Na+/H+ exchanger (NHE) isoforms NHE2 and NHE3, colocalized to the brush border membrane of the epithelial cells, exhibit differences in their pattern of tissue expression and regulation by various molecular signals. To investigate the mechanisms involved in regulation of NHE3 gene expression, the human NHE3 promoter region was cloned and characterized. Primer extension experiments located the transcription start site to a position 116 nucleotides upstream from the translation start codon. The 5′-flanking region lacked a CCAAT box but contained a TATA-like sequence. Nucleotide sequencing of the 5′-flanking region revealed the presence of a number of cis elements including Sp1, AP-2, MZF-1, CdxA, Cdx-2, steroid and nonsteroid hormone receptor half sites, and a phorbol 12-myristate 13-acetate-response element. Transient transfection experiments using C2/bbe cell line defined a maximal promoter activity in −95/+5 region. The regulatory response elements clustered within this region include a potential transcription factor IID (TF IID), a CACCC, two Sp1, and two AP-2 motifs. Deletion of a fragment containing the AP-2 and Sp1 motifs resulted in a drastic decrease in promoter activity. In gel mobility shift assays, an oligonucleotide spanning from −78 to −56 bp bound a recombinant AP-2, and the corresponding binding activity in nuclear extracts was supershifted with anti-AP2α antibody. Our studies suggest that the NHE3 expression is regulated by a combination of ciselements and their cognate transcription factors that include the AP-2 and Sp1 family members.
- nucleotide sequence
- transcription factor binding sites
- deletion analysis
the na+ /h+ exchangers (NHE) catalyze the electroneutral exchange of one extracellular Na+ for one intracellular H+across the plasma membrane and play a role in various important cellular functions that include sodium absorption, maintenance of intracellular pH and cell volume, and regulation of cell proliferation (for a review, see Refs. 1, 12,31, 41, and 42). To date, at least six members of the NHE gene family have been identified and characterized with regard to their localization to specific sites in the cell and function of the corresponding protein products (4, 8, 21, 30,32, 38, 40, 44). For example, the NHE1 isoform was shown to be ubiquitous in its expression and localized to the basolateral membrane of polarized epithelial cells, whereas the NHE2 and NHE3 isoforms are localized to the apical membrane. The NHE4 and NHE5 isoforms are expressed in the kidney and neural tissues, respectively, whereas the recently cloned NHE6 isoform is mitochondrial. Although extensive studies have recently been carried out investigating the regulation of the NHE1 and NHE3 protein products by phosphorylation and interaction with cytoplasmic regulatory proteins (37, 43, 45), only limited studies have been reported on the transcriptional regulation of the NHE isoforms. Studies focused on the NHE1 isoform have demonstrated the regulation of NHE1 gene at the transcriptional level by external acid, growth factors, serum, and retinoic acid (35, 48, 49). The involvement of AP-1, AP-2, and other cis elements in the 5′-flanking region of the NHE1 isoform has been reported (13, 14,26). Among the other NHE isoforms, the promoter sequences for the human and rat NHE2 (22, 28) and the rat NHE3 isoform (9, 19) have been identified, and transcriptional regulation of the rat NHE3 by glucocorticoids and thyroxin has been reported (9, 10, 19). However, transcriptional regulation of the human NHE2 and NHE3 isoforms has not been studied.
We have recently reported the genomic organization and cloning of the human NHE2 gene and its promoter (22, 23). Our studies described here report, for the first time, the molecular cloning of the human NHE3 promoter, and they demonstrate the presence of a number ofcis-acting elements that may be involved in regulation of the human NHE3 isoform. We defined a minimal promoter region that contains the maximal promoter activity in C2/bbe cell line and showed that deletion of a DNA fragment containing the binding sites for the transcription factors AP-2 and Sp1 results in a drastic loss of promoter activity. By footprinting experiments, we have shown that AP-2 binds to two regions in the NHE3 promoter. The AP-2 interaction with the proximal AP-2 binding site was confirmed with gel mobility shift assay and supershift analysis. This work provides the initial framework for further studies and understanding the molecular mechanisms responsible for regulation of NHE3 gene expression.
All chemicals were purchased from Fisher Scientific (Pittsburgh, PA) or Sigma (St. Louis, MO); restriction endonucleases and other modifying enzymes were from either New England Biolabs (Beverly, MA), Gibco-BRL (Gaithersburg, MD), or Promega (Madison, WI); polyclonal anti-human AP-2α, monoclonal anti-human Sp1, Sp2, and Sp3 antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA); plasmid PCR-II, a TA cloning vector, was from Invitrogen (San Diego, CA); JM109 competent cells and luciferase assay system were from Promega.
DNA manipulations, including restriction enzyme digestion, ligation, plasmid isolation, and transformation were carried out by standard methods (2).
RNA extraction and PCR analysis.
Total RNA was extracted by the RNAzol method (Tel-Test, Friendswood, TX) according to the manufacturer's directions. PCR was performed in a Perkin-Elmer/Cetus DNA cycler, with thermostable DNA polymeraserTth (Boehringer Mannheim, Indianapolis, IN) according to the manufacturer's directions.
Cloning of the 5′-flanking region.
The 5′-regulatory region of the NHE3 gene was cloned using the Genome Walker kit (Clontech Laboratories, Palo Alto, CA). PCR amplifications were performed with genomic DNA fragment pools as template, anchor primers AP1 or AP2 that hybridize to the 5′-end of the genomic fragments, and gene-specific primers GI-318 (5′-AGAGCGCGATGACGTAGGGATCC-3′) or GI- 358 (CCCGAGTCCCCACATTGCCGCCTGC) (8). This resulted in amplification of a 1.6-kb DNA fragment. This fragment was cloned in pCR-II cloning vector (Invitrogen) after gel purification and designated pJM1.6N3. Subsequently, with the use of primers synthesized based on the nucleotide sequences of the insert in pJM1.6-N3, a larger fragment (3.0 kb) of the 5′-flanking region was amplified and cloned in pCR-II (pJM3.0 N3). DNA nucleotide sequences at the 3′-end of the 3.0-kb fragment were determined using the dideoxy chain termination method (36) and compared with the hNHE3 cDNA. This confirmed that the 3.0-kb fragment represented the 5′-flanking region of the NHE3 gene.
Reporter plasmid construction.
Plasmids used for functional analysis of the NHE3 promoter activity were generated using pGL2-basic (Promega), that contains a promoterless luciferase reporter gene. To clone the 1.6-kb NHE3 promoter DNA fragment, the insert in pJM1.6N3 was released with restriction enzymeEcoR I, and the ends were filled in with Klenow. Following gel purification, this DNA fragment was cloned upstream from the luciferase structural gene in pGL2-basic that was blunt-ended at theHind III site. The clone that carries the promoter in forward direction was named p1.6 N3P. The plasmid carrying the promoter in reverse direction, p1.6N3P-Rev, was constructed by digesting the pJM1.6 N3 with restriction enzymes Xho I plusHind III and ligating with pGL2-basic digested with the same enzymes. With the use of restriction enzyme recognition sites from the NHE3 promoter sequence, chimeric plasmids containing progressive deletions of the NHE3 gene 5′-flanking region were generated. The plasmid containing the −1004/+131 insert was constructed by subcloning a Sac I-Xho I fragment from p1.6 N3P-Rev into pGL2-basic digested with the same enzymes. Plasmids harboring the −319/+131 and −95/+131 sequences were generated by digestion of p1.6 N3P-Rev with Pvu II plus Xho I and NruI plus Xho I, respectively, and cloning in pGL2-basic that was digested with Sma I plus Xho I. Plasmid containing the sequences from +2/+131 was constructed by deletion of aKpn I fragment from p-1004/+131. Plasmid p-95/+5 was generated by digestion of p-95/+131 with restriction enzymesAcc65 I plus Hind III, blunting the ends and religation. Plasmids p-43/+131 and p(Δ−187/−43) were constructed by deletion of Sac I-Sac II andSac II fragments from p-1004/+131, respectively.
Primer extension analysis.
The transcription initiation site of the human NHE3 gene was determined by primer extension using SuperScript II RT (Gibco-BRL). An antisense oligonucleotide complementary to nucleotides +40 to +67 (see Fig. 3) was synthesized and end-labeled with [γ-32P]ATP and T4 polynucleotide kinase. Free [γ-32P]ATP was removed by using mini Quick Spin Oligo Columns (Boehringer Mannheim). For primer extension reaction, 105 counts/min (cpm) of the end-labeled oligonucleotides and 15 μg of total RNA from C2/bbe cells were coprecipitated and dissolved in diethyl pyrocarbonate-treated water, heated at 75°C for 5 min, and brought to 42°C. The reaction was complemented with 200 μM dNTPs and reaction buffer to a final concentration of 100 mM Tris · HCl (pH 8.3), 150 mM KCl, 6 mM MgCl2, 10 mM 1,4-dithiothreitol (DTT), and 200 units of SuperScript II in a reaction volume of 20 μl. The primer extension was carried out for 60 min at 42°C. The extension products were phenol-chloroform extracted, ethanol precipitated, and pelleted nucleic acids were dissolved in stop solution (US Biochemicals). The samples were heated at 90°C for 3 min and analyzed on a 6% polyacrylamide, 7 M urea denaturing gel. The gel was dried and exposed to X-Omat AR film. A sequencing ladder was used as a size marker to determine the size of the extended primer.
Cell culture and transfections.
C2/bbe cell line, a subclone of the Caco-2 cells, was cultured and maintained as described (22). For transfection studies, cells (1.8 × 105) were seeded into 12-well plates and cotransfected 24 h later (80–90% confluent) with one of the NHE3-luc constructs and pRSV-βgal using Lipofectamine Plus reagent (Gibco-BRL). The latter plasmid served as an internal control for transfection efficiency. A total of 2.0 μg DNA/well, at a ratio of 4:1 for experimental vs. pRSV-βgal, was used for each transfection. After 48 h, cells were washed with phosphate-buffered saline and lysed using a kit from Promega. Luciferase activity was assayed using a luminometer (Promega) and normalized to β-galactosidase activity.
DNase I footprinting analysis.
To prepare probes for footprinting experiments, the plasmid p1.0N3P was digested with Kpn I restriction enzyme. DNA fragments were end-labeled after dephosphorylation with calf alkaline phosphatase using polynucleotide kinase and [γ-32P]ATP and then digested with NruI or Pvu II in two separte reactions. The labeled probes were purified on a 6% native polyacrylamide gel. DNase I footprinting analysis was performed using a kit from Promega, as per instructions. Briefly, 20,000 cpm of end-labeled probe were incubated in 25 μl of binding buffer (10 mM Tris · HCl, pH 7.5, 50 mM NaCl, 0.05% Nonidet P-40, 1 mM EDTA, 1 mM DTT, and 10% glycerol) for 10 min on ice in presence or absence of 2 μl AP-2 extract. Fifty microliters of a solution of Ca+2 and Mg+2 were added, and incubation was continued for 1 min at room temperature. The reaction mixtures were treated with 3 μl of diluted QRI (RNase-free DNase from Promega) for 1 min at room temperature, and further digestion was stopped by addition of 90 μl stop solution. The pellet was suspended in loading buffer after phenol/chloroform (1:1) extraction and ethanol precipitation, denatured by heating at 95 °C for 3 min, and analyzed on a 6% sequencing gel. The gel was dried and autoradiographed.
Gel mobility shift assay.
All oligonucleotides for gel mobility shift assay (GMSA) were synthesized by Gibco-BRL. Complementary oligonucleotides were made double-stranded by heating to 95°C for 5 min and slow cooling to 25°C in TE buffer (10 mM Tris · HCl, pH 7.5, 1 mM EDTA). The single-stranded sequence of the AP-2 probe was 5′-GGCTCCGCCCCGGGGCGGGAGGG-3′, where homology to consensus sequence is shown in bold letters. Nuclear proteins were prepared essentially as previously described (2). Protein/DNA binding reactions were performed in binding buffer [50 mM Tris · HCl (pH 7.5), 1 mM MgCl2, 0.5 mM EDTA, 0.5 mM DTT, 50 mM NaCl, 1 μg/sample poly(dI.dC)-poly(dI.dC), 4% glycerol] and 30,000 cpm of the probe. Reactions were initiated by addition of the nuclear proteins (3–5 μg) and incubation for 20 min at room temperature before electrophoresis on a native 4% polyacrylamide gel in 0.5× TBE running buffer (45 mM Tris borate, 1 mM EDTA, pH 8.3). Gels were dried and visualized by autoradiography. In competition assays, the unlabeled competitor oligonucleotide was added to the reaction 10 min before the addition of the labeled probe. Supershift assays were performed by addition of 1 μl of the AP-2α antibody (Santa Cruz Biotechnology) after the initial 20-min incubation with the labeled probe and then further incubation for 30 min at room temperature.
Cloning of the human NHE3 promoter.
With the use of the genome walking technique, we cloned a 3.0-kb DNA fragment upstream of the human NHE3 translation initiation codon. Nucleotide sequence analysis of the 3′-end of this clone confirmed that the cloned DNA fragment overlaps with the 5′-end of the hNHE3 cDNA reported previously (8).
The NHE3 promoter fragment directs luciferase activity in the C2/bbe cell line.
We have chosen to use the human intestinal epithelial cell line C2/bbe as a model to study the functional and molecular characteristics of the human NHE3 promoter. The C2/bbe cell line is a subclone derived from a heterogeneous population of the Caco-2 cells and has been shown to undergo spontaneous differentiation, as determined by exhibiting characteristics of the microvilli and the presence of the markers of differentiation (33). In a previous study (22), by using RT-PCR analysis, we have established that C2/bbe cells express NHE1, NHE2, and NHE3 mRNA. However, the endogenous level of NHE3 protein expression as assessed by Western blots seems to be very low in these cells, becuase NHE2 is readily detected but not NHE3 (25). Recent studies (29) from the same group indicated that treatment with short-chain fatty acids resulted in augmentation of the NHE3 activity and protein expression in C2/bbe cell line.
A 1.6-kb fragment of the promoter/enhancer region was cloned in both orientations upstream from the luciferase reporter gene, and the luciferase activity of the chimeric constructs, which is a direct measure of the promoter activity, was measured in transiently transfected C2/bbe cells. As shown in Fig.1, the forward orientation, pJM1.6N3P, reproducibly showed a marked increase in luciferase gene expression compared with the pGL2-basic vector. No luciferase activity over the background level was obtained for reverse construct, pJM1.6 N3P-Rev, indicating that the promoter activity was orientation dependent.
Identification of the transcription initiation site.
To map the transcription initiation site of the NHE3 gene, we employed primer extension analysis. A 29-mer oligonucleotide complementary to the sense strand at position +40 to +67 (Fig. 3) was used as an extension primer. Total RNA from the C2/bbe cells was hybridized with this end-labeled primer and subjected to reverse transcription. Extension products were analyzed on sequencing gel. A 67-nucleotide primer extension product was identified where C2/bbe RNA was used in the reaction (Fig. 2, lane 1), whereas no signal was found in control yeast transfer RNA (Fig. 2,lane 2). On the basis of this observation, the transcription initiation site of the human NHE3 gene was assigned to a deoxyguanosine residue, 116 nucleotides upstream from the translation initiation site. This site was designated +1, the start of transcription initiation.
DNA Sequence and characterization of the 5′-flanking region of NHE3.
The nucleotide sequence of the 1.6-kb NHE3 promoter fragment was determined and is shown in Fig. 3(GenBank accession no. AF282824). Between the ATG translation in start codon and the transcription initiation site lie the NHE3 minicistron (8) and the 5′-untranslated region of 83 nucleotides. Two sets of directly repeated sequences composed of penta- and decamer were observed downstream from the transcription initiation site (Fig. 3).
The 5′-flanking region of the NHE3 gene was highly GC-rich, especially the region surrounding the transcription initiation site from aPvu II restriction enzyme site at −319 to the ATG translation start site at +117 has an overall G+C content of 80%. The nucleotide sequence of the 1.6-kb promoter fragment was screened for potentially important cis elements. The putative response elements found within this region are indicated in Fig. 3. A TATA boxlike sequence, GATTAAA, is located at position −40, whereas no CCAAT box were found in close vicinity of the transcription initiation site. However, a cluster of binding sites for a number of transcription factors is present immediately upstream of the transcription start site. These include binding sites for Sp1 (−25 to −12, −69 to −72, −84 to −72, and −142 to −130); AP-2 (−17 to −28 and −56 to −67); CACCC (−32 to −27); MZF-1 (−22 to −14); TF IID (−41 to −35). Further upstream, several other potentially important sites were identified such as AP-4/E47/MyoD, PEA3, CCAAT, Cdx-2, and cAMP-response element binding site. Moreover, putative cis elements for TRE and 1/2 sites for GRE were also identified in this region.
The rat NHE3 promoter has been cloned and sequenced by two different groups (9, 19). Some of the cis elements identified in the human NHE3 promoter region are also present in the rat gene. Given these similarities, we compared the DNA nucleotide sequences of the 5′-nontranscribed regions of these promoters (Fig.4). The sequence of the human NHE3 proximal promoter region (−100 to +1) was 79% identical to that of the rat sequence. The consensus sequences for Sp1, TF IID, and CACCC were present in similar positions in the two species in this region. However, there were other cis elements for different transcription factor binding sites that are unique in each promoter (Fig. 4), suggesting that different mechanisms may be involved in the expression of NHE3 in these species. The overall sequence homology was 38%, and no obvious regional sequence identities were found with this homology search in the upstream region.
5′-Deletion analysis of the NHE3 promoter in C2/bbe cells.
To identify the promoter regions that were involved in directing NHE3 gene expression, luciferase reporter constructs carrying serially truncated segments of the 5′-flanking region were generated in pGL2-basic vector. Figure 5 shows a comparison of the luciferase activity between deletion constructs. The full-length promoter construct, p3.0N3P (−2900 to +131), exhibited a 20-fold activation of the luciferase activity compared with the promoterless vector. The deletion of sequences from −2900 to −1507 bp led to a 40-fold increase in promoter activity compared with promoterless vector, suggesting that an inhibitory element may be located at this region. Further deletion from −1507 to −1004 did not show a significant difference in the luciferase activity. However, more extensive deletions to positions at −319 and −95 bp resulted in ∼50- and 65-fold increases in promoter activity compared with the vector or 2.6- and 3.2-fold over the full-length promoter construct, respectively. The moderate increase in luciferase activity in plasmid containing −319/+131 bp compared with the previous truncated construct may also be attributed to the loss of a suppressor ciselement contained within the deleted DNA region. A number of putative response elements are located within this region (Fig. 3), and whether any of these regulatory factor(s) is involved in transcriptional repression of the NHE3 promoter is not clear at this point. As shown in Fig. 5, the construct containing sequences −95 to +5 bp contained the highest luciferase activity.
Transcriptional activity of the −1004/+131 construct after deletion of AP-2 and Sp1.
We examined the importance of the sequences in the proximal promoter region by introducing an internal deletion that removed a 144-bpSac II fragment immediately upstream from the putative TF IID-like binding site. A schematic drawing of the ciselements contained within the deleted region is shown in Fig.6, top, and includes an overlapping Sp1/Egr-1, Sp1, and an overlapping Sp1/AP-2 motif. Expression of this deletion construct (Δ−187 to −43) in C2/bbe cells resulted in a 75% decrease in the basal level of NHE3 transcription compared with the parental construct (−1004/+131; Fig.6), indicating a critical role for this region in NHE3 promoter expression. When the sequences upstream from nucleotides −43 (Fig. 6, −43/+131) or +2 (+2/+131; Fig. 6) were removed, the level of luciferase activity dropped to the background level, suggesting that sequences downstream from the −43 position alone do not contribute to transcription activity of the NHE3 gene. However, deletion of sequences between +5 to +131 had a positive effect on the reporter gene expression, as shown by a 25% increase in luciferase activity of the −95/+5 construct compared with −95/+131 construct (Fig. 6). Thus these observations indicate that the smallest promoter fragment containing sequences between −95 to +5 promotes the maximal luciferase activity and that all of the cis elements required for optimal NHE3 promoter activity are present in the first 95 nucleotides of the 5′-flanking region.
To investigate the involvement of the putative cis elements indicated by Sac II deletion (Fig. 6) in transcriptional regulation of the NHE3, DNase I footprinting experiments were undertaken. An Escherichia coli extract containing the cloned AP-2 protein, AP-2 extract, was used as the source of AP-2 transcription factor. Figure 7 shows an autoradiogram of the footprint seen with the AP-2 extract and two overlapping DNA probes from the NHE3 5′-regulatory region. In these experiments, two DNase I-protected regions were identified: a proximal AP-2 binding site was present at bp −51 to −71 and another site at −176 to −198. The role of the proximal AP-2 binding site in the NHE3 gene expression was explored further by GMSA.
The transcription factor AP-2 Binds to the NHE3 proximal promoter.
A perfect match to the consensus AP-2 binding site (5′-GCCCNNNGGC-3′) (47) is located in the proximal footprint shown in Fig. 7. Labeled oligonucleotides spanning this AP-2 element at −67 to −56 bp, N3-Ap2, were coupled with AP-2 extract (Promega) and analyzed by GMSA. As illustrated in Fig.8 A,lane 2, a single DNA/protein complex was detected (shown by an arrow). To determine the specificity of this DNA/protein complex, an excess of 100- and 200-fold unlabeled N3-AP2 oligonucleotide was used as a competitor. This resulted in the elimination of the DNA/protein complex (Fig. 8 A, lanes 3 and 4), whereas inclusion of an unlabeled nonspecific oligonucleotide as a competitor did not affect the complex formation (lanes 5 and6). With the use of a labeled oligonucleotide containing the consensus sequence for AP-2 transcription factor binding site as a probe (Promega), a complex was formed (lane 7) that migrated at the same position as that of the N3-AP2 probe. This DNA/protein complex was competed away with unlabeled N3-AP2 oligonucleotide (lanes 7 and 8). Gel mobility shift assays were also performed with nuclear extracts from three human intestinal cell lines: Caco-2, T84, and C2/bbe. The nuclear extracts of all three cell lines contained the same binding activity with N3-Ap2 probe, which migrated at the same positions. The result of GMSA with the C2/bbe nuclear extract is shown in Fig. 8 B, lanes 5–8. To confirm the identity of the protein in this complex, we performed supershift experiments using antibody against AP-2α. As shown in Fig. 8 B, lane 3, the DNA/protein complex formed by AP-2 extract was shifted completely, whereas the complex formed by C2/bbe nuclear proteins (lane 6) was shifted only partially. The identity of the remaining unshifted protein in not clear at this time, but it may be another protein interacting with this probe, or it could be the residual AP-2 that is not coupled with the antibody. These data confirm that Ap-2∝ is one of the transcription factors from C2/bbe nuclear proteins that interact with the NHE3 promoter.
Members of NHE gene family exhibit highly restricted temporal and spatial expression patterns. NHE2 and NHE3 isoforms are expressed at the apical membranes of the epithelial cells, where they play a major role in transepithelial Na+ absorption. To better understand the mechanisms underlying the regulation of the apical membrane NHE isoforms, we have focused our studies on the transcriptional regulation of the human NHE2 and NHE3. Previously we (22) have reported the cloning and preliminary characterization of the human NHE2 promoter. In the current study, we have cloned and identified the human NHE3 promoter region. Primer extension analysis identified a transcription initiation site 116 nucleotides upstream from the translation start codon. Nucleotide sequence analysis of the 1.6-kb promoter region revealed multiple potential cis-acting elements upstream from the transcription initiation site. These included Sp1, AP-2, Cdx-2, Mzf-1, MyoD, Egr-1, glucocorticoid, and thyroid hormone receptor binding sites. A CCAAT box sequence was not present in the immediate vicinity of the transcription initiation site, but a TATA-like sequence was found at approximately the −40 position. These features and also the presence of a highly GC-rich region surrounding the transcription initiation site in conjunction with multiple Sp1cis elements in this region may suggest a housekeeping role for the NHE3 gene. However, NHE3 promoter also contains features common in regulated genes. This is evident by the presence of the potential transcription factor binding sites that mediate tissue-specific and developmental regulation of the regulated genes. For example, the presence of Cdx-2, which is involved in gene expression and differentiation in the intestine (39), MyoD, which is implicated in myocyte differentiation (7), and Mzf-1, which is involved in erythrocyte-specific gene expression and regulation (27) would suggest developmental expression and tissue-specific regulation of this isoform.
Deletion of a 1.5-kb fragment from the 5′-flanking region of the NHE3 promoter showed that the deleted region had a marked inhibitory effect on the expression of the luciferase gene. Moreover, further deletions toward the transcription initiation site also resulted in augmentation of the luciferase activity of the corresponding reporter constructs (Fig. 4). Although the elements responsible for the inhibitory effect have not been defined, the increase in the luciferase activity of the 5′-deletion reporter constructs argues that the upstream region contains repressive functions. The physiological role of the negative regulatory elements in the NHE3 promoter remains to be determined. One possibility is that it could mediate the divergent levels of the NHE3 gene expression seen in various tissues (8). Further deletions revealed that the region between −95 and +131 contained the highest luciferase activity. This region contained three potential Sp1, two AP-2, and also an atypical TATA cis element. When the +5 to +131 DNA region was eliminated, the luciferase activity of the new construct was increased ∼25% (Fig. 5), suggesting that the nucleotide sequence between transcription initiation site and the translation start codon may be involved in the repression of NHE3 promoter activity. The exon-1 sequence in this region harbors the NHE3 minicistron (8), two sets of direct repeat elements (Fig.3), and a number of potential transcription factor binding sites (data not shown). Whether these motifs are responsible for the inhibitory effect residing in the +5/+131 region is not known at the present. A deletion from −95 to −43, which includes the Sp1 and AP-2 ciselements resulted in a total loss of reporter plasmid activity, suggesting the importance of the Sp1 and AP-2 transcription factors for the hNHE3 gene expression in C2/bbe cells.
A comparison of the human and the rat (19) NHE3 promoter sequences in the nontranscribed region exhibited that the sequence identity of the two promoters was confined to the vicinity of the transcription initiation site (Fig. 4). This region coincides with the human NHE3 promoter region (−95/+1) that drives the maximal luciferase activity in promoter-reporter transfection assays (Fig. 6). The consensus is that Sp1, AP-2, and TATA-like sequences appeared at the same positions in both species in this region. However, there were other potential cis elements for different transcription factor binding sites that are unique in each promoter (Fig. 4), suggesting that different mechanisms may be involved in the expression of NHE3 in these species and may offer a clue to the mechanisms by which the NHE3 gene is regulated.
Transcriptional regulation of the NHE3 gene by various agents is likely to be mediated by trans-acting factors that bind to the promoter region. Glucocorticoids have been shown to increase NHE3 mRNA levels (9, 19, 50). Thyroid hormone has also been reported to activate NHE3 transcription (10). Phorbol 12-myristate 13-acetate on the other hand showed inhibitory effects on expression of NHE3 gene (3, 16, 20). Ontogenic increase in Na+/H+ exchange activity that correlated with the increasing levels of NHE3 mRNA and protein abundance has been reported (5, 11). In addition, postnatal administration of glucocotricoids and thyroid hormone was shown to increase NHE3 mRNA and protein levels during maturation (5, 6). We have identified a potential binding site for T3R at position −211 to −198 and many half sites for glucocorticoid receptor binding on the NHE3 promoter region. These sites may be responsible for mediating the stimulatory effects of thyroid hormone and glucocorticoids shown in the studies mentioned above. Furthermore, it appears that at least two Sp1 and AP-2 cis-acting elements in the proximal promoter region control the basal transcription activity. AP-2 transcription factor modulates the expression of many genes as an activator (13, 18) or as a repressor (17). The ability of the recombinant AP-2 transcription factor to physically interact with the NHE3 promoter region was established by DNase I footprinting experiments (Fig. 7) and GMSA (Fig. 8). AP-2 transcription factor binds to at least two regions in the NHE3 promoter. The proximal binding site at the −51 to −71 position is contained within the −95/+1 promoter fragment. Interaction of AP-2 with this motif was shown by GMSA and confirmed by supershift experiments where an antibody against AP-2α was utilized. Deletion of this AP-2 cis element along with the upstream Sp1 binding sites resulted in the loss of promoter activity of the corresponding reporter construct, alluding to the potential functional importance of AP-2 and Sp1 in transcription activation of the NHE3 promoter. The second AP-2 binding motif at −179 to −199 was not readily identified by homology search due to its lower sequence identity to the AP-2 consensus sequence; however, this region was clearly protected from DNase I digestion by interactions with the AP-2 protein. Dyck et al. (13) have shown that the AP-2 transcription factor is involved in the activation of the mouse NHE1 gene. The expression of AP-2 is increased on cell differentiation and is regulated in a cell-type specific manner (46). Our RT-PCR analysis of RNA isolated from C2/bbe cells grown for 1, 3, 7, 14, and 21 days postplating exhibited increasing levels of NHE3 mRNA expression on cell differentiation (J. Malakooti and K. Ramaswamy, unpublished data). In addition, Janecki and co-workers (15) have reported that the NHE3 gene activity increases during Caco-2 cell differentiation. Therefore, in light of these observations, it is tempting to speculate that the increased expression of NHE3 gene might be linked to increased AP-2 levels; however, this theory remains to be proven experimentally.
In summary, to understand how the apical Na+/H+exchanger isoforms are regulated, we have located and isolated the human NHE3 gene promoter. Our studies showed that the NHE3 promoter has potential binding sites for a number of nuclear proteins. According to the transient transfection assays in a luciferase-reporter system, the first 95 nucleotides upstream from the transcription initiation site of the NHE3 promoter hold the majority of the transcriptional activity of the NHE3 promoter. The −95/+1 fragment contains two overlapping Sp1 sites, one on the sense and the other on the antisense strands, of which the latter also overlaps with the AP-2 site. Deletion of a portion of the promoter that includes the binding sites for the Sp1 and AP-2 motifs resulted in a drastic decrease in the promoter activity. We showed by footprinting and GMSA assays that AP-2 interacted with this promoter region. Additional studies are underway to better understand how Sp1 family members interact with the NHE3 promoter region and how the overlapping Sp1 and AP-2 binding sites contribute to the expression of the NHE3 gene. The overlapping Sp1 and AP-2 binding sites have been reported to be involved in gene regulation in other systems (24,34). Therefore, this region might prove to be important for expression of the NHE3 gene.
We thank R. Dahdal for technical help.
This study was funded by the National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-33349 and by the Dept. of Veterans Affairs.
Address for reprint requests and other correspondence: J. Malakooti, Univ. of Illinois at Chicago, Dept. of Medicine, Section of Digestive and Liver Diseases, M/C 716, 840 S. Wood St., Chicago, IL 60612 (E-mail address:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2002 the American Physiological Society