Why the brain’s GPS fails with age, and how some minds defy it

Now, researchers at Stanford Medicine and their colleagues are uncovering what goes awry in older brains when spatial memory falters and whether these changes can be prevented.

In a new study comparing young, middle-aged and old mice, the researchers found that activity in the medial entorhinal cortex — sometimes likened to the Global Positioning System of the brain — becomes less stable and less attuned to the environment in elderly animals. Those with the most impaired activity in this brain region were the most confused on a spatial memory test.

“You can think of the medial entorhinal cortex as containing all the components you need to build a map of space,” said Lisa Giocomo, PhD, professor of neurobiology and senior author of the study to publish Oct. 3 in Nature Communications.

“Before this study, there was extremely limited work on what actually happens to this spatial mapping system during healthy aging.”

Although, on average, elderly mice were noticeably worse than their younger counterparts at navigating their environments, there was wide variation among them — a sign that spatial memory decline may not be an inevitable part of advanced age.

Mental maps

The medial entorhinal cortex is an essential part of the brain’s navigation system. It contains a variety of cells that track different information, including the animal’s speed and head direction, as well as the dimensions and borders of a space. For the new study, the researchers focused on so-called grid cells, which create a map of the environment, almost like a longitude and latitude system.

They studied mice in three age categories: young mice approximately 3 months old, middle-aged mice approximately 13 months old and old mice approximately 22 months old. These ages roughly correlate to human 20-year-olds, 50-year-olds and 75- to 90-year-olds.

The researchers recorded the brain activity of slightly thirsty mice as they ran virtual reality tracks looking for hidden rewards — a lick of water. They ran on a stationary ball surrounded by screens that displayed the virtual environment, like a mouse-sized treadmill in a mouse-sized Imax theater.

Each mouse ran the tracks hundreds of times over six days. (Mice are naturally avid runners, the researchers noted.)

With enough repetition, mice in all age groups could learn the location of a hidden reward on a particular track. By day six, they stopped only to lick at the reward locations. Accordingly, the grid cells in their medial entorhinal cortex developed distinct firing patterns for each track, as if building custom mental maps.

Switching tracks

But on a more challenging task in which the mice were randomly alternated between two different tracks they had already learned, each with a different reward location, the elderly mice were stymied — seemingly unable to determine which track they were on.

“In this case, the task was more similar to remembering where you parked your car in two different parking lots or where your favorite coffeeshop is in two different cities,” Giocomo said.

Unsure of where they were, the old mice tended to sprint the rest of the track without bothering to stop and search for rewards. A few took a different tactic and tried licking everywhere.

Their grid cells reflected their confusion. Despite having developed distinct firing patterns for each track, their grid cells fired erratically when the tracks were alternated.

“Their spatial recall and their rapid discrimination of these two environments was really impaired,” said Charlotte Herber, PhD, an MD-PhD student and lead author of the study.

The findings seem to align with human behavior. “Older people often can navigate familiar spaces, like their home or the neighborhood they’ve always lived in, but it’s really hard for them to learn to navigate a new place, even with experience,” Giocomo said.

In contrast, both young and middle-aged mice understood the assignment by day six, and their grid cell activity swiftly matched whichever track they were on.

“Over days one through six, they have progressively more stable spatial firing patterns that are specific to context A and specific to context B,” Herber said. “The aged mice fail to the develop these discrete spatial maps.”

The middle-aged mice had somewhat weaker patterns in their brain activity, but they performed very similarly to the young mice. “We think this is a cognitive capacity that at least until about 13 months old in a mouse, or maybe 50 to 60 years old in a human counterpart, is probably intact,” Herber said.

Super-ager

Though young and middle-aged mice performed uniformly within their age groups, the oldest set showed more variability in spatial memory.

Male mice generally performed better than female mice, though the researchers do not yet know why.

One elderly male mouse stood out: It aced the test, remembering the hidden reward locations on alternating tracks just as well as, if not better than, the young and middle-aged mice.

“It was the very last mouse I recorded and, honestly, when I was watching it run the experiment, I thought, ‘Oh no, this mouse is going to screw up the statistics,'” Herber said.

Instead, the super-ager mouse turned out to confirm the link between grid cell activity and spatial memory. Its grid cells were as unusually sprightly as its behavior, firing clearly and accurately in each environment.

“The variability in the aged group allowed us to establish these correlative relationships between neural function and behavior,” Herber said.

The super-ager mouse also encouraged the researchers to look for genetic differences that might underly variability in aging. They sequenced the RNA of young and old mice and found 61 genes that were more expressed in mice with unstable grid cell activity. These genes could be involved in either driving or compensating for spatial memory decline, the researchers said.

The gene Haplin4, for example, contributes to the network of proteins that surround neurons, known as the perineuronal net, which could help shore up grid cell stability and protect spatial memory in aging mice.

“Just like mice, people also exhibit a variable extent of aging,” Herber said. “Understanding some of that variability — why some people are more resilient to aging and others are more vulnerable — is part of the goal of this work.”

Researchers at the University of California, San Francisco, contributed to the study.

The study received funding from the Stanford University Medical Scientist Training Program, the National Institute on Aging, the National Institutes of Health BRAIN Initiative (grant U19NS118284), National Institute of Mental Health (grants MH126904 and MH130452), the National Institute on Drug Abuse (grant DA042012), the Vallee Foundation and the James S. McDonnell Foundation.