Research Article: Acetylation: a new key to unlock tau’s role in neurodegeneration

Date Published: May 29, 2014

Publisher: BioMed Central

Author(s): Casey Cook, Jeannette N Stankowski, Yari Carlomagno, Caroline Stetler, Leonard Petrucelli.


The identification of tau protein as a major constituent of neurofibrillary tangles spurred considerable effort devoted to identifying and validating pathways through which therapeutics may alleviate tau burden in Alzheimer’s disease and related tauopathies, including chronic traumatic encephalopathy associated with sport- and military-related injuries. Most tau-based therapeutic strategies have previously focused on modulating tau phosphorylation, given that tau species present within neurofibrillary tangles are hyperphosphorylated on a number of different residues. However, the recent discovery that tau is modified by acetylation necessitates additional research to provide greater mechanistic insight into the spectrum of physiological consequences of tau acetylation, which may hold promise as a novel therapeutic target. In this review, we discuss recent findings evaluating tau acetylation in the context of previously accepted notions regarding tau biology and pathophysiology. We also examine the evidence demonstrating the neuroprotective and beneficial consequences of inhibiting histone deacetylase (HDAC)6, a tau deacetylase, including its effect on microtubule stabilization. We also discuss the rationale for pharmacologically modulating HDAC6 in tau-based pathologies as a novel therapeutic strategy.

Partial Text

The identification of tubulin as the first cytosolic protein to be modified by acetylation [1,2] challenged the traditional notion that acetylation only serves as a mechanism to regulate transcription through modification of histones. Since this discovery in 1985, researchers have sought to identify other proteins that undergo acetylation events and elucidate the effects of this post-translational modification on protein structure and function. Global proteomic studies allowed for the identification of hundreds of proteins that are acetylated on one or multiple lysine residues, as well as a myriad of lysine acetyltransferases and deacetylases, which respectively govern protein acetylation and deacetylation [1,3]. The discovery that the microtubule-associated protein tau is also a target of acetyltransferase and deacetylase enzymes [4,5] added a new layer of complexity, whereby the impact of phosphorylation or ubiquitination on tau function and biology will now need to be re-evaluated to include consideration of tau acetylation. The purpose of the current review is to discuss the recent findings associated with tau acetylation, a novel post-translational modification of tau, how it influences tau aggregation and function, and whether it could be exploited therapeutically as a treatment for tauopathies.

As lysine residues are unique in their ability to participate in electrostatic and hydrophobic interactions [6,7], and are also known to play a critical role in tau assembly and toxicity [8-10], we and others recently questioned whether tau acetylation of lysine residues would modulate its potential to aggregate [4,11]. Cohen and collagues [4] utilized the acetyltransferase CREB-binding protein (CBP) to acetylate a fragment of tau comprising the microtubule-binding domain (commonly referred to as K18), and observed an increase in aggregation of the K18 fragment. We subsequently performed a similar analysis but using full-length tau and the acetyltransferase p300; we detected a decrease in filament assembly following tau acetylation, the extent of which correlated with the concentration of p300 [11]. We also observed a complete reversal of p300-mediated acetylation and inhibition of tau assembly upon addition of the deacetylase histone deacetylase (HDAC)6 [11]. Furthermore, the modulation of tau assembly by acetylation was dependent on modification of tau’s KXGS motifs in the microtubule-binding domain, as evidenced by the fact pseudoacetylation of the four KXGS motifs generated a tau species that was assembly-incompetent and resistant to modulation by either p300 or HDAC6 [11]. The results from these two studies suggest that CBP and p300 may preferentially acetylate different residues in tau, thereby differentially impacting tau’s intrinsic propensity to aggregate.

The multitude of molecular and functional properties of the microtubule-associated protein tau are predominantly due to the protein’s natively unfolded structure, allowing tau to not only interact with a large number of other cellular proteins, but also undergo a variety of post-translational modifications [16]. The occurrence of several post-translational modifications on numerous proteins has been well described, and it has been postulated that the interaction of such modifications governs complex regulatory processes, which are essential for proper protein function and for the regulation of diverse cellular events [3]. While each post-translational modification is distinct and utilizes different chemical groups to modify a given protein on specific residues, a certain degree of overlap exists [3,17]. For instance, lysine residues are targets for acetylation events and other modifications, including ubiquitination, sumoylation and methylation [3]. Thus, a measure of rivalry between different post-translational modifications must exist, where the addition of one chemical group to a given residue precludes further modifications [3].

Following identification of the tau protein as a major constituent of NFTs in AD and other tauopathies, several lines of research focused on identifying the mechanism(s) responsible for tau accumulation in disease. Most conducted studies have focused on the effect of hyperphosphorylation on tau turnover, due to the fact that hyperphosphorylation has been the first and one of the most prominent post-translational modifications associated with tau pathology [33,39-42]. In particular, previous studies have demonstrated that the tau ubiquitin ligase, CHIP, is unable to bind and ubiquitinate tau species phosphorylated by Par-1/MARK2 on the 12E8 epitope (S262/356) [33], a p-tau species that is also resistant to degradation upon treatment with Hsp90 inhibitors [32,33]. Tau phosphorylated at the PHF1 epitope (S396/404) is still susceptible to degradation following Hsp90 inhibition and actually exhibits an enhanced interaction with Hsp90 [33]. These findings indicate that certain p-tau species, rather than normal tau, are a preferred client protein of Hsp90, while some phosphorylation events, in particular those mediated by Par-1/MARK2 on tau’s KXGS motifs, generate a p-tau species not recognized by the chaperone network. Phosphorylation by Par-1/MARK2 on KXGS motifs in the microtubule-binding domain of tau has been shown to be required for initiation of the pathogenic cascade of hyperphosphorylation, which is ultimately associated with NFT formation in tauopathies [29]. HDAC6 disrupts this cascade by potentiating Par-1/MARK2-mediated phosphorylation of KXGS motifs (detected by the 12E8 antibody), an effect that is eliminated by pseudoacetylation of KXGS motifs [11]. In addition, our recent findings indicate that HDAC6 directly modulates tau polymerization and acetylation, and this relationship is dependent on the ability of HDAC6 to deacetylate tau specifically on KXGS motifs [11]. These results support the hypothesis that decreased HDAC6 activity increases acetylation of KXGS motifs and, in so doing, prevents phosphorylation of serine residues within the same motif. As acetylation and phosphorylation of KXGS motifs act in a competitive manner, and phosphorylation of KXGS motifs generates a p-tau species that is resistant to degradation, future studies will be required to determine whether acetylation of tau on KXGS motifs impacts the ability of the chaperone network to recognize tau in a similar manner to phosphorylation on these sites. Given that progressive hypoacetylation and hyperphosphorylation of KXGS motifs is observed in rTg4510 mice with aging [11], the fact that tau turnover also decreases with aging in rTg4510 mice [43] may indicate that the relationship between acetylation and phosphorylation on KXGS motifs regulates tau turnover. The effects of other post-translational modifications on tau turnover are unknown; thus, it remains to be determined whether differentially modified tau species are degraded by the same mechanisms as hyperphosphorylated tau, or if they are preferentially targeted to alternative degradation pathways.

Based on recent evidence that HDAC6 regulates tau acetylation on KXGS motifs, it is of particular interest that, in a Drosophila model of tauopathy, loss of HDAC6 activity rescued tau-induced microtubule defects in both neuronal and muscle cells [45]. This finding provides the first in vivo evidence that reducing HDAC6 activity in a model of tauopathy is protective. Further emphasizing the therapeutic potential of HDAC6 inhibitors are results demonstrating that loss of HDAC6 expression/activity is also neuroprotective in other neurodegenerative diseases, including AD, Huntington’s disease and amyotrophic lateral sclerosis [46-48]. For instance, in a mouse model of AD, genetic ablation of HDAC6 alleviated cognitive impairment without impacting plaque burden, which may suggest that beneficial consequences of loss of HDAC6 expression are due to effects on endogenous tau, though this has not yet been assessed in this model [47]. Deletion of HDAC6 in a mouse model of mutant superoxide dismutase 1-linked amyotrophic lateral sclerosis is also neuroprotective, as reflected by the extended life span of mice and increased motor axon integrity [48].

There is now considerable evidence supporting the trans-cellular propagation and seeding of tau pathology in a variety of in vitro and in vivo models, ultimately demonstrating that extracellular tau filaments can be internalized by cells and function as seeds for the assembly of intracellular filaments [58-63]. While the precise mechanism(s) underlying trans-neuronal tau propagation has yet to be elucidated, recent work is beginning to provide insight into this pathway. Wu and colleagues [64] observe internalization of misfolded tau at the level of both dendritic and axonal terminals in neurons, after which pathologic tau species can be transported in either the antero- or retrograde direction, thereby leading to the spreading of pathology. In addition, injection of brain material from mice that express human mutant P301S tau into transgenic mice expressing human wild-type tau (ALZ17 model) was sufficient to induce tau pathology not only within, but also adjacent to, the injection site along anatomically connected pathways [58]. Furthermore, injection of brain extracts from patients with different tauopathies into either ALZ17 or non-transgenic mice was not only sufficient to drive inclusion formation, but actually effectively reproduced the classic hallmark lesions of the specific tauopathy characteristic of the inoculating brain extract [65]. These studies provide additional support for the concept that pathologically altered tau species possess a remarkable self-propagating and seeding capacity, and also indicate that seeding-competent tau species are somehow different and distinct across the class of tauopathies, such that the inoculating material acts as an exact template in the new host. The specific characteristics of pathological tau species that define and determine seeding capacity remain to be identified, and could be the result of a precise pattern of post-translational modifications that differentially impact conformation of the tau molecule and ultimately determine aggregate structure. Our recent findings, which demonstrate that acetylation within tau’s KXGS motifs generates a tau species that fails to polymerize [11], suggests that augmenting acetylation of the KXGS motifs would also decrease tau seeding capacity.

We review here the rationale supporting the utilization of HDAC6 inhibition to enhance tau acetylation as a novel therapeutic strategy for tauopathies. HDAC6 inhibitors simultaneously promote acetylation and prevent phosphorylation of tau on KXGS motifs, thereby interfering with tau’s propensity to aggregate. Decreasing HDAC6 activity also enhances microtubule stability and transport, which is expected to further stimulate neuronal function. As HDAC6 inhibitors are currently being evaluated in clinical trials for oncology indications, data will soon be available to assess the safety of pharmacologic modulation of HDAC6 in humans, which could expedite their repurposing for other diseases. Although additional research is needed to fully elucidate the cellular and molecular pathways associated with the neuroprotective consequences of HDAC6 inhibition, it is becoming increasingly apparent that modulating HDAC6 activity may offer a very promising avenue for the treatment of AD and associated tauopathies.

This article is part of a series on Tau-based therapeutic strategies, edited by Leonard Petrucelli. Other articles in this series can be found at

AD: Alzheimer’s disease; CBP: CREB-binding protein; CHIP: C-terminus of Hsc70 interacting protein; EpoD: Epothilone D; HDAC: Histone deacetylase; Hsp: Heat shock protein; NFT: Neurofibrillary tangle; PHF: Paired helical filament; p-tau: Hyperphoshorylated tau; SIRT1: Sirtuin 1.

The authors declare that they have no competing interests.