A number of studies already hinted that tau mediates Aβ toxicity, such as loss of dendritic spines and synapses, and reduced expression of synaptic markers (see, e.g., ARF related news story on Roberson et al., 2007 and ARF related news story on Roberson et al., 2011). As a result, scientists have been looking for molecules linking the two proteins. “If we know all the steps in the cascade, we will have a better idea of what to target for therapy,” said Erik Roberson at the University of Alabama in Birmingham, who also did not collaborate with Lu on the study.

One molecular mechanism that appears to play a role in the toxic cascade is the phosphorylation of tau by various kinases. In 2004, Lu’s group showed that the fruit fly MARK homologue, PAR-1 (partitioning defective-1), is needed for degeneration of photoreceptor cells in a Drosophila model of tauopathy (see ARF related news story on Nishimura et al., 2004). Overexpression of PAR-1 in these cells increased tau phosphorylation and cell degeneration in a dose-dependent manner. “We wanted to see whether these observations held up in a mammalian model, and if MARK/PAR-1 could be a target for therapy,” Lu told ARF.

To examine this question, they turned to cultured rat primary hippocampal neurons. Overexpression of MARK family member MARK4 resulted, as expected, in hyperphosphorylation of tau, as well as loss of dendritic spines and reduced synaptic marker expression in these cells. The researchers saw similar defects if they treated neurons with synthetic Aβ. But when Yu, Lu, and colleagues overexpressed a form of tau that cannot be phosphorylated by MARK4, neither overexpression of the kinase nor Aβ treatment produced synaptic defects in these cells. In this system, Aβ toxicity seemed to rely on tau phosphorylation by MARK4.

To definitively test this, the authors used a peptide inhibitor that blocks all four members of the MARK family (Nesić et al., 2010). When they overexpressed MARK4 in primary neurons, the inhibitor blocked the toxic effects of the kinase. The inhibitor also prevented the defects caused by Aβ. The findings suggest that Aβ leads to MARK activation, and that, in turn, causes tau hyperphosphorylation and synaptic dysfunction. However, it is not clear how Aβ activates MARK4. “Aβ does not directly interact with MARK,” said Lu. who suggested that there are probably other upstream kinases involved. “The puzzle is not going to be simple,” he said.

One piece of that jigsaw will be figuring out exactly how phospho-tau disrupts synaptic function. Studies by the groups of Eckhard and Eva-Maria Mandelkow at the Max Planck Research Institute in Hamburg, Germany (see ARF related news story on Li et al., 2011), and Dezhi Liao and Karen Hsiao Ashe at the University of Minnesota, Minneapolis (see ARF related news story), reported that tau, which normally resides in axons, moves into dendrites when hyperphosphorylated. This mislocalization of phospho-tau to dendrites is required for synaptic dysfunction. Researchers at Jurgen Goetz’s lab at the University of Sydney, Australia, also found that tau drives translocation of fyn kinase to dendrites, where it can mediate Aβ toxicity through glutamate receptors (see ARF related news story). In their study, Lu, Yu, and colleagues did not specifically focus on tau and phospho-tau localization.

One of the limitations of Lu’s work is that it was carried out in cells in culture, but “for an in-vitro system, it’s as good as it gets,” said Drewes. “They used primary neurons, they looked at synaptic changes rather than cell death as the primary readout, and they used Aβ as the toxic stimulus.” In their study Yu, Lu, and colleagues used synthetic Aβ42 at a concentration of 5 μM, which is generally regarded as much higher than found in the brain. Work from Dennis Selkoe’s group at Brigham and Women’s Hospital, Boston, reported that Aβ oligomers purified from the brains of AD patients induce toxic effects when applied to rat neurons at sub-nanomolar concentrations (see ARF related news story on Jin et al., 2011).

This study and earlier work suggest that the MARK family might be a therapeutic target for AD, and several groups are already looking at agents that modulate the function of these kinases. Next, Lu and colleagues plan to test the role of MARK in an AD animal model. “We want to see, if we give the MARK inhibitor to a mouse model of Alzheimer's, whether we can block some of the cognitive deficits,” said Lu.

Whether the normal physiological functions of the kinase will complicate animal or human studies remains to be seen. For example, in one set of experiments, Yu and colleagues found that Aβ treatment reduced the frequency, but not the amplitude, of AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptor mediated miniature excitatory post-synaptic potentials (mEPSC) in neurons. Treatment with the MARK inhibitor restored normal neurotransmission, but when it was applied to normal neurons, it led to reduced mEPSC frequency. This suggests that maintaining appropriate levels of MARK activity is important for neurotransmission. “Like most kinases MARK/PAR-1 is promiscuous. It is hard to specifically block one kinase-substrate interaction. When you inhibit one kinase, you usually inhibit other substrates,” said Roberson.—Laura Bonetta.

Reference:
Yu W, Polepalli J, Wagh D, Rajadas J, Melenka R, Lu B. A critical role for the PAR-1/MARK-tau axis in mediating the toxic effects of Aβ on synapses and dendritic spines.Hum Mol Genet. 2011 Dec 19. Abstract