P-tau217 and p-tau181 are some of the earliest known markers of Alzheimer’s disease, rising in blood and cerebrospinal fluid 20 years before symptoms appear. But what do these markers mean? In the December 28 Acta Neuropathologica, researchers led by Mathias Jucker at the University of Tübingen, Germany, tied them to amyloid deposits, but not tangles. In a mouse model of β amyloidosis, both p-taus rose in CSF in tandem with plaque formation in the brain. Curiously, p-tau217 also rose in a mouse model of familial Danish dementia. These mice develop brain deposits of a diffuse amyloid made from a totally different protein, BRI2. They have no fibrillar Aβ plaques. Neither mouse forms tau tangles.

  • In mice, amyloid plaque formation drives the rise in CSF p-tau217 and -181.
  • The effect was the same whether the amyloid was Aβ or Danish dementia deposits.
  • Despite high p-tau, the mice do not develop tangles.

“Our results raise the possibility that p-tau increases whenever amyloid is deposited in brain parenchyma, independent of the type of amyloid,” Jucker wrote to Alzforum.

Adam Boxer at the University of California, San Francisco, agreed. “The data are supportive of a model whereby amyloid leads to elevated production of soluble tau species,” he noted. He found it encouraging that transgenic mice mirror the same amyloid–then-p-tau dynamic seen in AD patients, which suggests they will be useful for mechanistic studies.

Amyloid Triggers P-Tau. CSF p-tau217 spikes with age in APPPS1 (red, left) and ADanPP (red, right) amyloidosis mice, but not in wild-type (green). [Courtesy of Kaeser et al., Acta Neuropathologica.]

In longitudinal patient data, elevated p-tau217 and -181 precede even amyloid PET as markers of amyloidosis (Aug 2019 conference news; Dec 2019 news). These two p-taus correlate tightly with plaque formation, though not as well with other AD pathologies such as tangles and atrophy (Mar 2020 news). 

To discern whether amyloid pathology alone is enough to boost p-tau, Jucker and colleagues turned to APPPS1 mice, which develop plaques by around 3 months of age, but no tangles. First author Stephan Kaeser measured CSF p-tau in the mice and their non-transgenic cousins at 1.5, 3, 6, 12, 18, and 20 months of age. In the transgenic mice, p-tau217 and p-tau181 paralleled the time course of plaque formation in the brain, rising from six to 18 months and then plateauing. P-tau217 had the more dramatic spike, ending up about 15-fold higher than in young mice, compared to a fourfold increase for p-tau181. In people, p-tau217 is more sensitive than p-tau181 at detecting AD (Apr 2020 conference news; Jul 2020 news). Total CSF tau also increased about fourfold with age. In the non-transgenic mice, by contrast, p-tau and t-tau remained unchanged.

The authors next examined ADanPP mice. They carry a mutant BRI2 gene expressing an elongated peptide that aggregates. BRI2 is a neuronal membrane protein that may play a role in differentiation and neurite extension. These mice accumulate diffuse BRI2 amyloid plaques in gray matter as well as vascular deposits. Surprisingly, CSF p-tau217 also rose with age in these animals, ending up 43-fold higher at 18 than at 3 months. Meanwhile, total tau increased about sevenfold with age. The authors did not have enough CSF to measure p-tau181.

The findings hint that p-tau217 is not specific for Aβ aggregation, but instead may act as a marker of amyloidosis more generally. If so, high p-tau217 would not necessarily mean someone is on the path to AD, Jucker speculated. They could, in theory, be developing a different type of secondary tauopathy.

In people, elevated p-tau predicts the formation of tangles (Mar 2020 news; Jan 2021 newsDec 2021 news). Why does this not happen in mice? Jucker noted that it could be due to differences between mouse and human tau, or could indicate that p-taus do not cause tangles.

To distinguish these possibilities, Kaeser plans to measure p-tau in tau transgenic mice expressing human mutant tau. He will also measure p-tau directly in the brains of amyloidosis models to find out if it increases in neurons as well as in CSF. Kaeser added that because amyloidosis mice model the p-tau increase seen in AD, they could also be useful for interpreting p-tau changes seen in trials of anti-amyloid agents.—Madolyn Bowman Rogers


  1. This interesting report suggests that CSF t-tau, P-tau181 and -217 are elevated in response to a variety of amyloid species, and that this elevation may occur similarly in mouse models and humans, even if the mice do not develop insoluble tau pathology. The data are supportive of a model whereby amyloid leads to elevated production of soluble tau species, which might potentially be a prerequisite for deposition of insoluble tau species in humans.

    It is encouraging that the relationship between amyloid and soluble tau species can be recapitulated in transgenic mice, thereby providing a model system for understanding the mechanisms by which elevated amyloid leads to elevated tau. 

    However, the relationship between soluble tau species and insoluble tau deposition (NFT and other aggregates) still remains unclear. While there are strong correlations between the two in humans, much work will be needed to understand how these different tau species relate to each other and to the clinical symptoms of tauopathies.  Understanding this relationship will be key for developing more efficient clinical trials and monitoring the response to novel AD therapeutics.

  2. Jucker and colleagues again make an important contribution to our understanding of the mechanisms of AD, and the relationships between amyloid and tau. Their results clearly indicate that amyloid is sufficient for p-tau217 and p-tau181 and, together with the time course of changes and effects seen in observational cohorts, begins to indicate the mechanistic links of how amyloid begets tau pathophysiology and ultimately, tau pathology.

    Several recent findings from human studies suggest that there is a sequence of tau changes, starting with amyloid plaques, then p-tau217 and 181, proceeding through p-tau205 and total tau, and then tau aggregation as measured by tau PET. However, it is critical to not conclude causality based on associations, as can happen in human studies without interventions. Longitudinal changes provide more support (e.g., increases in soluble tau concentrations with increasing tau pathology by tau PET); however, these associations are also limited. As seen in the DIAN study, p-tau217 and 181 phosphorylation occupancy decrease during the symptomatic phase of disease, when tau pathology is expected to be increasing. This paradoxical relationship further supports p-tau217 and 181 not being direct measures of tau aggregation pathology, even though overall tau concentrations may still increase.

    With the need to identify tau drugs that impact tau targets, the importance of interpretation of tau biomarkers is critical. For example, when p-tau217 or 181 are lowered with anti-amyloid drugs that remove amyloid, is this a direct downstream effect of amyloid removal, and/or evidence that tau pathology has been altered? The DIAN-TU recently launched a combination trial of an anti-tau and an anti-amyloid antibody, highlighting the need to understand how to measure tau biomarkers related to amyloid removal versus independent anti-tau biological impact.

    The rate of breakthroughs for amyloid and tau biomarkers is already fast and accelerating. We are in a golden age of tracking the molecular signatures of Alzheimer's pathophysiology, and understanding how to interpret these biomarkers will be essential for decision-making in trials and the clinic. This report provides an important insight into the relationships of p-tau to amyloid.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Move Over Aβ, CSF P-Tau Tells Us There’s Plaque in the Brain
  2. Blood Tests of Phospho-Tau, Aβ42, Track With Brain Amyloid
  3. Different CSF Phospho-Taus Match Distinct Changes in Brain Pathology
  4. 217—The Best Phospho-Tau Marker for Alzheimer’s?
  5. Plasma p-Tau217 Set to Transform Alzheimer’s Diagnostics
  6. A Phospho-Tau Plasma Assay for Alzheimer’s?
  7. Plasma P-Tau181 Predicts, Monitors Alzheimer’s Progression
  8. In Preclinical Alzheimer's, p-tau217 in Blood Best Predicts Tangles

Research Models Citations

  1. APPPS1
  2. ADanPP

Further Reading

Primary Papers

  1. . CSF p-tau increase in response to Aβ-type and Danish-type cerebral amyloidosis and in the absence of neurofibrillary tangles. Acta Neuropathol. 2022 Feb;143(2):287-290. Epub 2021 Dec 28 PubMed.