Limits...
p21(WAF1/CIP1) upregulation through the stress granule-associated protein CUGBP1 confers resistance to bortezomib-mediated apoptosis.

Gareau C, Fournier MJ, Filion C, Coudert L, Martel D, Labelle Y, Mazroui R - PLoS ONE (2011)

Bottom Line: We found that depleting CUGBP1 in cancer cells prevents bortezomib-mediated p21 upregulation.FISH experiments combined to mRNA stability assays show that this effect is largely due to a mistargeting of p21 mRNA in stress granules leading to its degradation.We propose that one key mechanism by which apoptosis is inhibited upon treatment with chemotherapeutic drugs might involve upregulation of the p21 protein through CUGBP1.

View Article: PubMed Central - PubMed

Affiliation: Department of Molecular Biology, Medical Biochemistry, and Pathology, Faculty of Medicine, Laval University, CHUQ Research Centre/St-François d'Assise Research Centre (CRCHUQ/CRSFA), Quebec, Canada.

ABSTRACT

Background: p21(WAF1/CIP1) is a well known cyclin-dependent kinase inhibitor induced by various stress stimuli. Depending on the stress applied, p21 upregulation can either promote apoptosis or prevent against apoptotic injury. The stress-mediated induction of p21 involves not only its transcriptional activation but also its posttranscriptional regulation, mainly through stabilization of p21 mRNA levels. We have previously reported that the proteasome inhibitor MG132 induces the stabilization of p21 mRNA, which correlates with the formation of cytoplasmic RNA stress granules. The mechanism underlying p21 mRNA stabilization, however, remains unknown.

Methodology/principal findings: We identified the stress granules component CUGBP1 as a factor required for p21 mRNA stabilization following treatment with bortezomib ( =  PS-341/Velcade). This peptide boronate inhibitor of the 26S proteasome is very efficient for the treatment of myelomas and other hematological tumors. However, solid tumors are sometimes refractory to bortezomib treatment. We found that depleting CUGBP1 in cancer cells prevents bortezomib-mediated p21 upregulation. FISH experiments combined to mRNA stability assays show that this effect is largely due to a mistargeting of p21 mRNA in stress granules leading to its degradation. Altering the expression of p21 itself, either by depleting CUGBP1 or p21, promotes bortezomib-mediated apoptosis.

Conclusions/significance: We propose that one key mechanism by which apoptosis is inhibited upon treatment with chemotherapeutic drugs might involve upregulation of the p21 protein through CUGBP1.

Show MeSH

Related in: MedlinePlus

CUGBP1 binds to p21 mRNA and promotes its accumulation and expression upon bortezomib treatment.(A–D) Seventy-two hours following transfection with CUGBP1, or control siRNAs, HeLa cells were treated with bortezomib (2 µM) for either 10 h (A) or 4 h (B). (A) Protein extracts were prepared and analyzed by western blot to detect CUGBP1, p21, and G3BP1 proteins (loading standards) using the appropriate antibodies. The percentages of CUGBP1 knockdown and p21 expression were determined by quantitation of the signal on films by densitometry using Adobe Photoshop as described in Figure 1C. Shown are typical results of three experiments. (B–D) p21 mRNA is quantitatively recruited into SG where it co-localizes with CUGBP1. (B) Cells were processed for FISH to detect p21 mRNA coupled to immunofluorescence to visualize SG using antibodies against CUGBP1 and G3BP1. The percentage of SG (>3 granules/cell) is indicated at the bottom of the right panel. The percentage of cells harboring SG positives for p21 mRNA is also indicated on the right of the figure. Shown are typical results from five different fields and three different experiments containing a total of more than 1000 cells. (C–D) Densitometry quantification of CUGBP1 immunofluorescence signal in SG versus nuclei (C) and in SG versus the cytoplasm (D) using Adobe Photoshop software as described in Figure 1. (E) Densitometry quantification of p21 mRNA FISH signal was done as described in Figure 2. (F–G) HeLa cells were treated with bortezomib for 4 h and their extracts were used to immunoprecipitate CUGBP1 with anti-CUGBP1 antibodies and with IgG as a control. IP: Immunoprecipitate; FT: flow-through following immunoprecipitation; Total: the input used for immunoprecipitation. (F) Proteins were analyzed by western blot for CUGBP1 immunoprecipiation. (G) mRNAs were isolated from each immunoprecipitate and quantified by qRT-PCR. The amounts of p21 mRNA and p27 mRNA (as control) were normalized against GAPDH mRNA. (H) HeLa cells were treated with CUGBP1-specific siRNA, or with control (non-specific) siRNA as described above, incubated with 2 µM bortezomib for either 4 or 10 h, lysed, and their total RNA isolated and analyzed for p21 mRNA levels by qRT-PCR. The amount of p21 mRNA was normalized against that of GAPDH mRNA as described in Figure 2.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3102688&req=5

pone-0020254-g003: CUGBP1 binds to p21 mRNA and promotes its accumulation and expression upon bortezomib treatment.(A–D) Seventy-two hours following transfection with CUGBP1, or control siRNAs, HeLa cells were treated with bortezomib (2 µM) for either 10 h (A) or 4 h (B). (A) Protein extracts were prepared and analyzed by western blot to detect CUGBP1, p21, and G3BP1 proteins (loading standards) using the appropriate antibodies. The percentages of CUGBP1 knockdown and p21 expression were determined by quantitation of the signal on films by densitometry using Adobe Photoshop as described in Figure 1C. Shown are typical results of three experiments. (B–D) p21 mRNA is quantitatively recruited into SG where it co-localizes with CUGBP1. (B) Cells were processed for FISH to detect p21 mRNA coupled to immunofluorescence to visualize SG using antibodies against CUGBP1 and G3BP1. The percentage of SG (>3 granules/cell) is indicated at the bottom of the right panel. The percentage of cells harboring SG positives for p21 mRNA is also indicated on the right of the figure. Shown are typical results from five different fields and three different experiments containing a total of more than 1000 cells. (C–D) Densitometry quantification of CUGBP1 immunofluorescence signal in SG versus nuclei (C) and in SG versus the cytoplasm (D) using Adobe Photoshop software as described in Figure 1. (E) Densitometry quantification of p21 mRNA FISH signal was done as described in Figure 2. (F–G) HeLa cells were treated with bortezomib for 4 h and their extracts were used to immunoprecipitate CUGBP1 with anti-CUGBP1 antibodies and with IgG as a control. IP: Immunoprecipitate; FT: flow-through following immunoprecipitation; Total: the input used for immunoprecipitation. (F) Proteins were analyzed by western blot for CUGBP1 immunoprecipiation. (G) mRNAs were isolated from each immunoprecipitate and quantified by qRT-PCR. The amounts of p21 mRNA and p27 mRNA (as control) were normalized against GAPDH mRNA. (H) HeLa cells were treated with CUGBP1-specific siRNA, or with control (non-specific) siRNA as described above, incubated with 2 µM bortezomib for either 4 or 10 h, lysed, and their total RNA isolated and analyzed for p21 mRNA levels by qRT-PCR. The amount of p21 mRNA was normalized against that of GAPDH mRNA as described in Figure 2.

Mentions: The SG components HuR and CUGBP1 have been shown to bind p21 mRNA, thereby regulating its expression [14], [54], [69]. While HuR promotes p21 mRNA stabilization through interaction with its AU-rich element in the 3′-untranslated region, CUGBP1 interacts with the GC-rich element in the 5′ region of p21 mRNA coding sequence and promotes its translation in senescent fibroblast cells. We investigated the role of these two proteins in the upregulation of p21WAF/CIP1 observed upon proteasome inhibition in HeLa cells. We found that depletion of HuR had no effect or rather slightly increased p21 protein expression following bortezomib treatment (data not shown). Depletion of CUGBP1 using two specific siRNAs had no effect on p21 protein expression under normal growth conditions but prevented bortezomib-mediated p21WAF/CIP1 upregulation as assessed by western blot analysis using anti-p21 antibodies (Fig. 3A and data not shown). These results indicate that CUGBP1 promotes p21 upregulation upon proteasome inhibition. As a first step to investigate how CUGBP1 up-regulates p21 upon proteasome inhibition, we performed co- localization studies between CUGBP1 and p21 mRNA. Under normal growth conditions, CUGBP1 is mostly nuclear [32], [56] (Fig. S4C, panel 2). Arsenite induces the localization of a fraction of CUGBP1 in SG [32], [56], and our immunofluorescence experiments showed that CUGBP1 was also recruited into bortezomib-induced SG (Fig. 3B). Quantification of these immunofluorescence data show that a significant fraction (∼45%) of total CUGBP1 is present in SG (Fig. 3B–C) where it co-localized with p21 mRNA (Fig. 3B and 3D–E) following treatment with bortezomib. Under these conditions, about 50% of CUGBP1 is nuclear (Fig. 3B–C) and only residual staining of CUGBP1 is detected in the cytoplasm (Fig. 3B and 3D). Prolonged treatment with bortezomib (10 h) resulted however in nuclear localization of most CUGBP1 (Fig. S4C). At this time we detected also some p21 mRNA FISH signal in the nucleus. However, and as mentioned above, the overall p21 mRNA FISH signal at 10-h bortezomib treatment is reduced (Fig. S4A and S4C), as compared to 4 h treatment (Fig. S4A; see also Fig. 2A) at which time point most of p21 mRNA (∼80%) co-localized with CUGBP1 in SG (Fig. 3B and 3D–E). Overall, our results (Fig. 3B) indicate that bortezomib treatment induces CUGBP1-p21 mRNA co-localization in SG. CUGBP1 might thus bind to p21 mRNA and recruit it to SG where it accumulates. The latter assumption would therefore predict that CUGBP1 depletion should trigger p21 mRNA degradation, which would explain the low amount of p21 protein following prolonged proteasome inhibition in CUGBP1-depleted cells (Fig. 3A). Indeed, we first confirmed that CUGBP1 binds to p21 mRNA under bortezomib conditions, as assessed by immunoprecipitating CUGBP1 (Fig. 3F) followed by quantitative RT-PCR (qRT-PCR) to detect associated p21 mRNA (Fig. 3G). This binding was specific, as suggested by the finding that control p27 mRNA was not recovered in CUGBP1 immunoprecipitates (Fig. 3G). Second, the steady-state p21 mRNA level significantly decreased in CUGBP1-depleted cells following proteasome inhibition (Fig. 3H). Depletion of CUGBP1 had however no effect on the steady-state level of p21 mRNA in untreated cells (Fig. S6A), suggesting that CUGBP1 is not required to maintain the basal level of the p21 mRNA. We then tested the possibility that the decreased level of p21 mRNA in CUGBP1-depleted cells upon bortezomib treatment might be consistent with a more rapid degradation of p21 mRNA. Indeed, the half-life of p21 mRNA in bortezomib conditions decreased from >3 h in mock-depleted cells (Fig. 4A) to 90 min ±15 min in CUGBP1-depleted cells (Fig. 4B). Depletion of CUGBP1 was efficient, as evidenced by both western blot analysis (Fig. 4C) and immunofluorescence (Fig. 4D), using anti-CUGBP1 antibodies. However, quantification of the localization of several SG markers (Fig. 4D, and data not shown) indicated that CUGBP1 depletion per se did not affect SG formation. This result ruled out the possibility that the destabilization of p21 mRNA observed in CUGBP1-depleted cells upon bortezomib treatment (Fig. 4A–B) was due to a cellular failure to form SG. One likely possibility of the destabilisation of p21 mRNA observed in CUGBP1-depleted cells is because it is no longer recruited to SG. We then assessed if CUGBP1 depletion affects the accumulation of p21 mRNA in SG using FISH. SG were visualized by immunofluorescence using antibodies to either Ras GTPase-activating protein-binding protein 1 G3BP1 (Fig. 4D) or FXR1 (data not shown). Whereas the major fraction of p21 mRNA (∼70%) is present in SG in over 60% of mock-depleted cells upon treatment with bortezomib (Fig. 4D–E), less than 30% of CUGBP1-depleted cells only exhibited such p21 mRNA localization in SG (Fig. 4D). Indeed, the large fraction of CUGBP1-depleted cells (∼80%) has a barely detectable FISH p21 mRNA signal in SG and in the cytoplasm (Fig. 4D and 4F). Under those bortezomib conditions, quantification of the intensity of p21 mRNA FISH signal showed that CUGBP1 depletion decreased this signal to 30–40%, as compared to mock-depleted cells where most p21 mRNA FISH signal was detected in SG (Fig. 4G). This effect seems to be specific since under the same conditions, depletion of CUGBP1 does not affect the association of the non-target GAPDH mRNA [47] with SG as assessed by immunofluorescence coupled with FISH (Fig. S7). Overall, our results show that CUGBP1 depletion significantly prevented bortezomib-induced p21 mRNA accumulation, which is likely due to its mistargeting in SG. As aforementioned, CUGBP1 depletion does not affect p21 mRNA steady-state levels in normal growth conditions (Fig. S6A) and we found that this is also the case under staurosporine treatment (Fig. S6B), a SG-free apoptotic condition [70] (Fig. S6D). These results suggest that CUGBP1 is not required to maintain the steady-state levels of p21 mRNA in the cytoplasm. On contrary, CUGBP1 seems to be required for bortezomib-induced p21 mRNA accumulation in SG (Fig. 3H, 4D, and S6C). We suggest that CUGBP1 promotes p21 mRNA accumulation by recruiting the transcript in SG, although we do not exclude that CUGBP1-mediated p21 mRNA accumulation involves additional mechanisms. Nevertheless, these results showed that CUGBP1 acts as a positive effector of p21 upregulation upon proteasome inhibition.


p21(WAF1/CIP1) upregulation through the stress granule-associated protein CUGBP1 confers resistance to bortezomib-mediated apoptosis.

Gareau C, Fournier MJ, Filion C, Coudert L, Martel D, Labelle Y, Mazroui R - PLoS ONE (2011)

CUGBP1 binds to p21 mRNA and promotes its accumulation and expression upon bortezomib treatment.(A–D) Seventy-two hours following transfection with CUGBP1, or control siRNAs, HeLa cells were treated with bortezomib (2 µM) for either 10 h (A) or 4 h (B). (A) Protein extracts were prepared and analyzed by western blot to detect CUGBP1, p21, and G3BP1 proteins (loading standards) using the appropriate antibodies. The percentages of CUGBP1 knockdown and p21 expression were determined by quantitation of the signal on films by densitometry using Adobe Photoshop as described in Figure 1C. Shown are typical results of three experiments. (B–D) p21 mRNA is quantitatively recruited into SG where it co-localizes with CUGBP1. (B) Cells were processed for FISH to detect p21 mRNA coupled to immunofluorescence to visualize SG using antibodies against CUGBP1 and G3BP1. The percentage of SG (>3 granules/cell) is indicated at the bottom of the right panel. The percentage of cells harboring SG positives for p21 mRNA is also indicated on the right of the figure. Shown are typical results from five different fields and three different experiments containing a total of more than 1000 cells. (C–D) Densitometry quantification of CUGBP1 immunofluorescence signal in SG versus nuclei (C) and in SG versus the cytoplasm (D) using Adobe Photoshop software as described in Figure 1. (E) Densitometry quantification of p21 mRNA FISH signal was done as described in Figure 2. (F–G) HeLa cells were treated with bortezomib for 4 h and their extracts were used to immunoprecipitate CUGBP1 with anti-CUGBP1 antibodies and with IgG as a control. IP: Immunoprecipitate; FT: flow-through following immunoprecipitation; Total: the input used for immunoprecipitation. (F) Proteins were analyzed by western blot for CUGBP1 immunoprecipiation. (G) mRNAs were isolated from each immunoprecipitate and quantified by qRT-PCR. The amounts of p21 mRNA and p27 mRNA (as control) were normalized against GAPDH mRNA. (H) HeLa cells were treated with CUGBP1-specific siRNA, or with control (non-specific) siRNA as described above, incubated with 2 µM bortezomib for either 4 or 10 h, lysed, and their total RNA isolated and analyzed for p21 mRNA levels by qRT-PCR. The amount of p21 mRNA was normalized against that of GAPDH mRNA as described in Figure 2.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3102688&req=5

pone-0020254-g003: CUGBP1 binds to p21 mRNA and promotes its accumulation and expression upon bortezomib treatment.(A–D) Seventy-two hours following transfection with CUGBP1, or control siRNAs, HeLa cells were treated with bortezomib (2 µM) for either 10 h (A) or 4 h (B). (A) Protein extracts were prepared and analyzed by western blot to detect CUGBP1, p21, and G3BP1 proteins (loading standards) using the appropriate antibodies. The percentages of CUGBP1 knockdown and p21 expression were determined by quantitation of the signal on films by densitometry using Adobe Photoshop as described in Figure 1C. Shown are typical results of three experiments. (B–D) p21 mRNA is quantitatively recruited into SG where it co-localizes with CUGBP1. (B) Cells were processed for FISH to detect p21 mRNA coupled to immunofluorescence to visualize SG using antibodies against CUGBP1 and G3BP1. The percentage of SG (>3 granules/cell) is indicated at the bottom of the right panel. The percentage of cells harboring SG positives for p21 mRNA is also indicated on the right of the figure. Shown are typical results from five different fields and three different experiments containing a total of more than 1000 cells. (C–D) Densitometry quantification of CUGBP1 immunofluorescence signal in SG versus nuclei (C) and in SG versus the cytoplasm (D) using Adobe Photoshop software as described in Figure 1. (E) Densitometry quantification of p21 mRNA FISH signal was done as described in Figure 2. (F–G) HeLa cells were treated with bortezomib for 4 h and their extracts were used to immunoprecipitate CUGBP1 with anti-CUGBP1 antibodies and with IgG as a control. IP: Immunoprecipitate; FT: flow-through following immunoprecipitation; Total: the input used for immunoprecipitation. (F) Proteins were analyzed by western blot for CUGBP1 immunoprecipiation. (G) mRNAs were isolated from each immunoprecipitate and quantified by qRT-PCR. The amounts of p21 mRNA and p27 mRNA (as control) were normalized against GAPDH mRNA. (H) HeLa cells were treated with CUGBP1-specific siRNA, or with control (non-specific) siRNA as described above, incubated with 2 µM bortezomib for either 4 or 10 h, lysed, and their total RNA isolated and analyzed for p21 mRNA levels by qRT-PCR. The amount of p21 mRNA was normalized against that of GAPDH mRNA as described in Figure 2.
Mentions: The SG components HuR and CUGBP1 have been shown to bind p21 mRNA, thereby regulating its expression [14], [54], [69]. While HuR promotes p21 mRNA stabilization through interaction with its AU-rich element in the 3′-untranslated region, CUGBP1 interacts with the GC-rich element in the 5′ region of p21 mRNA coding sequence and promotes its translation in senescent fibroblast cells. We investigated the role of these two proteins in the upregulation of p21WAF/CIP1 observed upon proteasome inhibition in HeLa cells. We found that depletion of HuR had no effect or rather slightly increased p21 protein expression following bortezomib treatment (data not shown). Depletion of CUGBP1 using two specific siRNAs had no effect on p21 protein expression under normal growth conditions but prevented bortezomib-mediated p21WAF/CIP1 upregulation as assessed by western blot analysis using anti-p21 antibodies (Fig. 3A and data not shown). These results indicate that CUGBP1 promotes p21 upregulation upon proteasome inhibition. As a first step to investigate how CUGBP1 up-regulates p21 upon proteasome inhibition, we performed co- localization studies between CUGBP1 and p21 mRNA. Under normal growth conditions, CUGBP1 is mostly nuclear [32], [56] (Fig. S4C, panel 2). Arsenite induces the localization of a fraction of CUGBP1 in SG [32], [56], and our immunofluorescence experiments showed that CUGBP1 was also recruited into bortezomib-induced SG (Fig. 3B). Quantification of these immunofluorescence data show that a significant fraction (∼45%) of total CUGBP1 is present in SG (Fig. 3B–C) where it co-localized with p21 mRNA (Fig. 3B and 3D–E) following treatment with bortezomib. Under these conditions, about 50% of CUGBP1 is nuclear (Fig. 3B–C) and only residual staining of CUGBP1 is detected in the cytoplasm (Fig. 3B and 3D). Prolonged treatment with bortezomib (10 h) resulted however in nuclear localization of most CUGBP1 (Fig. S4C). At this time we detected also some p21 mRNA FISH signal in the nucleus. However, and as mentioned above, the overall p21 mRNA FISH signal at 10-h bortezomib treatment is reduced (Fig. S4A and S4C), as compared to 4 h treatment (Fig. S4A; see also Fig. 2A) at which time point most of p21 mRNA (∼80%) co-localized with CUGBP1 in SG (Fig. 3B and 3D–E). Overall, our results (Fig. 3B) indicate that bortezomib treatment induces CUGBP1-p21 mRNA co-localization in SG. CUGBP1 might thus bind to p21 mRNA and recruit it to SG where it accumulates. The latter assumption would therefore predict that CUGBP1 depletion should trigger p21 mRNA degradation, which would explain the low amount of p21 protein following prolonged proteasome inhibition in CUGBP1-depleted cells (Fig. 3A). Indeed, we first confirmed that CUGBP1 binds to p21 mRNA under bortezomib conditions, as assessed by immunoprecipitating CUGBP1 (Fig. 3F) followed by quantitative RT-PCR (qRT-PCR) to detect associated p21 mRNA (Fig. 3G). This binding was specific, as suggested by the finding that control p27 mRNA was not recovered in CUGBP1 immunoprecipitates (Fig. 3G). Second, the steady-state p21 mRNA level significantly decreased in CUGBP1-depleted cells following proteasome inhibition (Fig. 3H). Depletion of CUGBP1 had however no effect on the steady-state level of p21 mRNA in untreated cells (Fig. S6A), suggesting that CUGBP1 is not required to maintain the basal level of the p21 mRNA. We then tested the possibility that the decreased level of p21 mRNA in CUGBP1-depleted cells upon bortezomib treatment might be consistent with a more rapid degradation of p21 mRNA. Indeed, the half-life of p21 mRNA in bortezomib conditions decreased from >3 h in mock-depleted cells (Fig. 4A) to 90 min ±15 min in CUGBP1-depleted cells (Fig. 4B). Depletion of CUGBP1 was efficient, as evidenced by both western blot analysis (Fig. 4C) and immunofluorescence (Fig. 4D), using anti-CUGBP1 antibodies. However, quantification of the localization of several SG markers (Fig. 4D, and data not shown) indicated that CUGBP1 depletion per se did not affect SG formation. This result ruled out the possibility that the destabilization of p21 mRNA observed in CUGBP1-depleted cells upon bortezomib treatment (Fig. 4A–B) was due to a cellular failure to form SG. One likely possibility of the destabilisation of p21 mRNA observed in CUGBP1-depleted cells is because it is no longer recruited to SG. We then assessed if CUGBP1 depletion affects the accumulation of p21 mRNA in SG using FISH. SG were visualized by immunofluorescence using antibodies to either Ras GTPase-activating protein-binding protein 1 G3BP1 (Fig. 4D) or FXR1 (data not shown). Whereas the major fraction of p21 mRNA (∼70%) is present in SG in over 60% of mock-depleted cells upon treatment with bortezomib (Fig. 4D–E), less than 30% of CUGBP1-depleted cells only exhibited such p21 mRNA localization in SG (Fig. 4D). Indeed, the large fraction of CUGBP1-depleted cells (∼80%) has a barely detectable FISH p21 mRNA signal in SG and in the cytoplasm (Fig. 4D and 4F). Under those bortezomib conditions, quantification of the intensity of p21 mRNA FISH signal showed that CUGBP1 depletion decreased this signal to 30–40%, as compared to mock-depleted cells where most p21 mRNA FISH signal was detected in SG (Fig. 4G). This effect seems to be specific since under the same conditions, depletion of CUGBP1 does not affect the association of the non-target GAPDH mRNA [47] with SG as assessed by immunofluorescence coupled with FISH (Fig. S7). Overall, our results show that CUGBP1 depletion significantly prevented bortezomib-induced p21 mRNA accumulation, which is likely due to its mistargeting in SG. As aforementioned, CUGBP1 depletion does not affect p21 mRNA steady-state levels in normal growth conditions (Fig. S6A) and we found that this is also the case under staurosporine treatment (Fig. S6B), a SG-free apoptotic condition [70] (Fig. S6D). These results suggest that CUGBP1 is not required to maintain the steady-state levels of p21 mRNA in the cytoplasm. On contrary, CUGBP1 seems to be required for bortezomib-induced p21 mRNA accumulation in SG (Fig. 3H, 4D, and S6C). We suggest that CUGBP1 promotes p21 mRNA accumulation by recruiting the transcript in SG, although we do not exclude that CUGBP1-mediated p21 mRNA accumulation involves additional mechanisms. Nevertheless, these results showed that CUGBP1 acts as a positive effector of p21 upregulation upon proteasome inhibition.

Bottom Line: We found that depleting CUGBP1 in cancer cells prevents bortezomib-mediated p21 upregulation.FISH experiments combined to mRNA stability assays show that this effect is largely due to a mistargeting of p21 mRNA in stress granules leading to its degradation.We propose that one key mechanism by which apoptosis is inhibited upon treatment with chemotherapeutic drugs might involve upregulation of the p21 protein through CUGBP1.

View Article: PubMed Central - PubMed

Affiliation: Department of Molecular Biology, Medical Biochemistry, and Pathology, Faculty of Medicine, Laval University, CHUQ Research Centre/St-François d'Assise Research Centre (CRCHUQ/CRSFA), Quebec, Canada.

ABSTRACT

Background: p21(WAF1/CIP1) is a well known cyclin-dependent kinase inhibitor induced by various stress stimuli. Depending on the stress applied, p21 upregulation can either promote apoptosis or prevent against apoptotic injury. The stress-mediated induction of p21 involves not only its transcriptional activation but also its posttranscriptional regulation, mainly through stabilization of p21 mRNA levels. We have previously reported that the proteasome inhibitor MG132 induces the stabilization of p21 mRNA, which correlates with the formation of cytoplasmic RNA stress granules. The mechanism underlying p21 mRNA stabilization, however, remains unknown.

Methodology/principal findings: We identified the stress granules component CUGBP1 as a factor required for p21 mRNA stabilization following treatment with bortezomib ( =  PS-341/Velcade). This peptide boronate inhibitor of the 26S proteasome is very efficient for the treatment of myelomas and other hematological tumors. However, solid tumors are sometimes refractory to bortezomib treatment. We found that depleting CUGBP1 in cancer cells prevents bortezomib-mediated p21 upregulation. FISH experiments combined to mRNA stability assays show that this effect is largely due to a mistargeting of p21 mRNA in stress granules leading to its degradation. Altering the expression of p21 itself, either by depleting CUGBP1 or p21, promotes bortezomib-mediated apoptosis.

Conclusions/significance: We propose that one key mechanism by which apoptosis is inhibited upon treatment with chemotherapeutic drugs might involve upregulation of the p21 protein through CUGBP1.

Show MeSH
Related in: MedlinePlus