Skip to Content

The Persistence of Memory

Professor Li-Huei Tsai’s years of hard work on Alzheimer’s point to the possibility of restoring what diseased brains have forgotten.

April 21, 2009

Li-Huei Tsai was four years old when she first saw the horrors of Alzheimer’s disease. She was living with her grandmother in a small seaport town north of Taipei. Late one morning, they were walking toward home when lightning cracked in the sky. That was frightening, but what followed was far worse. Her grandmother, in her early 50s at the time, became disoriented. She had no idea where they were, or how to get home; Tsai was too young to know the way herself. They were utterly lost. “It was a really, really scary experience,” says Tsai, MIT’s Picower Professor of Neuroscience in the Department of Brain and Cognitive Sciences. “She died two or three years later.”

Decades have passed, but that experience still drives the 49-year-old scientist. In the past few years she has uncovered new details about the devastating effects of Alzheimer’s on the brain and demonstrated that it’s possible for mice to retrieve memories that seemed to have been lost forever–a finding that one contemporary hailed as “game-changing.” At the same time, she’s spearheaded research into abnormal brain development, neuropsychiatric disorders like schizophrenia, and other brain diseases. Her results have been published in top journals such as Nature, Cell, and Neuron.

“She has tremendous depth of knowledge and experience in neuro­science, learning, and memory,” says Leonard Guarente, Novartis Professor of Biology, who has collaborated with Tsai. Beyond that, colleagues cite her knack for identifying the big, pressing questions in neuroscience, as well as for bringing in talented young ­researchers and managing their projects simultaneously. In chasing down those big questions, she’s shown remarkable persistence–but she didn’t originally set out to fight the disease that left her grandmother standing in the street, disoriented and helpless. That quest began with the discovery of a mysterious protein at the start of her career.

Tsai’s lab on the fourth floor of Building 46, which houses the Picower Institute for Learning and Memory, is a busy place: she oversees the research of a diverse mix of more than 20 graduate students and postdocs. During a recent visit, lab-coat-sporting scientists were grabbing plates full of cake from a conference room as Tsai hustled into her sunlight-filled office. A row of empty champagne bottles sat atop her bookshelf, the vestiges of postpublication celebrations. She tries to cultivate a family atmosphere in her lab, whether that means organizing birthday parties or popping champagne corks. Of course, managing all those research projects makes for a hectic life (she’s so busy she sometimes brings her 11-year-old daughter to conferences so they can spend more time together). She’s been known to realize on the way to the airport that she’s forgotten her passport; when it comes to such administrative details, she confesses with a laugh, “I’m a disaster.”

Tsai followed a circuitous career path to her gleaming corner office. She initially studied to be a veterinarian, coming to the University of Wisconsin-Madison from Taiwan in 1984 to pursue a master’s in the field. But after sitting in on a series of lectures delivered by the Nobel-winning cancer researcher Howard Temin, she found herself drawn to more basic research. “I was very inspired by his work, and realized I really liked lab work,” she recalls. Tsai dropped her childhood dream of becoming a vet and, following Temin’s lead, switched her focus to cancer. In 1990, she earned a PhD from the University of Texas Southwestern Medical Center in Dallas.

The following year, as a postdoc in cancer specialist Ed Harlow’s laboratory at the Massachusetts General Hospital Cancer Center, she stumbled on an odd protein. Tsai had been charged with identifying enzymes known as cyclin-dependent kinases, which typically play a role in cell division. Her task was to discern their function by analyzing and tracking cancer cells as they multiplied in petri dishes. But one of the molecules, CDK5, wasn’t behaving like the rest: it didn’t appear to be doing much of anything. “I became very intrigued by this particular kinase, simply because it wasn’t entirely easy to work with,” she recalls. “It was very peculiar.”

The molecule wasn’t relevant to her assignment, because it played no role in cell division–so it would have been easy to forget about it and move on. But not for Tsai. “I didn’t want to just give up and say, ‘Oh, this thing, it doesn’t matter,’” she says. “I decided to give it one last shot.”

Since CDK5 was dormant in the cancer cells, Tsai changed the medium. She pulled together a variety of tissue and organ samples, again checking to see whether CDK5 might be active. In most, it did nothing. Yet it was active in the brain. “That actually was the first time I seriously looked at the brain and started to discover all these fascinating things about the brain,” she says. Her days as a cancer researcher were coming to a close. CDK5 was leading her to a new calling.

While recounting her discovery of CDK5, Tsai laughs and says, “I was extremely lucky.” But the torrent of papers that followed this one finding had more to do with pure determination.

First, she found that the protein doesn’t act alone. To become active, CDK5 needs to bind with a protein she called p35, which is active only in the brain. To find out what this combo was up to, Tsai, then at Harvard Medical School’s pathology department, genetically modified mice so they couldn’t express p35. She and her colleagues shut down the gene that produced p35, halting CDK5’s activity, too. In these mice, she says, “we found an extremely intriguing defect in brain development.” The animals were prone to seizures, and in certain parts of their brains, their neurons were arranged differently from those in healthy mice. Without p35, and the associated activity of CDK5, their brains just didn’t develop properly.

Yet she soon learned that CDK5 wasn’t purely benevolent. As her group continued studying it, they noticed an odd, truncated version of that partner, p35. This molecule, dubbed p25, kept turning up in diseased or damaged brains in mice–and in tissue samples from deceased Alzheimer’s patients. “We found that this particular protein was more associated with neurotoxic conditions,” she says.

The p25 also drove the activity of CDK5, so Tsai developed a group of mice that overexpressed the new molecule when the antibiotic doxycycline was removed from their diet. This allowed her to crank up the activity of CDK5 instead of shutting it down. And when she did so, the mice developed Alzheimer’s-like effects in just a few weeks. Learning and cognition suffered, neurons died in massive numbers, and the tangled beta-amyloid fibers typically found in the tissue of deceased Alzheimer’s patients turned up in their brains too. Though Tsai had already shown that CDK5 is critical for proper brain development and function, the experiment proved that too much of the protein can be seriously detrimental. “When this [p25] is produced,” she says, “it drives CDK5 to the dark side. It makes it toxic to cells.”

Having shed light on the mechanisms driving the progress of Alzheimer’s disease, Tsai, who had come to MIT in 2006, wanted to figure out how to fight or even reverse some of the symptoms. She and postdoc Andre Fischer, now at the European Neuroscience Institute in Göttingen, Germany, knew of evidence from other studies that physical exercise and environmental enrichment–such as the addition of companions and toys–increases brain function in mice. So they decided to test what would happen if they tried this technique with their Alzheimer’s-like mice.

In one experiment, they trained mice to find and remember a platform submerged within a murky pool. Then they induced the Alzheimer’s-like effects. The mice swam aimlessly, unable to locate the spot. But when the researchers moved the mice to a more stimulating environment and then placed them back in the swimming pool, the rodents kicked directly to the platform. Those supposedly lost memories had returned.

Why this worked was a mystery, but Tsai thought environmental enrichment might have affected genes associated with learning and memory. She also knew of a set of enzymes called histone deacetylases, or HDACs, that were believed to suppress the activity of some cognition-related genes. Hoping to mimic the effects of environmental enrichment, Tsai and ­Fischer repeated the swimming experiment–this time injecting mice with drugs called HDAC inhibitors, which blocked these enzymes. They reported in a 2007 Nature paper that the drugs improved cognitive performance in Alzheimer’s-like mice, enabling them to recall the location of the platform.

The results imply that restoring seemingly lost memories might also be possible in people. “Even in those patients that seem to lose their memory, we don’t think the memory is really erased,” she says. Tsai suspects that the massive neuronal die-off damages the brain’s circuitry–the wiring that connects different regions. Rather than promoting neuron growth, she says, the new environment and the HDAC inhibitors strengthen synapses and dendrites, boosting connections between regions. In other words, they repair the circuits.

Though she’s still leading projects on brain development, as well as on the neuroscience of schizophrenia and other disorders, her ongoing work with HDAC inhibitors has her particularly enthusiastic, since it points to an entirely new way to fight Alzheimer’s. That 2007 paper just hinted at the possibilities. “We now have some very exciting observations of one particular HDAC that’s responsible for a negative regulation of learning and memory,” she says. Targeting that enzyme, she explains, could rewire the broken circuitry and improve cognition in Alzheimer’s patients.

“We’re very hopeful,” she says. “We may have something in the next few years that could be safe and beneficial enough to go into humans.” Basic research may remain her first love, she adds. “But if my work can do something for the community or society, I would be so overjoyed.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.