The central nervous system-restricted transcription factor Olig2 opposes p53 responses to genotoxic damage in neural progenitors and malignant glioma

Abstract

High-grade gliomas are notoriously insensitive to radiation and genotoxic drugs. Paradoxically, the p53 gene is structurally intact in the majority of these tumors. Resistance to genotoxic modalities in p53-positive gliomas is generally attributed to attenuation of p53 functions by mutations of other components within the p53 signaling axis, such as p14(Arf), MDM2, and ATM, but this explanation is not entirely satisfactory. We show here that the central nervous system (CNS)-restricted transcription factor Olig2 affects a key posttranslational modification of p53 in both normal and malignant neural progenitors and thereby antagonizes the interaction of p53 with promoter elements of multiple target genes. In the absence of Olig2 function, even attenuated levels of p53 are adequate for biological responses to genotoxic damage.

Figure 1. Olig2 promotes survival in irradiated neural progenitor cells

(A) Differential response of Olig2^+/+ and Olig2^−/− neural progenitors to ionizing radiation. Secondary neurosphere assays were counted at day 5 following treatment of Olig2^+/+ or Olig2^−/− cells with 2, 4 or 8 Gy of ionizing radiation. Scale bars = 100 µm. (B) Quantitation of data in panel A. The bars in the histogram represent the percentage of secondary neurospheres formed in irradiated samples relative to the untreated control samples. (C) Quantitation of percentage of viable cells in the irradiated samples relative to untreated control in the sample sets from panel A using trypan blue exclusion. (D) BrdU uptake (Olig2^+/+ and Olig2^−/−) in cells either untreated or treated with 2 Gy of IR. Cells were treated with 2 Gy of IR, 24 hours post radiation they were pulsed with 10 µM BrdU for 1 hr and then analyzed by FACS. Data shown here represents percentage uptake in irradiated samples relative to untreated samples. (E) Detection of apoptosis in Olig2^+/+ and Olig2^−/− cells after radiation treatment. Cells were treated with 2 Gy of IR and 24 hrs post treatment analyzed for activated caspase 3 by FACS analysis. The bar graph represents percentage of cleaved caspase 3-positive cells present in each sample. For all graphs, error bars indicate SEM. The data are representative of three independent experiments. (***) p < 0.001 (**) p<0.01 (*) p<0.05. See also Figure S1.

Figure 2. Olig2-mediated radiation resistance depends on p53 status

(A) Olig2-tva-cre^+/− driver mice (Schuller et al., 2008) and p53 conditional null (p53^fl/fl) mice were crossed to obtain neural progenitors that were null for p53 and either null or heterozygous for Olig2 function. The cells (neurosphere cultures) were then dissociated, treated with 2 Gy of IR and allowed to form secondary neurospheres for 5 days after treatment. Scale bars = 100 µm (B) Olig2^−/− cells were transduced with an expression vector encoding a dominant negative mutation of p53 (p53DD) or with a vector control. The cells were then irradiated and secondary neurosphere assays were conducted as per panel A above. Scale bars = 100 µm (C) Stabilizing p53 can radiosensitize Olig2 wild type cells. Olig2^+/+ cells were treated with 0.25 µM Nutlin or DMSO alone for 16 hours and then exposed to 2 Gy of IR. The cells were grown for 5 days to allow secondary sphere formation. (D) Quantitation of the percentage of neurospheres formed after radiation as compared to untreated control samples. Scale bars = 100 µm (E) Quantitation of percentage of viable cells after radiation treatment as compared to untreated controls. For all graphs, the data are representative of three independent experiments. (***) p < 0.001. Error bars indicate SEM..

Figure 3. Olig2 opposes cellular responses to radiation in normal and malignant neural progenitors

(A) Ablation of p19 attenuates the amount of activated p53 that is produced in response to radiation. Neurosphere cultures of the indicated genotypes were exposed to 2Gy of gamma irradiation. Cell lysates obtained 6 hrs post-treatment were analyzed by immunoblotting with antibodies recognizing phosphorylated p53 (Ser15), total p53 and β-actin, the quantification results are shown at right. (B–D) Neurosphere cultures of the indicated genotypes were exposed to 2Gy of gamma irradiation. Secondary neurosphere assays were counted at day 5 post-treatment. Scale bars = 100 µm. (E) Quantification of data in panels B–D. The data shown are the percentage of secondary neurospheres formed in irradiated samples relative to untreated control samples. The results shown are compiled from three independent experiments with three independent cell lines. (***) p < 0.001. Error bars indicate SEM.

Figure 4. Olig2 mediated radiation resistance in tumor progenitor cells is dependent on p53 status

(A) Suppression of p53 function. An expression vector encoding a dominant negative mutant of p53 (p53DD) was transduced into Olig2^−/− p16/p19^−/−; EGFRvIII tumor neurospheres as described in the text. These cells, together with vector controls were irradiated as shown. Secondary neurosphere assays were counted at day 5 post-treatment. Scale bars = 100 µm. (B) Enhancement of p53 function. Olig2^+/+ tumor neurospheres were treated with 0.25 µM Nutlin (an Mdm2 inhibitor) or DMSO control for 16 hours and then exposed to 2 Gy of radiation. Secondary neurosphere assays were counted at day 5 post-irradiation. Scale bars = 100 µm. (C) Quantitation of percentage of secondary neurospheres formed in treated samples as compared to untreated samples. (D) Quantitation of percentage of viable cells after radiation treatment as compared to untreated control samples. For both graphs, the data are compiled from three independent experiments. (***) p < 0.001, (**) p < 0.01. Error bars indicate SEM.

Figure 5. Olig2 promotes radioresistance in p53-positive human glioma cells

(A) BT37, a p53 wild-type human cell line was infected with either control or shOLIG2 containing lentivirus. After 48 hrs the cells were dissociated, exposed to 2 Gy of IR, and allowed to form secondary neurospheres for 5 days after treatment. Untreated cells served as control. Scale bars = 100 µm. (B) Same as panel “A” except that the line used (BT112) has amplified EGFR locus. (C) Same as panel “A” except that the line used (BT70) carries a mutant p53 (Arg273Cys). (D) Quantitation of percentage of secondary neurospheres formed in treated samples as compared to untreated controls. The difference in secondary neurosphere formation between the Olig2^+/+ (WT) and knockdown (KD) is significant to (***) p<0.001 (BT37) and (*) p<0.05 (BT112) for the p53 wild type cells. Error bars indicate SEM.

Figure 6. Olig2 requirement for tumorigenesis is p53-dependent

(A) Murine neural progenitors with the indicated Olig2 and p53 genotypes were transduced with EGFRvIII and injected into the brains of SCID mice. As indicated by the survival plots, mice injected with Olig2^cre/cre p53^fl/+ cells fail to form tumors in contrast to Olig2^cre/cre p53^fl/fl cells (p<0.0003). Note also that Olig2^cre/+ p53^fl/fl cells can form tumors with significantly shorter latency period than Olig2^cre/cre p53^fl/fl cells (median survival 46 days and 64 days respectively, p<0.003). (B) Comparison of tumors derived from Olig2^cre/+;p53^fl/fl and Olig2^cre/cre;p53^fl/fl cells. Note the greater proportion of GFAP-positive (astrocyte-like) cells and a near complete absence of PDGFRα positive (oligodendrocyte progenitor-like) and Tuj1-positive (neuron-like) cells in the Olig2 null tumors. Scale bars: the hEGFR panels 1mm and others 25µm. (C) Kaplan-Meier survival analysis of SCID mice intracranially implanted with human glioma cell lines transduced with OLIG2 shRNA or non-target shRNA (shNT). The differences in survival between two corresponding groups are p < 0.013 and 0.001 for p53 wild type BT37 and BT145 lines respectively, and p < 0.51 for the p53-mutant BT70 line. (D) Immunohistochemistry of tumors derived from injections of p53-positive BT37 human glioma cell line. The tumors formed by cells infected with shOLIG2 vector no longer express the GFP marker and express OLIG2 at levels comparable to that seen in tumors that arise from control cells (shNT). Scale bars for H&E staining are 1.25 mm and all others 50 µm. (E) Immunohistochemistry of tumors derived from injections of p53-mutant BT70 human glioma cell line. The tumors arising in these mice injected with shOLIG2 continue to express the GFP marker and show a significant knockdown of OLIG2 relative to tumors that arise from control cells (shNT). Scale bars for H&E staining are 1.25 mm and all others 50 µm.

Figure 7. Olig2 suppresses acetylation and DNA binding of p53

(A) Ablation of Olig2 does not affect basal expression of p53 or its phosphorylation upon DNA damage. Cell lysates from cultures either untreated (−) or treated (+) with 2 Gy of IR were obtained 6 hrs post-treatment and analyzed by immunoblotting with antibodies recognizing phosphorylated p53 (Ser15), total p53, and β-actin. (B) Olig2 suppresses DNA damage induced acetylation of p53. Cell lysates from cultures either untreated (−) or treated (+) with 2 Gy of IR in the presence of HDAC inhibitor were obtained 6 hrs post-treatment and analyzed by immunoblotting with antibodies recognizing acetylated p53 (lys379) or vinculin. Immunoblots from five independent experiments were quantitated. The bar graphs represent acetylated p53 levels after radiation in the indicated cell lines. (*) p < 0.05, (***) p < 0.001. (C) Quantitative ChIP analysis of p53 bound to its target promoters (Cdkn1a, Wig1, Bax and Mdm2). The bar graphs represent ratio of fold enrichment of p53 at target sites in Olig2^−/− cells over Olig2^+/+ cells. For all graphs the data is compiled from three independent experiments. For B and C, error bars represent SEM and (*) p < 0.05, (**) p < 0.01, (***) p < 0.001. (D) Model for Olig2-mediated negative regulation of p53 signaling pathway. DNA damage leads to stabilization and activation of p53 through posttranslational modifications (phosphorylation and acetylation). Activated p53 transactivates its downstream targets, which leads to either growth arrest or cell death (Barlev et al., 2001; Dornan et al., 2003; Horn and Vousden, 2007; Riley et al., 2008). As indicated, Olig2 suppresses p53 acetylation and thereby affects p53 association with target promoters. See also Figure S2.

Figure 8. Olig2 promotes radiation resistance in part through suppression of p21

(A) Cell lysates from indicated murine cell lines either untreated (−) or treated (+) with 2 Gy of gamma radiation were obtained 6 hrs post-treatment and analyzed by immunoblotting with antibodies recognizing p21 and β-actin. (B) Human glioma cell lines (BT37 and BT112) were transduced with a control hairpin (NT) or shOlig2 and treated with 2 Gy of gamma radiation. Cell lysates from untreated (−) or treated (+) samples were obtained 6 hrs post-treatment and analyzed by immunoblotting with antibodies recognizing p21 and β-actin. (C) Ablation of Cdkn1a in Olig2^−/− cells with wild-type EGFR restores radiation resistance. Olig2^−/− (left panels) or Olig2^−/− p16/p19^−/−; EGFRvIII (right panel) cells were infected with retrovirus expressing control shNT or shCdkn1a. After 48 hrs cells were dissociated, exposed to 2 Gy of IR and allowed to form spheres for 5d after treatment. Untreated cells served as control. Olig2^−/− p16/p19^−/−Cdkn1a^−/− cells were treated with 2 Gy of IR and allowed to form spheres for 5 days after treatment (middle panel). Untreated cells served as control. Scale bars = 100 µm. See also Figure S3.