Viewpoint Cite This: ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
pubs.acs.org/chemneuro
Bridging Multiple Dementias Hitoshi Okazawa* Department of Neuropathology, Medical Research Institute, Center for Brain Integration Research, Tokyo Medical and Dental University, Tokyo 113-8510, Japan ABSTRACT: Tau phosphorylation has come into the limelight again, this time as a critical player in the earliest stages of dementia pathology. Mislocalization of phosphorylated tau to dendritic spines and the resultant degeneration of synapses are observed across multiple neurodegenerative diseases, even in the absence of tau aggregation. Moreover, other molecules phosphorylated by the same kinases, such as MARCKS, might contribute to ultra-early phase pathology by promoting synapse dysfunction. KEYWORDS: Tau phosphorylation, dementia, MARCKS, synapse, Alzheimer's disease, frontotemporal lobar degeneration
■
NMDA receptor; and promotes excitotoxic signaling.1 By contrast, Hoover and colleagues2 showed that tau decreases the number of AMPA and NMDA receptors without decreasing spine number.2 Neither group performed a detailed investigation of the physical chemistry of tau in dendritic spines.
CONCEPTUAL CHANGE OF TAU’S ROLE IN ALZHEIMER’S DISEASE Along with extracellular amyloid beta (Aβ) aggregation, intracellular tau aggregation is a neuropathological hallmark of Alzheimer’s disease (AD). In pathological terms, tau aggregates are referred to as “neurofibrillary tangles” or “paired helical filaments,” whereas Aβ aggregates are “senile plaques.” The conformations of the monomeric peptides of these diseaseassociated proteins influence the speed with which they aggregate to form polymers, as well as the resultant structure of the aggregates; however, our understanding of the details of this process remains insufficient. Phosphorylation of tau, which induces a conformational change that could affect both its physiological function and the aggregation of the monomer, is therefore considered to be a critical step in the initiation of tau aggregation. However, protein phosphorylation can occur before human postmortem brain samples reach the laboratory, obscuring the significance of individual phosphorylation sites. Accordingly, the pathological significance of several phosphorylation sites detected in post-mortem AD human brains has been questioned in mouse models. Extracellular Aβ aggregates were once considered to be the toxic species that is both necessary and sufficient for AD. More recently, however, a new disease category was discovered, “tauopathy,” characterized by tau gene mutations and dementia with neurofibrillary tangles in the absence of senile plaques. This discovery led to the new hypothesis that tau aggregation is the downstream pathological process responsible for the toxicity of Aβ aggregates. According to this hypothesis, tau phosphorylation, which is invariably detected in the brains of patients with AD or tauopathy, plays a supporting role in tau aggregation. Today, however, tau phosphorylation is considered to be a major player in pathogenesis. Two groups using AD mouse models reported that nonaggregated phosphorylated tau is mislocalized to dendritic spines, where it impairs synapse function.1,2 The underlying mechanisms of tau spine pathology identified by these studies were similar, but not identical. Ittner and colleagues1 reported that tau carries Fyn, a nonreceptor tyrosine kinase, into spines, where it phosphorylates subunit 2 of the NMDA receptor; stabilizes PSD95, which anchors the © XXXX American Chemical Society
■
TAU PHOSPHORYLATION IN FRONTOTEMPORAL LOBAR DEGENERATION On the other hand, a similar scheme of pTau-mediated pathology at dendritic spines has not been examined in frontotemporal lobar degeneration (FTLD) either associated with tau aggregation (FTLD-tau) or with TDP43 aggregation (FTLD-TDP). We recently found that tau protein phosphorylated at Ser203 is mislocalized to dendritic spines in both post-mortem human FTLD-TDP brains and a FTLD-TDP mouse model, and eventually induces instability and loss of dendritic spines3 (Figure 1). FTLD-TDP, a major class of FTLD, is a human neuropathology associated with TDP43 aggregation but not tau
Figure 1. Relationship of the “ERK1/2-tau/MARCKS” core pathways to major dementias. Received: March 11, 2018 Accepted: March 13, 2018
A
DOI: 10.1021/acschemneuro.8b00119 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
Viewpoint
ACS Chemical Neuroscience
were initiated. Accordingly, clinical trials for preclinical or prodromal stages such as DIAN, A3, and A4 have already been initiated. The second possibility, not mutually exclusive with the first, is that other molecules had already initiated preaggregation pathology in these patients before extracellular Aβ aggregation took place. I refer to the concept underlying the second hypothesis as “ultra-early phase pathology.”
aggregation. The confirmation of pTau-mediated pathology in FTLD-TDP expands the territory of common pathology in which mislocalization of tau induces loss of dendritic spines and dysfunction of synapses to include AD, FTLD-tau, and FTLDTDP (Figure 2), which together account for more than 70% of cases of dementia.
■
PHOSPHORYLATION OF MARCKS BEFORE EXTRACELLULAR Aβ AGGREGATION Based on the second hypothesis, we performed comprehensive phosphoproteome analysis of brains obtained from AD mouse models or human AD patients, and found that the phosphorylation levels of three proteins were altered before extracellular Aβ aggregation was detectable by immunohistochemistry. Among them, phosphorylation of MARCKS at Ser465 appears to be most important, because this modification occurs specifically in neuronal processes undergoing degeneration (degenerative neurites), thereby weakening the interaction between the actin network and cell membrane. This destabilizes the neurite, as well as the postsynaptic membrane protrusion of the excitatory synapse called the dendritic spine5 (Figure 2). By screening for candidate kinases based on the sequence around the phosphorylation site, the kinase responsible for tau phosphorylation at Ser203 in FTLD-TDP was identified as Erk1/2,3 which is also implicated in phosphorylation of MARCKS at Ser46 in AD.5 Therefore, the pathway from Erk1/2 to tau and/or MARCKS (Figure 1) could be critical for initiating ultra-early phase pathology. Future work should seek to identify the upstream event that explains the connection between the disease-related proteins and Erk1/2 activation.
Figure 2. Implication of ERK1/2 activation in dendritic spine degeneration by phosphorylated tau and phosphorylated MARCKS.
The chronological relationship between tau phosphorylation, tau aggregation, and TDP43 aggregation was investigated in a FTLD-TDP mouse model carrying a progranulin (PGRN) gene mutation.3 This work revealed that, in this model, tau phosphorylation occurs first, before the onset of symptoms, followed by TDP43 aggregation, whereas tau aggregation does not occur even after the onset of symptoms. These data suggest that tau phosphorylation is one of the earliest pathological events prior to pathological protein aggregation, which is defined by formation of the detergent-insoluble ubiquitinated polymers.
■
FUTURE DIRECTIONS In this brief review, I summarized the emerging concept that phosphorylation, but not aggregation, of tau plays a critical role in the common pathology of multiple forms of neurodegenerative dementia, even including those not characterized by tau aggregation. In this view, phosphorylation precedes aggregation of tau, and phosphorylated tau is itself pathogenic. The underlying mechanism of pathogenesis could involve mislocalization of tau to dendritic spines. Erk1/2 phosphorylates not only tau, but also MARCKS, thereby inducing instability and loss of dendritic spines, which is directly linked to synapse dysfunction during ultra-early phase pathology, prior to protein aggregation. Future therapeutics based on this paradigm could delay or prevent the onset of various forms of dementia.
■
DEMANDS FOR NEW HYPOTHESES INDEPENDENT OF AGGREGATION Over the past few years, several clinical trials were performed to test the ability of anti-amyloid antibody to remove extracellular Aβ aggregates from the brains of human AD patients. These included small-scale phase II trials, such as that of aducanumab, performed by Biogen (URLs: https://www.biopharmadive. com/news/biogens-alzheimers-drug-fails-putting-focus-all-onaducanumab/513626/ and http://media.biogen.com/pressrelease/adaptive-phase-ii-study-ban2401-early-alzheimersdisease-continues-toward-18-month-end/), and large-scale phase III trials, such as that of solanetumab, performed by Lilly (URL: https://investor.lilly.com/releasedetail. cfm?ReleaseID=1000871). All of the antibodies tested could remove extracellular Aβ aggregation as expected, as demonstrated by PET imaging. Unfortunately, however, none of the clinical trials could confirm a therapeutic effect on cognitive symptoms. Consistent with this, a retrospective analysis showed that active immunization by Aβ removed extracellular Aβ aggregation in postmortem brains but did not alleviate cognitive impairment.4 The important lesson from the failure of these studies is that extracellular Aβ aggregation does not exactly correspond to cognitive symptoms. The failures of these clinical trials can be interpreted in two ways. The first possibility is that, in the trial subjects, extracellular Aβ aggregation had already induced irreversible effects on neurons before the anti-amyloid antibody injections
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Funding
H.O. is funded by Brain Mapping by Integrated Neurotechnologies for Disease Studies (Brain/MINDS) from the Japan Agency for Medical Research and Development (AMED), the Strategic Research Program for Brain Sciences (SRPBS), and a Grant-in-Aid for Scientific Research on Innovative Areas “Foundation of Synapse and Neurocircuit Pathology” (22110001, 22110002) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) B
DOI: 10.1021/acschemneuro.8b00119 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
Viewpoint
ACS Chemical Neuroscience and CREST from the Japan Science and Technology Agency (JST). Notes
The author declares no competing financial interest.
■
REFERENCES
(1) Ittner, L. M., Ke, Y. D., Delerue, F., Bi, M., Gladbach, A., van Eersel, J., Wolfing, H., Chieng, B. C., Christie, M. J., Napier, I. A., Eckert, A., Staufenbiel, M., Hardeman, E., and Gotz, J. (2010) Dendritic function of tau mediates amyloid-beta toxicity in Alzheimer’s disease mouse models. Cell 142, 387−397. (2) Hoover, B. R., Reed, M. N., Su, J., Penrod, R. D., Kotilinek, L. A., Grant, M. K., Pitstick, R., Carlson, G. A., Lanier, L. M., Yuan, L. L., Ashe, K. H., and Liao, D. (2010) Tau mislocalization to dendritic spines mediates synaptic dysfunction independently of neurodegeneration. Neuron 68, 1067−1081. (3) Fujita, K., Chen, X., Homma, H., Tagawa, K., Amano, M., Saito, A., Imoto, S., Akatsu, H., Hashizume, Y., Kaibuchi, K., Miyano, S., and Okazawa, H. (2018) Targeting Tyro3 ameliorates a model of PGRNmutant FTLD-TDP via tau-mediated synaptic pathology. Nat. Commun. 9, 433. (4) Holmes, C., Boche, D., Wilkinson, D., Yadegarfar, G., Hopkins, V., Bayer, A., Jones, R. W., Bullock, R., Love, S., Neal, J. W., Zotova, E., and Nicoll, J. A. (2008) Long-term effects of Abeta42 immunisation in Alzheimer’s disease: follow-up of a randomised, placebo-controlled phase I trial. Lancet 372, 216−223. (5) Okazawa, H. (2017) Ultra-Early Phase pathologies of Alzheimer’s disease and other neurodegenerative diseases. Proc. Jpn. Acad., Ser. B 93, 361−377.
C
DOI: 10.1021/acschemneuro.8b00119 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX