One Oncogene Creates Many Kinds of Senescent Liver Cells, Each With a Different Fate
In zebrafish livers, UHRF1 overexpression produces distinct senescent cell states. Tp53 loss frees some to proliferate, and Navitoclax clears only a subset.
Summary
Senescent cells can suppress or promote cancer, and it is unclear why. Researchers overexpressed the epigenetic regulator UHRF1 in zebrafish liver cells, which causes DNA methylation loss, retrotransposon activation, DNA double-strand breaks, cell cycle exit, and Atm- and Tp53-dependent senescence. Over time, several senescent populations emerged. Some had strong immune-signaling and anti-apoptotic profiles, and others also expressed proliferation genes. UHRF1 level and Tp53 status shaped each population's fate. When Tp53 was lost, cells with lower UHRF1 could re-enter the cell cycle, but cells with the highest UHRF1 stayed arrested. The senolytic Navitoclax cleared only some senescent cells, while Dasatinib had no significant effect. The results suggest senolytics could leave behind escape-prone cells in UHRF1-driven cancers, so senescence heterogeneity matters for therapy.
Detailed Summary
Senescence is often described as a single, permanent arrest state, but it is far more varied. Some senescent cells are cleared by the immune system and protect against cancer. Others resist apoptosis, persist, support the tumor microenvironment, or escape and proliferate. What generates this diversity is poorly understood, and it matters for senolytic drugs, which may only work on some senescent populations.
The team used a zebrafish model in which human UHRF1, an epigenetic regulator that maintains DNA methylation and is elevated in most cancers, is overexpressed in hepatocytes. This leads to senescence within days and liver cancer weeks later. They combined EdU labeling, γH2AX and p-Rpa staining, bulk RNA-seq at 80 and 120 hours post-fertilization (hpf), reduced-representation bisulfite sequencing, single-cell RNA-seq of 23,121 liver cells, confocal imaging, tp53 and atm mutants, and drug treatments.
Cell cycle withdrawal, DNA double-strand breaks (γH2AX) and LTR retrotransposon expression appeared as early as 80 hpf. By 120 hpf, about 15% of CpGs had lost methylation, including at LTR retrotransposons. The inflammatory secretory program (SASP), NF-kB induction and macrophage recruitment came later. SA-β-gal staining required both Atm and Tp53 but not Atr. Tp53 loss rescued senescence and SASP genes, yet cell-division genes stayed suppressed and early proliferation was not restored, so arrest at this stage was maintained upstream of Tp53.
Single-cell analysis showed four hepatocyte populations. Hep2 was most depleted and most senescence-enriched, with strong signaling to mesenchymal cells and macrophages. Transgene expression varied from cell to cell, and the highest-expressing cells showed the deepest senescence. At 10 days, tp53 loss tripled the number of proliferating EGFP-positive cells, but cells with the highest UHRF1 stayed EdU-negative whether or not Tp53 was present. Navitoclax reduced SA-β-gal staining and increased hepatocyte EdU incorporation, while Dasatinib had no significant effect.
The implication is that UHRF1 level and Tp53 status sort cells into deep senescence, Tp53-restrained senescence, or escape. Senolytic therapy in UHRF1-driven cancers could remove some cells while sparing or expanding escape-prone ones, which may help explain disappointing clinical results with Navitoclax in advanced liver cancer.
Caveats: this is a preprint in zebrafish larvae at an early precancer stage. The link between retrotransposon activation and DNA damage is speculative, and human relevance and long-term tumor outcomes after senolytic treatment were not directly tested. The full discussion text was truncated, so this summary relies on the methods and results.
Key Findings
- UHRF1 overexpression caused DNA methylation loss (~15% of CpGs), LTR retrotransposon activation, double-strand breaks and cell cycle exit within about 80 hours post-fertilization.
- Atm and Tp53, but not Atr, were required for UHRF1-induced SA-β-gal staining and SASP, yet Tp53 loss did not restore early proliferation.
- Single-cell RNA-seq revealed distinct hepatocyte senescent populations, with Hep2 showing the strongest senescence, signaling and senolytic-target enrichment.
- At 10 days, Tp53 loss let cells with low or medium UHRF1 proliferate, but the highest-UHRF1 cells stayed arrested.
- Navitoclax reduced SA-β-gal staining and raised hepatocyte EdU incorporation, while Dasatinib had no significant effect, suggesting only some senescent cells are eliminated.
Methodology
Transgenic zebrafish overexpressing human UHRF1 in hepatocytes were analyzed at 80–120 hpf and 10 dpf using EdU, γH2AX/p-Rpa staining, bulk RNA-seq, RRBS, scRNA-seq (23,121 cells), and confocal imaging. Tp53 and atm mutants, an ATR inhibitor, and the senolytics Navitoclax and Dasatinib were used to test pathway dependence and drug sensitivity.
Study Limitations
This is a non-peer-reviewed preprint in a zebrafish larval model, and the mechanism linking retrotransposon activation to DNA damage is speculative. Methylation profiling was only possible at 120 hpf, and the effect of senolytics on later tumor progression and in human tissue was not tested. The full discussion text was truncated in the source provided.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe:
