# Karyoptosis mediates cell death and neurodegeneration upon proteotoxic stress.

- **DOI:** [10.1038/s41467-026-73802-w](https://doi.org/10.1038/s41467-026-73802-w)
- **Rating:** Limited (2.9 out of 5)
- **Verdict:** Karyoptosis offers a consequential lamina-centred model of proteotoxic neuronal death, but non-independent analyses materially weaken its mechanistic and human-disease claims.
- **Framework version:** v5.9
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-10-08
- **Evaluation page:** https://www.nabu.science/paper/10.1038%2Fs41467-026-73802-w

## Key claims

1. karyoptosis, a distinct form of cell death, can be induced by proteotoxic stress and then develops through nuclear degeneration and cellular expulsion of nuclear material.
   - Support: Supported
   - Type: Primary finding
   - Basis: multiple stress models support the phenotype, but distinctness and death endpoints remain operationally limited
   - Location: Abstract
2. karyoptosis is regulated by the p38 kinase signalling pathway, which controls stability of the nuclear lamina protein LaminB1 via direct phosphorylation.
   - Support: Supported
   - Type: Secondary
   - Basis: biochemical and phospho-mutant evidence converge, although some reported rescue comparisons are non-significant
   - Location: Abstract; Results, Figs. 4–6
3. karyoptosis affects neurons in models of amyotrophic lateral sclerosis/frontotemporal dementia (ALS/FTD) pathology.
   - Support: Supported
   - Type: Secondary
   - Basis: rodent-neuron and fly DPR models converge, despite an incompatible statistic in one neuronal panel
   - Location: Abstract; Results, Figs. 6–7
4. we identify karyoptotic features in post-mortem frontal cortex of FTD and Alzheimer's disease (AD) patients.
   - Support: Supported
   - Type: Secondary
   - Basis: cross-sectional morphology identifies compatible features but cannot establish the underlying death process
   - Location: Abstract; Results, Fig. 8
5. karyoptosis appears to be responsible for a remarkable extra 18–20% of cell degeneration in FTLD or AD.
   - Support: Tentative
   - Type: Secondary
   - Basis: cross-sectional cell clustering cannot identify a causal share of degeneration and ignores case-level dependence
   - Location: Results, final paragraph before Discussion

## Strengths

- **Orthogonal tests of one mechanism** The study combines p38 inhibition, p38 RNAi, an in vitro kinase assay, and non-phosphorylatable LaminB1/LaminB mutants. Rescue of fly survival and lamina morphology gives the proposed pathway functional support beyond a single reagent.
  - Dimension: methodological rigour
  - Location: Results, Figs. 4–6
- **Evidence spans several biological systems** The proposed phenotype is examined in neuroblastoma cells, rat primary neurons, Drosophila, engineered human iPSC-derived neurons, and post-mortem cortex. This broadens the contribution beyond one model or assay.
  - Dimension: contribution
  - Location: Results, Figs. 1–8
- **Relevant alternatives receive direct testing** The paper evaluates apoptosis, necroptosis, ferroptosis, parthanatos, and senescence-associated markers rather than defining karyoptosis solely by morphology. It also discusses CREB3-associated karyoptosis and linker-cell death as related precedents.
  - Dimension: positioning
  - Location: Results, Fig. 3 and Supplementary Figs. 2–4; Discussion, paragraph 1

## Limitations

- **Central analyses use non-independent observations** Several imaging tests treat hundreds of cells as inferential observations, while the post-mortem χ² analysis treats 763 cells from 20 cases as independent. These analyses do not account for clustering within wells, experiments, flies, or human cases.
  - Dimension: methodological rigour
  - Location: Figs. 1b,e,h, 4b–f, and 8
- **Reported statistics conflict with interpretations** Fig. 6c contains an incompatible F/p pair, while the direct rescue comparisons in Figs. 4f and 6k are non-significant despite stronger narrative claims. These discrepancies create a material Internal Coherence concern.
  - Dimension: reporting
  - Location: Figs. 4f and 6c,k legends; corresponding Results sections
- **Human disease burden is over-attributed** Cross-sectional morphology and k-means clusters identify cells with compatible features but cannot show that karyoptosis caused their degeneration or quantify a causal 18–20% share. The case-level sampling structure is also absent from the inferential model.
  - Dimension: contribution
  - Location: Results, Fig. 8 and final paragraph before Discussion

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted
  - karyoptosis is regulated by the p38 kinase signalling pathway, which controls stability of the nuclear lamina protein LaminB1 via direct phosphorylation: "and reduced hallmarks of DNA damage and cell death in primary rodent neurons with STX17 knockdown (Fig. 4a–f)." (Results, p38 MAPK pathway section) vs "Stx17/− vs. Stx17/+ p > 0.9999" (Figure 4f legend, Dunn’s multiple comparisons). The direct comparison of STX17-knockdown neurons with and without p38 inhibition has adjusted p > 0.9999, so it does not establish the reduction attributed to inhibition in the narrative. → Not supported by the paper's own data
  - karyoptosis affects neurons in models of amyotrophic lateral sclerosis/frontotemporal dementia (ALS/FTD) pathology: "F = 0.4603, p = 0.0021 and Šídák’s multiple comparisons tests: non-transfected vs PR100 p = 0.0009, EGFP vs PR100 p = 0.0077, non-transfected vs GR100 p = 0.0495, non-transfected vs GA100 p = 0.7498 (c)." (Figure 6c legend). An upper-tail ANOVA F statistic of 0.4603 is incompatible with the reported p = 0.0021, but the degrees of freedom and underlying observations are not printed, preventing identification of the corrected statistic or p-value without reconstructing the analysis. → Effect on result unclear
  - karyoptosis is regulated by the p38 kinase signalling pathway, which controls stability of the nuclear lamina protein LaminB1 via direct phosphorylation: "Kruskal–Wallis statistic = 12.78, p < 0.0001 and Dunn’s multiple comparisons: PR8 + LamBT413A vs PR64 p = 0.0009, PR64 vs PR64 + LamBT413A p = 0.0401 (m)." (Figure 6m legend). Assuming the four PR8/PR64-by-phosphomutant groups implied by the panel description, the usual chi-square reference distribution with three inferred degrees of freedom gives P(χ² ≥ 12.78) ≈ 0.0051 rather than p < 0.0001. → Conclusion holds
- **Reporting and transparency:** Noted. Lifespan analyses have conflicting test names in Methods and figure legends, and Figure 3 assigns two statistical blocks to the same endpoint label; neither issue permits a numerical correction from the supplied text.
  - Conflict of interest: not stated · Ethics approval: declared · Preregistration: not stated · Data availability: declared
- **Reference integrity:** Clean
  - 69 references in manuscript
  - 69 of 69 checkable references found in an index
  - 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
