Sp1 and Sp3 control constitutive expression of the human NHE2 promoter by interactions with the proximal promoter and the transcription initiation site

Abstract

We have previously cloned the human Na+/H+ exchanger NHE2 gene and its promoter region. In the present study, the regulatory elements responsible for the constitutive expression of NHE2 were studied. Transient transfection assays revealed that the -40/+150 promoter region contains the core promoter responsible for the optimal promoter activity. A smaller fragment, -10/+40, containing the TIS (transcription initiation site) showed minimal activity. We identified a palindrome that overlaps the TIS and binds to the transcription factors Sp1 and Sp3. Mutations in the 5' flank of the palindrome abolished the Sp1/Sp3 interaction and reduced promoter activity by approx. 45%. In addition, a conserved GC-box centered at -25 was found to play a critical role in basal promoter activity and also interacted with Sp1 and Sp3. An internal deletion in the GC-box severely reduced the promoter activity. Sp1/Sp3 binding to these elements was established using gel-mobility shift assays, confirmed by chromatin immunoprecipitation and co-transfections in Drosophila SL2 cells. Furthermore, we identified two positive regulatory elements in the DNA region corresponding to the 5'-UTR (5'-untranslated region). The results in the present study indicate that Sp1 and Sp3 are required for constitutive NHE2 expression and that the positive regulatory elements of the 5'-UTR may co-operate with the 5'-flanking region to achieve the optimal promoter activity.

Figure 1. Deletion and mutational analysis of the human NHE2 promoter

Functional analyses were carried out with 5′- (A), 3′- (B) and internal- (C) deletion constructs. Luciferase reporter constructs containing the indicated promoter fragments of the NHE2 promoter along with pSV-βgal were transiently transfected into C2BBe1 cells. At 48 h post-transfection the cells were lysed and reporter gene activity was measured and normalized to β-galactosidase activity as indicated in the Experimental section. The luciferase activity of each deletion mutant is shown as a percentage of the activity of the parental plasmid p−85/+150, which was set at 100%. Values are the means±S.E.M., n≥4. *Significantly different from pGL2-basic.

Figure 2. Schematic representation of the human NHE2 promoter and core promoter region

The promoter region between nucleotide −1051 to the ATG codon at +316 (with respect to the TIS shown by a bent arrow) including the location of the potential regulatory elements is shown. The region corresponding to bp −85 to +150 is amplified and the sequences of the GC-box centred at position −25 (containing potential overlapping binding sites for three Sp1 and one EGR-1), the TIS, an overlapping AP2/Sp1 binding site at +60 and an E-box at +113 are shown. The arrows in the TIS oligonucleotide represent the palindromic sequence. The TIS is shown as +1. C/GCTGC/G repeats are underlined. The boxes drawn around nucleotide sequences mark the core sequence of the protein binding sites as indicated.

Figure 3. Sp1 and Sp3 bind to the NHE2 TIS

Gel-shift analysis was carried out using 5 μg of C2BBe1 cell nuclear proteins. The end-labelled probe used in these assays was the double-stranded oligonucleotide from −12 to +20, shown as TIS in Table 2. Binding reactions were performed in the presence or absence of competitors (50-fold molar excess) or in the presence of anti-Sp1 and -Sp3 antibodies (2 μg/lane) as indicated at the top of the Figure. All oligonucleotides used in the experiments are shown in Table 2. Four bands showed specific binding to the −12/+20 probe and are indicated by C1–C4. To separate the DNA–protein complexes the gel represented by numbers 4–9 was run for a longer time than the others that resulted in weaker interactions of C4. NS, non-specific nucloetide; N3, an oligonucleotide from the NHE3 promoter that we have shown previously binds Sp1 and Sp3. An arrowhead indicates the Sp1-supershifted band.

Figure 4. Mapping of the Sp1-binding site and functional analysis of the mutated TIS element

(A) GMSAs were performed with nuclear proteins from C2BBe1 cells and with the wild-type TIS oligonucleotide as the end-labelled probe. As expected, four nucleoproteins were formed and are indicated on the left-hand side. Nuclear proteins (5 μg) and a 50-fold molar excess of competitor oligonucleotides were used in each reaction with 50000 c.p.m. of the end-labelled probe. The oligonucleotides that were truncated or altered at specific nucleotides and used for competition assays or construction of mutated plasmids are shown in Table 2. (B) Mutated constructs were generated in the context of p−85/+150 by site-specific mutagenesis using primers m-TIS#1 and m-TIS#2 shown in Table 2 and as described in the Experimental section. C2BBe1 cells were transfected with the wild-type and mutated constructs along with pSV-βgal. Normalized luciferase activities were determined and compared with the wild-type promoter activity, which was set arbitrarily at 100%. All transfections were repeated three times each in triplicate.

Figure 5. Co-transfection experiments in C2BBe1 cells

C2BBe1 cells were co-transfected with 1.0 μg of test plasmid along with either 0.5 μg of pCMV-Sp1 or the empty vector pRC-CMV. Luciferase activity was measured 48 h post-transfection and normalized to total cellular protein. The results are shown as normalized activity relative to the activity of pGL2-basic, and the fold-increase by co-transfected Sp1 expression vector was calculated as the activity in the presence of the expression vector divided by the activity of reporter construct in the presence of the empty vector. The results represent at least three individual experiments performed in triplicate.

Figure 6. Characterization of the factors binding to the GC-box and their role in NHE2 promoter activity

(A) The sequences of the oligonucleotide probes or competitors used in DNA-binding studies are shown. (B) GMSA was performed with the wild-type end-labelled oligonucleotide spanning from nucleotide residues −37 to −14 (GC-WT) of the human NHE2 promoter and incubated with 5 μg of C2BBe1 nuclear proteins (lanes 1–8). Lane 1, no competitor; lane 2, 50-fold molar excess of an unlabelled competitor oligonucleotide containing the Sp1 consensus sequence; lanes 3–8 supershift assays, in the presence of 2 μg of antibodies specific for Sp1, Sp3 and EGR-1 (lanes 3–5) or a combination of those antibodies (lanes 6–8). (C) GMSA was performed with nuclear extracts from C2BBe1 cells and end-labelled wild-type probe (lanes 1–3 and 6–9), GC-mA (lanes 4 and 5), GC-mB (lane 10) and GC-mC (lane 11). The competitor oligonucleotides employed in these assays are shown in (A), and are indicated at the top the Figure. (D) Transfection analysis of the mutant constructs containing altered Sp1-binding sites. Transient transfections were performed with constructs harbouring base substitutions (GC-mB and GC-mC) or deletions (ΔSp1b-c) in the GC-box and the control p−85/+150. Each construct (1.6 μg) was co-transfected along with pSV-βgal (0.4 μg) into C2BBe1 cells and processed as described in the legend to Figure 1. The y-axis shows the normalized luciferase activity of each construct relative to the normalized luciferase activity of the promoter-less vector pGL2-basic. The values presented are means±S.E.M. of at least three independent experiments performed in triplicate.

Figure 7. Sp1 and Sp3 trans-activate the proximal promoter of the NHE2 gene

(A) Drosophila SL2 cells were co-transfected with 1.0 μg of p−85/+150 construct and concentrations of the Sp1 or Sp3 expression vectors as indicated. (B) SL2 cells were co-transfected with a constant concentration of pPacSp1 (0.5 μg) plus 0.5, 0.75 or 1.0 μg of pPacSp3. Transfected cells were harvested after 48 h and luciferase activities were determined using the luciferase assay system from Promega. Luciferase activity was normalized to cellular protein content. The activity indicated represents the fold-induction in luciferase activity relative to that obtained from co-transfection with the empty vector pPac0. These experiments were repeated three times each in duplicate and the results from one such experiment is shown. (C) ChIP assays with anti-Sp1 and -Sp3 antibodies. Cross-linked chromatin was isolated from C2BBe1 cells subsequent to formaldehyde treatment. ChIP assays were performed with anti-Sp1 and -Sp3 antibodies. Primers indicated in the text were used to amplify the NHE2 promoter region using the co-immunoprecipitated DNA as template. The PCR products were resolved on a 1.5% agarose gel. Ab, antibody; Ng Ab, negative control antibody.