Treatment with β-interferon induces a senescent cell cycle arrest in normal fibroblasts. (A) Growth curves. Cell numbers were followed for 12 d in IMR90 cells treated with β-interferon or vehicle. (B) Cells were treated with β-interferon or vehicle for 12 d and then they were pulsed with BrdU for 3 h and fixed, and BrdU incorporation was measured by indirect immunofluorescence. Cells on low serum (LS) were used as a control for cell cycle arrest. Data were quantified by counting 50 cells in different fields in three independent experiments and analyzed with the Student's t test. (C) Cells were treated with β-interferon or vehicle for 12 d; then they were pulsed with [³H] thymidine for 24 h, and thymidine incorporation was measured by scintillation counting and analyzed with the Student's t test. (D) Cell cycle analysis of cells treated with β-interferon. Cells were treated for 12 d with β-interferon or vehicle, stained with propidium iodide, and analyzed by flow cytometry. The data represent a typical profile that was reproduce in three independent experiments. (E) Senescence associated β-galactosidase at indicated times. (F and G) Reversibility of the senescence phenotype. Cells were treated with interferon for 18 d or hydroxyurea (HU) for 12 d and then washed and incubated in normal medium. Cells were fixed and stained for SA-β-Gal 0, 3, and 6 d after withdrawal of β-interferon or HU. The total amount of SA-β-Gal–positive cells per field in three independent fields of view was counted. Values represent the mean and SD of three independent measurements. In comparison to the control cells the number of SA-β-Gal–positive cells in samples treated with β-IFN of HU was statistically significant (p ≤ 0.001).

Two signaling stages from β-interferon to p53. (A) Immunoblots for several p53 modifications and HDM2 in cells treated with β-interferon or vehicle for the indicated amounts of time. (B) p53 and p53 modifications after 4, 12, or 36 h of β-interferon treatment. (C) Luciferase reporter assays with the HDM2 promoter in normal human fibroblasts transfected with p53 or treated with β-interferon. Data were analyzed with the Student's t test.

β-Interferon induces the DNA damage-signaling pathway. (A) Immunoblots of phospho-p53^S15 and phospho-CHK2^T68 in cells with an empty vector or its derivative expressing a shRNAs against ATM. (B) ATM levels in cells with an empty vector or its derivative expressing a shRNAs against ATM. (C) Luciferase reporter assays with the HDM2 promoter in normal human fibroblasts infected with a control vector or anti-ATM shRNA and treated with β-interferon or vehicle. (D) Relative percent of SA-β-Gal–positive cells bearing an empty vector or its derivative expressing shRNAs against ATM after treatment with β-interferon or vehicle for 12 d. Control cells treated with β-interferon were taken as 100%. Differences between each group and the control and between cells with the shRNA and control vector were statistically assessed with the Student's t test. (E) Representative fields of immunofluorescence for H2AX foci in cells treated with β-interferon for the indicated times or H₂O₂. (F) Immunoblots for phospho-ATM^S1981.

Constitutively active ERK2 sensitizes normal fibroblasts to β-interferon. (A) Cell expressing constitutively activated ERK2, RasV12, or vector control were treated with 1000 U of β-interferon for 12 d and then fixed and stained for of SA-β-Gal. Data were analyzed by the Student's t test: for (V; ERK2) p = 0.006 and for (V; RasV12) p = 0.003. (B) Growth curves. Cell numbers were followed for 12 d in IMR90 cells bearing a vector control or its derivative expressing constitutively active ERK2 and treated with β-interferon or vehicle. (C) Immunoblot for phospho-STAT1 in IMR90 cells with empty vector or its derivative expressing activated ERK2 and treated with β-interferon or vehicle for 12 d (Akiyama et al., 1999). (D) Luciferase reporter assays with the β-interferon promoter in normal human fibroblasts expressing constitutively active ERK2 or a vector control and treated with β-interferon or vehicle. Data were analyzed with the Student's t test.

Status of several members of the RB and p53 pathways in cells treated with β-interferon. (A) Immunoblots for RB and various members of the RB pathway. (B) Immunoblots for p53, its modifications, and targets. C+, positive control. (C) RT-PCR for PIG-3, p21, and β-actin in cells treated for 6 or 12 d with β-interferon, or cells treated with adryamicin (Ad).

Role of p53 in β-interferon–induced senescence. (A) SA-β-Gal: IMR90 cells infected with the retroviral vector pLPC or its derivative expressing p53 were treated with β-interferon or vehicle. Data were analyzed with the Student's t test. (B) Immunoblots from cells as above collected 6 d after retroviral infection. (C) SA-β-Gal: IMR90 cells were infected with the retroviral vector LXSN or its derivative expressing E6. Then the cells were treated with β-interferon or vehicle for 12 d and fixed for SA-β-Gal staining. (D) Immunoblots for cell cycle regulators and the p53 pathway in E6-expressing cells and control after treatment with β-interferon or vehicle for 12 d. (E) Colony formation assay at 39°C of TSP cells previously treated with α- or β-interferon for 8 d at 39°C or for 2, 4, or 8 d at 32°C. TSP cells expressing oncogenic ras were used as a positive control for induction of a permanent cell cycle arrest after activation of p53 at 32°C. Colonies were counted in three independent experiments, and the data were analyzed for significance with the Student's t test.

Interferon increases mitochondrial mass and accumulation of reactive oxygen species. (A) FACS measurement of mitochondrial mass with MitoFluor-Green in cells treated for 2 d with β-interferon or vehicle. (B) Immunofluorescence of cells labeled with MitoTracker-Red. (C) Immunofluorescence detection of cells labeled by the ROS sensor H₂DCFDA. (D) FACS detection of ROS with the H₂DCFDA probe. Control cells or cells treated with interferon for 12 d were also treated with 4 mM N-acetyl l-cysteine for 12 d. (E) Immunoblots for phospho-CHK2^Thr68 in cells as in C. (F) Immunofluorescence of H2AX foci in cells treated as in C. Cells with at least one foci were scored as positive and the mean and SD of three independent experiments is indicated at the bottom right of each panel. Cells with multiple foci were only seen in interferon-treated cells.

Proposed model for interferon-induced senescence. The immediate effects of interferon treatment are a reversible growth arrest, mitochondrial changes, and p53 modifications (acetylation at K320 and dephosphorylation at serine 392). These changes do not commit the cells to an irreversible withdrawal from the cycling pool. However, chronic stimulation results in an increase in ROS production and activation of the DNA damage signaling pathway. The ATM kinase is a central regulator of the DNA damage response induced by interferon and it activates p53 by promoting its phosphorylation at serine 15 and inducing its transactivation potential. Activation of p53 is then critical to trigger a senescent cell cycle arrest in normal fibroblasts.