fb-pixel Skip to main content

Petri-dish form of Alzheimer’s aids drug testing

MGH researchers’ work called a big step forward

Mass. General neurologists Doo Yeon Kim (left) and Rudolph Tanzi, in front of an image depicting Alzheimer’s growth.Dominick Reuter/New York Tiimes/NYT

NEW YORK — For the first time, and to the astonishment of many of their colleagues, researchers created what they call Alzheimer’s in a Dish: a petri dish with human brain cells that develop the telltale structures of Alzheimer’s disease.

In doing so, they resolved a longstanding problem of how to study Alzheimer’s and search for drugs to treat it. The best they had until now were mice that developed an imperfect form of the disease.

The key to their success, said the lead researcher, Dr. Rudolph E. Tanzi of Massachusetts General Hospital in Boston, was a suggestion by his colleague Doo Yeon Kim to grow human brain cells in a gel where they formed networks as in an actual brain. They gave the neurons genes for Alzheimer’s disease. Within weeks they saw the hard Brillo-like clumps known as plaques and then the twisted spaghetti-like coils known as tangles, the defining features of Alzheimer’s disease.

The work, which also offers strong support for an old idea about how the disease progresses, was published in the journal Nature on Sunday. Leading researchers said it should have a big effect.


“It is a giant step forward for the field,” said Dr. P. Murali Doraiswamy, an Alzheimer’s researcher at Duke University. “It could dramatically accelerate testing of new drug candidates.”

Of course, a petri dish is not a brain and the petri dish system lacks certain crucial components, like immune system cells, that appear to contribute to the devastation once Alzheimer’s gets started. But it allows researchers to quickly, cheaply, and easily test drugs that might stop the process in the first place. The crucial step, of course, will be to see if drugs that work in this system stop Alzheimer’s in patients.

The discovery, said Dr. Sam Gandy of the Icahn School of Medicine at Mount Sinai in New York, is “a real game changer” and “a paradigm shifter.” He added, “I’m really enthusiastic to take a crack at this in my lab.”


However, Karen Duff of Columbia University, while praising the work as “a tour de force,” cautioned that once Alzheimer’s gets started, tangles can take off on their own and may need to be attacked by drugs that strike them specifically in order to stop devastation in the brain.

Tanzi is starting an ambitious project to test 1,200 drugs on the market and 5,000 experimental ones that have finished the first phase of clinical testing, a project that is impossible with mice because each drug test takes a year. With their petri dish system, Tanzi said, “we can test hundreds of thousands of drugs in a matter of months.”

He already has used his system to look at drugs designed to prevent the formation of amyloid, the protein that clumps into plaques. The drugs, he reports, prevented both plaques and tangles in the petri dishes. Some are in clinical trials and it is not known if they work in people. One was tested in patients and failed because it was too toxic. One hope is t o find drugs for other diseases that are known to be safe and shown to work on Alzheimer’s in the petri dish.

He also found an enzyme needed to make tangles after plaques are present. When he blocked that enzyme, plaques form but not tangles. The enzyme is another potential drug target, he said.


Gandy wants to use the system to study the effects of genes that predispose a person to Alzheimer’s, especially the most powerful one, APOe4, which contributes to about half of all Alzheimer’s cases. No one really knows how or why it is linked to the disease, Gandy said.

“I think I would go after that to begin with,” he said.

Tanzi said that once his group got the idea of growing neurons in a gel, it was straightforward to set up their Alzheimer’s in a dish system.

They used human embryonic stem cells — those cells that can become any cell of the body — and grew them with a mixture of chemicals that made them turn into neurons. They gave those neurons Alzheimer’s genes and grew them in wells in petri dishes that were lined with a commercially available gel. Then they waited.

“Sure enough, we saw plaques, real plaques,” Tanzi said. “We waited and then we saw tangles, actual tangles. It looks like you are looking at an Alzheimer brain.”

All that was required to get the process started was the Alzheimer’s gene that made cells produce an excessive amount of a normal protein, beta amyloid. Previously, researchers had tried to grow the disease in a dish of liquid but the neurons did not connect or develop plaques and tangles.

The controversy over how and why Alzheimer’s gets going dates back three decades, when Dr. George G. Glenner proposed a simple process. Beta amyloid starts to accumulate in the brain. It turns into plaques. Neurons respond by making tangles. The combination proves fatal for brain cells and dementia sets in.


“He said, ‘This is how the disease starts,’ ” Tanzi said. “But for 30 years there was no proof that amyloid drives the rest of the disease.”

“The lack of a viable model for Alzheimer’s has been the Achilles’ heel of the field,” Doraiswamy of Duke said.

Some said the amyloid hypothesis was correct and the drugs failed because they were not potent enough or were given too late, when the disease was well established. But others asked if amyloid was the right target. Many proposed going after another protein instead, tau, a normal constituent of neurons that becomes deformed into tangles when a person has Alzheimer’s disease.

Even those who insisted on the amyloid hypothesis often elaborated on it, saying first amyloid accumulates and then a litany of other things go wrong — cell damage and inflammation and molecular stress — which finally lead to tau and tangles.

“There was a big black box of things going wrong,” Tanzi said.

But, he said, the more complex model was refuted by his study. Tangles formed with nothing but the presence of amyloid plaques.

And drugs that block beta amyloid prevent both plaques and tangles from forming, Tanzi and his colleagues report.