Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2000 Dec 15;28(24):4987-97.
doi: 10.1093/nar/28.24.4987.

CHOP gene expression in response to endoplasmic-reticular stress requires NFY interaction with different domains of a conserved DNA-binding element

M Ubeda  1 , J F Habener
Affiliations

CHOP gene expression in response to endoplasmic-reticular stress requires NFY interaction with different domains of a conserved DNA-binding element

M Ubeda et al. Nucleic Acids Res. .

Abstract

The transcription factor CHOP/GADD153 gene is induced by cellular stress and is involved in mediating apoptosis. We report the identification of a conserved region in the promoter of the CHOP gene responsible for its inducibility by endoplasmic reticulum (ER) stress. Deletion mutants of the human CHOP promoter identify a region comprising nucleotides -75 to -104 required for both constitutive and ER-stress-inducible expression. This region of the promoter, the ER-stress element (ERSE) is sufficient to confer both increased basal activity and ER-stress inducibility to an otherwise inactive heterologous promoter. The CHOP ERSE is a novel variant of the ERSE as it contains two different functional domains, and a GA- instead of GC-rich intervening sequence. The CCAAT-box domain occupied by the constitutive transcriptional activator nuclear factor Y (NFY) is required for constitutive activation whereas the variant GCACG 'inducible' domain uniquely mediates ER-stress inducibility. By UV-crosslinking analysis NFY makes contact not only with the constitutive activator CCAAT box but also with the inducible GCACG domain. Deletions and nucleotide substitutions in the CCAAT box as well as its replacement by an SP1 site failed to support ER inducibility. These findings support the notion that NFY is not only required for constitutive activation of CHOP gene transcription, but is also an active and essential element for the assembly of an ER-stress-inducible enhanceosome that activates CHOP gene expression in response to cellular stress.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Induction of the CHOP gene expression by ER-stress maps to a promoter region homologous to the ERSE elements identified in the GRPs, PDI and calreticulin genes. (A) Tunicamycin treatment (5 µg/ml) induces a robust induction of CHOP in NIH 3T3 fibroblasts. Western blot experiments using a specific CHOP and a control CREB antisera were performed on nuclear extracts from NIH 3T3 cells treated with tunicamycin for the indicated periods of time. (B) Transcriptional activities of different CAT reporters in which activity was controlled by different promoter deletions of the human CHOP gene. Tunicamycin treatment was initiated 16 h prior to harvesting the cells. Maps of the –104 bp CHOP-CAT, –75 CHOP-CAT and the mutated form (MUT CHOP-CAT) are also shown. The reporter activity was measured and data from three independent experiments performed in duplicate were plotted. Data shown correspond to the mean ± SEM. The specific mutated sequence in the MUT CHOP-CAT construct is shown in (C). (C) Sequence analysis of the –104 –75 region of the CHOP promoter identifies two overlapping elements matching the ERSE consensus. This consensus was previously defined based on the sequence of the GRPs, PDI and calreticulin promoters (29,30).
Figure 1
Figure 1
Induction of the CHOP gene expression by ER-stress maps to a promoter region homologous to the ERSE elements identified in the GRPs, PDI and calreticulin genes. (A) Tunicamycin treatment (5 µg/ml) induces a robust induction of CHOP in NIH 3T3 fibroblasts. Western blot experiments using a specific CHOP and a control CREB antisera were performed on nuclear extracts from NIH 3T3 cells treated with tunicamycin for the indicated periods of time. (B) Transcriptional activities of different CAT reporters in which activity was controlled by different promoter deletions of the human CHOP gene. Tunicamycin treatment was initiated 16 h prior to harvesting the cells. Maps of the –104 bp CHOP-CAT, –75 CHOP-CAT and the mutated form (MUT CHOP-CAT) are also shown. The reporter activity was measured and data from three independent experiments performed in duplicate were plotted. Data shown correspond to the mean ± SEM. The specific mutated sequence in the MUT CHOP-CAT construct is shown in (C). (C) Sequence analysis of the –104 –75 region of the CHOP promoter identifies two overlapping elements matching the ERSE consensus. This consensus was previously defined based on the sequence of the GRPs, PDI and calreticulin promoters (29,30).
Figure 1
Figure 1
Induction of the CHOP gene expression by ER-stress maps to a promoter region homologous to the ERSE elements identified in the GRPs, PDI and calreticulin genes. (A) Tunicamycin treatment (5 µg/ml) induces a robust induction of CHOP in NIH 3T3 fibroblasts. Western blot experiments using a specific CHOP and a control CREB antisera were performed on nuclear extracts from NIH 3T3 cells treated with tunicamycin for the indicated periods of time. (B) Transcriptional activities of different CAT reporters in which activity was controlled by different promoter deletions of the human CHOP gene. Tunicamycin treatment was initiated 16 h prior to harvesting the cells. Maps of the –104 bp CHOP-CAT, –75 CHOP-CAT and the mutated form (MUT CHOP-CAT) are also shown. The reporter activity was measured and data from three independent experiments performed in duplicate were plotted. Data shown correspond to the mean ± SEM. The specific mutated sequence in the MUT CHOP-CAT construct is shown in (C). (C) Sequence analysis of the –104 –75 region of the CHOP promoter identifies two overlapping elements matching the ERSE consensus. This consensus was previously defined based on the sequence of the GRPs, PDI and calreticulin promoters (29,30).
Figure 2
Figure 2
The CHOP ERSE element contains in itself the information required for ER-stress inducibility and constitutive activation. (A) CAT activity assayed in cellular extracts from NIH 3T3 cells transfected with the corresponding reporter constructs and treated with tunicamycin as indicated in Figure 1. ‘Empty reporter’ is the –41 TK-CAT alone. The wild-type (WT ERSE-CAT) and mutant (MUT ERSE-CAT) constructs contain inserts of the CHOP ERSE and CHOP ERSE MUT (sequences shown in Fig. 1C) upstream of the –41 TK-CAT. Each experiment was performed in duplicate. Tunicamycin-inducibility of the endogenous CHOP gene was determined by western blot assay of the same cell extracts in which the CAT activities were determined thereby serving as an internal control for ER-stress induction. (B) The level of CAT activity of each of the reporter constructs was calculated based on three independent experiments performed in duplicate and the results were plotted as the mean ± SEM.
Figure 2
Figure 2
The CHOP ERSE element contains in itself the information required for ER-stress inducibility and constitutive activation. (A) CAT activity assayed in cellular extracts from NIH 3T3 cells transfected with the corresponding reporter constructs and treated with tunicamycin as indicated in Figure 1. ‘Empty reporter’ is the –41 TK-CAT alone. The wild-type (WT ERSE-CAT) and mutant (MUT ERSE-CAT) constructs contain inserts of the CHOP ERSE and CHOP ERSE MUT (sequences shown in Fig. 1C) upstream of the –41 TK-CAT. Each experiment was performed in duplicate. Tunicamycin-inducibility of the endogenous CHOP gene was determined by western blot assay of the same cell extracts in which the CAT activities were determined thereby serving as an internal control for ER-stress induction. (B) The level of CAT activity of each of the reporter constructs was calculated based on three independent experiments performed in duplicate and the results were plotted as the mean ± SEM.
Figure 3
Figure 3
Identification of NFY as a transcription factor that binds the CHOP ERSE element in both basal and ER-stress stimulated conditions. (A) EMSA experiments performed using nuclear extracts obtained from control (–) and tunicamycin (+) treated NIH 3T3 cells. Nuclear extracts were incubated with the CHOP ERSE in the conditions described in the Materials and Methods section. Results show a similar DNA–protein-binding pattern in both control and tunicamycin-treated conditions. The most prominent complexes are called A and B, respectively. Super-shift experiments using specific antibodies for candidate factors identify the transcription factors NFY-A and NFY-B as the major component of A and B complexes (arrows). Nuclear extracts were incubated in the presence of specific antisera for 15 min prior to adding the DNA probe. (B) Competition experiments using increasing amounts of unlabeled CHOP ERSE (ERSE) and its corresponding mutant ERSE (MUT) oligonucleotides (the sequence of the ERSE and MUT is depicted in Fig. 1C). Unlabeled competitor oligonucleotides were used at 10- to 100-fold molar excess. Extracts are from cells treated with tunicamycin.
Figure 3
Figure 3
Identification of NFY as a transcription factor that binds the CHOP ERSE element in both basal and ER-stress stimulated conditions. (A) EMSA experiments performed using nuclear extracts obtained from control (–) and tunicamycin (+) treated NIH 3T3 cells. Nuclear extracts were incubated with the CHOP ERSE in the conditions described in the Materials and Methods section. Results show a similar DNA–protein-binding pattern in both control and tunicamycin-treated conditions. The most prominent complexes are called A and B, respectively. Super-shift experiments using specific antibodies for candidate factors identify the transcription factors NFY-A and NFY-B as the major component of A and B complexes (arrows). Nuclear extracts were incubated in the presence of specific antisera for 15 min prior to adding the DNA probe. (B) Competition experiments using increasing amounts of unlabeled CHOP ERSE (ERSE) and its corresponding mutant ERSE (MUT) oligonucleotides (the sequence of the ERSE and MUT is depicted in Fig. 1C). Unlabeled competitor oligonucleotides were used at 10- to 100-fold molar excess. Extracts are from cells treated with tunicamycin.
Figure 4
Figure 4
Of the two putative ERSE elements in the human CHOP promoter, only the one localized to the antisense DNA strand is functionally active. (A) Sequence comparison of the DNA sequences of the human, mouse and hamster promoters indicates conservation of the ERSE motif localized to the antisense DNA-strand but not of that localized to the sense strand. The human and hamster sequences were obtained from GenBank and the mouse sequence from our laboratory (39). (B) CAT reporter constructs were prepared by introducing the depicted oligonucleotides upstream of the TATA box in the –41 TK-CAT reporter. (C) Activity of the different constructs once transfected into NIH 3T3 cells. Tunicamycin treatment was as described in Figure 1. Results are expressed as the mean ± SEM of three independent experiments performed in duplicate. (D) The CHOP ERSE element belongs to a subgroup of ER-stress-inducible elements and differs from the reported consensus by having an AG- instead of a GC-rich spacer, and a G nucleotide substitution in the first position of the inducible domain. The domains required for constitutive activation ‘c’ and ER-stress inducibility ‘i’ were identified in our mutagenesis experiments (see also Fig. 5A).
Figure 4
Figure 4
Of the two putative ERSE elements in the human CHOP promoter, only the one localized to the antisense DNA strand is functionally active. (A) Sequence comparison of the DNA sequences of the human, mouse and hamster promoters indicates conservation of the ERSE motif localized to the antisense DNA-strand but not of that localized to the sense strand. The human and hamster sequences were obtained from GenBank and the mouse sequence from our laboratory (39). (B) CAT reporter constructs were prepared by introducing the depicted oligonucleotides upstream of the TATA box in the –41 TK-CAT reporter. (C) Activity of the different constructs once transfected into NIH 3T3 cells. Tunicamycin treatment was as described in Figure 1. Results are expressed as the mean ± SEM of three independent experiments performed in duplicate. (D) The CHOP ERSE element belongs to a subgroup of ER-stress-inducible elements and differs from the reported consensus by having an AG- instead of a GC-rich spacer, and a G nucleotide substitution in the first position of the inducible domain. The domains required for constitutive activation ‘c’ and ER-stress inducibility ‘i’ were identified in our mutagenesis experiments (see also Fig. 5A).
Figure 4
Figure 4
Of the two putative ERSE elements in the human CHOP promoter, only the one localized to the antisense DNA strand is functionally active. (A) Sequence comparison of the DNA sequences of the human, mouse and hamster promoters indicates conservation of the ERSE motif localized to the antisense DNA-strand but not of that localized to the sense strand. The human and hamster sequences were obtained from GenBank and the mouse sequence from our laboratory (39). (B) CAT reporter constructs were prepared by introducing the depicted oligonucleotides upstream of the TATA box in the –41 TK-CAT reporter. (C) Activity of the different constructs once transfected into NIH 3T3 cells. Tunicamycin treatment was as described in Figure 1. Results are expressed as the mean ± SEM of three independent experiments performed in duplicate. (D) The CHOP ERSE element belongs to a subgroup of ER-stress-inducible elements and differs from the reported consensus by having an AG- instead of a GC-rich spacer, and a G nucleotide substitution in the first position of the inducible domain. The domains required for constitutive activation ‘c’ and ER-stress inducibility ‘i’ were identified in our mutagenesis experiments (see also Fig. 5A).
Figure 4
Figure 4
Of the two putative ERSE elements in the human CHOP promoter, only the one localized to the antisense DNA strand is functionally active. (A) Sequence comparison of the DNA sequences of the human, mouse and hamster promoters indicates conservation of the ERSE motif localized to the antisense DNA-strand but not of that localized to the sense strand. The human and hamster sequences were obtained from GenBank and the mouse sequence from our laboratory (39). (B) CAT reporter constructs were prepared by introducing the depicted oligonucleotides upstream of the TATA box in the –41 TK-CAT reporter. (C) Activity of the different constructs once transfected into NIH 3T3 cells. Tunicamycin treatment was as described in Figure 1. Results are expressed as the mean ± SEM of three independent experiments performed in duplicate. (D) The CHOP ERSE element belongs to a subgroup of ER-stress-inducible elements and differs from the reported consensus by having an AG- instead of a GC-rich spacer, and a G nucleotide substitution in the first position of the inducible domain. The domains required for constitutive activation ‘c’ and ER-stress inducibility ‘i’ were identified in our mutagenesis experiments (see also Fig. 5A).
Figure 5
Figure 5
Two different domains are identified in the CHOP ERSE element. One is required for ER-stress inducibility and the other for constitutive activation. (A) Schematic representation of the constructs transfected into NIH 3T3 cells and CAT activity determined for each of the constructs. The Mut21 contains a gcac instead of CGTG substitution in its inducible domain. The MutAS has a ggtt instead of TTGG substitution in the functional inverted CCAAT motif. (B) EMSA experiments in which both the wild-type CHOP ERSE (WT) and the Mut21 (MUT21) oligonucleotides were labeled and incubated in the presence of nuclear extracts from control and tunicamycin-treated cells. Arrows indicate the CBF/NFY-containing band (NFY) and the band super-shifted by the NFY-A antibody (SS).
Figure 5
Figure 5
Two different domains are identified in the CHOP ERSE element. One is required for ER-stress inducibility and the other for constitutive activation. (A) Schematic representation of the constructs transfected into NIH 3T3 cells and CAT activity determined for each of the constructs. The Mut21 contains a gcac instead of CGTG substitution in its inducible domain. The MutAS has a ggtt instead of TTGG substitution in the functional inverted CCAAT motif. (B) EMSA experiments in which both the wild-type CHOP ERSE (WT) and the Mut21 (MUT21) oligonucleotides were labeled and incubated in the presence of nuclear extracts from control and tunicamycin-treated cells. Arrows indicate the CBF/NFY-containing band (NFY) and the band super-shifted by the NFY-A antibody (SS).
Figure 6
Figure 6
CBF/NFY binding to the CHOP ERSE element is required for ER-stress inducibility and is not functionally substituted by another constitutive activator such as SP1. Schematic representation of the constructs used in these experiments and CAT activity determined in three independent experiments. Data are expressed as the mean ± SEM. Each experiment was performed in duplicate. The full sequence of the different oligonucleotides is described in the Materials and Methods. CR-ERSE contains the inducible domain and all the flanking sequence with the exception of the CCAAT boxes of the human CHOP ERSE. The same sequence is repeated five times in the 5×-CR-ERSE. In the SP1-ERSE the flanking CCAAT boxes are substituted by respective SP1 binding sites. SP1 binding was tested in gel shift experiments (not shown).
Figure 7
Figure 7
The transcription factor CBF/NFY also interacts with the inducible domain of the CHOP ERSE. (A) A CHOP ERSE nucleotide containing a BrdU substitution in its inducible (‘i’) domain and an unsubstituted control oligonucleotide were used in these experiments. (B) Purified recombinant NFY/CBF generates a major EMSA shifting complex in both the presence of the BrdU substituted and unsubstituted probes. This complex is equivalent to the major A complex in NIH 3T3 nuclear extracts containing NFY as indicated by its specific recognition by an NFY antiserum (see Figs 3A and 5B). (C) After UV-crosslinking, the A complex was excised and its components were separated on a denaturing polyacrylamide gel (SDS–PAGE). Three different proteins crosslink specifically to the BrdU substituted probe (1) and not to the unsubstituted control probe (2). Both nuclear extracts (N.E.) and recombinant purified NFY/CBF (rNFY) show a similar crosslinking pattern that corresponds to the three NFY/CBF subunits with approximate molecular weights of 60 (NFY-B), 40 (NFY-A) and 32 kDa (NFY-C).
Figure 7
Figure 7
The transcription factor CBF/NFY also interacts with the inducible domain of the CHOP ERSE. (A) A CHOP ERSE nucleotide containing a BrdU substitution in its inducible (‘i’) domain and an unsubstituted control oligonucleotide were used in these experiments. (B) Purified recombinant NFY/CBF generates a major EMSA shifting complex in both the presence of the BrdU substituted and unsubstituted probes. This complex is equivalent to the major A complex in NIH 3T3 nuclear extracts containing NFY as indicated by its specific recognition by an NFY antiserum (see Figs 3A and 5B). (C) After UV-crosslinking, the A complex was excised and its components were separated on a denaturing polyacrylamide gel (SDS–PAGE). Three different proteins crosslink specifically to the BrdU substituted probe (1) and not to the unsubstituted control probe (2). Both nuclear extracts (N.E.) and recombinant purified NFY/CBF (rNFY) show a similar crosslinking pattern that corresponds to the three NFY/CBF subunits with approximate molecular weights of 60 (NFY-B), 40 (NFY-A) and 32 kDa (NFY-C).
Figure 7
Figure 7
The transcription factor CBF/NFY also interacts with the inducible domain of the CHOP ERSE. (A) A CHOP ERSE nucleotide containing a BrdU substitution in its inducible (‘i’) domain and an unsubstituted control oligonucleotide were used in these experiments. (B) Purified recombinant NFY/CBF generates a major EMSA shifting complex in both the presence of the BrdU substituted and unsubstituted probes. This complex is equivalent to the major A complex in NIH 3T3 nuclear extracts containing NFY as indicated by its specific recognition by an NFY antiserum (see Figs 3A and 5B). (C) After UV-crosslinking, the A complex was excised and its components were separated on a denaturing polyacrylamide gel (SDS–PAGE). Three different proteins crosslink specifically to the BrdU substituted probe (1) and not to the unsubstituted control probe (2). Both nuclear extracts (N.E.) and recombinant purified NFY/CBF (rNFY) show a similar crosslinking pattern that corresponds to the three NFY/CBF subunits with approximate molecular weights of 60 (NFY-B), 40 (NFY-A) and 32 kDa (NFY-C).
Figure 8
Figure 8
Model depicting the differential interaction of NFY with the α1 collagen promoter in which it serves only as a constitutive activator, and with the CHOP promoter in which it is required not only for constitutive activation but also ER-stress inducibility. It is proposed that the ER-stress inducible ‘i’ domain locates right to a DNA region in which NFY makes contact. Both the specific sequence of the ‘i’ domain and the region of NFY that makes contact with such a domain contribute to the creation of a new interaction surface for a putative ERSF, equivalent to the yeast Hac1p.
See this image and copyright information in PMC

References

    1. Bartlett J., Luethy,J., Carlson,S., Sollott,S. and Holbrook,N.J. (1992) J. Biol. Chem., 267, 20465–20470. - PubMed
    1. Chen Q., Yu,K., Holbrook,N.J. and Stevens,J.L. (1992) J. Biol. Chem., 267, 8207–8212. - PubMed
    1. Price B. and Calderwood,S. (1992) Cancer Res., 52, 3814–3817. - PubMed
    1. Wang X.-Z., Lawson,B., Brewer,J., Zinszner,H., Sanjay,A., Mi,L., Boorstein,R., Kreibich,G., Hendershot,L. and Ron,D. (1996) Mol. Cell. Biol., 16, 4273–4280. - PMC - PubMed
    1. Kaufman R.J. (1999) Genes Dev., 13, 1211–1233. - PubMed

MeSH terms