# Dysregulation of complement at the synapse in P301S mice and human tauopathies.

- **DOI:** [10.1186/s40478-026-02347-2](https://doi.org/10.1186/s40478-026-02347-2)
- **Rating:** Limited (2.6 out of 5)
- **Verdict:** Age-resolved mouse and human data extend synaptic complement research through CSMD1, but statistical incoherence prevents confidence in the reported mechanism.
- **Framework version:** v5.9
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-10-08
- **Evaluation page:** https://www.nabu.science/paper/10.1186%2Fs40478-026-02347-2

## Key claims

1. These findings demonstrate that complement dysregulation occurs in areas of tau pathology and may contribute to synapse loss in tauopathies.
   - Support: Tentative
   - Type: Primary finding
   - Basis: causal contribution is inferred from associative mouse and post-mortem comparisons without a complement intervention
   - Location: Abstract, final sentence; Conclusion
2. C1q protein levels were markedly increased in brain homogenates from P301S mice compared to WT, accompanied by C1q deposition on tau aggregates.
   - Support: Supported
   - Type: Secondary
   - Basis: large directional differences are reported, although incompatible ANOVA details reduce confidence in exact inference
   - Location: Abstract; Results, Figs. 3–4
3. Synapse loss was evident for both excitatory and inhibitory synapses and was accompanied by an increased percentage of C1q positive excitatory synapses, unaffected by proximity to tau aggregates.
   - Support: Tentative
   - Type: Secondary
   - Basis: Fig. 6 reports fewer C1q-positive synapses around tau-positive neurons, contradicting the proximity wording
   - Location: Abstract; Results, Figs. 5–6
4. The classic pathway regulator CSMD1 was present on synapses and decreased on C1q positive synapses in P301S mice, implying a loss of protection from complement attack.
   - Support: Tentative
   - Type: Secondary
   - Basis: CSMD1 differences are associative and absent at six months, with no functional perturbation of protection
   - Location: Abstract; Results, Fig. 7
5. in GGT cases there was a significant reduction in the percentage of total CSMD1 labelled synapses ... compared to healthy controls.
   - Support: Supported
   - Type: Secondary
   - Basis: the reported direction is GGT-specific and rests on four heterogeneous post-mortem cases
   - Location: Results, CSMD1 positive synapses are decreased in human tauopathies; Fig. 8

## Strengths

- **Age-resolved mouse disease series** Complement transcripts, proteins, synaptic markers, and tau pathology are examined at 2, 4, and 6 months, covering the model's reported progression window.
  - Dimension: methodological rigour
  - Location: Methods, Animals; Results, Figs. 1–7
- **CSMD1 adds a distinct angle** The study assesses a classical-pathway regulator at synapses in P301S mice and extends that measurement to CBD and GGT tissue, providing an incremental addition to prior C1q-focused work.
  - Dimension: contribution
  - Location: Results, Figs. 7–8; Discussion, CSMD1 paragraphs
- **Relevant prior studies are engaged** The article discusses earlier P301S complement-expression findings and C1q/C3 knockout studies, distinguishing the present descriptive work from existing functional evidence.
  - Dimension: positioning
  - Location: Introduction, paragraph 2; Discussion, paragraphs 1 and 5

## Limitations

- **Central statistics are internally inconsistent** Fig. 3 reports incompatible F and p values and anomalous factor degrees of freedom; Fig. 4 gives two correlation coefficients, and Fig. 2 conflicts with the Results on Cd11b timing. These defects require correction before the quantitative findings can be relied upon.
  - Dimension: methodological rigour
  - Location: Results and legends for Figs. 2–4
- **Conclusions exceed associative evidence** The study does not manipulate complement or CSMD1, yet the Conclusion attributes resultant synapse loss to classical-pathway dysregulation in mice and humans. Fig. 8 reports no significant human synapse-density difference.
  - Dimension: reporting
  - Location: Results, Fig. 8; Conclusion
- **Human validation is narrow and heterogeneous** The post-mortem groups are small and contain multiple co-pathologies, while the CSMD1 difference is significant only in GGT and is measured on total Bassoon-positive synapses rather than excitatory synapses.
  - Dimension: methodological rigour
  - Location: Methods, Human brain tissue; Table 1; Results and Fig. 8

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted
  - C1q protein levels were markedly increased in brain homogenates from P301S mice compared to WT, accompanied by C1q deposition on tau aggregates: "A correlation analysis comparing neuronal levels of aggregated tau and C1q staining by Pearson analysis showed a significant positive correlation (r = 0.31, p < 0.0001; Fig. 4f)." (Results, C1q is deposited on neurons with tau aggregates) vs "Pearson correlation analysis of the mean fluorescent intensity of tau and C1q on neurons showed significant positive correlation (r = 0.24, p < 0.0001)." (Fig. 4 legend). The same tau–C1q correlation is reported as 0.31 and 0.24; both are positive, but neither can be selected as correct from the supplied text. → Effect on result unclear
  - Synapse loss was evident for both excitatory and inhibitory synapses and was accompanied by an increased percentage of C1q positive excitatory synapses, unaffected by proximity to tau aggregates: "Synapse loss was evident for both excitatory and inhibitory synapses and was accompanied by an increased percentage of C1q positive excitatory synapses, unaffected by proximity to tau aggregates." (Abstract) vs "the percentage of C1q-positive synapses was ~ 20–30% lower around tau-positive neurons compared to tau-negative (NeuN-stained) neurons" (Results, Loss of excitatory and inhibitory synapses). The Abstract's unqualified absence of a proximity effect conflicts with the reported approximately 20–30% lower C1q-positive fraction around tau-positive versus tau-negative neurons, although this comparison does not itself establish a distance gradient. → Not supported by the paper's own data
  - in GGT cases there was a significant reduction in the percentage of total CSMD1 labelled synapses ... compared to healthy controls: "tissue from two primary tauopathy disease subtypes, CBD (5 cases) and GGT (4 cases), were selected and compared to healthy controls (6 donors)" (Results, CSMD1 positive synapses are decreased in human tauopathies) vs "Brain sections from corticobasal degeneration (CBD, n = 5) and Globular glial tauopathy (GGT, n = 4) cases were compared to healthy control (n = 5)." (Fig. 8 legend). The Results specify six controls while Methods and Fig. 8 specify five, so the analyzed control count and any resulting change in significance cannot be established. → Effect on result unclear
  - These findings demonstrate that complement dysregulation occurs in areas of tau pathology and may contribute to synapse loss in tauopathies: "We demonstrate that synaptic loss in both P301S mice and human tauopathy brain may be attributed to complement classical pathway dysregulation at the synapse, particularly in the proximity of tau pathology." (Conclusion) vs "In contrast to mouse brain, there was no significant difference in synaptic density; however, a non-significant trend towards decreased excitatory synapses was observed in GGT compared to control cases" (Results, CSMD1 positive synapses are decreased in human tauopathies). The Conclusion presents human synaptic loss as demonstrated, whereas the human Results explicitly report no significant synaptic-density difference and only a nonsignificant excitatory-synapse trend. → Not supported by the paper's own data
  - Minor reporting notes (3):
    - Descriptive: "F[2,37] = 3.864, p < 0.0001" (Fig. 3 legend, brain C3b/iC3b age effect). F[2,37] = 3.864 recomputes to p=0.030. → Conclusion holds
    - Descriptive: "Cd11b (F[1,42] = 4.9, p = 0.0309; WT-P301S 6 M p = 0.0325)" (Fig. 2 legend) vs "significantly higher expression levels were observed in P301S compared to WT for C3ar1 at 4M, Cd11b at 4 M and Cd11c at 2 M" (Results, Transcriptional changes in complement genes). The legend locates the significant Cd11b comparison at 6 months, whereas the Results and Discussion locate it at 4 months; the supplied text cannot identify the correct age. → Effect on result unclear
    - Descriptive: "Observations in human tauopathy brains demonstrated decreased CSMD1-labelled excitatory synapses, consistent with findings in P301S mice." (Abstract) vs "The percentage of synapses staining for CSMD1 and double-stained for CSMD1 and C1q were assessed only in total bassoon positive synapses because of detection availability" (Fig. 8 legend). The Abstract identifies a decrease specifically in excitatory CSMD1-labelled synapses, but the human CSMD1 analysis measured total Bassoon-positive synapses without excitatory classification. → Not supported by the paper's own data
- **Reporting and transparency:** Noted. The supplied text leaves ANOVA factor labels and analysis sizes unresolved, does not clearly identify the correlation's analysis unit or explain the Fig. 6 sample-size range, and asserts an unpresented sex comparison.
  - Ethics approval: declared
- **Reference integrity:** Noted
  - 27 references in manuscript
  - 27 of 27 checkable references found in an index
  - the reference list may be incompletely parsed; counts are indicative
  - Citation details diverge for 1 reference:
    - #24 Wanying M Lulu M Afang Z Huan C Yuguang H Astrocyte-microglia interaction through C3/C3aR pathway modulates neuropathic pain in rats model of chronic constriction injury Mol Pain 2025 18 17448069221140532 10.1177/17448069221140532 PMC9669679 36341694 Wanying M, Lulu M, Afang Z, Huan C, Yuguang H (2025) Astrocyte-microglia interaction through C3/C3aR pathway modulates neuropathic pain in rats model of chronic constriction injury. Mol Pain 18:17448069221140532. 10.1177/17448069221140532 10.1177/17 (DOI resolves to: “Astrocyte-microglia interaction through C3/C3aR pathway modulates neuropathic pain in rats model of chronic constriction injury” — Mou, 2022 (Molecular Pain). Differs from the DOI’s record on: year, authors) https://doi.org/10.1177/17448069221140532
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-10-17
