Now that we’ve gained enough insight, can we in a very rational way translate that into therapy? I think Rusty's point is a good one: It shouldn't be surprising, but somehow it is when you make the point that what you eat can influence your brain. If that's a truism, can we identify the components of diet that might enhance dentate gyrus function? That's really a question I want to pursue.

TKF: What do you think is achievable in the next ten years?

GAGE: I think we’ll have a better understanding of how the brain receives information and then encodes it within neurons and within circuits. We may not even use the same words we use today to describe it. But it's a serious challenge for the next generation of experimenters to get a handle on the so-called “engram” – the physical traces in the brain of memory acquisition, storage and retrieval. Am I underplaying something, Scott and Craig, or do you think that it’s better understood than it is?

STARK: No, I think that our field has been searching for the engram, as it were, for four decades at least. There've been a couple of spots where there are glimmering bits and pieces. We all hold this basic idea that memory is stored by changing the strength of the connections in networks. But in terms of understanding that code, and exactly how that works, we are still grappling with it. It's not for lack of trying, and it's not for lack of smart people working on this. But you're talking about a network of 10 billion neurons and a quadrillion synapses – and that’s just barely scratching the surface of the complexity. So it’s not an easy thing to make good sense out of.

SMALL: What I very much want is to have a neuroimaging tool that could more precisely track changes that are related to neurogenesis in living human beings. Just 20 years ago, the idea of neurogenesis in the adult was questioned, but now there are all these beautiful technologies that Rusty and his colleagues have developed in mice and in postmortem mouse brains that clearly demonstrate that it happens. So I’m optimistic that within the next ten years we’ll have in vivo human neuroimaging techniques that could measure neurogenesis with the kind of fidelity that we’ve shown in animals.