Limits...
Sumoylation delays the ATF7 transcription factor subcellular localization and inhibits its transcriptional activity.

Hamard PJ, Boyer-Guittaut M, Camuzeaux B, Dujardin D, Hauss C, Oelgeschläger T, Vigneron M, Kedinger C, Chatton B - Nucleic Acids Res. (2007)

Bottom Line: This protein is able to heterodimerize with Jun or Fos proteins and its transcriptional activity is mediated by interaction with TAF12, a subunit of the general transcription factor TFIID.In the present article, we demonstrate that ATF7 is sumoylated in vitro (using RanBP2 as a E3-specific ligase) and in vivo.Moreover, we show that ATF7 sumoylation affects its intranuclear localization by delaying its entry into the nucleus.

View Article: PubMed Central - PubMed

Affiliation: Institut Gilbert Laustriat, Ecole Supérieure de Biotechnologie de Strasbourg, UMR7175 CNRS-ULP, BP10413, 67412 Strasbourg Illkirch Cedex, France.

ABSTRACT
Over the past few years, small ubiquitin-like modifier (SUMO) modification has emerged as an important regulator of diverse pathways and activities including protein localization and transcriptional regulation. We identified a consensus sumoylation motif (IKEE), located within the N-terminal activation domain of the ATF7 transcription factor and thus investigated the role of this modification. ATF7 is a ubiquitously expressed transcription factor, homologous to ATF2, that binds to CRE elements within specific promoters. This protein is able to heterodimerize with Jun or Fos proteins and its transcriptional activity is mediated by interaction with TAF12, a subunit of the general transcription factor TFIID. In the present article, we demonstrate that ATF7 is sumoylated in vitro (using RanBP2 as a E3-specific ligase) and in vivo. Moreover, we show that ATF7 sumoylation affects its intranuclear localization by delaying its entry into the nucleus. Furthermore, SUMO conjugation inhibits ATF7 transactivation activity by (i) impairing its association with TAF12 and (ii) blocking its binding-to-specific sequences within target promoters.

Show MeSH

Related in: MedlinePlus

Sumoylation affects ATF7 activity. (A) ChIP analysis of genomic E-selectin promoter: HeLa-SUMO cells were transfected with px-ATF7WT(lanes 1,3) or px- ATF7K118R (lanes 2,4). Two days later, cells were subjected to ChIP assay with antibodies against ATF7 (lanes 1 and 2) or Flag (lanes 3 and 4). Endogenous E-selectin promoter DNA, coprecipitated with the indicated antibodies, was detected by PCR. (B) Recombinant pSG5-based vectors (0.5 µg) directing the expression of the various chimeras were transfected into HeLa-SUMO cells, and cell lysates were analyzed by EMSA: 20 µg of extracts were preincubated with either 50 ng of non-specific (mt) or specific (wt) competitor before addition of E-selectin CRE probe. Where indicated, reactions were further incubated (30 min, 0°C) with the anti-ATF7 antibody before loading on the gel. The black arrowhead points to the specific complex and the white one to the supershifted one. (C) Recombinant pX-based vectors (0.5 µg) directing the expression of the various chimeras were transfected into HUVEC-C-cells. Two days later, RNA was isolated and E-selectin mRNA expression was analyzed by semi-quantitative RT–PCR method using β-actin as internal control. Ethidium bromide-stained gel is shown in the figure.
© Copyright Policy - openaccess
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC1851647&req=5

Figure 6: Sumoylation affects ATF7 activity. (A) ChIP analysis of genomic E-selectin promoter: HeLa-SUMO cells were transfected with px-ATF7WT(lanes 1,3) or px- ATF7K118R (lanes 2,4). Two days later, cells were subjected to ChIP assay with antibodies against ATF7 (lanes 1 and 2) or Flag (lanes 3 and 4). Endogenous E-selectin promoter DNA, coprecipitated with the indicated antibodies, was detected by PCR. (B) Recombinant pSG5-based vectors (0.5 µg) directing the expression of the various chimeras were transfected into HeLa-SUMO cells, and cell lysates were analyzed by EMSA: 20 µg of extracts were preincubated with either 50 ng of non-specific (mt) or specific (wt) competitor before addition of E-selectin CRE probe. Where indicated, reactions were further incubated (30 min, 0°C) with the anti-ATF7 antibody before loading on the gel. The black arrowhead points to the specific complex and the white one to the supershifted one. (C) Recombinant pX-based vectors (0.5 µg) directing the expression of the various chimeras were transfected into HUVEC-C-cells. Two days later, RNA was isolated and E-selectin mRNA expression was analyzed by semi-quantitative RT–PCR method using β-actin as internal control. Ethidium bromide-stained gel is shown in the figure.

Mentions: We previously showed that the transcriptional activity of ATF7 is potentiated by expression of TAF12 and that the two proteins are interacting in vivo (14). We investigated the ability of the fusion SUMO-GA-ATF7 and TAF12 proteins to interact within a cellular context. To that end, HeLa-SUMO cells were transfected with plasmids expressing p-hsTAF12 together with px-ATF7, px-ATF7K118R and px-SUMO-GA-ATF7K118R (Figure 5B). Whole cell extracts were then submitted to immunoprecipitation (IP) with antibodies against ATF7, and the precipitated proteins were analyzed by western blotting using anti-TAF12 antibodies. As shown in Figure 5B (lanes 1–4), TAF12 was detected in the immunoprecipitates only when coexpressed with ATF7WT or ATF7K118R, but not when coexpressed with the constitutively sumoylated ATF7 construct. In HeLa-SUMO transfected cells, the majority of the overexpressed ATF7WT protein is not sumoylated. Therefore, these results strongly suggest that sumoylation may prevent the interaction between ATF7 and TAF12, and this loss of interaction would thus account for the transcriptional inhibitory effect of ATF7 sumoylation. To gain some insight into the molecular mechanism of this differential action on transcription, we first performed a chromatin immunoprecipitation (ChIP) experiment on the E-selectin gene. ATF7 has previously been shown to specifically bind to the NF-ELAM1 promoter element of the E-selectin gene and to interact with TAF12 (14). HeLa-SUMO cells were transfected with vectors expressing either ATF7WT, K118R ATF7, or SUMO-GA-ATF7K118R under conditions where equal levels of proteins were expressed, as verified by immunoblotting with specific antibodies (Figure 6A). We also checked by gel-shift assays that the three proteins were able to bind the NF-ELAM1 promoter element with very similar efficiencies (Figure 6B). As shown in Figure 6A, the E-selectin promoter was immunoprecipitated more efficently in ATF7K118R expressing cells than in ATF7WT or SUMO-GA-ATF7K118R transfected cells. These results clearly indicate that, in the context of a natural promoter (and in a steady-state level), non-sumoylated ATF7 form is more strongly associated with the chromatin. Therefore sumoylation prevents the association of ATF7 with the target promoter or facilitates its removal in vivo. In addition, a direct measurement of transcription was performed to analyze the changes in an ATF7 target gene expression upon the changes in ATF7 sumoylation. RT-PCR analysis clearly shows that transcription of the E-selectin gene is enhanced in the presence of ATF7K118R (Figure 6C) whereas the expression level of the control gene (β-actin) is unaffected. Altogether, these observations clearly confirm the inhibitory effect of sumoylation of ATF7 on the transactivation process: sumoylation not only impairs the interaction between ATF7 and TAF12 but blocks the access of ATF7 to promoters.Figure 6.


Sumoylation delays the ATF7 transcription factor subcellular localization and inhibits its transcriptional activity.

Hamard PJ, Boyer-Guittaut M, Camuzeaux B, Dujardin D, Hauss C, Oelgeschläger T, Vigneron M, Kedinger C, Chatton B - Nucleic Acids Res. (2007)

Sumoylation affects ATF7 activity. (A) ChIP analysis of genomic E-selectin promoter: HeLa-SUMO cells were transfected with px-ATF7WT(lanes 1,3) or px- ATF7K118R (lanes 2,4). Two days later, cells were subjected to ChIP assay with antibodies against ATF7 (lanes 1 and 2) or Flag (lanes 3 and 4). Endogenous E-selectin promoter DNA, coprecipitated with the indicated antibodies, was detected by PCR. (B) Recombinant pSG5-based vectors (0.5 µg) directing the expression of the various chimeras were transfected into HeLa-SUMO cells, and cell lysates were analyzed by EMSA: 20 µg of extracts were preincubated with either 50 ng of non-specific (mt) or specific (wt) competitor before addition of E-selectin CRE probe. Where indicated, reactions were further incubated (30 min, 0°C) with the anti-ATF7 antibody before loading on the gel. The black arrowhead points to the specific complex and the white one to the supershifted one. (C) Recombinant pX-based vectors (0.5 µg) directing the expression of the various chimeras were transfected into HUVEC-C-cells. Two days later, RNA was isolated and E-selectin mRNA expression was analyzed by semi-quantitative RT–PCR method using β-actin as internal control. Ethidium bromide-stained gel is shown in the figure.
© Copyright Policy - openaccess
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC1851647&req=5

Figure 6: Sumoylation affects ATF7 activity. (A) ChIP analysis of genomic E-selectin promoter: HeLa-SUMO cells were transfected with px-ATF7WT(lanes 1,3) or px- ATF7K118R (lanes 2,4). Two days later, cells were subjected to ChIP assay with antibodies against ATF7 (lanes 1 and 2) or Flag (lanes 3 and 4). Endogenous E-selectin promoter DNA, coprecipitated with the indicated antibodies, was detected by PCR. (B) Recombinant pSG5-based vectors (0.5 µg) directing the expression of the various chimeras were transfected into HeLa-SUMO cells, and cell lysates were analyzed by EMSA: 20 µg of extracts were preincubated with either 50 ng of non-specific (mt) or specific (wt) competitor before addition of E-selectin CRE probe. Where indicated, reactions were further incubated (30 min, 0°C) with the anti-ATF7 antibody before loading on the gel. The black arrowhead points to the specific complex and the white one to the supershifted one. (C) Recombinant pX-based vectors (0.5 µg) directing the expression of the various chimeras were transfected into HUVEC-C-cells. Two days later, RNA was isolated and E-selectin mRNA expression was analyzed by semi-quantitative RT–PCR method using β-actin as internal control. Ethidium bromide-stained gel is shown in the figure.
Mentions: We previously showed that the transcriptional activity of ATF7 is potentiated by expression of TAF12 and that the two proteins are interacting in vivo (14). We investigated the ability of the fusion SUMO-GA-ATF7 and TAF12 proteins to interact within a cellular context. To that end, HeLa-SUMO cells were transfected with plasmids expressing p-hsTAF12 together with px-ATF7, px-ATF7K118R and px-SUMO-GA-ATF7K118R (Figure 5B). Whole cell extracts were then submitted to immunoprecipitation (IP) with antibodies against ATF7, and the precipitated proteins were analyzed by western blotting using anti-TAF12 antibodies. As shown in Figure 5B (lanes 1–4), TAF12 was detected in the immunoprecipitates only when coexpressed with ATF7WT or ATF7K118R, but not when coexpressed with the constitutively sumoylated ATF7 construct. In HeLa-SUMO transfected cells, the majority of the overexpressed ATF7WT protein is not sumoylated. Therefore, these results strongly suggest that sumoylation may prevent the interaction between ATF7 and TAF12, and this loss of interaction would thus account for the transcriptional inhibitory effect of ATF7 sumoylation. To gain some insight into the molecular mechanism of this differential action on transcription, we first performed a chromatin immunoprecipitation (ChIP) experiment on the E-selectin gene. ATF7 has previously been shown to specifically bind to the NF-ELAM1 promoter element of the E-selectin gene and to interact with TAF12 (14). HeLa-SUMO cells were transfected with vectors expressing either ATF7WT, K118R ATF7, or SUMO-GA-ATF7K118R under conditions where equal levels of proteins were expressed, as verified by immunoblotting with specific antibodies (Figure 6A). We also checked by gel-shift assays that the three proteins were able to bind the NF-ELAM1 promoter element with very similar efficiencies (Figure 6B). As shown in Figure 6A, the E-selectin promoter was immunoprecipitated more efficently in ATF7K118R expressing cells than in ATF7WT or SUMO-GA-ATF7K118R transfected cells. These results clearly indicate that, in the context of a natural promoter (and in a steady-state level), non-sumoylated ATF7 form is more strongly associated with the chromatin. Therefore sumoylation prevents the association of ATF7 with the target promoter or facilitates its removal in vivo. In addition, a direct measurement of transcription was performed to analyze the changes in an ATF7 target gene expression upon the changes in ATF7 sumoylation. RT-PCR analysis clearly shows that transcription of the E-selectin gene is enhanced in the presence of ATF7K118R (Figure 6C) whereas the expression level of the control gene (β-actin) is unaffected. Altogether, these observations clearly confirm the inhibitory effect of sumoylation of ATF7 on the transactivation process: sumoylation not only impairs the interaction between ATF7 and TAF12 but blocks the access of ATF7 to promoters.Figure 6.

Bottom Line: This protein is able to heterodimerize with Jun or Fos proteins and its transcriptional activity is mediated by interaction with TAF12, a subunit of the general transcription factor TFIID.In the present article, we demonstrate that ATF7 is sumoylated in vitro (using RanBP2 as a E3-specific ligase) and in vivo.Moreover, we show that ATF7 sumoylation affects its intranuclear localization by delaying its entry into the nucleus.

View Article: PubMed Central - PubMed

Affiliation: Institut Gilbert Laustriat, Ecole Supérieure de Biotechnologie de Strasbourg, UMR7175 CNRS-ULP, BP10413, 67412 Strasbourg Illkirch Cedex, France.

ABSTRACT
Over the past few years, small ubiquitin-like modifier (SUMO) modification has emerged as an important regulator of diverse pathways and activities including protein localization and transcriptional regulation. We identified a consensus sumoylation motif (IKEE), located within the N-terminal activation domain of the ATF7 transcription factor and thus investigated the role of this modification. ATF7 is a ubiquitously expressed transcription factor, homologous to ATF2, that binds to CRE elements within specific promoters. This protein is able to heterodimerize with Jun or Fos proteins and its transcriptional activity is mediated by interaction with TAF12, a subunit of the general transcription factor TFIID. In the present article, we demonstrate that ATF7 is sumoylated in vitro (using RanBP2 as a E3-specific ligase) and in vivo. Moreover, we show that ATF7 sumoylation affects its intranuclear localization by delaying its entry into the nucleus. Furthermore, SUMO conjugation inhibits ATF7 transactivation activity by (i) impairing its association with TAF12 and (ii) blocking its binding-to-specific sequences within target promoters.

Show MeSH
Related in: MedlinePlus