Integrated human microglial tau pathway
The study follows tau from receptor-associated uptake through degradation, phosphorylation, secretion, EV packaging, and downstream seeding in human iPSC-derived microglia.
↳ Results, Figures 1–7
Abstract
INTRODUCTION: Microglia have been implicated in the templated spread of tau aggregates in tauopathies through mouse studies. However, it is unclear whether these findings translate to human disease.
METHODS: We challenged human induced pluripotent stem cell (iPSC)-derived microglia-like-cells (iMGL) with monomeric and fibrillar recombinant tau and tau purified from Alzheimer's patient brains, examining in detail the uptake, processing, release, and seeding of tau by microglia.
RESULTS: iMGL take up tau via lipoprotein receptor-related protein 1 (LRP)1 and heparan sulfate proteoglycans, with leucine-rich repeat kinase 2 affecting LRP1 trafficking. Monomeric tau is digested effectively with minimal effects on iMGL, but recombinant or brain-derived tau fibrils induce chemokine/interferon response subtypes, alongside downregulation of homeostatic genes. Fibrillar tau is degradation-resistant, can escape into the cytoplasm, and becomes phosphorylated on two specific residues. iMGL release partially digested fibrillar tau, including in extracellular vesicles, visualized by cryo-electron microscopy, that seed aggregation in neurons.
DISCUSSION: Our study reveals new insights into human microglial responses to tau, highlighting opportunities to limit pathogenic tau spread.
Key Claims
Primary findingSupported
iMGL release partially digested fibrillar tau, including in extracellular vesicles, visualized by cryo-electron microscopy, that seed aggregation in neurons.
Fragile — the claim's weakest requirement is that the effect exists at all. Based on how the study was designed and conducted, not on statistics extracted from the paper. Experimental. Every point below is traceable to a quoted passage, but this forecast has not yet been validated against known replication outcomes.
cryo-ET and neuronal seeding support the direction, but co-pelleted seeds and lipofection limit EV-specific attribution
Abstract; Results, Figures 6–7
SecondarySupported
iMGL take up tau via lipoprotein receptor-related protein 1 (LRP)1 and heparan sulfate proteoglycans, with leucine-rich repeat kinase 2 affecting LRP1 trafficking.
LRP1 has orthogonal perturbations, while HSPG and LRRK2 interpretations rely on less specific or partly non-isogenic comparisons
Abstract; Results, Figure 1 and Figures S4–S5
SecondarySupported
Monomeric tau is digested effectively with minimal effects on iMGL, but recombinant or brain-derived tau fibrils induce chemokine/interferon response subtypes, alongside downregulation of homeostatic genes.
RNA-seq and immunodepleted brain-tau comparisons support the pattern, though donor structure and some threshold wording limit certainty
Abstract; Results, Figures 2–3
SecondaryTentative
Fibrillar tau is degradation-resistant, can escape into the cytoplasm, and becomes phosphorylated on two specific residues.
degradation and phosphorylation are supported, but partial membrane compartments and negative leakage assays do not confirm cytoplasmic escape
Abstract; Results, Figures 4C and 5E
Derived from the full evaluation — not a separate score.
Strengths · Limitations
Strengths
Strong input and comparator controls
Recombinant tau was depleted of endotoxin, AD-brain material had an immunodepleted comparator, and release measurements included cell-free plastic-adherent tau controls.
↳ Methods, Recombinant tau production and Immunodepletion of hTau; Results, Figure 4A
Substantive engagement with divergent findings
The Discussion directly addresses why cytokine responses differ from earlier microglial studies, considering endotoxin, dose, and human-versus-rodent biology.
↳ Discussion, inflammatory-response paragraphs
Limitations
EV seeding attribution remains unresolved
Differential ultracentrifugation does not exclude co-pelleted free fibrils or seeds, while lipofection of EV fractions bypasses natural uptake. The experiments therefore establish seeding by the preparation more securely than seeding by intravesicular tau.
↳ Methods, EV purification and Tau seeding assays; Results, Figure 7
Cytoplasmic escape is overstated
Figure 4C shows fibrils in partial or damaged membrane-bound compartments, but the Discussion calls escape confirmed despite negative lysosomal-leakage assays that are not shown.
↳ Results, Figure 4C; Discussion, CLEM/endolysosomal-damage paragraph
Disease translation remains model-bound
The disease-level interpretation rests on iPSC-microglia monoculture, heparin-induced recombinant fibrils or pooled brain tau, and transfection-assisted seeding without in-vivo validation.
↳ Methods, fibril assembly and seeding assays; Discussion, concluding paragraphs
Nabu Commentary
The integrated human iMGL program and orthogonal assays support a clear advance over isolated uptake or release studies. Endotoxin control, brain-tau immunodepletion, receptor perturbations, multi-omics, and blinded cryo-ET assessment support a quality score in the competent-to-strong range. The main deductions arise because EV fractions may contain co-pelleted free seeds, seeding uses lipofection, and HSPG involvement is inferred from heparin alone. The Discussion also exceeds the direct Figure 4C evidence by treating cytoplasmic escape as confirmed despite negative, unshown leakage assays.
Check another paper
Author Note / Field Feedback
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Dimension breakdown
Study Quality
Strong3.8
Confidence mediumContribution
4.0
The study integrates tau uptake, processing, phosphorylation, secretion, EV imaging, and downstream seeding in human iPSC-derived microglia. Its clearest advance is the combined human-cell and cryo-ET evidence, although several uptake findings extend established rodent and neuronal work.
Introduction; Results, Figures 1–7
“Microglia secrete undegraded seeding-competent tau to the conditioned medium, including in EVs”
Methodological Rigour
3.7
Endotoxin-controlled tau, immunodepleted brain material, LRP1 competition and CRISPR perturbation, cell-free release controls, and orthogonal imaging and seeding assays provide a broad experimental foundation. Confidence is moderated by differential-ultracentrifugation EV isolation, lipofection-based seeding, heparin-only HSPG evidence, and incompletely visible independent-unit reporting.
Methods, Recombinant tau production, Immunodepletion of hTau, EV purification, and seeding assays; Results, Figures 1, 4, 6, and 7
“the proportion of tau packaged in EVs was substantially lower than the amount released freely into the CM”
Reporting
3.6
The paper follows a coherent progression from uptake to cellular response, processing, release, and seeding. Precision falls where the Discussion treats cytoplasmic escape as confirmed despite qualified CLEM evidence and negative leakage assays that are not shown.
Results, Figure 4C; Discussion, paragraph beginning “CLEM identified regions in tau-fibril-fed iMGL”
“confirms that tau can escape into the cytoplasm”
Positioning
3.9
The Discussion engages rodent, neuronal, microglial, inflammatory, and human-brain EV literature and offers concrete explanations for divergent cytokine findings. Model distance and assay-induced constraints are not fully traced into the disease-level conclusion about microglial EV contributions in AD brain.
Discussion, inflammatory-response and EV paragraphs
“microglia are likely contributing seeding-competent tau-fibril-laden EVs”
Reliability
Caveats4 of 4 checks
Internal CoherenceNoted
Several interpretations are stated more strongly than the reported evidence supports, most materially for cytoplasmic escape. These issues qualify specific claims but do not invalidate the overall tau-processing findings.
Research ConductNoted
Ethics approval and consent, conflicts of interest, RNA-seq data and code availability, and proteomics deposition are declared. No conduct concern is established from the supplied text. One or more identifiers named in the paper's availability or registration statements did not resolve when checked on 2026-10-08T13:39:54.784780+00:00: https://github.com/S‐Washer/Karabova_2025_Tracking_tau_and_cellular_responses_in_microglia_RNAseq/tree/main. Recorded as a discrepancy between the paper's claims and the cited sources; not an assessment of the underlying research.
Flags: 4 declared / 5 total
Reference IntegrityClean
107 references in manuscript 107 of 107 checkable references found in an index
Post-Publication RecordClean
No retraction notice found in Retraction Watch.
Last checked: 08 Oct 2026
Sources: Retraction Watch ✓
Readiness
Where this paper’s evidence sits on the path from initial observation to real-world use.
Mechanistic
The work characterizes mechanisms in controlled human-cell models and demonstrates seeding in engineered biosensors. It does not test a therapeutic intervention, natural EV uptake, in-vivo efficacy, or a clinical validation route.
Methods, Tau seeding assays; Results, Figure 7
“At DIV11, iNeurons were transfected using Lipofectamine200 with 2 µg of EVs”
AI-generated, human-governed. Something look off? Contact us to request a review.