BGPT: Paper Review: Absence of a Role for Phosphorylation in the Tau Pathology of Alzheimer’s Disease

Explore by Goal

Quick Explanation Copied

Central claim: phosphorylation in the assay context inhibits tau–tau repeat-domain binding, while becoming “aggregation-competent” requires a prior solid-phase polyanionic substrate capture; therefore, the paper argues disease-modifying efforts should not directly target phosphorylation as the initiating driver.

Long Explanation

Paper Review (Critical, Evidence-Forward): Absence of a Role for Phosphorylation in the Tau Pathology of Alzheimer’s Disease

DOI: 10.3390/biom6020019 • Published: 2016-04-08

What the experiments actually test

Two-phase binding geometry: a tau fragment (“dGA”, a repeat-domain fragment corresponding to phase-shifted repeats 2/3/4) is adsorbed to a PVC solid phase, while full-length tau is presented in the aqueous phase; binding is measured using immunochemical readouts with specific antibodies.
Phosphorylation manipulations: recombinant tau is hyperphosphorylated in vitro (T40P), and neonatal rat tau can be endogenously phosphorylated (NT) and further hyperphosphorylated in vitro (NTP).
Binding-parameter model: binding curves are fit with Langmuir-equation forms; the authors interpret parameters as dissociation constants (K_d) and maximal binding outputs.

Figure A — Aqueous-phase Kd values for tau binding to adsorbed dGA (phosphorylated vs non-phosphorylated)

Observed quantitative pattern (from the paper’s Table 2)

For binding of aqueous T40 to adsorbed dGA, the reported limiting K_d in the aqueous-phase context is ~21.1 ± 2.9 nM.
When the aqueous partner is phosphorylated (T40P), the reported K_d becomes ~236.5 ± 96.3 nM (≈11-fold inhibition).
The inhibition is much larger when phosphorylation is present in certain neonatal-derived contexts: e.g., Table 2 reports K_d values for binding involving further hyperphosphorylated neonatal tau (NTP) in the aqueous-phase context reaching ~756.4 ± 228 nM for capturing non-phosphorylated T40.

Figure B — Adsorbed-state limiting Kd values (solid-phase “capture” step)

Core mechanistic inference (and what it does / doesn’t prove)

The authors’ mechanistic centerpiece is a phase-geometry dependent “avidity” + conformational locking model: repeat-domain tau–tau binding becomes high-affinity when a tau species has first bound to a solid-phase polyanionic substrate (PVC here), yielding a repeat-domain configuration that supports saturable binding.

Skeptical boundary: this is still an in vitro binding assay. The strongest evidence here is about binding affinity changes under controlled phosphorylation states and phase presentation—not direct demonstration that phosphorylation is irrelevant in living brains. The authors themselves note the identity of the endogenous polyanionic “PAS” enabling nucleation remains unknown.

Figure C — Relative binding preference: tau–tau vs tau–tubulin (as ratio language in the paper)

Therapeutic implication: phosphorylation-only targeting may miss the initiating step

The paper argues that since phosphorylation inhibits binding in this system and does not appear to be required for establishing a high-affinity, aggregation-competent configuration, drug development should more directly target tau–tau binding inhibition and/or blockade of tau’s capture by the (unidentified) endogenous solid-phase substrate.

Critical appraisal (skeptical, mechanistic, and model-aware)

Strengths aligned with the paper’s claims

Quantitative parameterization: the binding data are fit with a saturation model and high correlation coefficients are reported.
Direct side-by-side comparison: tau–tau binding and tau–tubulin binding are assessed under comparable assay geometry, enabling consistent interpretation of phosphorylation effects in each interaction class.

Limitations / blind spots (what could break the inference)

Endogenous PAS identity is unknown: the “solid-phase substrate” is PVC in vitro, while the in vivo analogue is explicitly undetermined; mapping PVC-driven kinetics to brain substrates is therefore inferential.
Phosphorylation completeness is not characterized: hyperphosphorylation is produced using kinase extracts and phosphorylation-dependent antibodies label epitope-specific presence, but the paper does not appear to quantify the full site occupancy across all potential sites.
Assay readout captures binding, not necessarily aggregation morphology/kinetics: the system reports binding affinities consistent with Langmuir-like behavior; however, aggregation competence in vivo depends on multiple steps beyond pairwise binding. The paper’s conclusions are mechanistic extrapolations from binding affinity patterns.

Conflict-of-interest and inference caution

Two authors are officers in TauRx Therapeutics Ltd and co-inventors on tau-related patents; although the authors declare no conflict of interest, that relationship can plausibly bias framing of therapeutic implications. This doesn’t invalidate the binding measurements, but it does increase the need to separate “what the data show” from “what the data suggest for drug targeting.”

Sanity-check visualization: phosphorylation shifts Kd upward (log scale)

This view uses the reported limiting K_d values as a direct empirical proxy for “affinity” differences under the paper’s assay conditions.

What would disprove the paper’s main inference?

If, under more physiologically relevant substrates (not PVC) and more complete PTM profiling, phosphorylation states do not inhibit key tau–tau binding steps or do not re-shape aggregation competence as predicted.
If endogenous nucleation-capable polyanionic substrates are identified and show that phosphorylation is required for nucleation competency rather than merely modulating binding affinities.
If aggregation outcomes in vivo (not just binding) track differently from binding-affinity predictions across phosphorylation states.

Buttons: go deeper on the authors’ perspectives

Feedback:

Updated: April 07, 2026