Limits...
CSF1 is a novel p53 target gene whose protein product functions in a feed-forward manner to suppress apoptosis and enhance p53-mediated growth arrest.

Azzam G, Wang X, Bell D, Murphy ME - PLoS ONE (2013)

Bottom Line: This leads to an increase in p53-mediated growth arrest, along with a concomitant decrease in apoptosis.We show that cells expressing CSF1 and CSF1R are indeed radio-resistant, and further, that this effect requires p53.Finally, these data may explain the association of the P72 variant and the CSF1/CSF1R pathway with increased senescence and radio-resistance in some epithelial tumor types.

View Article: PubMed Central - PubMed

Affiliation: Program in Molecular and Cellular Oncogenesis, The Wistar Institute, Philadelphia, Pennsylvania, United States of America.

ABSTRACT
The p53 tumor suppressor gene has a common polymorphism at codon 72 that alters its function. We previously reported that the proline 72 polymorphic variant of p53 (P72) demonstrates increased ability to transactivate a subset of genes, relative to arginine 72 (R72); one of these genes is macrophage colony stimulating factor (CSF1). At present, the mechanism(s) underlying the increased transcriptional activity of P72 toward genes like CSF1 have not been completely elucidated. Additionally, the consequences of increased transcription of genes like CSF1 by the P72 variant to the downstream p53 pathway are unknown. In this report, we address these issues. We show that the CSF1 gene contains a conserved binding site for p53, and interestingly that the P72 variant shows increased ability to bind to this site. Moreover, we show that increased CSF1/CSF1R signaling in P72 cells feeds back on the p53 pathway to enhance p53 phosphorylation, levels, and transactivation of target genes, particularly the cyclin-dependent kinase inhibitor p21 (CDKN1A). This leads to an increase in p53-mediated growth arrest, along with a concomitant decrease in apoptosis. Notably, the CSF1/CSF1R signaling axis is overexpressed in several epithelial cancers, and there is clinical evidence that this pathway plays a role in radio-resistance of some cancers. We show that cells expressing CSF1 and CSF1R are indeed radio-resistant, and further, that this effect requires p53. These combined data are the first to implicate the CSF1/CSF1R pathway in the decision between p53-mediated growth arrest and apoptosis. They are also the first to highlight a cytokine as influential in cell fate determined by p53 in epithelial cells. Finally, these data may explain the association of the P72 variant and the CSF1/CSF1R pathway with increased senescence and radio-resistance in some epithelial tumor types.

Show MeSH

Related in: MedlinePlus

The NF-kB subunit p65RelA does not play a role in the transactivation of CSF1 by p53.A. Left panel: QRT-PCR of p65RelA RNA level in H1299 tet-inducible p53 (P72 variant) cells following infection with a short-hairpin to RelA (shRelA) or vector alone (Vector). Values shown are the average of 3 independent replicates, normalized to control. Right panel: QRT-PCR of RNA isolated from H1299 tet-inducible p53 cells (P72 variant), infected with shRelA or vector. Cells were treated with TNF-α (10 ng/mL) and analyzed for the level of LIF after the indicated time points (hours). Values shown are the average of 3 independent replicates, normalized to control. Error bars mark standard deviation. B. Left panel: Western analysis of lysates isolated from H1299 tet-inducible p53 (P72 variant) cells following infection with a short-hairpin to RelA (shRelA) or vector alone (Vector), after 24 hours of doxycycline treatment (0.75 ug/ml); lysates were analyzed for the level of p53, p65RelA and GAPDH. Right panel: QRT-PCR of RNA isolated from shRelA or Vector tet-on p53 H1299 cells, analyzed for CSF1 after no treatment (0), or treatment with doxycycline (0.75 ug/ml) for the indicated time points (hours). Values shown are the average of 4 independent replicates, normalized to control. Error bars mark standard deviation. C. QRT-PCR of RNA isolated from H1299 tet-inducible p53 cells (P72 variant), infected with shRelA or vector. Cells were untreated (0 hr) or treated with Bay11–7082 or DMSO just prior to the addition of TNF-α (10 ng/ml) for the indicated time points. RNA was analyzed for the levels of LIF and CSF1; values shown are the average of 3 independent replicates, normalized to control. Error bars mark standard deviation. D. Q-PCR of chromatin immunoprecipitation (ChIP) eluates isolated from shRelA or Vector tet-inducible p53 cells, untreated or treated with doxycycline (0.75 ug/ml) for 18 hours. Immunoprecipitations were performed using antibody to p53 (Ab6) or the equivalent amount of normal mouse Immunoglobulin G (IgG). Primers used for Q-PCR analysis flank the predicted p53 response element on the CSF1 promoter. Values shown are the average of 2 independent experiments repeated in duplicate, and are presented as percent bound normalized to input. Error bars mark standard error.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3760869&req=5

pone-0074297-g003: The NF-kB subunit p65RelA does not play a role in the transactivation of CSF1 by p53.A. Left panel: QRT-PCR of p65RelA RNA level in H1299 tet-inducible p53 (P72 variant) cells following infection with a short-hairpin to RelA (shRelA) or vector alone (Vector). Values shown are the average of 3 independent replicates, normalized to control. Right panel: QRT-PCR of RNA isolated from H1299 tet-inducible p53 cells (P72 variant), infected with shRelA or vector. Cells were treated with TNF-α (10 ng/mL) and analyzed for the level of LIF after the indicated time points (hours). Values shown are the average of 3 independent replicates, normalized to control. Error bars mark standard deviation. B. Left panel: Western analysis of lysates isolated from H1299 tet-inducible p53 (P72 variant) cells following infection with a short-hairpin to RelA (shRelA) or vector alone (Vector), after 24 hours of doxycycline treatment (0.75 ug/ml); lysates were analyzed for the level of p53, p65RelA and GAPDH. Right panel: QRT-PCR of RNA isolated from shRelA or Vector tet-on p53 H1299 cells, analyzed for CSF1 after no treatment (0), or treatment with doxycycline (0.75 ug/ml) for the indicated time points (hours). Values shown are the average of 4 independent replicates, normalized to control. Error bars mark standard deviation. C. QRT-PCR of RNA isolated from H1299 tet-inducible p53 cells (P72 variant), infected with shRelA or vector. Cells were untreated (0 hr) or treated with Bay11–7082 or DMSO just prior to the addition of TNF-α (10 ng/ml) for the indicated time points. RNA was analyzed for the levels of LIF and CSF1; values shown are the average of 3 independent replicates, normalized to control. Error bars mark standard deviation. D. Q-PCR of chromatin immunoprecipitation (ChIP) eluates isolated from shRelA or Vector tet-inducible p53 cells, untreated or treated with doxycycline (0.75 ug/ml) for 18 hours. Immunoprecipitations were performed using antibody to p53 (Ab6) or the equivalent amount of normal mouse Immunoglobulin G (IgG). Primers used for Q-PCR analysis flank the predicted p53 response element on the CSF1 promoter. Values shown are the average of 2 independent experiments repeated in duplicate, and are presented as percent bound normalized to input. Error bars mark standard error.

Mentions: In order to assess a possible mechanism for increased P72 binding, we next sought to test the hypothesis that NF-kB might influence p53 binding to the CSF1 promoter. We previously showed that the P72 variant of p53 demonstrates increased ability to interact with the p65RelA subunit of NF-kB, compared to R72. Further, we showed that this enhanced interaction was responsible in part for the increased ability of P72 to transactivate the gene for murine caspase-11, which contains both p53 and NF-kB consensus elements in its promoter [15]. In support of this premise, we previously reported that silencing p65RelA reduced the ability of the P72 variant to transactivate caspase-11 [15]. We noted a consensus NF-kB element in the promoter of CSF1 (Fig. 2C), and therefore sought to test the hypothesis that a similar cooperation between P72 and NF-kB might exist for the CSF1 gene, and thus further explain the mechanism for enhanced transactivation of this gene by P72. To address this hypothesis, we generated H1299 cells containing tetracycline-inducible P72 protein that were stably infected with control lentivirus (Vector) or a lentivirus expressing a short hairpin to p65RelA (shRelA). QRT-PCR analysis indicated that the short hairpin to p65RelA effectively silenced this gene (Fig. 3A). Consistent with this, cells with silenced RelA demonstrated reduced transactivation of the NF-kB target gene LIF following TNF-α treatment, compared to vector control (Fig. 3A). Notably, despite the significant decrease in p65RelA protein in RelA-silenced H1299 cells, there was no effect on the ability of the P72 variant of p53 to transactivate CSF1 (Fig. 3B). To confirm this result, we next sought to pharmacologically inhibit NF-kB and assess the impact on the ability of p53 to transactivate CSF1. Using the IKK inhibitor BAY 11–7082, we were able to show that incubation with this compound effectively inhibited the ability of LIF to be induced following treatment of H1299 cells with TNF-α. In contrast, this inhibitor had no effect on p53-mediated transactivation of CSF1 (Fig. 3C). Additionally, we performed chromatin immunoprecipitation for the p53 response element in CSF1 in H1299 tet-inducible p53 cells, stably infected with parental vector or shRelA. As in the transactivation analyses, ChIP experiments revealed there was no effect of silencing p65RelA on the ability of p53 to interact with the CSF1 promoter (Fig. 3D). Because the human orthologue of murine caspase-11 has not yet been identified, we were unable to use this gene as a control for these studies. However, the combined data suggest that the P72 variant uses a novel mechanism for increased transactivation of the CSF1 gene. This mechanism involves enhanced ability of P72 to bind to this promoter, and it does not involve increased cooperation of P72 with NF-kB.


CSF1 is a novel p53 target gene whose protein product functions in a feed-forward manner to suppress apoptosis and enhance p53-mediated growth arrest.

Azzam G, Wang X, Bell D, Murphy ME - PLoS ONE (2013)

The NF-kB subunit p65RelA does not play a role in the transactivation of CSF1 by p53.A. Left panel: QRT-PCR of p65RelA RNA level in H1299 tet-inducible p53 (P72 variant) cells following infection with a short-hairpin to RelA (shRelA) or vector alone (Vector). Values shown are the average of 3 independent replicates, normalized to control. Right panel: QRT-PCR of RNA isolated from H1299 tet-inducible p53 cells (P72 variant), infected with shRelA or vector. Cells were treated with TNF-α (10 ng/mL) and analyzed for the level of LIF after the indicated time points (hours). Values shown are the average of 3 independent replicates, normalized to control. Error bars mark standard deviation. B. Left panel: Western analysis of lysates isolated from H1299 tet-inducible p53 (P72 variant) cells following infection with a short-hairpin to RelA (shRelA) or vector alone (Vector), after 24 hours of doxycycline treatment (0.75 ug/ml); lysates were analyzed for the level of p53, p65RelA and GAPDH. Right panel: QRT-PCR of RNA isolated from shRelA or Vector tet-on p53 H1299 cells, analyzed for CSF1 after no treatment (0), or treatment with doxycycline (0.75 ug/ml) for the indicated time points (hours). Values shown are the average of 4 independent replicates, normalized to control. Error bars mark standard deviation. C. QRT-PCR of RNA isolated from H1299 tet-inducible p53 cells (P72 variant), infected with shRelA or vector. Cells were untreated (0 hr) or treated with Bay11–7082 or DMSO just prior to the addition of TNF-α (10 ng/ml) for the indicated time points. RNA was analyzed for the levels of LIF and CSF1; values shown are the average of 3 independent replicates, normalized to control. Error bars mark standard deviation. D. Q-PCR of chromatin immunoprecipitation (ChIP) eluates isolated from shRelA or Vector tet-inducible p53 cells, untreated or treated with doxycycline (0.75 ug/ml) for 18 hours. Immunoprecipitations were performed using antibody to p53 (Ab6) or the equivalent amount of normal mouse Immunoglobulin G (IgG). Primers used for Q-PCR analysis flank the predicted p53 response element on the CSF1 promoter. Values shown are the average of 2 independent experiments repeated in duplicate, and are presented as percent bound normalized to input. Error bars mark standard error.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3760869&req=5

pone-0074297-g003: The NF-kB subunit p65RelA does not play a role in the transactivation of CSF1 by p53.A. Left panel: QRT-PCR of p65RelA RNA level in H1299 tet-inducible p53 (P72 variant) cells following infection with a short-hairpin to RelA (shRelA) or vector alone (Vector). Values shown are the average of 3 independent replicates, normalized to control. Right panel: QRT-PCR of RNA isolated from H1299 tet-inducible p53 cells (P72 variant), infected with shRelA or vector. Cells were treated with TNF-α (10 ng/mL) and analyzed for the level of LIF after the indicated time points (hours). Values shown are the average of 3 independent replicates, normalized to control. Error bars mark standard deviation. B. Left panel: Western analysis of lysates isolated from H1299 tet-inducible p53 (P72 variant) cells following infection with a short-hairpin to RelA (shRelA) or vector alone (Vector), after 24 hours of doxycycline treatment (0.75 ug/ml); lysates were analyzed for the level of p53, p65RelA and GAPDH. Right panel: QRT-PCR of RNA isolated from shRelA or Vector tet-on p53 H1299 cells, analyzed for CSF1 after no treatment (0), or treatment with doxycycline (0.75 ug/ml) for the indicated time points (hours). Values shown are the average of 4 independent replicates, normalized to control. Error bars mark standard deviation. C. QRT-PCR of RNA isolated from H1299 tet-inducible p53 cells (P72 variant), infected with shRelA or vector. Cells were untreated (0 hr) or treated with Bay11–7082 or DMSO just prior to the addition of TNF-α (10 ng/ml) for the indicated time points. RNA was analyzed for the levels of LIF and CSF1; values shown are the average of 3 independent replicates, normalized to control. Error bars mark standard deviation. D. Q-PCR of chromatin immunoprecipitation (ChIP) eluates isolated from shRelA or Vector tet-inducible p53 cells, untreated or treated with doxycycline (0.75 ug/ml) for 18 hours. Immunoprecipitations were performed using antibody to p53 (Ab6) or the equivalent amount of normal mouse Immunoglobulin G (IgG). Primers used for Q-PCR analysis flank the predicted p53 response element on the CSF1 promoter. Values shown are the average of 2 independent experiments repeated in duplicate, and are presented as percent bound normalized to input. Error bars mark standard error.
Mentions: In order to assess a possible mechanism for increased P72 binding, we next sought to test the hypothesis that NF-kB might influence p53 binding to the CSF1 promoter. We previously showed that the P72 variant of p53 demonstrates increased ability to interact with the p65RelA subunit of NF-kB, compared to R72. Further, we showed that this enhanced interaction was responsible in part for the increased ability of P72 to transactivate the gene for murine caspase-11, which contains both p53 and NF-kB consensus elements in its promoter [15]. In support of this premise, we previously reported that silencing p65RelA reduced the ability of the P72 variant to transactivate caspase-11 [15]. We noted a consensus NF-kB element in the promoter of CSF1 (Fig. 2C), and therefore sought to test the hypothesis that a similar cooperation between P72 and NF-kB might exist for the CSF1 gene, and thus further explain the mechanism for enhanced transactivation of this gene by P72. To address this hypothesis, we generated H1299 cells containing tetracycline-inducible P72 protein that were stably infected with control lentivirus (Vector) or a lentivirus expressing a short hairpin to p65RelA (shRelA). QRT-PCR analysis indicated that the short hairpin to p65RelA effectively silenced this gene (Fig. 3A). Consistent with this, cells with silenced RelA demonstrated reduced transactivation of the NF-kB target gene LIF following TNF-α treatment, compared to vector control (Fig. 3A). Notably, despite the significant decrease in p65RelA protein in RelA-silenced H1299 cells, there was no effect on the ability of the P72 variant of p53 to transactivate CSF1 (Fig. 3B). To confirm this result, we next sought to pharmacologically inhibit NF-kB and assess the impact on the ability of p53 to transactivate CSF1. Using the IKK inhibitor BAY 11–7082, we were able to show that incubation with this compound effectively inhibited the ability of LIF to be induced following treatment of H1299 cells with TNF-α. In contrast, this inhibitor had no effect on p53-mediated transactivation of CSF1 (Fig. 3C). Additionally, we performed chromatin immunoprecipitation for the p53 response element in CSF1 in H1299 tet-inducible p53 cells, stably infected with parental vector or shRelA. As in the transactivation analyses, ChIP experiments revealed there was no effect of silencing p65RelA on the ability of p53 to interact with the CSF1 promoter (Fig. 3D). Because the human orthologue of murine caspase-11 has not yet been identified, we were unable to use this gene as a control for these studies. However, the combined data suggest that the P72 variant uses a novel mechanism for increased transactivation of the CSF1 gene. This mechanism involves enhanced ability of P72 to bind to this promoter, and it does not involve increased cooperation of P72 with NF-kB.

Bottom Line: This leads to an increase in p53-mediated growth arrest, along with a concomitant decrease in apoptosis.We show that cells expressing CSF1 and CSF1R are indeed radio-resistant, and further, that this effect requires p53.Finally, these data may explain the association of the P72 variant and the CSF1/CSF1R pathway with increased senescence and radio-resistance in some epithelial tumor types.

View Article: PubMed Central - PubMed

Affiliation: Program in Molecular and Cellular Oncogenesis, The Wistar Institute, Philadelphia, Pennsylvania, United States of America.

ABSTRACT
The p53 tumor suppressor gene has a common polymorphism at codon 72 that alters its function. We previously reported that the proline 72 polymorphic variant of p53 (P72) demonstrates increased ability to transactivate a subset of genes, relative to arginine 72 (R72); one of these genes is macrophage colony stimulating factor (CSF1). At present, the mechanism(s) underlying the increased transcriptional activity of P72 toward genes like CSF1 have not been completely elucidated. Additionally, the consequences of increased transcription of genes like CSF1 by the P72 variant to the downstream p53 pathway are unknown. In this report, we address these issues. We show that the CSF1 gene contains a conserved binding site for p53, and interestingly that the P72 variant shows increased ability to bind to this site. Moreover, we show that increased CSF1/CSF1R signaling in P72 cells feeds back on the p53 pathway to enhance p53 phosphorylation, levels, and transactivation of target genes, particularly the cyclin-dependent kinase inhibitor p21 (CDKN1A). This leads to an increase in p53-mediated growth arrest, along with a concomitant decrease in apoptosis. Notably, the CSF1/CSF1R signaling axis is overexpressed in several epithelial cancers, and there is clinical evidence that this pathway plays a role in radio-resistance of some cancers. We show that cells expressing CSF1 and CSF1R are indeed radio-resistant, and further, that this effect requires p53. These combined data are the first to implicate the CSF1/CSF1R pathway in the decision between p53-mediated growth arrest and apoptosis. They are also the first to highlight a cytokine as influential in cell fate determined by p53 in epithelial cells. Finally, these data may explain the association of the P72 variant and the CSF1/CSF1R pathway with increased senescence and radio-resistance in some epithelial tumor types.

Show MeSH
Related in: MedlinePlus