Protein Found That Turns Brain Enzyme Into Possible Trigger For Alzheimer's Tangles

Reported as a full article in the December 9 Nature, the study builds a systematic case against p25. This small protein, the researchers claim, can divert an enzyme called cdk5 from its day-to-day work in the brain and turn it into a menace that ultimately destroys neurons.

"We believe that the production and accumulation of p25 in the Alzheimer's brain very likely plays a role in the pathogenesis of the disease," says senior author Li-Huei Tsai, associate professor of pathology at Harvard Medical School.

The study supports one of the two major attempts at explaining this complex condition. The hypothesis holds that a long-sought but still mysterious enzyme attaches many phosphate groups to a normal protein called tau, which stabilizes the neuron's inner skeleton, helping the cell maintain its extended connections with neighboring and distant cells. This hyperphosphorylation is known to set in motion a deadly neurodegenerative cascade that shows up in postmortem exams as tangles of intertwined tau filaments cluttering the cells' inside.

In recent years, this hypothesis languished as another line of thought, which centers on extracellular beta-amyloid plaques, has made rapid progress identifying protein components, enzymes, and the biochemical process of how plaques form.

It was not for lack of trying. Many tau laboratories have for years sought the enzyme, or kinase, that hyperphosphorylates tau. Indeed, many kinases do so when nudged in biochemical tests, but none appeared overly active in early Alzheimer's disease until now. Tsai's paper "is a major advance in this search," writes tangle expert Eckart Mandelkow of the Max-Planck-Unit for Structural Molecular Biology in Hamburg, Germany, in an accompanying News and Views article.

Yet while her study gives a lift to the tangle view of Alzheimer's, Tsai does not promote one theory at the expense of the other. Rather, she considers this study the first chapter in a fast-evolving story that may soon unite both ideas.

That is because a new trend is afoot as developmental neurobiologists realize that cdk5, their "favorite" molecule - whose precise role in brain formation they have been trying to parse-might also function in neurodegeneration. Take Tsai, for example. In previous work, she discovered that when controlled by its protein partner p35, the kinase cdk5 helps newly generated neurons deep inside the embryonic brain migrate outwards past their older cousins, thus enabling the cerebral cortex's characteristically layered pattern to form properly.

Trying to confirm other scientists' data that cdk5 can also work with p25 - a fragment of p35 - Tsai tried hard to find p25 but failed. That is, until first author Gentry Patrick decided to look in diseased human brain. There it was, at last: p25 proved to be accumulated 20- to 40-fold in cortical neurons of people with Alzheimer's. Its amount correlated with the progression of the disease in that later stages harbored more p25.

"That was very exciting. It really opened our eyes and initiated this whole investigation," recalls Tsai.

From then on, things fell into place. P25, it turned out, misses the piece of its parent p35 that anchors p35 to the neuron's cell membrane, where it belongs. What's more, p25 is stable, whereas p35's half-life is but 20 minutes because it gets degraded in the cell's waste disposal organelle. Taken together, this means that the careful controls on space and time that p35 exerts over cdk5 all fall away with p25. Once made, it can hang around the cytoplasm and send cdk5 on a "phosphorylation binge."

Tsai says she was struck by how readily p25/cdk5 managed to undo the neurons. First, several different experiments indicated a disintegration of the cultured neurons' microtubules - the main struts of the cell's inner skeleton and its major transport rails. Second, the neurons then withdrew their processes, and shriveled to being mere rounded cell bodies.

The second major implication of this study results directly from this apparent degeneration: within three days, 90 percent of the cultured neurons expressing p25/cdk5 were dead. What's more, they died by programmed cell death, or apoptosis, the authors show. Scientists at Harvard and elsewhere are pursuing intensely the question of how neurodegeneration eventually seals a cell's fate, and this study offers a new handle on the underlying molecular pathways.

Whether p25/cdk5 actually wreaks the kind of havoc in people's brains that it causes in cultured neurons remains to be proven. But now that Tsai and her collaborators have established this phenomenon, others can test its significance, for example, by trying to slow or reverse neurodegeneration in Alzheimer's animal models with inhibitors of cdk5 or p25.

The work also raises the question whether cdk5 plays a general role in neurodegeneration. Other "tauopathies" exist, including progressive supranuclear palsy and a type of dementia called FTDP-17. Moreover, cdk5 is known to accumulate in motor neurons with Lou Gehrig's disease.

For her part, Tsai is intrigued by the fact that the suspect she nabbed is not an entirely novel molecule found only in a disease. Instead, it is a workhorse of an enzyme that - properly reined in - functions in indispensable ways in the developing and the adult human. But unleash p25, and cdk5 turns destructive. "The most pressing questions right now are, "What is the mechanism converting p35 into p25 and What conditions induce this conversion?" she adds. Stay tuned.