What’s New in Neurogenesis
An Interview with Fred H. Gage, Ph.D.


by Brenda Patoine

March, 2007

Q: Your 1998 Nature Medicine article reporting that new neurons are regularly born in the adult human hippocampus helped overturn a long-held central tenet of neuroscience—and set off a wave of follow-on research that is just beginning to crest. What has surprised you most in the ensuing research?

A: I admit being surprised at the flood of articles, and by the sheer number of individuals who have gone into the field. One reason for that is because the tools that are available now for investigating neurogenesis are repeatable in the laboratory, and better techniques are being developed all the time. The repeatability of the findings and the consensus that developed quickly in the field have been encouraging.

I remember presenting our findings on adult neurogenesis at the Society for Neuroscience annual meeting 10 years ago, when there were only about 10 posters on the subject. There was a lot of skepticism about it, and a lot of debate going on. Contrast that with the 2006 meeting, where there were aisles and aisles of posters on neurogenesis. Investigators are now looking at this from many different directions, not just as a phenomenon itself but also at its role in epilepsy, stroke, spinal cord injury, etc. And there are reports of neurogenesis occurring in multiple brain areas, outside of the hippocampus and the olfactory bulb where much of the work has focused. The controversy remains in some domains, but the core observations about neurogenesis are solid.

Q: You’ve recently developed a complex computer simulation of hippocampal neurocircuitry to track the developmental progression of newly born neurons. Describe what your aims are and why a computational model is the best approach.

A: It has become clear over the last 10 years that neurogenesis is a process, not a single event. It encompasses a series of phenomena in the adult brain, one of which is self-renewal or maintenance of the stem cells within the niche in the brain where these cells originate. But neurogenesis as a process also involves the migration of those stem cells from the niche; the initiation of differentiation into various types; polarization, in which dendrites form and axons extend out from the cell body in a polar manner; and connectivity, which is how the inputs from, say, interneurons in the entorhinal cortex or inhibitory neurons in the hylus make contact with these new cells. The process as a whole takes time—up to a month or more from the cells’ emergence to their integration into the system. Over the course of their maturation, the cells change their pattern of activity, and there is a transient period of hyperexcitability during which the cells are particularly sensitive to input signals. Finally, there is the question of what role these cells play in the normal functioning of the intact hippocampus and dentate.

For each of these steps, there is a plethora of new data being generated. So, in trying to decide what the next most important questions were and what experiments to do next, we wanted to take into consideration as much of the existing information as possible. Our goal was not so much to model the hippocampal system as to describe the current state of knowledge about neurogenesis mathematically.

Because adult neurogenesis is a newly recognized phenomenon, it doesn’t easily fit within the current understanding of hippocampal function, in terms of physiology, anatomy, or behavior. In order to better understand how it may fit in, we need to put together as much information as we can about what is actually happening throughout the process of neurogenesis relative to the existing circuitry. From there, we can propose theories that can be tested to see whether we are right or wrong.

Q: Where are you with that research right now?

A: The modeling has led my colleagues Brad Aimone, Janet Wiles, and me to generate certain hypotheses about the function of newly born cells. We recently published a completely theoretical article (in Nature Neuroscience, June 2006) on a potential role that the newly born hippocampal granule cells could play in information processing and memory, specifically in what we’re calling “time-coding” of events.

Our theory is that the new neurons link existing events. So, you have an event that occurs at a certain time, then you have another event, which occurs at a second time point. The current thinking about the dentate gyrus is that part of its role is to keep these two events separate—what we call sparsification. By looking at a model encompassing all of the existing data on what these cells are hypothesized to do in the context of the existing neurocircuitry, we can deduce that they seem to be linking these two independent events.

When you remember a past experience—let’s say a summer vacation—you may pick up a very specific memory of say, a dinner you had and the people who were there at the table. As you draw that memory up, it will at the same time open up other memories of things that happened at the same time but were not directly related to that dinner and the specific events that occurred at it. Our brains have a way of linking things together that occurred generally close in time; we hypothesize that perpetually generating new cells is a mechanism by which the brain accomplishes this. While we may mostly think of memory associations in terms of very short time frames—seconds or minutes—this is more of an extended linkage related to the time course in which these cells remain in the process of integration into the circuitry.

Q: How are these new understandings contributing to an evolving view of hippocampal function and plasticity?

A: In the theory that we’re currently working with, the general idea is that these new cells provide an added level of plasticity, of dynamic action within a hard-wired circuit that is pretty dynamic to begin with. This is really another level of plasticity beyond what we see in existing hippocampal neurons, which are known to change synaptic affinity in response to activity. With neurogenesis, new synapses and whole new neurons are actually being added into the circuitry. So it is the level of plasticity that is important.

Originally, the hippocampus may have been thought of as a structure involved in learning and memory, but it’s now clear that it’s less involved in long-term memory. So the earlier view was that it would be unreasonable to have newborn cells in a circuitry that involves long-term memories because that might disrupt existing memories. That is less of a concern now because the current conceptualization of the hippocampus is that it is involved more in the formation of memories as opposed to memory storage.

Answering questions related to the function of these cells is important, and we will have a much better handle on this over the course of time. I suspect many of these questions will be resolved in the next five years. What’s driving the excitement and the theories is the acquisition of basic knowledge about the sequence of events underlying the cells’ maturation. This is a great example of how basic biological research drives applications and drives the ability to extrapolate about what is going on.

Q: What would you say is the focal point of neurogenesis research right now?

A: There are a few things. One is understanding the cellular/molecular events that constitute the progression from an adult stem cell in vivo to an integrated functioning neuron. For many of us, that’s enough—just to understand that process of fate/maturation in an adult context. A separate set of questions is: why does this area of the brain allow this to happen, when it doesn’t really happen anywhere else? What’s so important about this part of the hippocampus?

The third layer is applying this to understanding diseases. In many disease states it looks like there are changes occurring in the rate and the function of these cells relative to the normal progression: either they’re not developing as much, as in aging and depression, or they’re developing too much, as in epilepsy and stroke. For some of these diseases, this is actually the first time there has been an anatomical locus to pay attention to. This is particularly true in the affect disorders. This area of the brain (the hippocampus) is known to be involved in depression, schizophrenia, and others, but now there is evidence of dynamic changes in neurogenesis occurring in correlation with changes in disease states.

Q: What does all of this mean to the average person? Is there a “neurogenic” lifestyle that will help us ramp up the volume of new neurons in our brain?

A: Based on the experimental evidence that currently exists, there are several things that one could conclude. Physical exercise, environmental complexity, and specific types of learning are three conditions that have been robustly shown to increase neurogenesis. On the other side of that, both acute and chronic stress decrease neurogenesis, so the implication is that decreasing the pathogenic properties of stress should have less of a detrimental effect on neurogenesis.

Given those observations, my guess is that it’s probably a pretty good idea to globally decrease stress, increase physical activity and environmental enrichment, and seek to continually acquire new information within your environment—to continue to be stimulated, in other words. I don’t think that’s going out on a limb.