Part 1 of a two-part story. Click here for Part 2.
Tau comes in many forms, and calls many locations home, both inside and outside of cells. Which is the one that wreaks damage to neurons, and where exactly does it reside? Researchers grappled with these questions at back-to-back meetings in San Diego, the 5th RNA Metabolism in Neurological Disease Conference and the Society for Neuroscience annual meeting, on November 1–7.
- Both tau and Aβ oligomers bungle synaptic firing, in different ways.
- Tau oligomers propagate between neurons with help from TIA1.
- Structural biology ties specific shapes of tau oligomers to toxicity.
Oligomers on the Move
One of tau’s infamous tricks is its ability to travel from neuron to neuron in the brain, by jumping the synaptic cleft (Feb 2012 news). First suggested by the characteristic spread of tangles in postmortem brain, and later backed up in mouse models, tau’s trans-synaptic spread has been considered a property of fibrils; however, in San Diego, Benjamin Wolozin of Boston University said tau oligomers travel in this way. He reiterated that oligomers are the most toxic species of tau.
Previously, Wolozin had reported that the RNA-binding protein TIA1 stabilized tau oligomers within stress granules, membraneless organelles that sequester mRNAs from translation during a stress response. Halving TIA1 expression reduced tau oligomers and their harmful effects in tau transgenic PS19 mice, even though these mice had even more tau fibrils than PS19 controls (Nov 2017 news). In San Diego, Wolozin added that TIA1 also facilitated propagation of tau oligomers throughout the brain.
In side-by-side preparations, Wolozin’s group generated tau fibrils and oligomers, then injected them into the entorhinal cortices of PS19 mice expressing either one or two copies of TIA1. Three months later, both oligomers and fibrils had seeded the spread of their kind to synaptically connected regions beyond the injection site. However, in the TIA1-deficient mice, the oligomers did not trigger spreading, whereas fibrils did, regardless of TIA1 copy number. Using antibodies specific for TIA1 and tau, Wolozin spotted TIA1 and oligomeric tau comingling in the neuronal cytoplasm, while fibrillar tau did not appear to interact with TIA1. Wolozin saw more neuronal loss in the entorhinal cortices of mice injected with oligomers than fibrils, suggesting that oligomeric tau is more toxic than fibrillar. The study was published on November 21 in Acta Neuropathologica (Jiang et al., 2018).
Wolozin proposed a bidirectional cascade, in which tau binding TIA1 promotes formation of stress granules in the cytoplasm, which shut down protein synthesis. In turn, the granules stabilize tau oligomers, furthering the cycle. Notably, Wolozin said that he observed oligomeric tau more commonly within the soma of neurons, while fibrillar tau accumulated in dendrites.
Oligomers, Take the Good with the Bad
Tau oligomers come in many shapes and sizes. Which are the worrisome ones? At SfN, Amritpal Mudher of the University of Southampton, U.K., described using biophysical techniques to track the emergence of different tau oligomer conformations throughout disease. Mudher uses flies to model tauopathy. These insects do not develop neurofibrillary tangles, but they do make tau oligomers, which is exactly what Mudher wants to study. Mudher reported that flies expressing three-repeat or four-repeat isoforms of human tau falter at climbing up glass vials and die younger than controls. The 3R isoforms are more toxic than the 4R (Sealey et al., 2017). Comparing the tau species present in fly neurons at early and late stages of this phenotype, Mudher found that oligomers emerge when symptoms first arise, while larger oligomers and ultimately fibrils appear only after the disease phenotype is entrenched. Mudher concluded that tau oligomers are primarily responsible for the deficits in the insects.
Are all tau oligomers equally harmful, or only certain strains? When Mudher treated the flies with lithium chloride, an agent that lessens tau phosphorylation and relieves damaging effects of tau pathology in flies (Mudher et al., 2004), she rescued the climbing and survival deficits, but found even more oligomers in the insects (Cowan et al., 2015). Reasoning that these oligomers must be of a distinct, and more benign, type than those found in sick flies, Mudher compared their conformation using Raman spectroscopy. By blasting molecules with a laser that jiggles molecular bonds, this technique produces a spectral signature specific to the secondary structure of a molecule. Mudher found tau oligomers in lithium chloride-treated flies to be distinct from those in untreated, diseased flies. Specifically, oligomers in the latter had a much higher β-sheet content than those in treated insects.
In collaboration with Martin Margittai of the University of Denver, Mudher also used Raman spectroscopy to compare several kinds of tau, including 3R, 4R, P301S, and delta-K18. Thus far she has only looked at fibrils, not oligomers, of these different types, but early findings already reveal striking differences, where each has a distinct combination of α-helical, β-sheet, and disordered content.
Mudher believes Raman spectrometry could identify the most toxic conformations of tau in disease models. That technique might even track tau oligomers, and disease progression, in diagnostic tests. Responding to audience questions, Mudher conceded that Raman requires grams of protein, so tau oligomers in CSF would need to be amplified to generate enough material for analysis. It is unclear whether the original oligomer conformations would be conserved after rounds of amplification.
Outside In, or Inside Out?
The field’s current emphasis on tau’s travels through the brain might easily obscure the fact that tau is an intracellular protein. Its normal post is on microtubules, and its abnormal accumulation in synapses correlates with synaptic dysfunction. How do extracellular tau oligomers relate to tau wreaking havoc within neurons? One possibility is that neurons take up these soluble forms of tau, which then travel to synapses. At SfN, Claudio Grassi of Università Cattolica del Sacro Cuore in Rome proposed another player. Grassi had previously reported that in cell culture, astrocytes more readily gobbled up tau oligomers than did neurons, and that tau accumulation in astrocytes interfered with release of astrocyte “gliotransmitters” that support synaptic function of nearby neurons. Grassi showed that tau accumulation in astrocytes blocked their release of ATP, which in turn stifled neuronal firing (Piacentini et al., 2017).
At SfN, Grassi described what happened when he blocked tau uptake into astrocytes with an antibody against GPC4, a heparin sulfate proteoglycan expressed only on this glial cell type. Such HSPGs have been implicated in the internalization of tau (Holmes et al., 2013; May 2018 news). Grassi reported that blocking astrocytic uptake of tau in hippocampal slice cultures prevented synaptic deficits otherwise caused by tau oligomers. He also reported that astrocytes can spit out tau, which is then readily internalized by neighboring astrocytes. He proposed that the indirect effects of astrocytic tau might explain a proportion of tau-dependent synaptic malfunction observed in neurons.
Eckhard Mandelkow of the German Center for Neurodegenerative Diseases (DZNE) in Bonn questioned Grassi after his talk, noting that the concentrations of tau oligomers Grassi used in his slice cultures far exceeded those found in the brain. He also challenged Grassi’s finding that the reduction in ATP released by tau-laden astrocytes was responsible for the synaptic defects in neighboring neurons. Previously, Eckhard and Eva Maria Mandelkow had reported that blocking the effects of adenosine, a breakdown product of ATP, with the adenosine A1 receptor antagonist rolofylline restored memory function in tau transgenic mice (Oct 2016 news). The Mandelkows had previously proposed that release of ATP by tau-laden neurons might increase extracellular levels of its breakdown product, which inhibits synaptic function in neurons. Grassi replied that he has not yet studied whether neuronal ATP or its breakdown products might contribute to his findings.
Mandelkow further pointed to a body of evidence implicating tau within presynaptic terminals of neurons (Apr 2015 news; Apr 2017 conference news). Researchers have recently reported that tau tethers to synaptic vesicles there, holding back their release (Feb 2018 news; Jul 2018 conference news). Mandelkow finds the hypothesis that tau harms synapses directly from within terminals more congruent with current data than the idea that it does so indirectly, via astrocytes. Grassi responded that he believes tau oligomers damage synapses via a combination of effects within astrocytes and the neurons themselves.
In synapses in the AD brain, tau is not the only troublemaker. Aggregated Aβ is commonly spotted at the scene, as well. For more on how synapses fare when both culprits are present, see Part 2 of this report.—Jessica Shugart
- Tau Silences, Aβ Inflames; Hitting Excitatory Synapses Hardest
- Mice Tell Tale of Tau Transmission, Alzheimer’s Progression
- Stress Granule Protein Stabilizes Tau Oligomers, Hastens Neurodegeneration
- To Deliver Itself From Cell to Cell, Phospho-Tau Uses UPS
- Could a Failed Heart Drug Treat Tauopathies?
- Not All About Dendrites: Presynaptic Tau Harms Plasticity, Too
- Location, Conformation, Decoration: Tau Biology Dazzles at AD/PD
- Tau Uses Synaptogyrin-3 to Clump Synaptic Vesicles
- Synaptic Tau Clangs the Dinner Bell for Hungry Microglia
Research Models Citations
- Jiang L, Ash PE, Maziuk BF, Ballance HI, Boudeau S, Abdullatif AA, Orlando M, Petrucelli L, Ikezu T, Wolozin B. TIA1 regulates the generation and response to toxic tau oligomers. Acta Neuropathol. 2018 Nov 21; PubMed.
- Sealey MA, Vourkou E, Cowan CM, Bossing T, Quraishe S, Grammenoudi S, Skoulakis EM, Mudher A. Distinct phenotypes of three-repeat and four-repeat human tau in a transgenic model of tauopathy. Neurobiol Dis. 2017 Sep;105:74-83. Epub 2017 May 11 PubMed.
- Mudher A, Shepherd D, Newman TA, Mildren P, Jukes JP, Squire A, Mears A, Drummond JA, Berg S, MacKay D, Asuni AA, Bhat R, Lovestone S. GSK-3beta inhibition reverses axonal transport defects and behavioural phenotypes in Drosophila. Mol Psychiatry. 2004 May;9(5):522-30. PubMed.
- Cowan CM, Quraishe S, Hands S, Sealey M, Mahajan S, Allan DW, Mudher A. Rescue from tau-induced neuronal dysfunction produces insoluble tau oligomers. Sci Rep. 2015 Nov 26;5:17191. PubMed.
- Piacentini R, Li Puma DD, Mainardi M, Lazzarino G, Tavazzi B, Arancio O, Grassi C. Reduced gliotransmitter release from astrocytes mediates tau-induced synaptic dysfunction in cultured hippocampal neurons. Glia. 2017 Aug;65(8):1302-1316. Epub 2017 May 18 PubMed.
- Holmes BB, DeVos SL, Kfoury N, Li M, Jacks R, Yanamandra K, Ouidja MO, Brodsky FM, Marasa J, Bagchi DP, Kotzbauer PT, Miller TM, Papy-Garcia D, Diamond MI. Heparan sulfate proteoglycans mediate internalization and propagation of specific proteopathic seeds. Proc Natl Acad Sci U S A. 2013 Aug 13;110(33):E3138-47. Epub 2013 Jul 29 PubMed.