Enzyme Traced to Two Alzheimer’s Pathways

Loss of Pin1 Implicated in Development of Both Tangles and Plaques

In the late stages of Alzheimer’s disease the brain becomes clogged with amyloid plaques and neurofibrillary tangles. The plaques, found in the extracellular milieu, are made up mostly of beta-amyloid peptide, while the tangles, which gather inside neurons, comprise the large microtubule-binding protein tau. Despite the distinct localizations of the two aggregates, scientists have long suspected that these pathologies may share a common cause. But after decades of research, a single lesion that drives both plaques and tangles has not emerged, until now.

Photo by Liza Green, HMS Media Services; inset courtesy of Lucia Pastorino
	Kun Ping Lu and colleagues have found that in mice the Pin1 enzyme is needed to prevent both plaques and tangles, the quintessential hallmarks of Alzheimer’s disease. The team includes, clockwise from left, Xiao Zhen Zhou, Jormay Lim, Martin Balastik, Greg Finn, and Lu. Lucia Pastorino, one of the joint first authors, appears above.

In the March 23 Nature, Kun Ping Lu, HMS associate professor of medicine at Beth Israel Deaconess Medical Center, and colleagues report that Pin1 slows production of Abeta and prevents its aggregation. The finding seems to nail Pin1 as a crucial player in both Abeta and tau pathology because just a few years ago, Lu and colleagues showed that Pin1 also protects against neurofibrillary tangles.

“Since plaques and tangles were first documented by Alois Alzheimer almost exactly 100 years ago, scientists have struggled to isolate the cause of these aggregates,” said Lu. But a single reason for both was elusive. “For the first time, we have been able to manipulate a single molecule and generate plaques, tangles, and neurodegeneration in a mouse model of the disease.”

Catalytic Flip-flop
The job of that single molecule, Pin1, is to catalytically flip proline amino acids between cis and trans isoforms. Because Pin1 carries out this isomerization on polypeptides, the enzyme can profoundly alter protein structure. In fact, when Lu was a postdoc at the Salk Institute, he was the first to identify Pin1 and recognize that its activity is essential for mitosis.

Pin1 does not catalyze the isomerization of just any old proline; it recognizes only those that follow a phosphorylated serine or threonine. When Lu realized that such motifs are present in both the Abeta precursor protein (APP) and tau, and that phosphorylation of the tau motifs are enhanced in mitotic cells where Pin1 function is elevated, he reasoned that Pin1 might play some role in Alzheimer’s. His prediction bore out in 2003 when researchers in his lab discovered that the enzyme isomerizes tau prolines and protects mice against tau pathology. Pin1 appears to do this in two ways. It restores functionality to tau, which is required to stabilize microtubules. And it accelerates dephosphorylation of the serine or threonine that precedes proline, making tau less likely to dissociate from microtubules and form aggregates.

Cis- and Trans-formation
Now, Lu’s team has been able to demonstrate that in a similar fashion, Pin1 can prevent the formation of Abeta (see diagram). Research fellows Lucia Pastorino and Pei-Jung Lu and instructor Anyang Sun, joint first authors on the Nature paper, have discovered that isomerization of proline 669 in the C-terminal tail of APP alters how the full molecule is processed.

The researchers first determined that Pin1 does, in fact, isomerize proline 669, which lies in a Pin1 motif. In collaboration with Linda Nicholson at Cornell University, they obtained ROESY (rotating-frame Overhauser effect spectroscopy) spectra, which clearly demonstrate that a small amount of Pin1 can convert the proline from cis to trans with dramatic speed. “This is the first time that anyone has actually visualized a cis to trans isomerization catalyzed by Pin1,” noted Lu.

Next, the researchers looked to see how this activity might affect APP biology. The transmembrane APP can be processed sequentially by several proteases. Both alpha- and beta-secretases snip off parts of the N-terminal that lie on the extracellular side of the cell membrane, leaving the rest of the protein susceptible to cleavage by the intramembrane protease gamma-secretase. While the alpha- and gamma-secretase combination generates innocuous protein fragments, the beta- and gamma-secretase combination releases the notorious Abeta. So, in effect, two competing processes occur in the cell, only one being linked to Alzheimer’s pathology.

The researchers examined both of these proteolytic pathways in the context of Pin1 activity, first in cell cultures, then in mice. Because neurons are hard to transfect, they chose to work in vitro with Chinese hamster ovary (CHO) cells, which express Pin1, and breast cancer cells, which do not express the protein. They found that overexpression of Pin1 in the CHO cells reduced the amount of Abeta formed, while in breast cancer cells obtained from Pin1-negative mice, the amount of Abeta produced was around eightfold higher than normal. Curiously, in those same Pin1-negative cells, alpha-secretase processing of APP fell more than twofold. The results suggested that in the absence of Pin1, more APP is processed via the amyloidogenic pathway that leads to Abeta production.

Mouse Tales
Lu and colleagues tested this theory in vivo using Pin1-negative normal and transgenic mice—the transgenics produce human APP with mutations that cause early-onset, familial forms of Alzheimer’s. The researchers found that in both sets of animals, absence of Pin1 leads to selectively increased production of Abeta 42, the notoriously toxic species of Abeta. The transgenic animals had the greatest increase, about 50 percent, in keeping with their propensity to make large amounts of Abeta. And the Pin1 effect was age-dependent, which fits with the timing of disease pathology in both humans and animal models. Pastorino and colleagues also found that the Pin1 knockouts generated less of the non-amyloidogenic alpha-secretase–cleaved APP, supporting the idea that Pin1 steers APP processing away from Abeta formation.

Diagram adapted by Rachel Eastwood from original courtesy of Kun Ping Lu

Pinning Alzheimer’s on Pin1. Phosphorylation of threonine–proline motifs in tau and APP contributes to Alzheimer’s disease by driving the formation of plaques and neurofibrillary tangles. Though dephosphorylation can be protective, it is restricted to motifs that have proline in the trans conformation. By converting spontaneously formed cis isomers back to trans, Pin1 helps drive dephosphorylation of both proteins, preventing plaques and tangles. But loss of Pin1 through genetic mutation, stress, or downregulation, for example, would have the opposite effect. A single lesion in Pin1 could, therefore, explain why both plaques and tangles are found in the brains of those with Alzheimer’s disease.

How does the action of Pin1 at the C-terminal tail of APP prevent formation of Abeta? Lu predicts that the isomerization of proline helps to restore the local structure of the tail that is altered by phosphorylation. This restoration, he believes, changes the affinity for some binding partners that influence processing of APP at the cell surface or in the endocytic pathway. One possible candidate that Lu is investigating is a protein called Fe65, which has been shown to bind near threonine 668 and also affects Abeta formation.

All told, the work from Lu’s lab suggests that Pin1 plays a pivotal role in preventing both plaques and tangles. Whether this will hold true in humans remains to be seen, but there are some tantalizing clues that it will. Work from Lu’s and other labs, for example, has revealed Pin1 deficiencies in the brains of those with Alzheimer’s and other dementias, while polymorphisms in the Pin1 promoter have also been linked to increased risk for developing Alzheimer’s.

—Tom Fagan

top