Identification and characterization of a novel promoter for the human ANO1 gene regulated by the transcription factor signal transducer and activator of transcription 6 (STAT6)

Abstract

Anoctamin-1 (Ano1) is a widely expressed protein responsible for endogenous Ca²⁺-activated Cl⁻ currents. Ano1 is overexpressed in cancer. Differential expression of transcriptional variants is also found in other diseases. However, the mechanisms underlying regulation of Ano1 are unknown. This study identifies the Ano1 promoter and defines a mechanism for regulating its expression. Next-generation RNA sequencing (RNA-seq) analysis in human gastric muscle found a new exon upstream of the reported exon 1 and identified a promoter proximal to this new exon. Reporter assays in human embryonic kidney 293 cells showed a 6.7 ± 2.1-fold increase in activity over empty vector. Treatment with a known regulator of Ano1 expression, IL-4, increased promoter activity by 1.6 ± 0.02-fold over untreated cells. The promoter region contained putative binding sites for multiple transcription factors including signal transducer and activator of transcription 6 (STAT6), a downstream effector of IL-4. Chromatin immunoprecipitation (ChIP) experiments on T84 cells, which endogenously express Ano1, showed a 2.1 ± 0.12-fold increase in binding of STAT6 to P0 after IL-4 treatment. These results were confirmed by mutagenesis, expression, and RNA interference techniques. This work allows deeper understanding of the regulation of Ano1 in physiology and as a potential therapeutic target in a variety of diseases—Mazzone, A., Gibbons, S. J., Bernard, C. E., Nowsheen, S., Middha, S., Almada, L. L., Ordog, T., Kendrick, M. L., Reid Lombardo, KM., Shen, K. R., Galietta, L. J. V., Fernandez-Zapico, M. E., and Farrugia, G. Identification and characterization of a novel promoter for the human ANO1 gene regulated by the transcription factor signal transducer and activator of transcription 6 (STAT6).

Ano1 (TMEM16A, anoctamin1) is one of a family of genes encoding membrane proteins with 8 transmembrane spanning domains known as anoctamins. Ano1 and several other anoctamins encode Ca²⁺-activated Cl⁻ channels (1–3). Expression of Ano1 has been described in a broad range of tissues, including a variety of epithelia (1, 3–6), sensory cells (3, 7, 8), and smooth muscles (9–12). In the muscle layers of the human and mouse gastrointestinal tract, Ano1 is not present in smooth muscle cells but is expressed exclusively in the interstitial cells of Cajal (13).

Functionally, Ano1 plays a critical role in many physiologic processes, including chloride transport in airways (1, 9), salivary glands (3, 4), and gastrointestinal epithelial cells (14); rhythmic contraction (15, 16) and proliferation (17, 18) in the gastrointestinal tract; and heat sensation in sensory neurons (19). Thus, it is not surprising that Ano1 knockout mice do not survive long past weaning (9).

Aberrant expression and activity of Ano1 have also been implicated in the pathophysiology of several diseases, including cancer (20–23), cystic fibrosis (24), and hypertension (25–27). It has been identified as a possible therapeutic target in cystic fibrosis and asthma (24, 28), as well as hypertension-associated cardiovascular diseases such as stroke (25, 26) and pulmonary hypertension (27).

The widespread expression and diverse functions of the Ano1 protein indicate that expression of the gene must be controlled by several mechanisms. For example, in the lung, IL-4 is a potent inducer of Ano1 expression (1) and may be important in the pathophysiology of asthma (29). In several cancers, Ano1 expression is also highly up-regulated by mechanisms that are not fully established (20–23).

An additional complication in understanding the transcription of the Ano1 gene is the variety of splice variants distributed across different tissues and in different physiologic and pathologic situations. The published Ano1 gene comprises 26 exons coding for multiple splice variants in both human (1, 30) and mouse (31). We have reported on alternative splicing of the 5′ end of Ano1 and described 3 isoforms, including 1 with a shorter 5′ end. These isoforms had different electrical properties, and changes to the expression of these different transcripts were found in the disease gastroparesis, characterized by delayed stomach emptying (32). Although there is abundant evidence supporting the role of Ano1 dysregulation in the pathogenesis of different diseases, the mechanisms underlying the differential expression of this ion channel and its splicing variants in health and in disease are still elusive.

The goal of this study was to identify and characterize the human Ano1 promoter to further understand the regulation of Ano1 in diverse cell types and changes in expression associated with disease. By use of a combination of RNA-seq and rapid amplification of cDNA ends (RACE) experiments, we identified a novel promoter for the human Ano1 gene upstream of a newly described exon. Bioinformatics, luciferase, expression, and ChIP assays demonstrate that the promoter drives Ano1 expression and is modulated by IL-4, the best characterized activator of Ano1, via the STAT6 transcription factor.

MATERIALS AND METHODS

RNA sequencing experiments and analysis

Strips of human gastric smooth muscle were obtained from 8 nondiabetic patients undergoing duodenal switch gastric bypass surgery for obesity following institutional review board-approved protocols. Total RNA was extracted using the RNA-Bee solution (Tel-Test Incorporated, Friendswood, TX, USA) according to the manufacturer’s instructions, followed by an in-column cleanup with an RNeasy kit (Life Technologies, Grand Island, NY, USA). RNA-seq experiments were performed by the Mayo Clinic Medical Genome Facility Gene Expression Core using the Illumina TruSeq v2 library prep kit with 3 samples in each lane using a paired end index read with 51 base reads. Analysis of RNA-seq data was performed by the Mayo Clinic Division of Biomedical Statistics and Informatics.

Rapid amplification of cDNA ends

The RACE technique was used to isolate full-length 5′ ends of Ano1 mRNA, and it was performed using the SMARTer RACE cDNA amplification kit (Clontech, Mountain View, CA, USA) as described previously (32).

Bioinformatics analysis

Approximately 5 kb of sequence upstream of the newly identified exon 0 of human ANO1 was analyzed using 2 bioinformatics tools for the recognition of human RNA Polymerase II (PolII) promoter regions and start of transcription. Proscan Version 1.7 (www-bimas.cit.nih.gov/molbio/proscan/) predicts promoter regions based on scoring homologies with putative eukaryotic PolII promoter sequences. TSSG (linux1.softberry.com/berry.phtml?topic=tssg&group=programs&subgroup=promoter) predicts potential transcription start positions by combining characteristics describing functional motifs and oligonucleotide composition of these sites. To facilitate the investigation of binding sites on the P0 promoter, Gene Promoter Miner (GPMiner) was used (gpminer.mbc.nctu.edu.tw/index.php). Neural Network Promoter Prediction (www.fruitfly.org/seq_tools/promoter.html) is a method that identifies eukaryotic and prokaryotic promoters in a DNA sequence. Alignment of the protein sequences was performed using the multiple sequence alignment web tool ClustalW2 at EMBL-EBI (www.ebi.ac.uk/Tools/msa/clustalw2/) (33, 34).

Cloning and site-directed mutagenesis

For the full-length P0 reporter vector and all the truncation reporter plasmids, the fragments were amplified from human genomic DNA by PCR using specific primers (Table 1) and cloned into the pMetLuc vector included in the Ready-To-Glow secreted luciferase reporter system (Clontech) using the EcoRI restriction site. Binding sites for transcription factors of interest on P0 were modified using QuikChange Lightning Site-Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA, USA) according to the manufacturer instructions. The integrity of the constructs and the presence of the desired mutation were verified by DNA sequencing. The primers used are listed in Table 2.

TABLE 1.

Sequences of primers used for cloning of P0 promoter and deletion constructs into luciferase reporter vector

Fragment	Primer sequence (5′. . . 3′)
P0—forward	AGCCTGCCGCCCACGGAGAAT
−750—forward	GGTCGGGGACAGGCTGTGCAT
−591—forward	GTGGCTTGCGGGAGGGAACC
−342—forward	TCTCTCGGGGGCTGGCTCAAG
−250—forward	CCCCACTGCCCGCCCCAACT
−44—forward	GGGACCCCCAGACCAAAAAG
+13—forward	GCCAAGTTTGGCTACAGCAC
Common reverse	CTGGAGGTCTGCGGCGGCTCT

TABLE 2.

Sequences of primers used for site directed mutagenesis

Transcription factor	Forward (5′ . . . 3′)	Reverse (5′ . . . 3′)
STAT6	GCCTCATGGGCTCAGGGCAAAAAAAAGGGATTGCCCCTCTCCAGC	GCTGGAGAGGGGCAATCCCTTTTTTTTGCCCTGAGCCCATGAGGC
CEBP	CCTTGGTTTCTAAACCCCACAGGGCTGGAGATTTTTTTTCCCGGCTTCCTGCCC	GGGCAGGAAGCCGGGAAAAAAAATCTCCAGCCCTGTGGGGTTTAGAAACCAAGG
SMAD	AGCCCTGTGGGGTTTAGAAACCAAGAAAAAGACAATTCAGCATCCTCAGAGACAG	CTGTCTCTGAGGATGCTGAATTGTCTTTTTCTTGGTTTCTAAACCCCACAGGGCT
E2F/SP1	GGGGTGTGAGGTTTGTGGGTACACAAGGATTTTTCGGCACTGCAGTTCC	GGAACTGCAGTGCCGAAAAATCCTTGTGTACCCACAAACCTCACACCCC

Cell lines, culture conditions, and transfections

T84 cells (ATCC, Manassas, VA, USA), a human epithelial cell line derived from a lung metastasis of colon carcinoma, have been shown to endogenously express Ano1 (14); therefore, they are used here as an in vivo model to study the ANO1 promoter. For IL-4 treatment, cells were grown in complete medium containing 10 ng/ml human recombinant IL-4 (eBioscience, San Diego, CA, USA) for 24 h and then harvested for ChIP analysis as described below. Human embryonic kidney (HEK) 293 cells (ATCC) were used for the luciferase assays. They were transiently transfected with plasmids of interest using Lipofectamine 2000 (Life Technologies) according to the manufacturer’s instruction. After 4 h, the medium was changed to serum containing medium, and the cells were let rest for an additional 2 h. They were then lifted and divided into 3 wells for each condition. After 2 h rest for attachment, 100 µl media were collected as time 0 and then at defined time points for each condition. For IL-4 treatment, after dividing the cells in 3 wells, they were left for 16 h in medium containing 2% serum. Then the medium was changed to fresh medium ± 10 ng/ml IL-4. After 2 h (time 0) and then after 24 h, 100 µl of media was collected for analysis.

Luciferase assay

To assess the activity of P0 as a promoter, the Ready-To-Glow secreted luciferase reporter system (Clontech) was used, according to the manufacturer instructions. This system uses secreted Metridia luciferase as a reporter molecule to monitor the activity of promoters and enhancers by sampling media supernatant, without the need for cell lysis. To normalize for transfection efficiencies, the cells were cotransfected with pSEAP2-Control (Clontech), a plasmid with the constitutive SV40 early promoter inserted upstream of the SEAP gene and the SV40 enhancer inserted downstream. This vector expresses secreted embryonic alkaline phosphatase (SEAP), a secreted form of human placental alkaline phosphatase, as a reporter molecule. Luciferase assays were performed in HEK293 cells, because T84 cells had endogenous high expression of human placental alkaline phosphatase that made them unsuitable for the SEAP assay used for normalization.

Methylation assay

Analysis of methylation of the promoter was performed using the EZ DNA Methylation Kit (Zymoresearch, Irvine, CA, USA) based on the bisulfite conversion method. The reaction was performed according to the manufacturer’s instruction on DNA extracted from human stomach muscle strips using DNeasy kit (Qiagen, Valencia, CA, USA). After bisulfite conversion and desulfonation, the DNA was subjected to PCR using the following protocol: 94°C for 1 min, followed by 45 cycles at 94°C for 30 s, 58°C for 30 s, and 68°C for 30 s, finishing with 68°C for 7 min. The primers used were designed using the Bisulfite Primer Seeker Program (www.zymoresearch.com/tools/bisulfite-primer-seeker) and are listed in Table 3. After PCR, the bands were gel purified, cloned using the TA cloning kit (Life Technologies), and sequenced to analyze the presence of converted cytosines; 20–50 clones per sample were analyzed.

TABLE 3.

Sequences of primers used for methylation analysis

P0 region	Forward (5′ . . . 3′)	Reverse (5′ . . . 3′)	Size (bp)
1	TTTTTGTAGTTAATATTTTTTTTTTGGGGTTTTGGTTG	CRTACTATAACCAAACTTAACTACCCTC	350
2	GTTTTTYGGGGGTTGGTTTAAGGTTTTTTTTTGTAG	TCCTCCCCCTCRACTTTTTAATCTAAAAATCCC	332
3	GGTTTTGTYGTTTGGGGGAGGGGGTTAG	AAAACAACRACTAATCCCCAAATCCC	292
4	ATTTAGATTTTTATATTTTGTATAGGTTGTGGTTTG	TACACCCCCAACCAAAACCCCAAAAAAAAAATATTAAC	349
5	TGGGTAGAGGYGAATTTTTGGATTTAAGTTTAGG	ACCCTACRAAATACACCCCCAACCAAAACC	295
6	TGATTTGTAATTTTTAGTTTTTTTTTTTATTTTTGGG	ACCCTACRAAATACACCCCCAACCAAAACC	254
7	AATTTTYGGGGGGTGGGGAGGTTTG	AACCTAACTCRAAAAAAAACTATCCCCTCTC	166
8	AATTTYGGTTTTTTGTTTTGAGTTTATGAGG	ACACRAAAAAACTAAAACCCAAAAATAAAAAAAAAAAC	339
9	AGGAAGGGTYGGGGATAGGTTGTGTATTAGAGTTTTTG	ACACRAAAAAACTAAAACCCAAAAATAAAAAAAAAAAC	300

ChIP Assay

Twenty-four hours after IL-4 treatment, DNA/proteins from T84 cells were cross-linked with 1% formaldehyde, followed by cell lysis. DNA was sheared by 30 repeated cycles of 30 s of sonication followed by 30 s of rest using a Bioruptor 300 (Diagenode, Denville, NJ, USA). ChIP was performed as described previously (35). For the immunoprecipitation, 1 µg each of the following antibodies was used: nonimmune rabbit IgG and mouse IgG, anti-PolII and anti-PolII, S5-phosphorylated (Millipore, Billerica, MA, USA), anti-STAT6 (Cell Signaling, Danvers, MA, USA). Quantitative PCR was performed for each sample as described below and experiments were repeated 3 times. All the primers used are listed in Table 4.

TABLE 4.

Sequences of primers used for quantitative PCR and ChIP experiments

Gene	Forward (5′ . . . 3′)	Reverse (5′ . . . 3′)	Size (bp)
Ano1	GATCCCATCCAGCCCAAAGTG	CGGGTTTTGCTGTCGAAAAAGGA	131
Exon 0	CCGTGGATGGGGAGGGC	ATGATGTGGACGCTGCGG	266
STAT6	TGGAGGCCTTCCAGCACCGT	GAGAGCCATCCTGGCCCCGG	285
ChIP-ST6-1	AAAGCTGGGCACTAGGTGAAG	AGAATGCTAAGTGGCTCCGTG	187
ChIP-ST6-2	GAGGTTTGTGGGTACACAAGGA	GTCCCCGACCCTTCCTCTTC	195
ChIP-ST6-3	GAAGAGGAAGGGTCGGGGAC	AGTTAACTGCGGCTAAGAGGAC	93
ChIP-PolII-1	CTGCTCCTCTGCAGCGTC	GGTCCCGGGCAGACCT	112
ChIP-PolII-2	AGGTCTGCCCGGGACC	TCCCCTCTCTCCTCCCCA	126
ChIP-PolII-3	TGGGGAGGAGAGAGGGGA	AGACTAGGGAGCGACGAAGT	101

siRNA knockdown

ON-TARGET Plus SMARTpools against human STAT6 and nontarget control pools were obtained from Thermo Fisher Scientific (Waltham, MA, USA). Small interference RNAs at a final concentration of 300 nM were transiently transfected into T84 cells using X-tremeGENE HP DNA Transfection Reagent (Roche, Indianapolis, IN, USA). The cells were incubated for 48–72 h after siRNA transfection and then treated in the presence or absence of 10 ng/ml IL-4 for 24 h. Finally, RNA was extracted and used for quantitative RT-PCR experiments.

Quantitative RT-PCR

Reverse transcription reactions were conducted using the SuperScript VILO cDNA Synthesis Kit (Life Technologies) and 500 ng RNA for each sample. cDNA was then used for real-time PCR using the LightCycler 480 system with SYBR Green I Master Mix (Roche). The program used for all reactions was 95°C for 10 min, followed by 45 cycles at 95°C for 10 s, 60°C for 10 s, and 72°C for 20 s. Beta-actin was used as the housekeeping gene expression control (Qiagen). All ChIP samples were normalized to 2% input. The results were calculated following the ΔΔCt method. The sequences of all the primers used for the quantitative PCR experiments are listed in Table 4.

Electrophysiology

An expression vector bearing the novel isoform of human Ano1, indicated as Ano1(+0), was constructed by inserting the sequence for exon 0 into a construct containing the previously considered full-length Ano1 [here indicated as Ano1(−0)] using PCR and restriction digestion. Currents were recorded by standard whole-cell voltage clamp recordings from HEK293 cells transfected with either of the Ano1 isoforms examined together with the fluorescent marker green fluorescent protein (GFP). Experiments were performed as previously described (18). Significance was determined by unpaired t tests with Welch’s correction. A P value less than 0.05 was considered significant.

RESULTS

Identification of a new exon in the human ANO1 gene

RNA-seq experiments were performed on RNA extracted from the tunica muscularis of normal human stomach, a tissue that showed robust Ano1 expression and a diseased state associated with altered expression of Ano1 isoforms (32). By use of a whole-transcriptome sequencing approach, a new exon was identified upstream of the published exon 1. Figure 1A shows a representative sample of RNA-seq reads mapped on the 5′ end of Ano1. In addition to the sequencing results mapped onto the published exon 1 and its 5′-UTR region (between base pairs 69,924,408 and 70,035,651 on chromosome 11), there were several reads that spanned exon-exon junctions and continued approximately 93 kb upstream of exon 1 (between base pairs 69,831,997 and 69,832,273). Figure 1B represents a schematic of the 5′ end of the human ANO1 gene including mapping of the newly identified exon. Given its localization upstream of the known exon 1, we called this new exon “exon 0.” Interestingly, the mouse Ano1 gene (gene ID: 101772) presents a similar conformation with a first exon located 60 kb upstream of the second one. The mouse cDNA sequence has also been recently updated to include this exon (NM_178642.5), and 5′-RACE experiments confirmed the presence of an Ano1 variant with an additional 261 bp at the 5′ end (Fig. 1C, green). The new sequence had an ATG (Fig. 1C, green, underlined) that was in frame with the ATG in exon 1 (Fig. 1C, gray, underlined) and would produce a longer transcript with additional 120 bp at 5′ end and a protein that is 40 aa longer at the N terminus. The whole nucleotide sequence of the new exon has been deposited in the NCBI GenBank (accession number: KC577595). The sequencing results from the RACE experiment also demonstrated that the transcriptional start site (TSS) for the novel isoform is located 202 nucleotides upstream from the translation start site, and it corresponds to the TSS predicted by the Neural Network software. All the numbering relative to truncation and transcription factors’ binding sites in this report are enumerated considering this TSS as +1. Analysis of the human chromosome 11 sequence (NC_000011.9) showed canonical consensus sequences at the donor (GT) and acceptor (AG) splice sites between exons 0 and 1. These data confirm the presence of a previously unidentified exon located 93 kb upstream of the published exon 1 of the human ANO1 gene. Analysis of the published protein sequences for Ano1 showed that the mouse (NP_848757.4) and rat (NP_001101034.1) sequences but not of other mammals have been recently updated to include an extra exon at the 5′ end. Alignment of the predicted Ano1 protein with bovine (NP_001179646.1) and rodent species showed that the sequence of the new exon is not highly conserved (the percent identity of human vs. rodents is about 37%), but there are some similarities (e.g., the DIGLEGL in position 5 for human and 8 for rodents, Fig. 1D). This sequence does not contain any previously described protein domains.

Figure 1.

Identification of a new exon for human ANO1 upstream of the published exon 1. A) Alignment visualization of RNA-seq reads from a representative sample mapped to the hg19 human genome using IGV genome viewer. Top: Localization of human ANO1 gene on chromosome 11 and the coverage graphs. B) Schematic of the 5′ end of the human ANO1 gene where the boxes represent the exons and the lines are the introns. C) RACE results showing the sequence of the newly described exon 0 (green) upstream of exon 1 (gray), with the ATG in exon 0 (green, underlined) in frame with the ATG in exon 1 (gray, underlined). In red is exon 2 and in black, the 5′ end of exon 3. D) Multiple sequence alignment of the human, mouse, cow, and rat Ano1 protein N-terminal region. Asterisk (*) and colon (:) represent conserved and partially conserved nucleotides, respectively. Dashes indicate nucleotides absent either in the human, mouse, cow, and/or rat sequence.

Characterization of human ANO1 promoter

Prediction of possible promoter regions for the human ANO1 gene was performed using a 5000 bp sequence upstream of exon 0. The TSSG program predicted a promoter with a TSS at −45 bp from the previously published TSS, with a linear discriminant factor score of 19.71 (threshold = 4.00). ProScan predicted a putative promoter region between 250 and 40 bp upstream of the previously published TSS, with a score of 63.68 (cutoff = 53.00). Both analyses indicate that this promoter may be TATA-less. Based on these data, 1800 bp of genomic DNA upstream of exon 0 (referred to as P0) were amplified by PCR, cloned into the luciferase reporter vector, and transiently transfected into HEK293 cells. Twenty-four hours after transfection P0 resulted in a 6.7 ± 2.1-fold increase in activity when compared with the empty vector (Fig. 2A, n = 5, P < 0.05, t test). To further characterize the promoter region, a set of luciferase reporter gene constructs, containing successive 5′-deletion sequences of the P0 promoter ranging from −751 to +13, were engineered. Reporter gene assays showed that the deletion constructs resulted in a significant increase in promoter activity up to 10.2 ± 2.2-fold in the −250 construct, indicating the presence of repressing elements in the upstream region. Further deletion of the 5′ sequence resulted in progressive loss of activity and indicated that the −44 construct (∼150 bp) represented the minimal functional promoter (Fig. 2B, n = 5, P < 0.05, 1-way ANOVA with Dunn’s multiple comparison test).

Figure 2.

Characterization of putative promoter activity. A) Bar graph showing the results of the activity analysis by reporter gene luciferase assays of the construct bearing the putative promoter P0. The construct was transfected into HEK293 cells together with pSEAP vector for normalization. The results show a 6.7 ± 2.06-fold increase in activity compared with the empty vector (n = 5, *P < 0.05, t test). B) Activity analysis of 5′-deletion constructs of the P0 promoter by gene reporter assays. Left: Schematic representation of the promoter-reporter constructs containing the promoter fragments of different lengths cloned into the reporter vector. The arrow represents the TSS identified by RACE. Right: Normalized luciferase activity values for each construct (n = 5, *P < 0.05, 1-way ANOVA with Dunn’s multiple comparison test). Values are mean ± se. CTRL, control.

Next, bioinformatics analyses were performed using GPMiner on the minimal P0 sequence, a fragment that included 239 bp upstream of the TSS previously identified by RACE. This sequence (Fig. 3A) contained core promoter elements including several potential GC boxes (in positions −104, −103, and −102) and a CAAT box consensus (in position −35). As predicted, no TATA box was identified, but an initiator (INR) element (with sequence: CCAGACC) was found located 30 bp upstream of the TSS identified by RACE. This sequence was compatible with the consensus of INR as YYANWYY (IUPAC nucleotide code) but we did not obtain RACE products or RNA-seq data containing this sequence. ChIP experiments in T84 cells that endogenously express Ano1 showed recruitment of total PolII, RNA polymerase II (PolII) in this region of P0 (4.7 ± 0.3-fold, n = 3, P < 0.05, 1-way ANOVA with Dunn’s multiple comparisons posttest). Furthermore, occupancy of PolII with serine 5-phosphorylated C-terminal repeat domain (PolIIS5p), which predominates near the beginning of genes and marks initiating PolII (36), is increased in the region of P0 containing the TSS and the INR, showing that they may be involved in actual transcription of Ano1 (Fig. 3B, 17.4 ± 1.1-fold, n = 3, P < 0.05, 1-way ANOVA with Dunn’s multiple comparisons post-test). Confirmation of the presence of a transcript containing exon 0 came from quantitative RT-PCR experiments on RNA extracted from T84 cells showing detection of exon 0 by cycle 23.3 ± 0.4 (Fig. 3C, n = 3). The minimal promoter sequence presented an elevated GC content (73.4%) and analysis of the full-length P0 showed a lower but still elevated GC content (64.3%). Examination of the full-length P0 also showed a possible CpG island measuring 750 bp in length (from –620 to +130 bp) that included the TSS. Methylation analysis was carried out from DNA extracted from muscle strips isolated from human stomach. As shown in Fig. 3D, the total percent methylation was 15.3 ± 5.8% and was not statistically different from 0 (P = 0.08, t test, n = 4 human samples). There were variable levels of methylation in different clones, ranging from many with 0 to others with variable percentages up to maximum of 53%. This is consistent with expression of Ano1 in a subset of cells within the muscle layer. Therefore, P0 is an active and functional promoter for ANO1 with multiple regulatory elements.

Figure 3.

PolII binds to the minimal P0 promoter region and drives transcription of Ano1. A) Sequence of the minimal functional P0 promoter. Putative core promoter elements identified by GPMiner are indicated by boxes. An arrow in position +1 indicates the TSS of human ANO1 as determined by 5′-RACE experiments. B) Quantitative PCR results showing the increased occupancy of PolII phosphorylated on serine 5 (PolII S5p) vs. total PolII on the region of P0 with TSS and putative INR. Above the graphs is a schematic of the minimal P0 promoter with the promoter regions examined by ChIP identified by pattern, where Δ represent the GC boxes, ▾ is the INR, and the arrow is the TSS (n = 3, * P < 0.05, 1-way ANOVA with Dunn’s multiple comparison test). Values are mean ± se. C) Quantitative RT-PCR on RNA extracted from T84 cells showing expression of exon 0. Values are mean ± se, n = 3. D) Analysis of methylation using bisulfite conversion reaction shows that the minimal P0 promoter area in human stomach muscle strips has a low rate of methylation that is not statistically different from 0 (P = 0.08, t test, n = 4 samples).

The ANO1 promoter is induced by IL-4 in a STAT6-dependent manner

IL-4 has been shown to positively regulate expression of Ano1 (1). To test whether the activity of the P0 promoter could be modulated by IL-4, HEK293 cells transfected with the luciferase reporter vector were treated for 24 h with IL-4 at a final concentration of 10 ng/ml. In these conditions, the promoter activity of P0 was significantly increased compared with untreated controls (Fig. 4A, 1.6 ± 0.02-fold, n = 3, P < 0.05, t test). Furthermore, we detected an increased occupancy of active PolII (PolIIS5p) on the region of the TSS (Fig. 4B, 7.7 ± 0.7-fold, n = 3, P < 0.05, t test). Cells transfected with the promoter deletion constructs showed that treatment with IL-4 was not able to stimulate truncation constructs smaller than −750 (Fig. 4C, n = 4, P < 0.05, t test). Together these results indicate that the IL-4-induced gene transcription of Ano1 is mediated by transacting factors binding in the region of P0 that is upstream of the −750 truncation. GPMiner was again used to analyze the region of P0 between −1610 and −750 bp. The consensus sites for transcription factors identified in this region of P0 are indicated in Table 5. The most likely candidate was STAT6, a transcription factor whose role as a downstream effector of human IL-4 signaling has been extensively studied (see elsewhere for reviews) (37, 38). Also analyzed were transcription factors that have been implicated in modulating IL-4 signaling, such as E2F (39) and CCAAT/enhancer-binding protein (CEBP) (40), that are regulated by IL-4 such as specificity protein 1 (SP1) (41) or that are not directly involved in IL-4 signaling such as Mothers Against DPP Homolog 1 (SMAD) but were highlighted by GPMiner. To test the role of those transcription factors in the up-regulation of Ano1 in response to IL-4, the sequence of the putative binding site on the P0 promoter vector for each of them was disrupted by site-directed mutagenesis. The results show that the reporter vector fails to respond to IL-4 with an increased activity only in the case of the disruption of the STAT6 putative binding site (Fig. 5A, n = 3, P < 0.05, t test). Under basal conditions, the activity of each of the mutated construct was not different from the wild-type P0 construct (data not shown). To confirm binding of STAT6 on the promoter region of interest, ChIP experiments were performed on T84 cells cultured in presence or absence of IL-4. Treatment with IL-4 increased recruitment of STAT6 on the region of the promoter previously identified as the possible binding site (Fig. 5B, 2.1 ± 0.12-fold, n = 4, P < 0.05, 1-way ANOVA with Dunn’s multiple comparisons test). Together, these results suggest that STAT6 is the transcription factor mediating the up-regulation of Ano1 by IL-4.

Figure 4.

P0 promoter activity is up-regulated by IL-4. A) Treatment with IL-4 up-regulated the activity of P0 promoter by 1.6 ± 0.02-fold when compared with untreated cells. B) Quantitative PCR results showing the active PolII vs. total PolII enrichment on the region of interest of P0 after IL4 treatment (n = 3, * P < 0.05). C) 5′ deletions of P0 promoter abolished its modulation by IL4. Left: Schematic representation of the promoter-reporter constructs containing the promoter fragments of different lengths cloned into the reporter vector. Right: Normalized luciferase activity values for each construct (n = 3, *P < 0.05, t test). Values are mean ± se. CTRL, control.

TABLE 5.

Putative binding sites for transcription factors identified on P0 promoter sequence using GPMiner

Transcription Factors	Location from TSS
AML1	−1233
AREB6	−830
CEBP	−807
E2F	−916
ETF	−916 −1054
ETS	−791
MYB	−908
MZF1	−809 −1309
SMAD	−842
SP1	−916
STAT6	−792
PU1	−761
TGIF	−1080

Figure 5.

STAT6 mediates P0 promoter induction by IL-4. A) Luciferase assay performed with reporter constructs for P0 where the indicated putative transcription factor binding sites were deleted by site-directed mutagenesis. The top sequence is from the P0 region with the binding sites for transcription factors of interested underlined. The bar graph shows that only the disruption of the STAT6 binding site rendered the vector unresponsive to IL-4 stimulation. Values are expressed as mean ± se (n = 3, *P < 0.05, t test). B) Quantitative PCR results for ChIP performed on T84 cells showing the fold increase in occupancy of STAT6 on P0 in the region surrounding the position −792 (in green) when the cells were treated with IL4 compared with untreated cells. The patterns of the histograms represent the regions of the 5′ end of P0 as shown in the schematic on the right. Values are mean ± se (n = 4, *P < 0.05, Friedman test with Dunn’s multiple comparisons posttest). The arrow represents the TSS and the ▾ is the INR.

The role of STAT6 in this pathway was confirmed by testing the effects of up-regulation and knockdown of STAT6 on IL-4-induced Ano1 expression. Cotransfection of the P0 reporter vector with an expression vector for STAT6 significantly increases the activity of the promoter by 2.4 ± 0.4-fold (Fig. 6A, n = 3, P < 0.05, t test). When we repeated the same experiment using the −750 truncated form of P0 or the construct with a mutated consensus sequence for STAT6 binding to make it inactive (P0ΔST6), the promoter construct was unresponsive (Fig. 6A, 0.9 ± 0.04 and 0.9 ± 0.1 respectively, n = 3, P > 0.05, t test). Overexpression of STAT6 in T84 cells in absence of IL-4 stimulation resulted in a 2.3 ± 0.2-fold increase in the expression of mRNA for Ano1 (Fig. 6B, n = 3, P < 0.05, t test). Conversely, siRNA-mediated knockdown of STAT6 in the same cell line (0.35 ± 0.08-fold decrease of STAT6 compared with nontarget controls, n = 3,P < 0.05, t test), showed that, unlike cells transfected with nontarget siRNA (1.94 ± 0.15-fold increase in Ano1 expression, n = 3, P < 0.05, t test), there was no increase in the Ano1 mRNA expression after treatment with IL-4 (Fig. 6C, 1.02 ± 0.15-fold increase in Ano1 expression, n = 3, P > 0.05). These results suggest that STAT6 is a transcription factor mediating the up-regulation of Ano1 by IL-4.

Figure 6.

Manipulation of STAT6 reveals its critical role in the regulation of Ano1 expression in response to IL-4. A) Bar graph showing the analysis of the activity of P0 promoter reporter construct, the “−750” truncation construct, and the mutated construct with disrupted STAT6 consensus site (P0ΔST6) cotransfected with an expression vector for human STAT6. The results are normalized to a control performed by cotransfection with the empty vector used for STAT6 construct. Values are mean ± se (n = 3, *P < 0.05, t test). B) Bar graph showing in vivo data with the quantification of Ano1 mRNA by quantitative PCR in T84 cells transfected with an expression vector for STAT6. Values are expressed as mean fold change over mock-transfected cells ± se (n = 3, *P < 0.05, t test). C) Bar graphs showing the effect of STAT6 knockdown by siRNA on the expression of Ano1 mRNA in T84 cells after treatment with IL4. Values are expressed as mean fold change ± se (n = 3, *P < 0.05, t test). CTRL, control.

The novel ANO1 isoform results in greater current density

To test whether the addition of 40 amino acids to the N terminus of Ano1 due to the translation of the exon 0 sequence changes the function of the channel, we examined the current density of the 2 isoforms by whole-cell electrophysiology. We found that in response to a 1 s step to +100 mV from a −100 mV holding potential, Ano1(+0) Cl⁻ currents were greater than Ano1(−0) with free intracellular Ca²⁺ buffered at 500 nM (Fig. 7A). The average current for Ano1(−0) was 58 ± 8 pA/pF (n = 5) and for Ano1(+0) was 124 ± 20 pA/pF (n = 16, Fig. 7B, P < 0.05, unpaired t test with Welch’s correction).

Figure 7.

The novel Ano1 isoform results in greater current density. Currents were recorded from HEK293 cells transfected with GFP and either Ano1(+0) in black, Ano1(−0) in gray, or no channel, dotted line. A) Representative whole-cell Cl⁻ current traces were elicited by a 1 s depolarization from −100 to +100 mV. Intracellular free Ca²⁺ concentrations were 500 nM. B) Bar graph showing the average density of Cl⁻ currents over the final 0.5 s of the depolarizing step. ^†P < 0.05 compared with control, *P < 0.05 compared with Ano1(−0) by unpaired t tests with Welch’s correction. CTRL, control.

DISCUSSION

In the present work, we report on a novel promoter region for ANO1. We show that the human Ano1 full sequence includes the presence of an additional, hereto not described exon (exon 0) at the 5′ end of the human ANO1 gene located in the genome about 93 kb upstream of the previously published exon 1, suggesting that the previously published sequence was incomplete. Expression of the new isoform resulted in larger current density compared with the previously described full-length Ano1. In addition, we identify and characterize a promoter for ANO1 proximal to this exon. By reporter assays, deletion construct analysis, and ChIP, we found that the activity of the promoter is up-regulated by IL-4 through the STAT6 transcription factor.

Bioinformatics analyses showed that the promoter region of the human ANO1 gene lacks a canonical TATA box sequence within the expected proximity to the transcription start site but has a possible INR and a CAAT box. ChIP data on PolII accumulation showing occupancy of PolIIS5p on P0 provide assessment of the presence of an active TSS and/or INR. It is possible that the actual 5′ end of the transcript is the INR but the extension phase performed during the RACE experiment did not reach it because of the high GC-rich content. However, our RNA-seq data are consistent with the TSS site we identified by RACE. Given that we could not resolve the exact location of the TSS using ChIP experiments for PolII, we kept our numbering based on the TSS obtained by RACE and RNA-seq and identified it as +1. Interestingly, it has been generally accepted that a correlation exists between TATA-less promoters and multiple TSS usage (42) and ANO1 indeed presents a wide array of canonical (32) and noncanonical TSSs (43). In this context, it is possible that alternative TSSs can add flexibility to the ways in which Ano1 expression is regulated representing an extended transcriptional platform that can accommodate diverse transcriptional cofactors with different requirements for cell types or physiologic context. Thus, expression of the ANO1 gene in the various contexts where it plays a role can be modulated by these diverse mechanisms. These data are of interest given the widespread distribution of Ano1 and the increasing recognition that this channel is responsible for the endogenous Ca²⁺-activated Cl⁻ current found in several cell types (1–3).

Alternative splicing is a regulated process by which a single gene can encode for multiple proteins. Of particular interest in the current study is the presence of a very long intron between the exon 0 and exon 1, with the intron between exon 1 and exon 2 also being long (7543 bp). Intron length plays a crucial role during intron recognition and splicing. Recent work by Farlow and colleagues suggests that during evolution, as intron length increases, more potential cryptic splice sites can be generated by mutation. In fact, the longer the intron, the more elaborate the signals associated with exon definition. It is conceivable that, as a trade-off, within longer introns a lower filtering system is used to purge cryptic splice sites (44).

Previous studies have demonstrated up-regulation of Ano1 gene expression via the IL-4 signaling pathway (1); however, its mechanism was not known. In the present study, we provide evidence that significant induction of expression was mediated through regulatory elements located within the first 860 nucleotides in the P0 promoter. Within this region, the STAT6 binding site was the likely candidate for mediating the IL-4 stimulatory effects. We showed that disruption of the STAT6 binding site by truncation or by site-directed mutagenesis resulted in loss of transcriptional up-regulation of Ano1 after IL-4 treatment, and up-regulation of Ano1 promoter activity was seen on cotransfection with STAT6. Finally, ChIP assays confirmed the role of STAT6 in this pathway. The activity of STAT transcription factors is influenced by their association with other proteins, either through direct interaction on a promoter or through posttranslational modification (38). Therefore, other transcription factors may also contribute to the modulation of the IL-4 stimulation pathway. For example, SP1 binds to GC-boxes with great affinity and can regulate TATA-less or TATA-containing gene promoters by direct protein-protein interactions or by recruiting cofactors and other transcription factors (45), and SP1 has been previously shown to be up-regulated by IL-4 (41). Furthermore, it has been demonstrated that the STAT6 transcription factor specifically activated in response to IL-4 recognizes a subset of target sites for E2F family of transcription factors (identified in position −916 on P0, overlapping with SP1 consensus site) and can stimulate gene expression by binding to these DNA elements (39). In human keratinocytes, E26 transformation-specific (that has a possible binding site in position −1054 on P0) and STAT6 are physically associated within cells, and this interaction has been implicated in the regulation of Socs-1 mRNA expression (46). Our data suggest these transcription factors are not directly involved in the up-regulation of Ano1 by IL-4, but interplay of these transcription factors with STAT6 cannot be excluded and may also be important in the modulation of Ano1 response to IL-4.

The P0 promoter also had a CpG island located in the region from –620 to +130 bp. Previous studies suggested that CpG islands are, on average, 1000 bp long and show an elevated G+C base composition (47), and CpG islands containing promoters often lack TATA boxes and display multiple TSS (48) as is the case for P0. CpG islands represent a fraction of the genome with obvious regulatory potential. They frequently show absence of DNA methylation thus forming open regions of DNA that are highly accessible to transcription initiation complexes (49), which we showed is the case for Ano1 in human stomach muscle strips. However, several studies have shown that hypermethylation of promoter CpG islands result in stable transcriptional repression (see elsewhere for a review) (50), and aberrant methylation of CpG island can be more pronounced in pathologic conditions such as some tumor types (51). The presence of a CpG island on P0 could therefore add another layer of modulation to the transcriptional regulation of Ano1 expression, creating pockets of methylation that make P0 inaccessible to transcription factors, in particular physiologic conditions.

The diversity of putative response elements is consistent with widespread and highly regulated expression of the various transcript of Ano1. Bulk transport of chloride across epithelia requires high levels of protein expression in the membrane, whereas sensory function can be achieved by movement of small quantities of ions through the channel, as has been demonstrated for Ano1 in neurons (52). Varying the nature of the transcript will likely change the subcellular distribution of the protein (53) as well as the biophysical properties of the channel including chloride conductance and calcium sensitivity (32).

The newly identified isoform of Ano1 has 40 additional amino acids at the N terminus of the protein. We previously described alternative splicing as a versatile mechanism to regulate Ano1 function (32). By electrophysiology, the novel isoform, Ano1(+0), resulted in greater current density compared with Ano1(−0). Further detailed characterization is necessary to understand the exact role of the longer N-terminus in altering the electrophysiological properties of the Ano1 channel.

In conclusion, we described a novel exon for Ano1 located approximately 93 kb upstream of the published sequence. The novel Ano1 isoform resulted in greater current density compared with controls. We also identified a promoter for Ano1 and showed its regulation by STAT6. The association of STAT6 with P0 supports a critical role for this transcription factor in IL-4 stimulation of Ano1. Given that Ano1 has been identified as a therapeutic target for a wide variety of diseases, this work opens up the opportunity to now further understand the regulation of Ano1 and its splicing in vivo and changes in expression associated with disease.

Glossary

Ano1

anoctamin-1

ChIP

chromatin immunoprecipitation

CEBP

CCAAT/enhancer-binding protein

ETS

E26 transformation-specific

INR

initiator element

RACE

rapid amplification of cDNA ends

PolII

RNA polymerase II

PolIIS5p

RNA polymerase II phosphorylated on serine 5

RNA-seq

next-generation RNA sequencing

SEAP

secreted embryonic alkaline phosphatase

SMAD

Mothers Against DPP Homolog 1

SP1

specificity protein 1

STAT6

signal transducer and activator of transcription 6

TSS

transcriptional start site