Fig 2: p16Ink4a-expressing cells recruit neutrophils during aging ARNA-sequencing data showing expression of genes associated with neutrophil recruitment during liver aging.B, CQuantification of (B) neutrophil infiltrates (NIMP-positive) and (C) mean number of TAF-positive hepatocytes in liver of young (3 months) and old (24 months) mice. Data are mean ± SEM of 3–5 mice per group.DQuantification of mean number of TAF in hepatocytes as a function of the distance from a Ly6G labeled neutrophil. Data are from five mice aged 29 months of age; line indicates linear relationship between mean TAF and distance (linear regression and F test: R 2 = 0.35; P < 0.0001).ERepresentative micrograph showing hepatocyte from aged mouse containing a TAF located in close proximity to a neutrophil (labeled with Ly6G).FSchematic depicting experimental design.G% of neutrophil migration toward conditioned media (CM) obtained from young (early passage) and replicatively senescent human fibroblasts. Media alone and fMLP were used as negative and positive controls respectively. Data are mean ± SEM of three independent experiments.HSchematic depicting experimental design for panels I–K.IComparison between p16Ink4a mRNA levels of young (3 months old) and old INK-ATTAC mice (28–29 months old) treated with vehicle or AP20187. Data are mean ± SEM of n = 5–8 mice per age group.J% of TAF-positive hepatocytes in old INK-ATTAC mice (28–29 months old) treated with vehicle or AP20187. Data are mean ± SEM of n = 5–6 mice per group.KLiver neutrophils (Ly6G) in old INK-ATTAC mice (28–29 months old) treated with vehicle or AP20187. Data are mean ± SEM of n = 5–7 mice per group.LIntrahepatic leukocyte profiling by mass CyTOF. CyTOF was performed on young (6 months) and old (24 months) INK-ATTAC mice (n = 5 mice per group); 23 unique clusters of intrahepatic leukocytes were defined by a 24-cell surface marker panel using the Rphenograph clustering algorithm and were visualized on a tSNE plot (t-distributed stochastic neighbor embedding plot).M(left) tSNE plots comparing the distribution of young and old intrahepatic leukocytes. Red indicates high frequency categorization of cells to a cluster while blue indicates low frequency; (right) tSNE plots comparing p16Ink4a expression in young and old intrahepatic leukocytes. Red indicates high p16Ink4a expression while blue indicates low expression. Data are from n = 5 mice per group.N% of p16Ink4a-positive cells (detected by CyTOF) in different intrahepatic leukocyte populations isolated from young and old mice. Data are mean ± SEM of n = 5 mice per group.OSchematic depicting experimental design for panels P and Q. 3-month-old wild-type mice were injected with murine Cxcl1 for 7 days.PQuantification of liver neutrophils (Ly6G) in mice injected with vehicle or Cxcl1. Data are mean ± SEM of n = 7–10 mice per group.QQuantification of the % of TAF-positive hepatocytes in mice injected with vehicle or Cxcl1. Data are mean ± SEM of n = 6–7 mice per group.RCorrelation between TAF-positive hepatocytes and neutrophils (linear regression and F test: r 2 = 0.70; P = 0.0001). Data information: Statistical analysis was performed using one-way ANOVA (Holm–Sidak method) for multiple comparisons and two-tailed t-test for single comparisons. **P < 0.01; *P < 0.04.

Fig 3: Cell cycle analysis of CDKN2A variants.G1 percent for 45 CDKN2A variants (9 pathogenic variants, 7 benign variants, 29 variants of uncertain significance [VUSs]) in PANC-1 cells. Significant inhibition of G1 arrest function of CDKN2A by benchmark pathogenic variants and functionally deleterious or potentially functionally deleterious VUSs. *p < 0.01. For raw data in this figure, please refer to Figure 4 and Table 1. Figure 4—source data 1.Raw data in Figure 4 and Figure 4—figure supplement 5.

Fig 4: Neutrophils induce telomere dysfunction and senescence in human precision-cut liver slices (PCLS) ASchematic depicting experimental design.BRepresentative micrographs of neutrophil marker CD66B.CQuantification of neutrophil counts per field.DLactate dehydrogenase (LD) activity as a function of time in bioreactor.ERepresentative Immuno-FISH micrographs (using telomere-specific (CCCTAA) peptide nucleic acid probe and anti-?H2A.X antibody). Arrow indicates co-localization between telomeres and ?H2A.X.FMean number of TAF in hepatocytes.G% of TAF-positive hepatocytes.HNumber of p21-positive hepatocytes per field.INumber of p16Ink4a-positive hepatocytes per field. Data information: Data are mean ± SEM from liver biopsies obtained from seven male patients aged between 49 and 72 years. Neutrophils were isolated from healthy individuals aged between 21 and 30 years (with purity above 90%). Statistical analysis was performed using one-way ANOVA (Holm–Sidak method) for multiple comparisons and two-tailed t-test for single comparisons. **P < 0.01; *P < 0.04.

Fig 5: Rejuvenating effects of Box A in D-gal-induced and natural aging animal models. Box A-produced DNA gaps revitalized natural aging and D-gal-induced aging rats. Schematic diagrams for (A) D-gal-induced and natural aging rats intraperitoneally injected with Box A plasmid/Ca-P nanoparticle (100 µg of plasmid/kg rat body weight) once a week for 8 weeks compared to PC or p. F38G plasmid treatment. Serum AST levels of the D-gal aging model (n = 6) were determined before D-gal induction (baseline; rats were 3 months old), after 8 weeks of D-gal induction, and after Box A treatment (A, right panel). For the natural aging model, the levels of serum AST were measured before (the age of 28 m) and after Box A treatment (n = 6) (A, middle panel). Serum AST levels of 7 m, D-gal, and 30 m treated with PC or Box A plasmids: N = 4–6; 30 m with p.F38G: N = 3 (A, left panel). SA-ß-gal staining in rat liver sections quantification (B) is indicated (n = 3–4). (C) The number of DNA gaps (DNA-GAP number) were quantitated in rat WBCs (n = 4–6, p. F38G; n = 2). (D) Visceral fat (g) per 100 g of body weight (n = 5–7, p. F38G; n = 3). Morris water maze tests in D-gal-induced and natural aging models were assessed, including the acquisition training test (E and F, left panel) and the probe test (E and F, right panel), (n = 4–6, p. F38G; n = 3). For the natural aging model, (G) immunoblotting levels of ?H2A.X and ß-actin proteins in 30 m rat livers (n = 2–3) and (H) percentage of relative protein level of ?H2A.X, p16INK4A, and p21 (n = 3–6). The number of DNA gaps in rat WBCs of PC-treated 7 m rats (D) and the relative protein level in rat liver from the PC-treated group (H) were normalized to 100%. Paired t-test (A), data represent means ± SEM. *p = 0.05, **p = 0.01, ***p = 0.001, ****p = 0.0001 one-way ANOVA followed by post hoc analysis, unpaired t-test, and not significance (NS) at 30 m p. F38G versus PC