In Alzheimer’s disease, the axonal protein tau accumulates aberrantly in dendritic spines, where it contributes to excitotoxicity. What causes this mislocalization? Previous studies have linked hyperphosphorylation or overexpression of the protein to its accumulation in dendrites. At the Mechanisms of Neurodegeneration conference held June 14-17 in Heidelberg, Germany, Jürgen Götz of the University of Queensland in Brisbane, Australia, proposed a different explanation. Tau levels spike in dendrites because the protein is synthesized on the spot, he said. Aβ oligomers trigger this local production through a pathway that involves the kinase Fyn. “This alternative mechanism may explain the somatodendritic accumulation of tau induced by Aβ,” Götz told Alzforum. He believes the findings support therapeutic strategies that target tau production, rather than phosphorylation.
Other speakers in Heidelberg focused on specific ways for getting rid of dendritic tau. Karen Duff of Columbia University, New York, discussed how to use the neuropeptide PACAP to activate the proteasome in dendrites, thus chewing up unwanted tau. This approach might help prevent the spreading of tau from cell to cell early in disease, she noted. Researchers in attendance found the idea intriguing, peppering her with questions. Lennart Mucke of the Gladstone Institute of Neurological Disease, San Francisco, called the talk impressive. “This data is in line with other studies suggesting that enhancing the proteasomal degradation of tau could be of therapeutic benefit in AD and other tauopathies,” he told Alzforum.
In the last few years, several studies have tied dendritic tau to excitotoxicity and synaptic damage (see Sep 2010 news; Jan 2011 news). Götz previously worked out one mechanism for this, finding that tau escorts Fyn to the synapse, where the kinase then mediates excitotoxic signaling by Aβ (Jul 2010 conference news).
These results led Götz and Chuanzhou Li at Queensland to further study how tau and Fyn interact. They transfected both genes into HEK293 cells, and noticed a curious thing: Transfection of tau alone produced but a faint signal from the exogenous protein, whereas co-transfection of both genes boosted tau expression roughly 30-fold. Investigating this, they found that the increase was due to greater translation of tau, rather than transcription. The interaction was specific for tau and Fyn, as Fyn did not jack up translation of any other proteins they examined.
To dissect the mechanism, the researchers examined various components of the translational machinery, and manipulated them with activators and inhibitors. They delineated a pathway that runs from Fyn kinase through ERK and S6 kinase to pump up tau translation. Turning to mouse hippocampal primary neurons, they found that this same pathway acted in these cells, and that it could be kicked off by adding synthetic oligomeric Aβ to the culture. Because the kinases in this pathway also phosphorylate tau, the excess protein becomes hyperphosphorylated, Götz noted.
Does this happen in vivo? The researchers found that neurons cultured from Fyn knockout mice made 60 percent less tau, while animals with constitutively active Fyn had excess tau. In APP23 mice, Fyn, ERK, and S6 were all highly activated. Injecting Aβ oligomers into wild-type mice also activated these kinases, and raised tau levels in the affected neurons, Götz said. Meanwhile, Fyn inhibitors turned down production induced by Aβ.
To find out where in the cell this tau translation was occurring, Götz and Li used a proximity ligation assay that allowed them to label newly synthesized tau. They found the new protein only in the somatodendritic compartment, not in axons. The researchers fractionated lysates from APP23 brain and confirmed high tau levels in the cytosolic and synaptosomal fractions. By comparing older and younger mice, the researchers concluded that tau rises first in the soma, and then later spreads to dendrites.
The findings intrigued Christian Haass, Ludwig Maximilians University in Munich, who pointed out that cerebrospinal fluid tau levels rise steeply in AD, but not in frontotemporal dementia, even though tau pathology accumulates in many forms of that disease (Hampel et al., 2004; Olsson et al., 2005). Perhaps this discrepancy occurs because the FTD brain does not contain Aβ oligomers, and therefore does not turn up tau production, Haass suggested.
What are the consequences of local tau synthesis? Because dendritic tau can corral more Fyn, leading to excitotoxicity, while Fyn further boosts tau production, the pathways may form a vicious circle, Götz noted. He believes the data support therapeutic strategies that lower total tau levels. The tau immunotherapy field has been moving in this direction already. Other approaches would be to silence tau expression using antisense oligonucleotides or microRNA, Götz suggested. Targeting ERK or S6 kinase would not be an option, because these enzymes affect so many cellular processes, but Fyn might be a viable target for turning down tau synthesis, he added. Fyn does play a key role in myelination, but this process is finished in adults.
“The findings shed new light on well-established links among Aβ, Fyn, and tau, and further solidify the rationale for inhibiting Fyn and related signaling pathways in AD,” Mucke wrote to Alzforum. However, he noted that his own group has not found elevated tau levels in the J20 mouse line, which, like APP23, expresses human mutant APP. “Additional studies will likely resolve whether these differences are due to the age at which the mice were analyzed, the methods used, or other variables,” he wrote.
For her part, Duff’s talk covered a different approach for lowering tau. She had previously found that tau aggregates clog the proteasome, causing tau clearance to grind to a halt. Raising cyclic AMP levels counteracts this by switching on protein kinase A, which in turn activates the proteasome to mop up aggregates. Drugs that boost cAMP levels, such as rolipram and cilostazol, lowered tau levels and improved learning and memory when given to young rTg4510 mice (Dec 2015 news).
In Heidelberg, Duff described a more selective way to locally raise cAMP in dendrites. The neuropeptide PACAP activates the G-protein coupled receptor, PAC1, which elevates cAMP levels and is expressed in neuronal cell bodies and dendrites but not axons (Joo et al., 2004). Using an osmotic pump to administer a constant trickle of PACAP to rTg4510 mice lowered tau specifically in the postsynaptic fraction, not in presynapses or total cell lysate, Duff said. Proteasome activity ramped up, and memory in the Morris water maze and novel object recognition test improved in the treated mice.
The experiment acts as a proof of concept for this approach, Duff suggested. She believes it may be possible to target other receptors, such as muscarinic or adrenergic ones, to affect cAMP. Activating the proteasome could also help clear other aggregating proteins, for example α-synuclein or FUS, she added. In future work, she will also look at whether this strategy limits the spread of pathologic tau. The approach does not work for mice at advanced stages of diseases, she noted. “Once the proteasome is dead, you cannot rescue it,” she said. Götz noted that data suggest PACAP and the PAC1 receptor play a role in post-traumatic stress disorder and psychiatric conditions. “There may be a broader application of this therapeutic strategy,” he wrote to Alzforum.
Bart De Strooper of the U.K. Dementia Research Center at UCL urged caution, noting that the full picture may be more complicated since G-protein coupled receptors have been linked to various aspects of amyloid pathology. The PAC1 receptor, for example, also regulates ADAM10, which is responsible for the α-cleavage of APP, De Strooper pointed out, adding that there is a web of complicated GPCR systems. Duff acknowledged this, but suggested that specific pharmacological interventions might be worked out.—Madolyn Bowman Rogers
- The Plot Thickens: The Complicated Relationship of Tau and Aβ
- Tau’s Synaptic Hats: Regulating Activity, Disrupting Communication
- Honolulu: The Missing Link? Tau Mediates Aβ Toxicity at Synapse
- Protecting Proteasomes from Toxic Tau Keeps Mice Sharp
Research Models Citations
- Hampel H, Buerger K, Zinkowski R, Teipel SJ, Goernitz A, Andreasen N, Sjoegren M, DeBernardis J, Kerkman D, Ishiguro K, Ohno H, Vanmechelen E, Vanderstichele H, McCulloch C, Moller HJ, Davies P, Blennow K. Measurement of phosphorylated tau epitopes in the differential diagnosis of Alzheimer disease: a comparative cerebrospinal fluid study. Arch Gen Psychiatry. 2004 Jan;61(1):95-102. PubMed.
- Olsson A, Vanderstichele H, Andreasen N, De Meyer G, Wallin A, Holmberg B, Rosengren L, Vanmechelen E, Blennow K. Simultaneous measurement of beta-amyloid(1-42), total tau, and phosphorylated tau (Thr181) in cerebrospinal fluid by the xMAP technology. Clin Chem. 2005 Feb;51(2):336-45. Epub 2004 Nov 24 PubMed.
- Joo KM, Chung YH, Kim MK, Nam RH, Lee BL, Lee KH, Cha CI. Distribution of vasoactive intestinal peptide and pituitary adenylate cyclase-activating polypeptide receptors (VPAC1, VPAC2, and PAC1 receptor) in the rat brain. J Comp Neurol. 2004 Aug 30;476(4):388-413. PubMed.
- Li C, Götz J. Somatodendritic accumulation of Tau in Alzheimer's disease is promoted by Fyn-mediated local protein translation. EMBO J. 2017 Sep 1; PubMed.