14 Nov 2011

As far as Alzheimer’s disease pathology goes, growing evidence puts tau on roughly equal footing with Aβ. Not only do tau neurofibrillary tangles correlate better with cognitive decline than do Aβ plaques (see Arriagada et al., 1992), but recent evidence suggests that each is needed for the other's toxic effects (see ARF related news story on Zempel et al., 2010 and Vossel et al., 2010). Since the AD field has gradually realized that tackling Aβ in symptomatic individuals may be too little too late (see ARF related news story), there is renewed interest in tau as a therapeutic target. Researchers are taking a close look at tau to understand its normal function, figure out what goes awry in disease, and explore ways to prevent its malfunction. Some recent findings lift the curtain a little higher on tau’s role. Eckhard Mandelkow and colleagues from the Max-Planck Unit for Structural Molecular Biology, Hamburg, Germany, report on why tau is restricted to the axon unless abnormally phosphorylated. A collaborating group, led by Eva-Maria Mandelkow, identifies new kinase inhibitors that could aid in the study of tau phosphorylation. Meanwhile, a kinase inhibitor with therapeutic potential reportedly alleviates Alzheimer's disease symptoms in transgenic mice, and a team has identified other kinases that target tau. Taken together, the findings provide a snapshot of the state of tau research and may have implications for the treatment of Alzheimer's and other tauopathies.

Eckhard Mandelkow’s work is an extension of findings from his and other labs indicating that tau, which is normally axonal, shifts to the soma and dendrites of neurons when hyperphosphorylated (see Braak et al., 1994; ARF related news story on Hoover et al., 2010; ARF related news story on Zempel et al., 2010). Writing in the October 18 European Molecular Biology Organization (EMBO) Journal, Mandelkow and colleagues report that, under normal circumstances, tau moves forward and backward in the axon but not into the soma or dendrites. However, when microtubules break down or tau becomes hyperphosphorylated and detached from intact microtubules, it seems free to diffuse into the cell body and dendrites.

To track the shifty protein, first author Xiaoyu Li and colleagues tagged neuronal tau with Dendra2, a green fluorescent label that converts to red when hit by ultraviolet light. Li zapped a small portion of the middle axon of cultured rat cortical neurons with UV light to turn the marker red, then followed it with a confocal laser scanning microscope. The team saw that red-labeled tau moved both in an anterograde direction, toward the end of the axon, and retrogradely toward the soma, but retrograde movement past the axon initial segment never happened (see video below). However, when the team hyperphosphorylated tau—which detaches it from microtubules—or broke down the microtubules, tau diffused past the axon initial segment into the somatodendritic compartment. The team concluded that there is some barrier that arises from tau-axon microtubule association and keeps tau from travelling along microtubules back into the cell soma.

"As long as tau is on the microtubules, it's well behaved; but if it comes off, it becomes mischievous," said Mandelkow. "If we could find ways to keep it in the axon on the microtubule, we'd probably be in good shape."

A diffusion barrier at the axon initial segment keeps photo-converted Dendra2-labeled tau from leaving the axon. Image credit: Li et al., 2011

"I think it makes perfect sense," said Hana Dawson, Duke University in Durham, North Carolina, who was not involved in the work. "There was always the question of why tau is mostly distributed to the axon and there weren't any good answers."

In light of the important role of tau phosphorylation in tau biology and disease, inhibitors that block the responsible kinases could prove useful as therapies or for studying mechanisms. Researchers in companies, but also academics such as Eva-Maria Mandelkow, also at the Max-Planck Unit for Structural Molecular Biology, are looking for such inhibitors. In the October 7 Journal of Biological Chemistry online, Mandelkow and colleagues reported several compounds that inhibit MARK kinases, which are known to phosphorylate tau. The team screened the ChemBioNet library, a collection of 17,591 small molecules, and plans to screen more libraries in search of other compounds. Four of the identified inhibitors share a similar functional group—the 9-oxo-9H-acridin-10-yl functional group—which may provide a basis for developing inhibitors that can be human therapeutics. One other compound they found was extremely specific for MARK, and did not affect other tested kinases.

In the paper, the authors also describe a reporter they created that allows them to measure MARK activity in living cells. Phosphorylation by MARK brings about a conformational change in the reporter that brings the two ends of a fluorophore together, giving off a fluorescent signal. The team finds that MARK kinases are most active in the growth cones of neurons. "A biosensor that works in primary neurons is a potentially very powerful tool for drug discovery," said Gerard Drewes of Cellzome in Heidelberg, Germany. With it, one "can study the interplay of an activation of the target with a pharmacological antagonist," he added.

Given the role of tau phosphorylation in Alzheimer's disease and other tauopathies, one potential avenue of treatment is to block the kinases that modify the protein in vivo. For years, researchers have targeted GSK-3β kinase. Lithium, widely used in bipolar disease, inhibits GSK-3β, but has complex actions, prompting researchers to look for more specific compounds. Pharmaceutical companies, such as AstraZeneca in Sweden and Noscira in Madrid, Spain, are among those that have pursued GSK-3 as a target.

Masayuki Takizawa and colleagues from the Takeda Pharmaceutical Company in Osaka, Japan, reported a highly selective GSK3 inhibitor, 2-methyl-5-(3-{4-[(S)-methylsulfinyl]phenyl}-1-benzofuran-5-yl)-1,3,4-oxadiazole (MMBO). It blocks hippocampal tau phosphorylation and tau pathology in mice. First author Tomohiro Onishi and colleagues gave MMBO to male 3xTg-AD mice for up to 33 days, and then evaluated them on Y-maze and novel object recognition tests. They found that, although Aβ pathology was unaffected, tau phosphorylation fell in a dose-dependent manner at residues regulated by GSK-3. Treated mice reportedly performed better on memory and cognitive tests than control triple-transgenic mice.

"These results indicate that pharmacological GSK-3 inhibition ameliorates behavioral dysfunction...and that MMBO might be beneficial for AD treatment," the authors wrote.

Allon Therapeutics in Vancouver, Canada, is developing another tau phosphorylation inhibitor called davunetide, or NAP (see ARF related news story). This eight-amino-acid peptide gets snipped out of activity-dependent neuroprotective protein (ADNP), which protects against cognitive deficits, tauopathy, and neuron death. On its own, NAP protects against tau hyperphosphorylation in ADNP-deficient mice, seems safe in humans (see ARF related news story), and is now being tested for efficacy in a Phase 2/3 trial) in patients with progressive supranuclear palsy, a pure tauopathy." If we show benefit of a tau drug in a pure tau disease, it's our hypothesis that that might predict greater success in a more common tauopathy like Alzheimer's disease," said Adam Boxer, University of California, San Francisco, who is conducting the davunetide trials for Allon Therapeutics.

Researchers also continue to hunt for other tau kinases that might one day be therapeutic targets. Michel Goedert, Hirotaka Yoshida, and colleagues at the Medical Research Council Laboratory of Molecular Biology in Cambridge, U.K., have recently identified other candidates in the AMPK (5' adenosine monophosphate-activated protein kinase) family, which includes the previously mentioned MARK kinases. In the October 10 Journal of Neurochemistry, the team reported that salt-inducible kinase (SIK), brain-specific kinase (both BRSK1 and 2), and maternal embryonic leucine-zipper kinase (MELK) phosphorylate tau in a region known to contribute to tau toxicity.

Recent setbacks of Aβ treatments in clinical trials (see Holmes et al., 2008 and ARF related news story) have re-energized the tau field, and researchers are taking other approaches to targeting it. For instance, researchers are examining ways to inhibit tau aggregation (see ARF related news story) or use immunotherapy to clear tau (see ARF related news story).

But despite the growing attention on tau-based therapies, no inhibitor has gotten very far in clinical trials. The field has to overcome certain challenges, many of which amyloid researchers faced 10 years ago, says Ratan Bhat at AstraZeneca R&D in Södertälje, Sweden, who has worked on tau as a drug target for many years. For instance, investigators must figure out which proteins make the best therapeutic targets, which biomarkers to measure in the clinic, and how much and how long tau phosphorylation or aggregation inhibition is necessary and still safe. Despite these challenges, scientists are optimistic about where the research might lead, Bhat says, and his company is still pursuing tau as a therapeutic target, he told ARF.—Gwyneth Dickey Zakaib