Scientists debuted two new tauopathy models recently, and they are unlike other tau mice. Developed in parallel by two laboratories, these mice point to excitotoxicity as a possible mechanism by which the alanine-152-threonine tau mutation boosts risk for various neurodegenerative diseases. Tau aggregation was scant in the mice, which nonetheless exhibited age-related neuroinflammation, neurodegeneration, and symptoms mimicking human dementias. The study authors suspect that soluble, oligomeric tau-A152T is to blame.
“These are important and useful mouse models of tauopathy for both mechanistic studies and future anti-tau drug testing,” commented Yadong Huang of the Gladstone Institute of Neurological Disease in San Francisco, who was not involved in either paper.
Most tau mutations that cause neurodegeneration are found in or near the protein’s microtubule-binding repeats, and they typically cause the protein to aggregate. In contrast, tau-A152 sits away from the repeats in a proline-rich area that might be involved in intracellular signaling. Unlike the majority of tau mutations, which lead to frontotemporal dementia in people, A152T seems to boost risk for a variety of other conditions as well, including Alzheimer’s, and dementia with Lewy bodies.
One of the research groups led by Lennart Mucke at the Gladstone Institute used the CaMKII promoter to overexpress, by about fourfold, mutant or wild-type tau in the forebrain. First author Sumihiro Maeda and colleagues describe the animals in the March 1 EMBO Reports online. The other group, led by Eva-Maria Mandelkow at the German Center for Neurodegenerative Disease (DZNE) in Bonn, placed the mutant human transgene under control of the Thy1.2 promoter, expressing it in all neurons. These animals made about as much mutant tau as they did endogenous tau. Mandelkow and colleagues described the mice in two papers: in the same issue of EMBO Reports, co-first authors Jochen Decker and Lars Krüger of DZNE report excitotoxicity results, while co-first authors Astrid Sydow and Katja Hochgräfe of the Max-Planck-Institute for Metabolism Research in Hamburg, Germany, report on the neuroinflammatory and cognitive abilities of the mice in the February 25 Acta Neuropathologica Communications.
Despite the expression system differences, mice from both labs displayed more or less the same phenotypes. “The most striking thing was the similarity between the papers,” Mucke said. Though the mice lived a normal lifespan, they displayed plenty of interesting biochemical and neurological quirks.
In Mucke’s lab, Maeda and colleagues found that four- to 10-month-old A152T mice expressed only half as much human tau mRNA as did the wild-type line, yet made about the same amount of tau protein. The A152T animals also had fewer tau protein fragments, suggesting that the mutation might stabilize tau. “That is one way A152T could contribute to disease, by making tau harder to get rid of,” Mucke speculated.
The Mandelkow group observed that as the A152T mice aged from three to 20 months, their brains accumulated more hyperphosphorylated, insoluble tau, and more tangles than wild-type mice. The aggregation was much less pronounced than in other tauopathy models based on different tau mutations, Mandelkow said. Maeda saw hyperphosphorylation in both the mutant and control tau mice, as well, but did not observe much aggregation of tau, leading Mucke’s group to conclude that soluble, possibly oligomeric tau was the cause of neurodegeneration. Previously, researchers observed high levels of tau oligomerization in vitro as well (Coppola et al., 2012).
In Germany, Sydow and Hochgräfe examined brain sections from 10-month-old mice for signs of neuroinflammation. They saw proliferation of both astrocytes and microglia, compared with wild-type control animals. To measure the phenomenon in living mice, they crossed the A152T line with mice expressing luciferase under control of the astrocyte-specific GFAP promoter. By injecting the mice with luciferin, the authors could see the astrocytes glow in a special imaging chamber (see image above). GFAP expression in 14-month-old mice was double that of three-month-olds. This assay could make a convenient longitudinal marker for tau toxicity, said Mandelkow. Mucke’s group also observed astrocytosis in their A152T mice.
All of this soluble tau accumulation and inflammation added up to bad news for mice. The Mandelkow group detected decreasing expression of the neuronal marker NeuN in the hippocampus starting at 12 months of age. Mucke and colleagues found fewer NeuN-positive neurons in the dentate gyrus and hippocampus by about 20 months in their A152T mice, with no degeneration in the wild-type tau controls.
What caused the cell death? One thing that kills neurons, Mucke noted, is hyperexcitability. Both sets of investigators studied neural activity in their models. They focused on the mossy fiber synapses between the dentate gyrus and the CA3 pyramidal neurons in the hippocampus, where antibodies indicated abnormal tau accumulated. At Gladstone, Biljana Djukic stimulated brain slices from the tau model mice, and observed stronger transmissions in both the A152T and wild-type lines, compared to control nontransgenic mice. This started as early as four months of age and strengthened further by 20 months. EEGs also indicated that the neural circuits of the A152T mice, but not the wild-type version, tended to fire in synchrony, in a seizure-like pattern—possibly due to the excess synaptic activity. At DZNE, Decker and Krüger found higher rates of synchronous synaptic transmission in the mossy fibers of brain slices as well. The heightened synaptic activity distinguishes these A152T mice from most other tauopathy models, in which synaptic transmission drops, noted Maeda and Krüger.
The Mandelkow group reasoned the hyperexcitability could lead to excitotoxicity. Over-firing neurons would release too much glutamate into synapses, leading downstream neurons to take up too much calcium, activating cell death processes. In hippocampal slices from the A152T mice, they observed excess extracellular glutamate, abnormally high uptake of calcium into neurons, and amplified release of the cell death marker lactate dehydrogenase, supporting this theory.
The neurodegeneration and hyperexcitability created noticeable behavioral changes. The mice in Mucke’s lab struggled with nest-building, as well as the learning and memory in a water maze. The maze difficulties worsened as the animals aged. Serious impairment was evident starting at around 17 months. Sydow and Hochgräfe reported their A152T mice struggled to learn a water maze once they reached 16 months of age
Since tau-A152T is a risk factor for Alzheimer’s and other conditions, Mucke hypothesized that it acts by sensitizing neurons to other problems. To test this idea, the San Francisco researchers crossed their A152T mice to a line expressing mutant APP. Most of the resulting mutants did not make it past the embryo stage, but those that did exhibited more synchronized neuronal firing than the A152T or APP mice. “We think that one way A152T tau might increase the risk of AD is to synergize with the Aβ-dependent network hyperactivity,” said Mucke. In the case of other conditions, such as FTD or dementia with Lewy bodies, tau-A152T might sensitize neurons to other pathologies, he surmised.
Overall, the new models indicate that the A152T tau mutant differs from most other tau mutations studied, in particular because it does not aggregate much tau. “However, even if we do not have aggregation, we still can have toxicity,” said Eckhard Mandelkow, a co-lead author on the Decker and Krüger paper. He noted that the proline-rich domain where A152 is located likely controls unidentified cellular pathways that might be altered by the mutation. The Mandelkows have developed a nematode model for tau A152T in which they plan to screen for such disruption. Alternatively, Huang suggested, the A152T mutation might alter the structure of tau.
“Just blocking aggregation may not be enough as a tau-targeted treatment,” said Mucke. “Our papers also underline that fact that one also needs to worry about the pathogenicity of soluble tau.” Both research groups now plan to try treating their mice, for example by blocking hyperexcitability. Decker and Krüger have already tried the antibiotic ceftriaxone in their brain slices because it increases astrocyte uptake of glutamate. It normalized the concentrations of calcium and glutamate, and prevented neuron death.—Amber Dance
1. Maeda S, Djukic B, Taneja P, Yu GQ, Lo I, Davis A, Craft R, Guo W, Wang X, Kim D, Ponnusamy R, Gill TM, Masliah E, Mucke L. Expression of A152T human tau causes age-dependent neuronal dysfunction and loss in transgenic mice. EMBO Rep. 2016 Mar 1. [PubMed].
2. Decker JM, Krüger L, Sydow A, Dennissen F, Siskovà Z, Mandelkow E, Mandelkow EM. The Tau/A152T mutation, a risk factor for frontotemporal-spectrum disorders, leads to NR2B receptor-mediated excitotoxicity. EMBO Rep. 2016 Mar 1. [PubMed].
3. Sydow A, Hochgräfe K, Könen S, Cadinu D, Matenia D, Petrova O, Joseph M, Dennissen FJ, Mandelkow EM. Age-dependent neuroinflammation and cognitive decline in a novel Ala152Thr-Tau transgenic mouse model of PSP and AD. Acta Neuropathol Commun. 2016 Feb 25;4(1):17.[PubMed].
To view commentaries, primary articles and linked stories, go to the original posting on Alzforum.org here.
Copyright 1996–2016 Biomedical Research Forum, LLC. All Rights Reserved.