Tau protein undergoes a bewildering variety of modifications, from phosphorylations and acetylations to the addition of sugar groups. What roles do these modifications play, and which most likely associate with disease? To tackle this question, researchers led by Lennart Mucke at the Gladstone Institute of Neurological Disease, San Francisco, cataloged and quantified changes to tau in wild-type and transgenic mice. Their results, published in the July 20 Nature Neuroscience, provide the most complete picture yet of how the adult brain regulates tau. The data confirmed 41 known modifications and uncovered 22 new ones, particular acetylations and ubiquitinations. Many tau amino acids appeared to be competitively modified by two or more processes, implying a high degree of regulatory control over tau. Notably, although the transgenic mice had tau-dependent memory and connectivity problems, their tau did not differ from those in control animals. “The data fit with the notion that a physiological form of tau enables Aβ-induced network abnormalities,” Mucke told Alzforum.
Commentators were enthusiastic, praising the thoroughness and rigor of the study. “This is a great paper and will be discussed at much length in tau circles,” Peter Davies at the Feinstein Institute for Medical Research in Manhasset, New York, wrote to Alzforum (see full comment below). He said the work is a heroic effort to characterize post-translational modifications of tau. Gail Johnson at the University of Rochester Medical Center, New York, expressed another common sentiment, writing, “This paper provides an excellent resource for future studies on the role of tau in physiological as well as pathological processes.”
Previously, Mucke and colleagues reported that reducing tau protein protected transgenic J20 mice against excitotoxicity and memory defects caused by a mutant human APP (see May 2007 news; Jan 2011 news on Roberson et al., 2011). They wondered if physiological forms of tau mediated Aβ’s synaptotoxic effects, or if specific pathological species accumulated in the transgenics.
The Big Picture. Competing acetylations and ubiquitinations dot the microtubule-binding region of tau (orange segments), while phosphorylations are largely confined to the proline-rich region (turquoise segment). [Courtesy of Morris et al., Nature Neuroscience.]
To address this, joint first authors Meaghan Morris and Giselle Knudsen isolated endogenous tau from whole hippocampal and cortical samples of wild-type and J20 mice, then subjected the protein to quantitative mass spectrometry. Improvements in this technology in recent years allow researchers to pin down the location of modifications, even when phosphorylation sites are highly clustered, Knudsen noted (see Beausoleil et al., 2006; Baker et al., 2011). The authors confirmed 27 previously reported phosphorylation sites, mostly in the proline-rich region or the C-terminal region (see image above). They also found three sites of arginine methylation, two of which are new. Most other modifications occurred on lysine residues, where the authors recorded 15 sites of ubiquitination, 14 of acetylation, and three of methylation. Only 12 of these 32 alterations had been reported previously. Many modifications appeared to compete for the same amino acid. “That suggests an incredibly high level of regulation, maybe by different pathways,” Mucke said.
Li Gan, who is also at Gladstone but was not involved in this work, found the implied competition between acetylation and ubiquitination particularly intriguing. “This is exciting for us since we showed that reducing tau acetylation elevates tau clearance in vitro and in vivo,” she wrote to Alzforum (see full comment below, and Sep 2010 news).
In recent years, some researchers have reported that the covalent addition of a particular sugar, N-acetylglucosamine (GlcNAc), via the hydroxyl group on serine and threonine residues of human tau curtails phosphorylation, perhaps by directly competing with the latter modification for the same residues. Enhancement of this O-GlcNAcylation has been touted as a therapeutic strategy (see Liu et al., 2004; Yuzwa et al., 2008). However, some studies cast doubt on this theory, reporting only one N-acetylglucosamine on rodent tau (see Wang et al., 2010). In agreement with this, Mucke and colleagues found only a single site (serine 400) of O-GlcNAcylation. Moreover, the level of O-GlcNAcylated tau was quite low, only detectable after running extracts on lectin chromatograpy to enrich for glycosylated proteins. “That surprised us. We had expected to find a lot of it,” Knudsen told Alzforum. The results agree with some studies that have proposed that O-GlcNAcylation reduces tau pathology by hindering its aggregation rather than competing with other modifications, or that it limits tau's interactions with downstream proteins by modifying them instead (see Mar 2012 news; Nov 2012 conference news).
Some researchers cautioned that the data does not reflect what might happen if removal of N-acetylglucosamine was blocked. Inhibiting the glycoside hydrolase O-GlcNAcase, for example, ramps up levels of tau O-GlcNAcylation. “It is worth bearing in mind that an O-GlcNAcase inhibitor could potentially enhance tau O-GlcNAcylations that are not detectable under native conditions,” Dirk Beher at Asceneuron, Lausanne, Switzerland, wrote to Alzforum (see full comment below).
Khalid Iqbal and colleagues at the New York State Institute for Basic Research in Developmental Disabilities, Staten Island, pointed out that mouse models may not match what happens in human brains, where tau is spliced and regulated differently. “Hopefully [the authors] will use the mass spectrometry methodology they have developed to analyze post-translational modifications of tau in AD and normal human brains,” Iqbal wrote (see full comment below).
On this point, the authors noted that despite some differences in mouse and human tau, the sequences are 89 percent identical, and all the new modifications identified in this study occurred on conserved sites, suggesting they will be relevant to human disease. However, the authors did not find 15 tau modifications previously reported in human studies in their mouse samples. Several of these occur in poorly conserved regions.
This mouse study did not identify a clear pathological species of tau. Mucke noted that it remains possible that pathological tau accumulates only in certain cellular components or cell types, or in such small quantities that it cannot be detected in whole lysates. In particular, Mucke and colleagues wondered if pathological tau might be confined to post-synaptic densities. The authors isolated tau from these structures in wild-type and transgenic mice and found 16 isoforms. Again, they found no difference between mouse strains. However, they did see altered amounts of two types of phosphorylated tau in post-synaptic densities compared to whole lysate. This suggests a potential role for these species at synapses, Mucke noted.
Regarding the lack of pathological tau, Mucke pointed out that this study did not address other ways tau might cause problems, such as through mislocalization to dendrites, aggregation, conformational changes, or via structural features such as prolyl isomerization (see Jul 2015 news). In future work, Mucke plans to further compare tau distribution among different cellular compartments, as well as look for tau mislocalization in transgenic mice.
Some researchers said that because the J20 mice do not develop tau pathology, their tau modifications may not reflect what happens in Alzheimer’s disease. Numerous studies have reported excessive acetylation or phosphorylation of tau in different mouse models, as well as in AD. However, Davies wrote, “Much of the talk about tau modification has been as much wishful thinking as anything. Especially with immunocytochemistry, it is possible to detect changes in tau phosphorylation that might be entirely insignificant in quantitative terms. We really need the kind of rigorous analysis the authors have done here.”
Overall, commentators agreed that the study may help expand the field’s focus beyond tau phosphorylation. “It is likely that the coordinated effects of different modifications play an important role in the regulation of tau function, localization, and turnover, and thus we really need to look beyond just phosphorylation events,” Johnson wrote.—Madolyn Bowman Rogers