Molecular Dissection of the Human Ubiquitin C Promoter Reveals Heat Shock Element Architectures with Activating and Repressive Functions

Abstract

The promoter of the polyubiquitin C gene (UBC) contains putative heat shock elements (HSEs) which are thought to mediate UBC induction upon stress. However, the mapping and the functional characterization of the cis-acting determinants for its up-regulation have not yet been addressed. In this study, the sequence encompassing 916 nucleotides upstream of the transcription start site of the human UBC gene has been dissected by in silico, in vitro and in vivo approaches. The information derived from this analysis was used to study the functional role and the interplay of the identified HSEs in mediating the transcriptional activation of the UBC gene under conditions of proteotoxic stress, induced by the proteasome inhibitor MG132. Here we demonstrate that at least three HSEs, with different configurations, exist in the UBC promoter: two distal, residing within nucleotides -841/-817 and -715/-691, and one proximal to the transcription start site (nt -100/-65). All of them are bound by transcription factors belonging to the heat shock factor (HSF) family, as determined by bandshift, supershift and ChIP analyses. Site-directed mutagenesis of reporter constructs demonstrated that while the distal elements are involved in the up-regulation of UBC in response to proteasome inhibition, the proximal one appears rather to function as negative regulator of the stress-induced transcriptional activity. This is the first evidence that an HSE may exert a negative role on the transcription driven by other HSE motifs on the same gene promoter, highlighting a new level of complexity in the regulation of HSFs and in the control of ubiquitin levels.

Fig 1. Schematic representation of the cloned UBC promoter region.

The UBC promoter region (black box), the first non coding exon and the unique intron (grey box, the dashed box corresponds to exon 1), as they have been previously cloned into the pGL3-basic firefly reporter vector to generate the constructs named P1 (-916/+878) and P3 (-371/+878) (23). The transcription start site is marked as +1. Depicted to scale are also the PCR-generated DNA fragments used in EMSA. Promoter probes are indicated as FR1-6 (in black); exon/intron probes as probe I-VI (in grey). Position of the DNA segments is indicated. LUC, luciferase.

Fig 2. EMSA analysis of the UBC DNA fragments generated by PCR.

(A) Protein/DNA complexes detected upon incubation of promoter/intron probes with nuclear proteins derived from DMSO- or MG132-treated cells (20 μM, 4h). Nuclear extracts (5 μg) were submitted to EMSA using as probes the twelve PCR-generated DNA fragments encompassing the cloned UBC promoter sequence (FR1-6), the first exon and the unique intron (I-VI). Protein-DNA complexes were separated on 5% polyacrylamide gels and the radiographic signal was detected in a GS-250 Molecular Imager. Arrows indicate the position of complexes specifically formed upon MG132 treatment (B) Specificity of MG132-induced protein/DNA complexes was assessed by competition experiments where nuclear extracts from MG132-treated cells were directly incubated with the labeled probes (FR1 to FR6) or pre-incubated with a 50-fold excess of an ODN containing the canonical HSF consensus sequence (ODN-cHSE), before probe addition. The image was cropped to show only the gel region of interest.

Fig 3. Binding of HSFs to FR1 and FR6 UBC DNA regions in vivo.

ChIP analysis was performed on DMSO- and MG132-treated cells (20 μM, 4h) using specific antibodies against HSF1 and HSF2. Chromatin submitted to the immunoprecipitation procedure in the absence of added antibody (No Ab) was used as internal IP control. RealTime PCR was carried out on chromatin before (input) and after immunoprecipitation using specific primers which amplify the UBC promoter regions FR1 (-916/-759) and FR6 (-196/-96). Primers flanking the HSE contained in the human HSP70 promoter or amplifying the UBC intron probe V (+608/+766) were used to generate HSF positive and negative ChIP controls, respectively. Each sample was tested in triplicate. Data were analyzed according to the 2^−ΔCT method, normalized to the input DNA and expressed as % input. The average value ± SE was calculated from six independent ChIP analyses. Asterisks indicate statistical significance versus the corresponding No Ab, unless otherwise highlighted in the graphs. Statistical analysis was performed with two-way ANOVA. *p<0.05; **p<0.01.

Fig 4. Dissection of the UBC FR1, FR2 and FR6 segments.

Wild-type and mutant ODNs containing the putative UBC HSEs were submitted to EMSA, competitive EMSA and supershift analysis. Nuclear extracts from MG132-treated cells were directly incubated with the indicated ³²P-ODN probe (lanes 1A, 3A, 5A, 1B, 1C, 1D, 1E, 1F and 1G) or pre-incubated with a 50-fold excess of unlabeled competitor oligonucleotide (lanes 2A, 4A, 6A, 2-4B, 2C, 2-5D, 2-5E and 2-5F) or antibodies against HSF1 and HSF2 (lanes 5-6B, 6-7D and 2-3G), before addition of the radiolabeled probe. ODN-cHSE: ODN containing the classical HSE consensus sequence. The images were cropped to show only the gel region of interest.

Fig 5. Reporter construct gene expression analysis.

(A) The UBC promoter-pGL3 constructs P1 (-916/+878) and P3 (-371/+878) containing FR1-2-6 and only FR6, respectively, were transfected in HeLa cells. Fourty-eight hours post-transfection, cells were treated with the DMSO vehicle or MG132 (20 μM). After 8h, luciferase mRNA levels were determined by RealTime PCR. Expression data, normalized to the housekeeping GAPDH gene, were analyzed by the 2^-ΔΔCT method and referred to the value obtained in DMSO-treated cells. The data represent the mean fold induction ± S.E. of 5 independent experiments. Asterisks indicate statistical significance versus cells receiving only the vehicle (*p<0.05; **p<0.01). Statistical analysis was performed on ΔCT values with one-way ANOVA. (B) The wild-type UBC promoter-pGL3 construct P1 (FR1-2-6) and its mutant counterparts P1mut FR1-2 and P1mut FR6, where HSF binding motifs were selectively mutated within FR1 and FR2 altogether or FR6, respectively, were transfected in HeLa cells. Forty-eight hours post-transfection, cells were treated with the DMSO vehicle or MG132 (20 μM). After 8h, luciferase mRNA levels were determined by RealTime PCR. Expression data, normalized to the housekeeping GAPDH gene, were analyzed by the 2^-ΔΔCT method and referred to the value obtained in DMSO-treated cells. The data represent the mean fold induction ± S.E. of 7 independent experiments. Asterisks indicate significant differences of the MG132-induced LUC mRNA fraction between cells transfected with wild-type and mutant reporter constructs (*p<0.05; **p<0.01). Statistical analysis was performed on ΔΔCT values with one-way ANOVA.