Alone with her thoughts, Dr. Karen Ashe couldn't help wondering if the answer was staring her in the face.

Garlic. Fish oil. Cabernet sauvignon.

If you believe the research, they all contain something that's good for the memory.

So does green tea. And pomegranate juice.

As a neurologist, Ashe was skeptical. But when she took a closer look, she found that dozens of substances appeared to prevent memory loss in studies of lab mice.

Something in her gut told her they couldn't all be wrong.

That sparked an idea. One that could draw tens of thousands of Minnesotans into an ambitious experiment to try to turn the tide against Alzheimer's, one of the most feared diseases of our time.

For 20 years, Ashe has been quietly unraveling the mysteries of Alzheimer's -- which can rob even the most brilliant mind of its memories -- from her lab at the University of Minnesota. Her discoveries have made headlines around the world. At 57, she's won almost every major award in her field short of the Nobel Prize, and her admirers say that, too, is just a matter of time.

But time, she fears, is running out. With no cure in sight, the Alzheimer's epidemic is poised to overwhelm the health care system in this country, soaring from 5 million souls to more than 13 million by 2050. Ashe believes we still have a chance to prevent it. For that, the notoriously reclusive scientist has started to step out of the shadows.

Enlisting supporters

The Minneapolis skyline glitters through the penthouse windows as Beverly Grossman welcomes Ashe with a warm hug. Grossman, an effervescent woman in her 70s, is expecting some friends for a fundraiser at her home overlooking the Minneapolis Sculpture Garden.

The invitation read: "What if Alzheimer's Disease could be prevented? ... A conversation with Dr. Karen Ashe."

Ashe has traded her white lab coat for a stylish black dress with a Mandarin collar. Waiters swirl by with trays of wine and canapes, past an eye-popping Andy Warhol portrait of the hostess.

Ashe stands primly to one side as guests approach her. More often than not, they want to tell her about someone they once knew.

A few steps away, a photo of Bud Grossman, Beverly's husband, rests on a piano by a spray of white orchids.

"Did you know Bud?" a family friend asks.

"No," Ashe replies softly, "I didn't."

"We knew him," the woman says, "in his glory days."

Bud Grossman, who made a fortune in the auto and oil business, was part-owner of the Minnesota Vikings and chair of the Minnesota Orchestra. Even now, his wife can't bear to talk about his struggle with Alzheimer's, which ended his life at 88. "He was a brilliant man," Beverly says.

In 2007, she decided to do "something meaningful" to honor him. She donated $5 million to the university for Ashe's program, now called the N. Bud Grossman Center for Memory Research and Care. She uses evenings like this to enlist others in the cause.

At 6:30 p.m., Ashe takes center stage in front of the Warhol painting, in the soft glow of a spotlight.

Up to now, she tells the guests, the search for a cure has turned up empty. But scientists think they know why: It may be too late to reverse dementia once it sets in. "We need to treat people before the symptoms appear," during a "silent phase" of the disease that seems to start years earlier, she said.

Now, the search is on for something to prevent dementia -- much like baby aspirin cuts the risk of heart attacks.

Ashe has a two-step plan. First, to pick the most promising substances, based on previous research, and scrutinize them in her lab. Second, to give them to large groups of people with healthy memories and track their dementia rates over time. The goal is to find something cheap, safe and readily available -- not the kind of research that a drug company eager for profits would fund.

If she can pull it off, this would be one of the biggest experiments of its kind, involving tens of thousands of volunteers, tens of millions of dollars and at least ten years.

"What we're trying to do has never been done," Ashe says. "It is a big risk, a big gamble."

Just the kind of challenge Ashe has always found utterly irresistible.

Genesis of a scientist Some people find their calling early in life. Karen Hsiao Ashe was 3.

"I've wanted to be a scientist as long as I can remember," she says.

It was probably something in the gene pool.

Her dad, C.C. Hsiao, was a popular professor of aerospace engineering at the University of Minnesota. Her mom, Joyce Hsiao, trained as a biochemist. They left China to earn Ph.D.s in the United States, settling in Arden Hills to raise their family.

Karen, the eldest of four, was the kind of kid who "could close off the rest of the world and be very happy," says Caroline Van, a younger sister. Clearly, she had a methodical mind. In the third grade, Ashe wrote a school project on how the brain works, then took scissors and carefully trimmed it into the shape of a brain.

One of the few Asian kids at St. Paul Academy in the 1970s, the petite girl with jet-black hair stood out. She was a whiz at math, gymnastics, music. "Both left brain and right brain," as her sister puts it. "Whatever she did, she would do it very intensely."

Often, Ashe was the only girl in the top math and science classes. Once, a male math teacher pulled her aside, Ashe recalled, assuming she would need extra help. "He said, 'in the whole history of the world, there hasn't been a good female mathematician.'" By the final exam, she out-scored all the boys. "At some level, you get used to being challenged."

Ashe was "super modest" about her achievements, her sister remembers. Awards were tossed under the bed or in the waste basket. It was partly a "Chinese thing," says Van. "Modesty is a virtue, and you do not brag."

At 17, Ashe skipped her freshman year of college and entered Harvard as a sophomore. By 27, she had a medical degree from Harvard and a Ph.D. from MIT.

Her childhood fascination with the brain had turned into a career.

First thrill of discovery Ashe was still a medical resident when she got her first taste of science in the big leagues.

One day over lunch, her boss confided his latest discovery: a genetic clue to mad cow disease.

The news, Ashe remembers, "just made my heart race."

She was working with Dr. Stanley Prusiner, who would go on to win the Nobel Prize, at the University of California, San Francisco.

At the time, in the mid-1980s, much of the scientific world was skeptical of Prusiner's theory -- that an abnormal protein, which he called a prion, was the culprit in a family of fatal brain-wasting diseases. The most notorious, mad cow, was killing cattle and people in Europe.

After obsessing about it for days, Ashe went to Prusiner with an idea: She wanted to study people with a rare inherited form of the disease, to see if she could find the genetic link.

Ashe picked a family with GSS (Gertsmann-Straussler-Scheinker) syndrome, which causes dementia and death at an early age.

She was transfixed as she closed in on an answer. Pregnant with her second child, Ashe had just one more task to finish -- a Southern blot DNA test -- when she went into labor.

Four days after her son William was born, Ashe was running back and forth to the lab every two hours, between feedings, to complete the experiment. "If I timed it right," she said, "I could get everything done."

Her exhaustion was swept away by euphoria as she stood in a darkroom, holding a film strip up to a red light.

There it was: a mutant gene in this unfortunate family, causing prions to run amok.

The study helped prove Prusiner's theory. It also firmly established Ashe as a rising star.

"She's a brilliant woman who has a lot of imagination and an extraordinary amount of talent," says Prusiner, who credited her work in his Nobel bio. "I think that what drives her is what drives many scientists: the excitement of discovering something new, which nobody has known before."

The forgetful mice Not much has changed about Room 697 of Diehl Hall since the day Ashe walked in 20 years ago. Empty. Drab. Reform-school windows.

Fresh off her work with Prusiner, she had been wooed back home by the University of Minnesota for what she calls her first real job, as an assistant professor of neurology at the Medical School.

Now, with a lab of her own, she had to decide what to study.

Alzheimer's disease was ripe for the picking. It was 1992, almost 90 years after Dr. Alois Alzheimer, a German psychiatrist, identified the disease. For decades, it was considered a rare condition, while "senility" was seen as a normal part of aging. But by the 1970s, scientists realized it was "the exact same disease," said Dr. William Thies, chief medical officer at the Alzheimer's Association. Trying to lure young scientists into the field, his group gave Ashe a grant to jump-start her work. As it turns out, Thies said, "we made a good investment."

Ashe leapt into the hottest research area -- a race to create an animal model of the disease. Shortly before, scientists had discovered the first Alzheimer's gene, which is found in a small number of people.

Back then, she said, no one knew if mice could get anything like Alzheimer's.

In her new lab, Ashe did most of the grunt work herself. Washing windows. Cloning genes.

She began the delicate task of inserting the human gene into fertilized mouse eggs. It took three years to get it right.

In 1995, Ashe got a call from a fellow scientist who was helping analyze the mice.

"He said, 'Karen, Karen, I can't believe this!'"

One group of mice had a large amount of amyloid protein, or plaques -- a telltale sign of Alzheimer's -- in the brain.

The next question: Were they getting dementia?

At first, Ashe had no idea how to test a mouse's memory. Paul Chapman, a psychology professor, suggested a water maze. So Ashe bought a large metal trough from a farm supply store and set it in the basement of Diehl Hall.

In the maze, mice swim around until they find a small submerged platform to rest on. Normally, after a few tries, they head in that direction pretty quickly. In other words, they remember. "The mouse equivalent," Ashe says, "of finding your car in the parking lot."

Would the genetically engineered mice have more trouble? A student videotaped them swimming, timing them with stopwatches.

Ashe's suspicions were confirmed when Chapman, the psychologist, looked at the data. Her mice were swimming in circles, unable to remember how to find the platform. Her reaction? Relief. "I thought, this will help me get tenure," she said.

In fact, it made her -- and her forgetful mice -- world-famous.

Today, her mice number in the hundreds of thousands, in labs around the globe. At one point, a breeding set sold to businesses for as much as $1 million. But Ashe insisted they be given free to universities, in the interest of science.

The forgetful mouse was "absolutely pivotal," says Thies, of the Alzheimer's Association. It opened up whole new ways to study the disease and experiment with treatments.

Creative tension crackles To get to her lab today, Ashe must pass a security desk at the Wallin Biosciences Building, a gleaming new glass structure on the edge of campus, and swipe her ID card at a door marked: "STOP: This is a Secure Building." There is a second security door upstairs.

This is the legacy of an infamous 1999 break-in, when animal rights activists "liberated" dozens of her mice. It wasn't much of a setback -- most of Ashe's mice were in another building. But it made security a top priority in research facilities like hers.

The new lab is state-of-the-art -- a long way from her original digs at Diehl Hall. Nor is Ashe the lone researcher anymore. She is maestro of a team of scientists who have come from all over the world to work with her.

On a recent morning, she slipped into a conference room near her office, ready for battle.

Across the table, Sylvain Lesne, a French scientist with George Clooney hair and a chic gray sweater, clasped a glass of espresso.

They were about to go head-to-head over the meaning of dots on a chart. For 10 years, they have been the closest of collaborators. Lesne, who joined her lab as a junior scientist, discovered the tiny molecule that Ashe calls A beta star -- which may, she believes, be the underlying cause of Alzheimer's.

Now, they're testing the theory in multiple ways. That can be tricky.

"Depending on how you analyze it," she says, "you can get opposite results." Ashe insists she wants people around who are willing to challenge her. A flawed theory, she points out, "is not going to lead to a cure."

In this case, she is questioning Lesne's decision to include two groups of people in a study on which they are co-authors. She has invited two fellow scientists to act as referees.

Staring at an overhead chart, Ashe purses her lips. "The underlying pathology of these people are quite different," she says. Some are probably on the road to Alzheimer's, others are not. Is it right to mix them?

"Well, that's one point of view," Lesne parries.

They lapse into a language of their own -- of dimers and rho values and plaque loads.

Ashe takes on a professorial tone: "That correlation doesn't look very convincing."

Lesne leans back and digs in. "That's your point of view," he repeats. "I don't agree."

They pepper each other with citations from obscure studies. An hour into it, they are clashing over raw numbers.

"That has 25," Ashe says.

"There's not 25," Lesne replies.

"I just counted the dots," she retorts. "Maybe there's lots of things I cannot do, but I can count."

Then she counts again. "Actually, there's 24," she admits with an impish grin, and her laughter fills the room.

Food, music, but no yoga Ashe, who travels the world talking about Alzheimer's, has an informal rule: no more than one trip a month away from home.

She lives in a lakeside house in North Oaks with her husband, Dr. James Ashe, a neurologist at the Minneapolis Veterans Medical Center, and their daughter, Sorcha, 11. Her sons from her previous marriage -- Philip, 25, and William, 23 -- are grown and gone. But family meals remain a cherished ritual, including Sunday dinner with her mother.

On a chilly Sunday evening in mid-December, James Ashe --who is writing a book on healthful eating -- takes charge of the stir-fry, while his wife prepares egg rolls and winter melon soup from scratch.

In her spare time, she's been practicing Beethoven's "Kreutzer" Sonata on the piano to serenade her husband's book club, which is reading the Tolstoy novella of the same name.

A question comes to mind: Is there anything Karen Ashe does not do well?

"Karen is not good at tennis," her husband deadpans. Or yoga, their daughter chimes in.

Karen Ashe laughs. "I tried yoga to relax," she confesses. But she had trouble with the concept of making her mind go blank. "I just keep thinking about all the things that I have on my mind."

She is "very persistent," her husband says -- that's one of the things that makes her an extraordinary scientist. She is not afraid, he adds, to go "after big, important questions."

Big risk, big potential When she came up with the idea for the prevention project, there was one nagging question: How would she find the volunteers?

Then she heard a colleague, Dr. Riley McCarten, talk about a memory test called the "mini-cog," which is used to help identify people on the road to dementia. At the Minneapolis Veterans Medical Center, McCarten gave it to veterans 70 and older, and discovered about 25 percent failed the test.

Ashe saw the potential flip side. The same test could find middle-aged or older people with healthy memories -- the very ones she would want to recruit for her experiment.

HealthPartners, one of the partners in her project, started using the mini-cog last summer at one clinic, in a pilot project.

Eventually, Ashe would like to see the test used routinely, to track people's memories over time. It's fast and simple -- memorize three words and draw a clock. But the implications can be tricky, as became clear at a December meeting of her planning committee, the newly formed Twin Cities Consortium for Alzheimer's Research (T-CAR).

"Our primary care providers got very nervous," said Dr. Michael Rosenbloom, a HealthPartners neurologist. They're worried they won't have the expertise, or time, to help everyone who fails the test.

"How many people are you going to see fail?" Ashe asked. "Are we talking 10,000 a year? 50,000?"

"That's a great question," Rosenbloom replied.

Jim Rickert, a business consultant on the project, admits "it makes my brain hurt to talk with all these people sometimes."

But he's grateful they're making an effort to bring Ashe's project to life.

"Alzheimer's runs in my family like crazy," says Rickert, 55. He watched his aunt, and then his mother, die from the disease. He would not be surprised if he develops it as well. "It might already be too late for me," he said. But he's pinning his hopes on Ashe, "if not for my generation, then for the next."

It may be five years before they are ready to start testing treatments in people. For now, Ashe is heading back to her lab, to see what happens when her forgetful mice start feasting on cabernet sauvignon or garlic or something else that might short-circuit the disease.

"What I would love to see come out of this," says Ashe, "is a very quiet realization, 50 years from now, that the epidemic of Alzheimer's disease that seems so threatening now, never really happened."

Maura Lerner • 612-673-7384