Differential p53 Engagement in Response to Oxidative and Oncogenic Stresses in Fanconi Anemia Mice

Research on Fanconi anemia (FA) has recently generated great interest because the disease serves as an excellent model for hematopoietic failure and leukemic evolution. FA is a genetic disorder associated with bone marrow (BM) failure and cancers, particularly leukemia (1–3). FA is genetically heterogeneous, with 13 complementation groups identified thus far. The genes encoding the groups A (FANCA), B (FANCB), C (FANCC), D1 (FANCD1/BRCA2), D2 (FANCD2), E (FANCE), F (FANCF), G (FANCG), -I (FANCI/KIAA1794), J (FANCJ/BRIP1), L (FANCL), M (FANCM), and N (FANCN/PALB2) have been cloned (4–18). The biological function of FA proteins has been the subject of intense investigation in recent years.

Because cells deficient for FA genes are hypersensitive to DNA cross-linking agents, such as mitomycin C and diepoxybutane, it has been proposed that the FA proteins may be involved in the sensing and/or repair of DNA interstrand cross-links. This model is supported by evidence that shows that an intact FA nuclear complex is required for biochemical modification in the form of monoubiquitination of FANCD2 and FANCI as mutations in any one of the FA proteins that disrupt the complex prevent the activation of FANCD2 and FANCI, leading to cross-linker hypersensitivity and the characteristic broken and radial chromosome formation in FA cells (3, 18). The common damage to DNA in vivo is oxidative stress, and ample evidence has suggested that FA cells are in an in vivo prooxidant state (19) and that the FA proteins may play important roles in cellular responses to oxidative stress. For example, the FANCC protein has been found to interact with NADPH cytochrome P450 reductase and glutathione S-transferase P1-1 (20, 21), two enzymes involved in either triggering or detoxifying reactive intermediates, including reactive oxygen species. In addition, mice with combined deficiencies of the antioxidative enzyme, Cu/Zn superoxide dismutase, and the Fancc genes showed a defective hematopoiesis (22). Another FA protein, FANCG, interacts with cytochrome P450 2E1 (23) and mitochondrial peroxiredoxin-3 (24), suggesting a possible role of FANCG in protection against oxidative DNA damage. Recently, Saadatzadeh and colleagues (25) showed that oxidant hypersensitivity of Fancc^−/− cells was due to an altered redox regulation and hyperactivation of the serine-threonine kinase apoptosis signal-regulating kinase 1, an important kinase involved in oxidant-induced apoptosis. Moreover, oxidative stress induces complex formation by two major FA proteins, FANCA and FANCG (26). These observations corroborate a critical role for oxidative stress in FA phenotype and disease progression.

Cells from patients with FA gene mutations have high predisposition to leukemia and other cancers. However, little is known about whether these mutant cells have high susceptibility to oncogenic transformation. In response to oncogenic activation, normal cells induce genetically encoded programs that prevent deregulated proliferation and, thus, protect multicellular organisms from cancer progression. Two such programs induced by oncogenic activation are apoptosis and senescence, which are normally triggered by DNA damage or other stresses. Studies have shown that overexpression of antiapoptotic proteins, such as Bcl-2, or deletion of apoptosis-associated and senescence-associated proteins, such as p53, accelerates oncogene-induced tumorigenesis (27, 28). The theory that has risen from these findings is that oncogene-driven proliferation must be associated with inhibition of apoptosis and senescence to allow malignant outgrowth.

The tumor suppressor p53 is a key transcription factor that activates vital damage containment procedures to restrict aberrant cell growth in response to DNA damage, oncogene activation, and loss of normal cell contacts (29, 30). p53 restricts cellular growth by inducing senescence, cell cycle arrest, or apoptosis (31). Thus, p53 plays a major role in the prevention of cancer. Consistent with this, emerging evidence suggests that p53 deficiency may increase cancer development in patients with FA and FA mice. For example, studies have indicated a higher proportion of human papillomavirus-positive squamous cell carcinomas in patients with FA than in healthy controls, indicating that loss of functional p53 facilitates tumor development (32, 33). Mice deficient for Fancd1 or Fancd2 have been shown to have accelerated tumor development in Trp53-deficient background (34, 35). In addition, Fancc deficiency accelerates the development of certain blood and solid tumors in mice heterozygous at Trp53 (36). Moreover, these studies found that the FA proteins and p53 cooperate in apoptosis and cell cycle checkpoint control after DNA damage (35, 37, 38). In this study, we show that p53 is engaged differentially in response to oxidative and oncogenic stresses in FA cells in vitro and in vivo.

Mice. Wild-type (WT), Fanca^−/− mice and double knockout mice (p53^−/−; Fanca^−/−) were generated by interbreeding the heterozygous Fanca^+/− and p53^+/− mice. The genetic background of the mice was C57BL/6 mice, which were used at ∼6 to 10 wk of age. All experimental procedures conducted in this study were approved by the institutional animal care and use committee of Cincinnati Children's Hospital Medical Center.

Isolation of BM lineage-depleted cells and differentiation assay. The femora and tibiae were harvested from the mice immediately after their sacrifice with CO₂. BM cells were flushed from bones into Iscove's modified Dulbecco's medium (IMDM; Invitrogen) containing 10% FCS, using a 21-guage needle and syringe. Low-density BM mononuclear cells (BMMNC) were separated by Ficoll Hypaque density gradient (Sigma) and washed with IMDM. BMMNC were depleted of lineage-committed cells using a lineage cell depletion kit (Miltenyi Biotec, Inc.) in accordance with the manufacturer's instruction. For myeloid and lymphoid differentiation, BM Lin− cells were cultured with various cytokine combinations known to support myeloid [100 ng/mL stem cell factor (SCF), 10 ng/mL granulocyte-monocyte colony-stimulating factor (GM-CSF)] or lymphoid [100 ng/mL SCF, 10 ng/mL each of interleukin-3 (IL-3) and IL-7] differentiation. After 5 d, cells were stained with antibodies against myeloid markers Gr-1 and Mac-1 or lymphoid markers B220 and CD3e, followed by flow cytometric analysis using a FACSCalibur (Becton Dickinson).

Mouse embryonic fibroblast isolation and culture. Mouse embryonic fibroblasts (MEF) were prepared from day 13.5 embryo derived from crosses of p53^+/− with Fanca^+/− mice. The head and the red organs were removed, and the torso was minced and dispersed with 0.25% trypsin-EDTA by using 20-gauge needle. Single-cell suspension obtained was cultured in DMEM containing 10% fetal bovine serum, glutamine, MEM nonessential amino acid, penicillin/streptomycin, and gentamicin under 5% CO₂ for 3 d until the genotypes were confirmed. Cells were then trypsinized and expanded for experimental use. MEFs were further maintained on the basis of 3T3 protocol.

Construction of retroviral expression vectors. The full-length human FANCA cDNA was subcloned into the NotI site of retroviral vector MIEG3 to create MIEG3-FANCA. The WT mouse p53 (mp53wt), a temperature-sensitive mutant p53 (tsp53^V143A), and H-ras V12 cloned in MIEG3 were kindly provided by Dr. Yi Zheng (Cincinnati Children's Medical Center). The plasmids (10 μg each) were used to produce retroviral supernatant.

Retroviral-mediated gene transfer. Helper virus–free Phoenix packaging cells (kindly provided by Dr. Gary Nolan, Stanford University) were plated at the density of 3 × 10⁶ in 10-cm tissue culture dish and incubated for 24 h. Cells were then transfected by calcium phosphate precipitation with 10 μg of retroviral plasmids (15 h at 37°C). Retroviral supernatant was collected at different time points (24, 36, and 48 h), respectively. After transfection, the virus-containing medium was filtered (0.45 μm filter, Millipore) and aliquoted at −80°C. WT, Fanca^−/−, p53^−/−;Fanca^+/+, and p53^−/−;Fanca^−/− MEFs were plated at the density of 8 × 10⁵ cells per 10-cm dish and incubated overnight at 37°C. For infections, culture medium was replaced by appropriate retroviral supernatant in the presence of 4 μg/mL polybrene (Sigma) for 12 h. Infection process was repeated twice. At 24 h after infection, EGFP-positive cells were used for different assays, as indicated in Results.

Colony and cell proliferation assay. Low-density BMMNCs were plated in a 35-mm tissue culture dish in 4 mL of semisolid medium containing 3 mL of MethoCult M 3134 (Stem Cell Technologies) and the following growth factors: 100 ng/mL SCF, 10 ng/mL IL-3, 100 ng/mL GM-CSF, and 4 units/mL erythropoietin (Peprotech). On day 10 after plating, erythroid and myeloid colonies were enumerated. Hematopoietic clonal growth results were expressed as means (of triplicate plates) ± SD of three experiments. For proliferation, cells were cultured for 24 h in normal growth medium supplemented with 10 μmol/L BrdUrd (Sigma), harvested, and fixed in 70% ethanol. BrdUrd-labeled cells fixed in 70% ethanol were treated with 2 N HCl (20 min at room temperature), followed by addition of two volumes of 0.1 mol/L sodium borate (pH 8.5). The cells were incubated with an anti-BrdUrd mouse monoclonal antibody, washed, and incubated with FITC-conjugated antimouse antibody. Cells were counterstained overnight with 5 μg/mL of propidium iodide containing 40 μg/mL of RNase. The stained cells were analyzed by flow cytometry.

Apoptosis assay and cell cycle analysis. Cells were stained with Annexin V and 7AAD using BD ApoAlert Annexin V kit (BD Pharmingen) in accordance with the manufacturer's instruction. Apoptosis was analyzed by quantification of Annexin V–positive cell population by flow cytometry. For cell cycle analysis, cells were fixed with 0.25% formaldehyde in PBS and permeabilized with 0.3% Nonidet P-40. Cells were then stained with propidium iodide containing 1 mg/mL RNase A, followed by fluorescence-activated cell sorting (FACS) analysis for G₀-G₁, S, and G₂-M populations using a FACS caliber.

Immunocytochemistry. Cells were plated on 18-mm diameter glass coverslips and allowed to recover for 24 h before treatment with H₂O₂. Treated cells were fixed with 4% formaldehyde in PBS for 20 min at room temperature. Cells were then washed with PBS and permeabilized with 0.2% Triton X-100 in 3% bovine serum albumin (BSA)/PBS for 10 min at room temperature. The cells were then washed with PBS and blocked with 10% BSA/PBS solution for 1 h at room temperature. Cells were then incubated with primary antibodies against p53, p53^Ser20, or p21^WAF1 (Cell Signaling Technology) in 10% BSA/PBS solution at room temperature for 2 h. After extensive washes, cells were incubated with Rhodamine Red X–conjugated goat anti-rabbit IgG or Rhodamine Red X–conjugated goat anti-mouse IgG (Jackson Immuno Research). DNA was then labeled with 4′,6-diamidino-2-phenylindole(Sigma). Slides were finally mounted in a mounting medium (Vector) and were visualized under Carl Zeiss invert Axiovert 200M microscope with OpenLab 4.0.3 software (Improvision).

Temperature shifts and cell proliferation analysis. For the temperature shift experiment, 1 × 10⁵ tsp53 transduced p53^−/−;Fanca^+/+ and p53^−/−;Fanca^−/− MEF cells were cultured in triplicate in six-well tissue culture plates. Plates were incubated at 37°C or 32°C. Every third day, cells were trypsinized and counted with a hemacytometer, and the same number of cells was replated in six-well tissue culture plates.

Senescence-associated β-galactosidase. Senescence-associated β-galactosidase (SA-β-gal) activity was determined using a SA-β-gal staining kit (Cell Signaling Technology) according to the manufacturer's instructions. Briefly, cells were washed once with PBS and fixed with 0.5% glutaraldehyde. Cells were further washed with PBS and stained with X-gal solution (1 mg/mL X-gal) overnight at 37°C. The percentage of positive-stained cells was quantified under a phase-contrast microscope.

Immunohistochemistry. Spleen, thymus, and bone were fixed with formalin and embedded in paraffin for sectioning. After deparaffinization and hydration, sections were blocked with 10% serum in PBS for 30 min and then incubated with 1:50 diluted anti-p53 or 1:50 diluted anti-p21^WAF1 (all from Santa Cruz Technologies) primary antibodies overnight at 4°C, followed by three washing with PBS at 5-min interval. Detection of p53 and p21^WAF1 staining was performed using the biotin peroxidase complex method (Vectastain ABC kits, Vector Laboratories). Color was developed with diaminobenzidie tetrahydrochloride, and nuclei were stained with hematoxylin.

Preparation of cell extracts and immunoblotting. To prepare cell protein, single-cell suspension was prepared from BMMNCs, spleenocytes, or trypsinized MEFs, washed with ice-cold PBS, and resuspended in ice-cold lysis buffer containing 50 mmol/L Tris-HCL (pH 7.4), 0.1% NP40, and 1mol/L NaCl supplemented with protease and phosphatase inhibitors [10 μg/mL aprotinin, 25 μg/mL leupeptin, 10 μg/mL pepstatin A, 2 mmol/L phenylmethylsulfonyl fluoride, 0.1mol/L NaP₂O₄, 25 mmol/L NaF, and 2 mmol/L sodium orthovandate] for 30 min on ice. Cell debris was removed from the lysates by centrifuging them at 14,000 rpm for 30 min. Protein concentration was quantified by using Bio-Rad reagent. Cell lysates (100 μg) were resolved on (12%) SDS-PAGE and transferred onto nitrocellulose membranes. Immunoblots were then incubated with primary antibodies specific for p53, p53^Ser20, and p21^WAF1 (Santa Cruz Biotechnologies), γ H₂AX (Upstate Biotechnology), and phosphorylated pRB, phosphorylated p38, phosphorylated extracellular signal-regulated kinase 1/2 (ERK1/2), and the pan kinases p38 and ERK1/2 (all from Cell Signaling) or β-actin (Sigma) for 12 to 16 h at 4°C. Signals were revealed after incubation with antimouse or antirabbit secondary antibodies.

Statistics. Data were analyzed statistically using a Student's t test. The level of the statistical significance stated in the text was based on the P values. P < 0.05 was considered statistically significant.

Hematopoietic cells and MEFs from Fanca^−/− mice are hypersensitive to oxidants. Because FA patients have severe defects in hematopoietic function and because it is not clear whether the FA proteins function in oxidative stress response, we used mice deficient for the FA complementation group A (Fanca) gene to examine the sensitivity of primary FA cells to oxidative stress. We first treated BMMNCs from Fanca^−/− or their WT littermates with increasing concentrations of the oxidant hydrogen peroxide (H₂O₂) and analyzed cell proliferation using the BrdUrd incorporation assay. We observed that treated Fanca^−/− cells proliferated much slower than their WT counterparts (Fig. 1A,, left). Because FA hematopoietic failure is a direct consequence of the defect in the hematopoietic stem cell/progenitor (HSC/P) compartment, we tested whether the FA HSC/P cells were hypersensitive to oxidative damage. Indeed, the number of colonies formed by Fanca^−/− HSC/P cells was markedly decreased with 100 μmol/L of H₂O₂, whereas WT HSC/P cells showed little effect with H₂O₂ treatment (Fig. 1A , right).

We also established MEFs from Fanca^−/− or WT mice and used this well-established cell type as a cellular model system for the evaluation of cellular and molecular alterations in oxidative response in FA mutant cells. Similar to BMMNCs, the growth of Fanca^−/− MEFs was severely inhibited even at low concentrations of H₂O₂, which had little effect on the proliferation of WT cells (Fig. 1B). In addition, a sublethal dose (10 μmol/L) of H₂O₂ induced premature senescence, as tested by SA-β-gal staining in Fanca^−/− MEFs (Fig. 1C). Together, these results indicate that loss of FA function results in cellular hypersensitivity to oxidants.

Oxidative stress induces premature senescence by activating several major signaling pathways, including the p16^INK4A/pRB and ADP ribosylation factor/p53 pathways (39). Oncogene Ras has also been shown to induce cell growth arrest after oxidative injury by activating the mitogen-activated protein kinase (MAPK) pathway (40). We, therefore, examined the expression and activities of proteins and enzymes in the Ras, p53, and pRB pathways. We did not find significantly elevated Ha-Ras expression or the downstream activation of the MAPK kinases ERK and p38 in senescent Fanca^−/− cells treated with 10 μmol/L H₂O₂ (Fig. 1D,, left), suggesting that the Ras pathway was not responsible for the 10 μmol/L H₂O₂-induced premature senescence in Fanca^−/− cells. There was only a marginal increase in the expression of p16^INK4A and p19^Arf, which was correlated with a decrease in phosphorylated pRB, in response to 10 μmol/L H₂O₂ in both WT and Fanca^−/− cells as determined by reverse transcription–PCR (RT-PCR) or Western blotting, respectively (Fig. 1D,, middle and right). However, the activity of p53, as reflected by its phosphorylation at Ser²⁰ (p53^Ser20, which is a specific indicator of oxidative DNA damage; ref. 41) and p21^WAF1 expression, was significantly increased in senescent Fanca^−/− cells treated with 10 μmol/L H₂O₂ (Fig. 1D , right). These results suggest that the p53 pathway may play a predominant role in H₂O₂-induced premature senescence in Fanca^−/− cells.

H₂O₂ induces G₂-M cell cycle arrest and DNA double-strand breaks in Fanca^−/− BM cells. One hallmark of FA cells in response to genotoxic stress is increased retardation of the cells at the G₂-M phase of the cell cycle (1–3). We thus examined whether H₂O₂ induced abnormal G₂-M cell cycle arrest in primary FA BM cells. Indeed, H₂O₂-treated Fanca^−/− BM cells displayed a significant increase in the percentage of cells in G₂-M compared with H₂O₂-treated WT cells (Fig. 2A). Because G₂ cell cycle arrest is a major cellular response to DNA damage preceding the decision to repair or enter mitosis and because FA cells are defective in the repair of DNA double-strand breaks (DSB; refs. 1–3), we determined whether H₂O₂-treated Fanca^−/− BM cells accumulated high levels of DNA damage by examining the well-established DSB marker, phosphorylated histone H2AX (γH2AX; ref. 42). As shown in Fig. 2B, there was no significant accumulation of DSB in WT BM cells, indicating that either the H₂O₂ treatments have little effect on WT BM cells or DNA damage induced by H₂O₂ had been mostly repaired in these WT BM cells. Strikingly, H₂O₂ induced significantly more DSB in Fanca^−/− BM cells than in untreated Fanca^−/− or H₂O₂-treated WT BM cells (Fig. 2B). Thus, like mitomycin C, H₂O₂ induces abnormal G₂-M cell cycle arrest in primary FA BM cells, possibly in a DNA damage-dependent manner.

H₂O₂ overactivates DNA damage response in Fanca^−/− cells in vitro and in vivo. Reasoning that the abnormal G₂-M accumulation induced by H₂O₂ in Fanca^−/− BM cells might have resulted from the prolonged activation of the oxidative DNA damage and DNA DSB checkpoints, we examined the expression of several oxidative damage response molecules p53, p53^Ser20, and p21^WAF1. We found increased expression of both the total and activated forms (p53^Ser20) of p53 in H₂O₂-treated Fanca^−/− cells compared with WT cells (Fig. 3A). Consistent with its activation, p21^WAF1, the target of p53, was up-regulated in H₂O₂-treated Fanca^−/− cells.

Next, we examined oxidative DNA damage response in vivo. For this purpose, we injected (i.v.) H₂O₂ into WT and Fanca^−/− mice and examined the expression of p53, p53^Ser20, and p21^WAF1. Western blot analysis showed that all these DNA damage response molecules were elevated in response to H₂O₂ treatment in BM and spleen harvested from Fanca^−/− mice (Fig. 3B). Similar results were noticed in the paraffin-embedded tissue sections of BM, spleen, and thymus prepared from H₂O₂-treated mice (Fig. 3C). These results suggested that loss of FA function causes strong oxidative DNA damage response, possibly through up-regulation of the p53 pathway.

Kinetics of H₂O₂-induced DNA damage response in Fanca^−/− mice. The observation that H₂O₂ induced strong DNA damage response in Fanca^−/− cells and tissues prompted us to study the kinetics of H₂O₂-mediated p53 signaling in Fanca^−/− mice. We treated (through tail vein injection) WT and Fanca^−/− mice with a single dose (250 μmol/kg) of H₂O₂ and sacrificed the injected animals 2, 4, 6, 12, and 24 hours after H₂O₂ injection. The kinetics of DNA damage response in the BM and spleen of the treated mice was analyzed for the expression of p53^Ser20 and γH₂AX. As shown in Fig. 4, p53 activation (in the form of p53^Ser20) was short-lived in both marrow (Fig. 4A) and spleen (Fig. 4B) of WT mice, which peaked at 4 hours but was decayed rapidly to background level by 6 hours after H₂O₂ treatment. In Fanca^−/− mice, in contrast, high level of p53 activation was evident at 2 hours and persisted for 24 hours postinjection. Although the expression kinetics of γH₂AX, a well-established DSB marker (42), varied somewhat among the mouse tissues, Fanca^−/− mice showed prolonged expression of the DNA damage marker compared with WT mice, in which γH₂AX expression was undetectable by 12 to 24 hours after injection. Therefore, DNA damage response initiated by oxidative stress is persistent in Fanca^−/− mouse hematopoietic tissues in vivo.

The observation that H₂O₂-induced expression of the DSB marker γH2AX seemed to be more pronounced and sustained in the BM than in spleen raised the question as to whether there was a lineage-specific effect on oxidative DNA damage in Fanca^−/− mice. To address this, we determined H₂O₂-induced DSBs in lymphoid and myeloid cells (Fig. 4C) from the BM of Fanca^−/− mice. Whereas increased and prolonged expression of γH2AX was seen in Fanca^−/− cells compared with WT cells, no significant difference in H₂O₂-induced DSBs was evident between lymphoid and myeloid cells (Fig. 4D).

The dependency and persistence of p53-mediated response to oxidative and oncogenic stresses in Fanca^−/− cells. To explore the requirement for p53 activity and potential FA-p53 interaction in oxidative DNA damage response, we generated Fanca^−/− mice with the deletion of the Trp53 gene, isolated primary MEFs from Fanca^+/+;Trp53^−/− and Fanca^−/−;Trp53^−/− mice, and expressed in these cells the temperature-sensitive p53 mutant tsp53^V143A, which can be reverted to functional state at the permissive temperature of 32°C (43). The functional reversibility of the mutant tsp53^V143A was verified by Western analysis, demonstrating that H₂O₂ induced increased p21^WAF1 expression in all the tsp53^V143A-expressing cells cultured at 32°C (Fig. 5A). A prolonged time course analysis of cell growth indicated that, at nonpermissive temperature (37°C), loss of p53 function almost completely abolished growth response to H₂O₂ in cells with or without the Fanca protein (Fig. 5B), suggesting that H₂O₂-induced damage response requires a functional p53. Indeed, when the function of p53 was restored at 32°C, H₂O₂-induced growth inhibition was clearly evident in both Fanca^+/+ and Fanca^−/− cells (Fig. 5C). However, we observed differential responses to H₂O₂-induced stress in cells with different Fanca status. In Fanca^+/+ cells, the p53 response to oxidative stress seemed to be relatively short (3–6 days) after the insult; whereas Fanca^−/− cells underwent prolonged growth inhibition (Fig. 5C). Interestingly, TUNEL assay indicated that p53 deficiency increased H₂O₂-induced apoptosis in Fanca^−/− cells, which could be inhibited by functional tsp53^V143A (Fig. 5D). Thus, although p53 caused prolonged proliferative suppression in Fanca^−/− cells, a functional p53 seemed to be required for the survival of the mutant cells under oxidative stress.

We next used the same cell system to explore the requirement for p53 function in response to oncogenic stress, specifically activated ras, which is a well-studied p53-antagonizing oncogene (44). We infected tsp53^V143A-expressing Fanca^+/+;Trp53^−/−and Fanca^−/−;Trp53^−/− MEFs with an H-Ras V12 retrovirus and monitored cell proliferation at 32°C or 37°C for a period of 15 days. Growth kinetics showed that at 37°C, wherein p53 was inactivated, all cell lines proliferated well regardless of the Fanca genotype (Fig. 6A). When the cells were cultured at 32°C, wherein p53 was functional, Fanca^+/+ cells exhibited persistent response to activated ras-induced growth arrest (Fig. 6A). Interestingly, Fanca^−/− cells initially responded to activated ras-triggered growth arrest and then started to proliferate like p53-null cells after day 6. These arrested cells expressed SA-β-gal and seemed morphologically senescent (Fig. 6B).

The interesting observation that Fanca-deficient cells failed to respond to ras-induced growth arrest after 6 days of culture at permissive temperature (32°C) suggested that Fanca^−/− cells might have lost their ability to engage p53 during constitutive ras activation. To test this hypothesis, we cultured H-Ras V12-transduced tsp53^V143A-expressing cells at 37°C for either 3 or 12 days and then restored p53 function by shifting the cultures to 32°C. Cell proliferation and induction of p21^WAF1 were analyzed by BrdUrd incorporation and Western blot, respectively. Both cell lines shifting to 32°C at day 3 underwent approximately same level of growth arrest with few cells in the S-phase population (Fig. 6C), which was accompanied by strong induction of p21^WAF1 (Fig. 6D). Whereas p53 restoration at day 12 could still triggered immediate growth arrest in Fanca^+/+;Trp53^−/− cells; Fanca^−/−;Trp53^−/− cells showed lack of activated ras-induced growth arrest and p21^WAF1 expression (Fig. 6C and D). Therefore, loss of Fanca impairs p53 function during prolonged ras activation.

Cellular stresses, most significantly oxidative DNA damage and oncogene insults, have been shown to activate p53 (45). Given that p53 activation can lead to cell growth arrest or apoptosis, cancer cells, in particular, must have developed strategies to regulate p53 activity. In addition to the p53-negative regulator Mdm2, several p53-interacting proteins have been shown to regulate p53 activity. The interaction of p53 with other cellular proteins is known to influence the activity of the tumor suppressor (46). Our finding that the Fanca protein coordinates with p53 to regulate oxidative and oncogenic stress responses supports the notion that a fine control of p53 activity is fundamentally important for cellular homeostasis and cancer prevention.

Patients with FA have an increased susceptibility to cancer, and cells from FA patients are deficient in repair of DNA damage induced by certain genotoxic agents. This suggests that the FA proteins may interplay with other tumor suppressor pathways that are involved in DNA damage repair and oncogenic signaling. Encouraged by recent reports that p53 deficiency increases cancer development in patients with FA and FA knockout mice (32–36), we formally tested the hypothesis that the FA proteins may functionally interact with the p53-activating signals in response to oxidative and oncogenic stresses. Our results suggest that the major FA protein, Fanca, may coordinate with p53 in regulation of oxidative and oncogenic stress responses. This notion is supported as follows: (a) the hypersensitive response to oxidative stress in tissues in vivo and cells in vitro derived from Fanca^−/− mice is correlated with a persistent overactivation of the tumor suppressor p53; (ii) the loss of Fanca function causes prolonged oxidative DNA damage response through up-regulation of the p53 pathway; and (c) the functional status of p53 dictates the kinetics and persistence of response to oxidative and oncogenic stresses in these FA cells.

The mechanistic link between p53 signaling and FA has not been well defined. The involvement of p53 in FA pathophysiology has been highlighted by recent studies that show accelerated tumor development in Fancd1, Fancd2, or Fancc mice, also deficient for the Trp53 gene (34–36). Furthermore, developmental defects and increased apoptosis in Fancd2-deficient zebrafish could be corrected by p53 knockdown, suggesting that p53-dependent apoptosis may be an underlying mechanism for the developmental defect in the Fancd2^−/− fish (38). The current study establishes the requirement for p53 signaling in oxidative and oncogenic stress responses in Fanca-deficient cells. We have used primary MEFs from Fanca^+/+;Trp53^−/− and Fanca^−/−;Trp53^−/− mice combined with a functionally switchable p53 mutant to define the kinetics, dependence, and persistence of p53-mediated response to oxidative and oncogenic stresses in FA cells. One important finding of our study is that we observed different pattern of p53-dependent response to H₂O₂-induced and deregulated oncogene-induced stress in cells with different Fanca status. For instance, restoration of p53 function elicited a transient response to oxidative stress in Fanca^+/+ cells, whereas Fanca^−/− cells underwent prolonged p53-dependent growth inhibition (Fig. 5). This persistent p53 engagement may be due to the fact that Fanca-deficient cells accumulated high levels of oxidative DNA damage as a consequence of impairment in DNA damage repair. We previously reported that BM progenitor cells from Fancc^−/− mice contained high levels of oxidative DNA damage and exhibited persistent DNA damage response (47). Thus, the prolonged p53 response in Fanca^−/− cells might have been induced by the unrepaired DNA damage, leading to persistent DNA damage response.

Another interesting finding in the current study is that, although activation of p53 by oxidative stress causes increased proliferative suppression in Fanca^−/− cells compared with WT cells (Fig. 1), studies with the Fanca^−/−;Trp53^−/− double knockout cells indicated that p53 deficiency increased H₂O₂-induced apoptosis in Fanca^−/− cells, which could be inhibited by functional tsp53^V143A (Fig. 5). Thus, a functional p53 seems to be required for the survival of the mutant cells under oxidative stress. Circumstantial evidence indicates that the selectivity of p53 to regulate the prosurvival or proapoptotic signaling is affected by the growth environment, the type of stress used, the cellular context, and differential expression of the target genes (48, 49). In response to stress, p53 induces cell growth arrest, thus preventing the damaged cells from further replication (45, 50). More importantly, p53 can activate cell cycle checkpoint in cells with damaged DNA, which allows cells to repair the damage before reentering cell cycle. Indeed, it has been reported that p53-null cells are more sensitive to mitotic catastrophe induced by genotoxic drug (51, 52). Because H₂O₂-induced arrest in Fanca^−/− cells with a functional p53 occurs at the G₂-M cell cycle checkpoint, inactivation of p53 would lead to the impairment of the checkpoint. Consequently, failure of the damaged Fanca^−/− cells to arrest at G₂ would result in entry into mitosis and potential death through mitotic catastrophe and apoptosis.

Patients with FA have high predisposition to leukemia and other cancers. Whether FA cells have high susceptibility to oncogenic transformation is not known. We have used an in vitro system in which a functional p53 could be restored in Fanca^−/−;Trp53^−/− cells to investigate the dynamic interplay between Fanca and p53 in response to oncogenic stress. Whereas WT cells exhibit prolonged response to oncogene activation, the p53-activating signals induced by oncogenic ras were short-lived in Fanca^−/− cells. Indeed, when p53 function was restored 12 days after ras activation, Fanca^−/−;Trp53^−/− cells showed lack of activated ras-induced growth arrest and p21^WAF1 expression (Fig. 6). This suggests that a functional Fanca may be required for the cell to engage p53 during constitutive ras activation. In contrast to oxidative DNA damage, which triggers p53 activation only transiently, activated Ras elicits persistent p53-activating signals in WT cells. Furthermore, our results show that the Fanca function is required to maintain p53 response to Ras-induced growth arrest, as restoration of p53 function after 12 days of constitutive ras activation failed to suppress proliferation of cells deficient in Fanca.

In conclusion, our results derived from both in vivo and in vitro studies using the restorable p53 model shed new light on the potential interplay between p53 and the FA pathway during cellular response to oxidative and oncogenic stresses. Our studies also define the important kinetic difference between WT and FA cells during prolonged response to oxidative DNA damage and activated oncogene signals. Additional studies into functional interaction between the p53 and FA pathways in DNA damage and oncogenic stress response may aid us in better understanding on how cells can bypass the normal checkpoints and continue to proliferate in the presence of damaged DNA and oncogenic activation. In the context of FA, new insights on the role of the FA proteins in oxidative DNA damage response/repair and oncogene activation can suggest new pathways and proteins to target for therapeutic prevention of cancer progression of the disease.

No potential conflicts of interest were disclosed.

Grant support: NIH grants R01 CA109641 and R01 HL076712. Q. Pang is supported by a Leukemia and Lymphoma Scholar award.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Dr. Markus Grompe (Oregon Health and Science University) and Dr. Madeleine Carreau (Laval University) for the Fanca^+/− mice, Dr. Yi Zheng (Cincinnati Children's Hospital Medical Center) for the tsp53^V143A and H-Ras V12 retroviral vectors, Dr. Xiaoling Zhang for technical assistance and discussion, and the Vector Core of the Cincinnati Children's Research Foundation (Cincinnati Children's Hospital Medical Center) for the preparation of retroviruses.