Molecular Vision 2004; 10:351-360 <http://www.molvis.org/molvis/v10/a44/> Received 22 January 2004 | Accepted 15 May 2004 | Published 18 May 2004

Download Reprint

Identification of the promoter region of the human βIGH3 gene

Ching Yuan, Mei-Chuan Yang, Emily J. Zins, Christopher S. Boehlke, Andrew J. W. Huang
 
(The first two authors contributed equally to this publication)
 
 

Department of Ophthalmology, University of Minnesota, Minneapolis, MN

Correspondence to: Andrew J. W. Huang, M.D., Department of Ophthalmology, University of Minnesota, PWB Room 9-241, 516 Delaware Street S.E., Minneapolis, MN, 55455; Phone: (612) 625-6914; FAX: (612) 626-4455; email: huang088@umn.edu

Abstract

Purpose: To isolate and characterize the promoter of the human βIGH3 gene.

Methods: Primer extension and CapSite Hunting methods were used to determine the transcription start sites (TSS) of the human βIGH3 gene. Putative transcription factor-binding sites and potential promoter regions were identified by online tools. Two clones containing 3 Kb and 1 Kb of the 5'-flanking region of the βIGH3 gene were isolated and their respective promoter activities were characterized. Various fusion constructs of βIGH3 promoter-luciferase reporter were made to transfect A549 cells. The responses of these fragments to TGF-β1 were also measured after being treated with TGF-β1 at different concentrations. Several human and nonhuman cell lines were also transfected with the 1 Kb βIGH3 promoter-reporter construct to compare the activity of the βIGH3 promoter in these cells.

Results: The transcription start site of human βIGH3 mRNA was determined to be 65 bp upstream of the ATG start codon. Both the 3 Kb (-3011 to -1) and 1 Kb (-1000 to -1) fragments displayed strong and comparable promoter activity in transfected cells. Truncation analyses in A549 cells identified the nucleotide region from -336 to -1 as having high promoter activity (minimal promoter). The results also indicated that the nucleotide fragment from -1000 to -646 contained negative regulatory elements. Twenty ng/ml TGF-β1 upregulated the activity of the 1 Kb construct, but did not upregulate the activity of the -336 to -1 construct, suggesting that TGF-β1 responsive elements existed in the region from -1000 to -336. The 1 Kb construct universally demonstrated promoter activity in all cell lines tested.

Conclusions: We identified the βIGH3 gene promoter with a distinct regulatory pattern in the 1 Kb region upstream of the ATG start codon. Further elucidation of the functions of this promoter region may facilitate understanding of βIGH3 and its related corneal dystrophies.

Introduction

βIGH3 (TGF-β-induced gene-human, clone 3) is an essential constituent of the extracellular matrix responsible for cell adhesion and cell-matrix interactions. The encoding gene of βIGH3 was first discovered from a subtraction library screening in human adenocarcinoma cell line A549 treated with transforming growth factor-β1 (TGF-β1) [1,2]. It has various names such as βIGH3, BIGH3, βig-h3, beta ig-h3, keratoepithelin, RGD-CAP (in chicken and pig) [3,4], and MP78/70 [5]. In humans, the βIGH3 gene is located at chromosome 5q31 and encodes a 683-amino acid polypeptide which is highly conserved among species (human, mouse, chicken, and pig). βIGH3 is found widely in human tissues such as cornea, skin, lung, bone, bladder, and kidney [6]. Primary sequence analysis reveals that the encoded protein contains an N-terminus secretory signal peptide sequence. In βIGH3, there are four homologous domains designated as "fas-1" domains (about 140 amino acids for each domain) that are highly conserved among many secretory and membrane proteins from bacteria to mammals, such as mycobacterium MBP70, algae Algal-CAM, osteoblast-specific factor 2, periostin, and fasciclin I [7].

Native and recombinant βIGH3 have been shown to bind to matrix proteins such as laminin, collagen types I, II, IV, VI [3,8,9], and fibronectin [9]. βIGH3 and type VI collagen were co-localized to the fine microfibrils surrounding collagen fibers [5]. βIGH3 can enhance the spreading of osteoblasts [10], chondrocytes [11], and dermal fibroblasts [12] via integrin α₁β₁, and the spreading of corneal epithelial cells via integrin α₃β₁ [13]. Recently, we have demonstrated that the attachment of corneal keratocytes was also facilitated by βIGH3 [14]. In addition, it has been shown that the second or fourth fas-1 domain alone is sufficient to mediate cell spreading via integrin [13]. Therefore, the RGD motif of βIGH3 may not be required for mediating cell spreading since truncated recombinant βIGH3 still retains the capability for enhancing cell attachment. This RGD motif, however, can exert an apoptotic effect on corneal epithelial cells and other cell lines as revealed by two recent studies [15,16]. Besides promoting cell adhesion, βIGH3 was found to support the migration and proliferation of dermal keratinocytes [17], alter cell morphology, and inhibit cell cycling in CHO cells [2]. It also affects cell differentiation [13], reduces tumorigenesis in nude mice [2], and alters hormone response to dexamethasone [18]. This evidence strongly indicates that, in addition to acting as the "molecular glue" for cell adhesion to extracellular matrices, βIGH3 can potentially exert its effect on other cellular functions via interaction with integrins.

βIGH3 has been purported to play a role in maintaining the integrity of the normal cornea as well as during corneal wound healing, as it is upregulated with fibronectin in injured corneas [19]. It has been detected at the subepithelium/stroma junction and at the stroma/Descemet's membrane interface of the human cornea [11]. Not only is the βIGH3 gene expressed by corneal epithelial cells, but corneal endothelium can also synthesize βIGH3 following cell loss from surgical and pathological processes [8]. Previous reports failed to detect βIGH3 mRNA transcripts in normal adult keratocytes, therefore the βIGH3 protein in adult corneal stroma was thought to diffuse from the epithelium [8,19,20].

Direct evidence linking corneal dystrophies to βIGH3 mutations was first reported by Munier and co-workers [21]. Several 5q31-linked autosomal dominant stromal corneal dystrophies such as granular corneal dystrophy types I, II (Avellino corneal dystrophy), and III (Reis-Bucklers dystrophy), lattice corneal dystrophy types I, IIIA, and IV, and Thiel-Behnke dystrophy are correlated with permutations in the βIGH3 gene [21-23]. The genotype/phenotype correlation associated with βIGH3 mutations is succinct: R124H in Avellino corneal dystrophy, R124C in lattice corneal dystrophy type I, and R555W in granular corneal dystrophy [24-27]. To date, at least eight different types of βIGH3-related corneal dystrophies attributed to approximately 24 missense mutations have been identified [22,23].

The βIGH3-related inherited corneal dystrophies, such as lattice and granular dystrophies, are characterized by the presence of corneal opacities composed of amyloid or non-amyloid/fibril deposits and recurrent epithelial erosions [28]. Abnormal protein products are deposited in the corneal epithelia and stroma of these clinical entities (depending on the specific mutations) as accumulations of rod-shaped crystalloid, amyloid, or curly fibers. These protein deposits are immunoreactive with antibodies against βIGH3 proteins, suggesting that βIGH3 is a major component of these deposits [28,29]. A recent study [30] revealed that recombinant βIGH3 proteins could form fibrillar structures in vitro. An in vitro model using native and mutated peptides to produce amyloid fibrils has also been established [31].

To better understand the role of βIGH3 in corneal wound healing and related corneal dystrophies, it is prudent to conduct further studies of the βIGH3 gene, such as identifying promoter elements that control gene expression. This paper describes the cloning, sequencing, and characterization of promoters of the βIGH3 gene. We have identified the transcription start site and putative transcription binding sites that are essential for basal expression of the gene. In addition, the responsiveness of βIGH3 promoters to TGF-β1 was also evaluated.

Methods

Primer extension experiment

To determine transcription start site(s), primer extension experiments were conducted utilizing non-radioactive, infrared (IR) dye-labeled primers for non-radioactive automated primer extension analysis (NAPE) according to the Li-Cor (Lincoln, NE) protocol. The sensitivity of NAPE has been shown to be comparable to the conventional radioisotope-labeling method for primer extension experiments by various researchers [32-34]. Two IR-700 dye labeled antisense primers corresponding to the sequence of +7 to +26 (IR700-HβIGH3-A.3) and +141 to +160 (IR700-HβIGH3-B.3) in the βIGH3 coding region were synthesized by Li-Cor and used for extension reactions. The later primer is located at the beginning of the second exon of the βIGH3 gene and both primers amplify βIGH3 cDNA fragments by RT-PCR. Total RNA from A549 cells was isolated with the RNAeasy mini kit (Qiagen, Valencia, CA). The primers (0.1 to 1 μM final concentration) were annealed with total RNA (5 to 20 μg) in a 10 μl reaction volume by first heating to 90 °C for two min and then slowly cooling to 30 °C within 30 min in a thermocycler. Reverse transcriptions were performed at 42 °C for one h with 20 units of Superscript II in the presence of RNaseOut RNase inhibitor (Invitrogen, Carlsbad, CA). EDTA, pH 8.0, was added to stop the reactions (5 mM final concentration) and DNase-free RNase A was also used to remove remaining RNA. First strand cDNA was then precipitated with 10 mM spermine tetrahydrochloride and resuspended in sample buffer containing 95% formamide, 10 mM EDTA and 0.1% bromophenol blue. The products were analyzed on 5.5% acrylamide/7 M urea gels and scanned with the Odysse image system (Li-Cor) for their IR signals.

CapSite Hunting experiment

The CapSite Hunting method [35,36] was used to conclusively determine the transcription start site(s) of human βIGH3 mRNA. A human heart CapSite cDNA^TM library (Nippon Gene, Tokyo, Japan) was used to clone the 5'-UTR region of βIGH3 according to the manufacturer's instruction. In brief, the 5'-terminal m⁷GpppN cap structure of mRNA was removed by tobacco acid phosphatase and a synthetic RNA oligo (rOligo) was recapped by T4 RNA ligase. The recapped mRNA was then reverse-transcribed into first strand cDNA, after which the 5'-UTR region in the gene of interest was amplified by nested PCR. The first PCR reaction was performed using the rOligo-specific primer 1RC and HβIGH3-B.3. The second round of PCR was performed using another rOligo-specific primer 2RC and HβIGH3-A.3. The reaction conditions were as follows; 95 °C for 5 min, followed by 25 to 35 cycles of 95 °C for 20 s, 60 °C for 20 s, and 72 °C for 45 s, with a final extension of 72 °C for 10 min. The PCR products were separated on an agarose gel and DNA fragments were isolated and cloned into a pCR-Blunt vector (Zero-blunt cloning kit, Invitrogen). The identities of individual clones were confirmed by the standard automated sequencing method at the Microchemical Facilities at the University of Minnesota.

Generation of constructs containing βIGH3 gene promoters

A human genome BLAST homology search of the public database was performed using human βIGH3 cDNA (BE206112). DNA sequences of the 5'-flanking region of the βIGH3 gene were amplified by PCR using human genomic DNA with specific primers designed for human βIGH3 gene from the BLAST search. A clone containing approximately 3 Kb of the upstream sequence was amplified with two primers; βIGH3-KpnI-3Kb.5, aaa agg tac cgt gag gtc cag tga act tg and βIGH3-NheI-3Kb.3, gtt tgc tag cgg agc ggg acg acg cgc acc. The amplified sequence covered the nucleotides from -3011 to -1 (the base pair before the ATG start codon). Another clone encoding the -1000 to -1 region of βIGH3 was PCR-amplified by βIGH3-KpnI-1Kb.5, gaa tgg tac cct tca tgg aac atc att ggc ttg gg and βIGH3-HindIII-1Kb.3, gtt taa gct tgg agc ggg acg acg cgc acc. Owing to the high GC content of the sequence immediately preceding the ATG start codon, DMSO (dimethyl sulfoxide) was added to a final concentration of 5% to facilitate amplification by Pfu polymerase (Stratagene, La Jolla, CA). The cycling conditions were as follows: Initial denaturation at 95 °C for 2 min, followed by thirty-two cycles of denaturation at 95 °C for 45 s, annealing at 60 °C for 45 s, elongation at 72 °C for 1 min (for 1 Kb PCR product) or 3 min (for 3 Kb PCR product), and then followed by 10 min of final extension at 72 °C. The PCR-amplified products were gel-purified and subcloned into pCR-Blunt vectors according to the manufacturer's protocol. Resulting clones were sequenced by standard automated sequencing and their sequences were confirmed to be identical with the results of our BLAST search.

Analyses of the promoter region and putative transcription factor-binding sites using online tools

To identify putative transcription factor-binding sites and potential promoter regions, the 3 Kb and 1 Kb upstream DNA sequence of the βIGH3 gene was analyzed with several promoter and transcription factor analysis tools available online; PROSCAN (BioInformatics & Molecular Analysis Section, NIH, Bethesda, MD), PromoterInspector and MatInspector (Genomatix Software GmbH München, Germany), and TFSEARCH (Computational Biology Research Center, Tokyo, Japan).

Construction of βIGH3 promoter-luciferase reporter plasmids

The corresponding 3 Kb and 1 Kb fragments were excised by Kpn I/Nhe I or Kpn I/Hind III digestion and ligated into the multiple cloning site of the pGL3-Basic vector (Promega, Madison, WI) using luciferase as the reporter. Since our initial data revealed that the efficacies of promoting βIGH3 transcription were comparable between the 3 Kb and 1 Kb fragments, the 1 Kb fragment was chosen as the working template to generate various truncated constructs to study the regulatory functions of this promoter. For 5'-deletion constructs, double digestions by Kpn I/BstX I, Kpn I/Stu I, or Kpn I/Sma I were used to remove various 5' portions of the potential promoter region. The digested vector arms were further blunt-ended with Pfu polymerase (Stratagene) at 72 °C for 1 h, gel-purified and then self-ligated at 16 °C overnight to generate constructs corresponding to -646 to -1, -336 to -1, and -87 to -1 bases upstream of the βIGH3 gene. For 3'-deletion constructs, double digestions by BstX I/Hind III, Stu I/Hind III, or Sma I/Hind III were used to remove various 3'-portions of the potential promoter region to generate constructs corresponding to -1000 to -646, -1000 to -336, and -1000 to -87 bases upstream of the βIGH3 gene. The individual constructs were sequenced to confirm their integrity. The activity of each truncated construct was evaluated by measuring the activity of luciferase expressed in the A549 cell line.

Cell culture and transient transfection

Human corneal epithelial cell primary culture and other cell lines (A549, 293, 3T3, COS-7, and HCE-T; ATCC, Manassas, VA) were used to test the efficacies of various constructs of the βIGH3 promoter. Human cornea buttons were obtained from the Minnesota Lions Eye Bank with an exemption from the Institutional Review Board, University of Minnesota. After removal of the endothelium, the dissected corneal explants were cultivated in DMEM (Dulbecco's modified essential medium, Invitrogen)/F-12 medium (Invitrogen) containing 0.5% DMSO, 30 ng/ml of cholera toxin (Sigma, St. Louis, MO), 2 ng/ml EGF (Sigma), 5% fetal bovine serum (FBS; Hyclone, Logan, UT), 1X ITS medium supplement (Sigma, containing insulin, transferrin and selenium) and antibiotics. Cultured corneal epithelial cells were trypsinized and seeded at 3,500 cells/cm², then grown to 50-70% confluence before transfection. A human corneal epithelium cell line, HCE-T, was cultured in KGM-2 medium (Clonetics, San Diego, CA). A549 cells were cultured in F-12K medium containing 10% FBS. 3T3 (mouse embryonic fibroblasts), COS7 (monkey embryonic kidney fibroblast cells), and 293 (human embryonic kidney epithelial cells) were cultured in DMEM containing 10% FBS. All cells were incubated at 37 °C in a humidified atmosphere containing 5% CO₂. Cells were seeded in 12-well plates and grown for at least 24 h before transfection to 50 to 70% confluence.

Lipofectamine (Invitrogen) was used for transfection according to the manufacturer's protocol. The pGL3 (containing firefly luciferase as a reporter) and pRL (containing Renilla luciferase as an internal control) plasmids were co-transfected in order to normalize the variability of transfection efficiency in each cell line. Prior to transfection, 0.2 μg of DNA-pGL3 plasmid and 0.04 μg of pRL were added to 24 μl of serum-free OPTI-MEM1 (Invitrogen), and 2 μl of Lipofectamine Reagent were added to 24 μl of OPTI-MEM1. The diluted DNA and Lipofectamine were combined and incubated for 15 min at room temperature before transfection. Another 450 μl OPTI-MEM1 were added to the DNA/liposome mixtures in each well. The cells were incubated at 37 °C in a CO₂ incubator for 4 h and the medium was replaced with fresh culture medium. In HCE-T cells and primary cultured corneal epithelial cells, KBM-2 medium (basal medium without growth supplement) instead of OPTI-MEM1 was used for transfection because the high calcium concentration in OPTI-MEM1 was known to cause the differentiation of corneal epithelial cells.

Luciferase assay

Cells were harvested 48 h after transfection. After aspirating the media and washing cells with 1 ml of 1X PBS, cells were lysed by adding 250 μl of 1X passive lysis buffer (Dual Luciferase Reporter System, Promega), and the culture plates were gently rocked/agitated on a rotating platform for 15 min at room temperature. The above lysate (20 μl) was used for the measurement of luciferase activity in the Lumat LB 9507 Luminometer (Berthold Technologies USA, Oak Ridge, TN) according to the manufacturer's instructions. The expression of firefly and Renilla luciferase was measured sequentially from a single sample, and the activity of each promoter fragment was obtained directly by the ratio of firefly to Renilla luciferase.

The pGL3-Promoter vector containing SV40 promoter sequence was used as a positive control, and pGL3-Basic vector without promoter or enhancer was used as a negative control. The data from the luciferase experiments were then compared with the activity of pGL3-Basic vector, and promoter activity was expressed as an n-fold increase over the pGL3-Basic vector activity. Statistical analyses were performed with Analysis of Variance (ANOVA). A p value less than 0.05 was considered as statistically significant.

Effects of TGF-β1 on promoters

To investigate the effect of TGF-β1 on the βIGH3 promoter, A549 cells (about 60 to 80% confluence) were transfected with -1000 to -1 or -336 to -1 fragment-luciferase constructs. After four h of transfection, cells were washed and allowed to recover in serum-containing medium for 8 h, and then serum-starved for another 12 h. Cells were then treated with 0, 2, and 20 ng/ml of recombinant TGF-β1 (Sigma) in the absence of serum for another 48 h. The activity of the promoter fragments in response to TGF-β1 stimulation was evaluated.

Results

Transcription starting sites of βIGH3

Our initial analysis did not identify any TATA or CAAT box in the proximal region of the βIGH3 gene. Therefore, it is possible that the human βIGH3 gene has multiple transcription starting sites (TSS). We adopted two approaches, primer extension and CapSite Hunting techniques, in order to determine TSS of the human βIGH3 gene conclusively. One major and two minor bands (Figure 1A, lane 2), corresponding to 91, 88, and 87 bp were identified from primer extension experiments. The major product corresponds to a potential TSS site 65 bp upstream of the ATG start codon (since the primer used in this reaction, HβIGH3-A.3, starts 26 bp downstream of the ATG site). When using a different primer, HβIGH3-B.3, corresponding to a region that is 134 bp further downstream of HβIGH3-A.3 for primer extension, one major and two minor bands with higher shift mobility (about 140 bp) were also observed (data not shown). The primer extension experiments have been repeated five times from three different RNA preparations and displayed the same patterns. These results indicated the transcription of βIGH3 mRNA starts at 65, 62, and 61 bp upstream of the start codon with the -65 bp location as the major TSS.

We further used the CapSite Hunting method to identify the TSS of the human βIGH3 gene (Figure 1B). Although the first round of PCR did not generate any visible amplified products (Figure 1B, lane 2), the 2RC/HβIGH3-A.3 primers-mediated nested PCR (the second round of PCR, Figure 1B, lane 3) revealed a DNA band with a size corresponding to about 100 bp. The resulting amplicons were cloned into pCR-Blunt vectors and sequenced extensively. The boundary of 2RC sequence was used to determine where the transcription of mRNA starts. Our results clearly showed that the TSS was located 65 bp upstream of the ATG start codon (Figure 1C), reconfirming the results obtained from the primer extension experiments. Thereby, we assigned the major TSS of the human βIGH3 gene to be located 65 bp upstream of the ATG start codon (Figure 2A, indicated by the asterisk). We could not confirm the -62 and -61 bp TSS from the CapSite Hunting results, as indicated by our primer extension results (Figure 1A, two minor bands corresponding to -88 and -87 bp).

Transcription factor-binding sites within the proximal 5'-flanking region of the βIGH3 gene

Although no TATA box or CAAT-like motif could be identified by our initial investigation, further analysis of the 1 Kb sequence of the 5'- flanking region of the βIGH3 gene by Proscan and PromoterInspector predicted a potential promoter to be located within the -270 to -1 region. Proscan, TFSEARCH and MatInspector also revealed numerous putative transcription factor-binding sequences for MZF1, MYOD, CEBPB, IK2, GATA1, E2F, STAT, Sp1, and VBP within this 1 Kb region (Figure 2A). From our search, there are five putative Smad binding elements (denoted SBE-1 to SBE-5, Figure 2A) identified within the -1000 to -380 bp region. Several transcription factor-binding sites were clustered within the -139 to -125 region (MZF1 and Sp1) and the -107 to -75 region (VBP, CEBPB, and Sp1; Figure 2A). These transcriptional factors might play a role in regulating TGFβ1 responsiveness and the promoter activity in the βIGH3 gene. The proximal regions of the 5'-UTR in βIGH3 gene are highly conserved between mouse and human, as shown in Figure 2B.

PCR amplification of the βIGH3 promoter

Our initial attempt to amplify the 5'-flanking region of βIGH3 from -1000 to -1 bp using a pair of βIGH3-Kpn I-1Kb.5 and βIGH3-Hind III-1Kb.3 primers did not produce any DNA fragments. After inspection of the 5' flanking sequence, a GC-rich (77%) region was noted extending from -170 bp to the ATG start codon. However, we were able to amplify the corresponding region successfully using βIGH3-Kpn I-1Kb.5 and a different 3'-primer that is upstream of this GC-rich region (data not shown). Therefore, DMSO (final concentration at 0, 1, and 5%) was added to the subsequent PCR reaction mixtures to improve amplification. As shown in Figure 2C, 5% DMSO greatly facilitated the PCR amplification of the 5'-flanking region (from -1000 to -1) of βIGH3.

Truncation analysis of the βIGH3 promoter

To identify sequences essential for basal transcription, constructs with various fragments of the βIGH3 promoter fused to luciferase gene were generated (Figure 3A), and their activities to promote transcription of the reporter gene were evaluated in the A549 cell line. The 1 Kb construct (-1000 to -1) had a 6.4 fold increase of activity over the pGL3-Basic vector (Figure 3B). The 3 Kb construct (-3011 to -1) displayed comparable promoter activity to 1 Kb with no significantly statistical difference (p>0.05). This result suggested that the absence of the nucleotide sequence from -3011 to -1001 is not critical for basal promoter activity. In the 5'-deletion series, the fragments -646 to -1 and -336 to -1 expressed higher levels of luciferase activity (27.5 fold and 23.6 fold that of the pGL3-Basic vector, respectively; and around 2 fold greater than pGL3 containing SV40 promoter). Comparison between the -646 to -1 and -336 to -1 constructs showed no statistically significant difference in their promoter activities (p>0.05). A shorter fragment containing only the -87 to -1 bps had significantly lower activity and exhibited a promoter activity that was only 1.6 fold higher than that of pGL3-Basic vector. Our results suggest that -336 to -1 fragment is the minimal promoter by definition. Since -646 to -1 and -336 to -1 constructs displayed much higher activity than that of 1 Kb, one possible explanation is that the region of -1000 to -646 bp contains negative regulatory elements which suppress transcriptional activity. In the 3' deletion series, three fragments -1000 to -646, -1000 to -336 and -1000 to -87 were generated. None of these constructs showed significant promoter activity in A549 cells (0.998, 0.993, and 0.88 fold that of pGL3-Basic vector, Figure 3B).

βIGH3 promoter response to TGF-β1 in A549 cells

To investigate transcription regulation of the βIGH3 gene by TGF-β1, we used the 1 Kb (the working template) and -336 to -1 fragments (minimal promoter) for the comparison study in A549 cells. The activity of the 1 Kb fragment treated with 2 ng/ml and 20 ng/ml TGF-β1 was 1.2 times and 2.3 times the baseline, respectively (Figure 4). TGF-β1 up-regulated the activities of the 1 Kb construct in a dose-dependent manner. The minimal promoter (-336 to -1), on the other hand, was not affected by treatment with TGF-β1. These results suggested that the TGF-β1-response element was located outside the minimal promoter and was within the -1000 to -336 region.

Expression of the βIGH3 promoter in different cell lines

Transient transfections of the βIGH3 1 Kb construct into A549, 293, 3T3, COS-7, HCE-T, and human corneal epithelial primary cultures (HCE-P) showed different transcriptional activities in different cell lines (Figure 5). The 1 Kb promoter in all transfected cells universally increased the expression of luciferase, and ranged from an 8.5 fold increase in A549 to a 62.6 fold increase in 293 cells. The enhancement of the luciferase level by βIGH3 gene promoter in primary corneal epithelial culture was similar to that in HCE-T cell culture (10.8 and 13.7 fold increase, respectively), and there was no statistically significant difference between these two cell lines (p>0.05).

Discussion

βIGH3 was first identified from TGF-β1-treated A549 cells [1]. The mRNA transcript was detected as a 3.4 Kb band from northern hybridization [12]. There is currently no report regarding alternative splicing of βIGH3 mRNA. The βIGH3 protein is usually expressed at modest levels in most cells and is significantly increased after TGF-β1 treatment [1,2]. In human patients, mutations of βIGH3 correlate well to several types of autosomal dominant corneal dystrophies linked to chromosome 5q [21]. The expression of βIGH3 has been noted among many tissues and organs such as kidney, heart, skin, and cornea [6,12]. Although the roles of βIGH3 in cell attachment and tumorigenesis have been investigated by many researchers, the expression mechanism of the βIGH3 gene and its regulation by TGF-β1 remains unclear.

To facilitate the understanding of gene expression of βIGH3, we first set out to identify the promoter of βIGH3 by constructing and characterizing the 5'-flanking region of the βIGH3 gene. We further determined the TSS of βIGH3 mRNA in A549 cells. By both primer extension and CapSite Hunting methods, we found that the major TSS of βIGH3 mRNA is located at -65 bp (Figure 1 and Figure 2A) upstream from the ATG start codon. Since the 5'-flanking region of the βIGH3 gene does not contain any TATA box, multiple TSS were expected to be found. Our primer extension results did reveal a major band of 91 bp and two minor bands with sizes of 88 and 87 bp (Figure 1A). They corresponded to the -65, -62, and -61 bp upstream from the ATG start codon in the βIGH3 gene, respectively. The major TSS was also identified from the CapSite human heart cDNA library using mRNA from A549 cells. It is known that TSS could very well be cell-dependent and vary among tissues. Genes such as nNOS [37] and dystrophin [38] display tissue-specific TSS-dependent promoter activities. In nNOS, the diversity of the 5'-UTR can also have major impacts on the translation of mRNA [39]. When we searched the published cDNA clones containing βIGH3 cDNA, TSS in the 5'-UTR regions were highly variable among different tissues: -72 bp for uterus (AK094055), -65 bp for trachea (AK093916), -47 bp for lung adenocarcinoma (M77349), -49 bp and -54 bp for renal adenocarcinoma (BC004972 and BC000097), -65 bp for hypothalamus (BC026352), and -105 bp for amygdala (AK094581). Whether the expression of βIGH3 can be modulated post-transcriptionally or translationally via TSS-dependent mechanisms awaits further investigation. The TSS we identified herein likely originates from a common major transcription version of βIGH3 mRNA in human heart and A549 cells. This TSS site fits the sequence requirements for the eukaryotic mRNA initiator: PyPy(A+1)N(A/T)PyPy, where Py is C or T (pyrimidines), A+1 is the start site of transcription and N is any of the 4 bases.

We further demonstrated that the immediate upstream region of βIGH3 gene possesses a strong intrinsic promoter activity in vitro, suggesting its potential role as the promoter for βIGH3 gene in vivo. Although no obvious TATA box or CAAT motif was found, we identified a number of putative transcription factor consensus sites within the 1 Kb sequence in the 5'-flanking region of the βIGH3 gene such as MZF1, IK2, GATA-1, CEBPB, and Sp1 (Figure 2A). Since MZF1, IK2, and GATA-1 were reported to be involved in the differentiation and maturation of blood cells [40-42], we surmise the expression of βIGH3 in hematopoietic cells may be subjected to the regulation of these transcription factors [43]. When various 5'-truncation constructs (which still contain the immediate region upstream of the ATG start codon) were transfected into A549 cells, they displayed either comparable or significantly higher activity than the SV40 viral promoter activity. A high sequence homology was conserved between human βIGH3 and murine βigm3 genes in this proximal promoter region (Figure 2B), further suggesting its critical role in gene expression and regulation. Truncation analyses of various fragments of this βIGH3 promoter region showed that the -336 to -1 bp fragment had a high promoter activity, commonly known as the minimal promoter or basic promoter. This finding is consistent with the potential promoter site within the -270 to -1 region as predicted by Proscan and other tools. We also noted that several transcription factor-binding sites cluster within this region (MZF1, CEBPB, VBP, and Sp1, Figure 2A), further supporting its potential role in gene regulation. Sp1 sites are especially interesting since they are often present in TATA-less promoters [44] and were shown to activate transcription via interaction with basic transcription machinery (such as TFIID factor) [45]. There are three Sp1 sites located within the -336 to -1 region in βIGH3 gene and the functional significance of these Sp1 sites awaits further study.

Deletion at the 3'-end of the promoter region eliminated the entire promoter activity (Figure 3B). One possible explanation is that the -87 to -1 fragment is a crucial component of the aforementioned -336 to -1 fragment. This region contains the Sp1 transcription factor binding site located between -85 and -75 bp upstream of the βIGH3 gene (Figure 2A) and could be essential for the promoter activity. Alternatively, since we found that the TSS of the βIGH3 is located at -65 bp and the pGL-basic vector used by us does not contain promoter or TSS elements, deletion of this region (-87 to -1) would completely abolish the transcription of the reporter gene (as shown in Figure 3B). We also found that longer constructs (such as -1000 to -1) had less promotional activities toward βIGH3. This finding suggests that there is an inhibitory segment for βIGH3 gene activation between -1000 and -336.

The expression of βIGH3 is primarily modulated by TGF-β1. Several groups have reported the up-regulation of βIGH3 by TGF-β1 during tumorigenesis or transformation of cultured cells [2,46-49]. To further elucidate the molecular mechanism of TGF-β1-dependent expression of βIGH3, we tested the promoter activities of various constructs in response to TGF-β1 treatment. The differential response between the -1000 to -1 and -336 to -1 fragments to TGF-β1 indicates that deletion of the -1000 to -336 region abolished the responsiveness of the βIGH3 promoter to TGF-β1. We therefore speculate that a TGF-β1 responsive element resides within the -1000 to -336 bp region. Upon activation by TGF-β1, this element negates the inhibitory effect of -1000 to -336 (in the absence of TGF-β1) and upregulates the expression of βIGH3. Interestingly, we found five putative Smad protein binding elements (SBE-1 to SBE-5) within this region. Since signal transduction of TGF-β1 is mediated via the Smad and/or MAPK pathways, these SBEs could play essential roles in regulating the promoter activity in βIGH3. Co-operations of Smad with other transcription factors are often necessary for efficient promoter activities [50] such as Sp1 in the case of the α2(I) collagen promoter [51,52]. We speculate that a similar concerted mechanism between SBEs and Sp1 (or other elements) could exist for the regulation of βIGH3 promoter. Further study by DNA footprinting and mutagenesis of these SBE and Sp1 sites should help identify transcription factors involved in the TGF-β1 mediated process. In addition, other regulatory elements which may be present in the 3'-UTR region or downstream of the ATG start codon should be further explored.

Since the expression of βIGH3 is widely distributed in tissues, we further investigated whether the promoter-mediated βIGH3 expression is cell-specific or tissue-specific. Primary corneal epithelial cells, as well as several other cell lines transfected with the 1 Kb construct, displayed a universally enhanced expression of reporter gene. We have also noted that a cell line such as 293 that does not have endogenous βIGH3 expression (data not shown) also displayed prominent reporter activity. The observed upregulation of βIGH3 expression by the 1 Kb promoter construct is likely to be mediated through a universal TGF-β1-dependent mechanism in the cells we tested.

To the best of our knowledge, the βIGH3 promoter has never been investigated in patients with βIGH3-related corneal dystrophy. In addition to mutations in coding regions with resultant mutant βIGH3 protein, augmented expressions of βIGH3 protein via the gene promoter may also be another potential pathogenic mechanism. It remains unclear whether altered βIGH3 expression could contribute to the pathogenesis of the βIGH3-related corneal dystrophies, possibly by mutation in its promoter region or by altered responsiveness to TGF-β1. Characterizing the promoter region and its activation mechanism should help to elucidate the pathophysiology of these corneal dystrophies.

In summary, we have identified a putative βIGH3 promoter within the 1 Kb upstream region of the βIGH3 gene, which includes a minimal promoter of 336 bp (from -336 to -1). The transcription starting site for βIGH3 mRNA in A549 cells was at the -65 bp location. We also observed the presence of putative regulatory element(s) for βIGH3 expression in the region between -1000 to -336 bp. Upon adding TGF-β1, the inhibitory effect exerted by the regulatory elements was neutralized and the promoter activity was upregulated. Further studies will be needed to better delineate the modulatory mechanisms of TGF-β1 on βIGH3 and its promoters.

Acknowledgements

This study was supported in part by Minnesota Medical Foundation Grants 3180-9927-02, a grant from Eye Bank Association of America, and an unrestricted grant from the Research to Prevent Blindness. The authors thank Professor Cliff Steer for assisting in data interpretation.

References

1. Skonier J, Neubauer M, Madisen L, Bennett K, Plowman GD, Purchio AF. cDNA cloning and sequence analysis of beta ig-h3, a novel gene induced in a human adenocarcinoma cell line after treatment with transforming growth factor-beta. DNA Cell Biol 1992; 11:511-22.

2. Skonier J, Bennett K, Rothwell V, Kosowski S, Plowman G, Wallace P, Edelhoff S, Disteche C, Neubauer M, Marquardt H, Rodgers J, Purchio AF. beta ig-h3: a transforming growth factor-beta-responsive gene encoding a secreted protein that inhibits cell attachment in vitro and suppresses the growth of CHO cells in nude mice. DNA Cell Biol 1994; 13:571-84.

3. Hashimoto K, Noshiro M, Ohno S, Kawamoto T, Satakeda H, Akagawa Y, Nakashima K, Okimura A, Ishida H, Okamoto T, Pan H, Shen M, Yan W, Kato Y. Characterization of a cartilage-derived 66-kDa protein (RGD-CAP/beta ig-h3) that binds to collagen. Biochim Biophys Acta 1997; 1355:303-14.

4. Ohno S, Doi T, Tsutsumi S, Okada Y, Yoneno K, Kato Y, Tanne K. RGD-CAP ((beta)ig-h3) is expressed in precartilage condensation and in prehypertrophic chondrocytes during cartilage development. Biochim Biophys Acta 2002; 1572:114-22.

5. Gibson MA, Kumaratilake JS, Cleary EG. Immunohistochemical and ultrastructural localization of MP78/70 (betaig-h3) in extracellular matrix of developing and mature bovine tissues. J Histochem Cytochem 1997; 45:1683-96.

6. Schorderet DF, Menasche M, Morand S, Bonnel S, Buchillier V, Marchant D, Auderset K, Bonny C, Abitbol M, Munier FL. Genomic characterization and embryonic expression of the mouse Bigh3 (Tgfbi) gene. Biochem Biophys Res Commun 2000; 274:267-74.

7. Kawamoto T, Noshiro M, Shen M, Nakamasu K, Hashimoto K, Kawashima-Ohya Y, Gotoh O, Kato Y. Structural and phylogenetic analyses of RGD-CAP/beta ig-h3, a fasciclin-like adhesion protein expressed in chick chondrocytes. Biochim Biophys Acta 1998; 1395:288-92.

8. Hirano K, Klintworth GK, Zhan Q, Bennett K, Cintron C. Beta ig-h3 is synthesized by corneal epithelium and perhaps endotheliumin Fuchs' dystrophic corneas. Curr Eye Res 1996; 15:965-72.

9. Billings PC, Whitbeck JC, Adams CS, Abrams WR, Cohen AJ, Engelsberg BN, Howard PS, Rosenbloom J. The transforming growth factor-beta-inducible matrix protein (beta)ig-h3 interacts with fibronectin. J Biol Chem 2002; 277:28003-9.

10. Kim JE, Kim EH, Han EH, Park RW, Park IH, Jun SH, Kim JC, Young MF, Kim IS. A TGF-beta-inducible cell adhesion molecule, betaig-h3, is downregulated in melorheostosis and involved in osteogenesis. J Cell Biochem 2000; 77:169-78.

11. Ohno S, Noshiro M, Makihira S, Kawamoto T, Shen M, Yan W, Kawashima-Ohya Y, Fujimoto K, Tanne K, Kato Y. RGD-CAP ((beta)ig-h3) enhances the spreading of chondrocytes and fibroblasts via integrin alpha(1)beta(1). Biochim Biophys Acta 1999; 1451:196-205.

12. LeBaron RG, Bezverkov KI, Zimber MP, Pavelec R, Skonier J, Purchio AF. Beta IG-H3, a novel secretory protein inducible by transforming growth factor-beta, is present in normal skin and promotes the adhesion and spreading of dermal fibroblasts in vitro. J Invest Dermatol 1995; 104:844-9.

13. Kim JE, Kim SJ, Lee BH, Park RW, Kim KS, Kim IS. Identification of motifs for cell adhesion within the repeated domains of transforming growth factor-beta-induced gene, betaig-h3. J Biol Chem 2000; 275:30907-15.

14. Yuan C, Reuland JM, Lee L, Huang AJ. Optimized expression and refolding of human keratoepithelin in BL21 (DE3). Protein Expr Purif 2004; 35:39-45.

15. Kim JE, Kim SJ, Jeong HW, Lee BH, Choi JY, Park RW, Park JY, Kim IS. RGD peptides released from beta ig-h3, a TGF-beta-induced cell-adhesive molecule, mediate apoptosis. Oncogene 2003; 22:2045-53.

16. Morand S, Buchillier V, Maurer F, Bonny C, Arsenijevic Y, Munier FL, Schorderet DF. Induction of apoptosis in human corneal and HeLa cells by mutated BIGH3. Invest Ophthalmol Vis Sci 2003; 44:2973-9.

17. Bae JS, Lee SH, Kim JE, Choi JY, Park RW, Yong Park J, Park HS, Sohn YS, Lee DS, Bae Lee E, Kim IS. Betaig-h3 supports keratinocyte adhesion, migration, and proliferation through alpha3beta1 integrin. Biochem Biophys Res Commun 2002; 294:940-8.

18. Dieudonne SC, Kerr JM, Xu T, Sommer B, DeRubeis AR, Kuznetsov SA, Kim IS, Gehron Robey P, Young MF. Differential display of human marrow stromal cells reveals unique mRNA expression patterns in response to dexamethasone. J Cell Biochem 1999; 76:231-43.

19. Rawe IM, Zhan Q, Burrows R, Bennett K, Cintron C. Beta-ig. Molecular cloning and in situ hybridization in corneal tissues. Invest Ophthalmol Vis Sci 1997; 38:893-900.

20. Akhtar S, Meek KM, Ridgway AE, Bonshek RE, Bron AJ. Deposits and proteoglycan changes in primary and recurrent granular dystrophy of the cornea. Arch Ophthalmol 1999; 117:310-21.

21. Munier FL, Korvatska E, Djemai A, Le Paslier D, Zografos L, Pescia G, Schorderet DF. Kerato-epithelin mutations in four 5q31-linked corneal dystrophies. Nat Genet 1997; 15:247-51.

22. Klintworth GK. Advances in the molecular genetics of corneal dystrophies. Am J Ophthalmol 1999; 128:747-54.

23. Munier FL, Frueh BE, Othenin-Girard P, Uffer S, Cousin P, Wang MX, Heon E, Black GC, Blasi MA, Balestrazzi E, Lorenz B, Escoto R, Barraquer R, Hoeltzenbein M, Gloor B, Fossarello M, Singh AD, Arsenijevic Y, Zografos L, Schorderet DF. BIGH3 mutation spectrum in corneal dystrophies. Invest Ophthalmol Vis Sci 2002; 43:949-54.

24. Mashima Y, Yamamoto S, Inoue Y, Yamada M, Konishi M, Watanabe H, Maeda N, Shimomura Y, Kinoshita S. Association of autosomal dominantly inherited corneal dystrophies with BIGH3 gene mutations in Japan. Am J Ophthalmol 2000; 130:516-7.

25. Sakimoto T, Kanno H, Shoji J, Kashima Y, Nakagawa S, Miwa S, Sawa M. A novel nonsense mutation with a compound heterozygous mutation in TGFBI gene in lattice corneal dystrophy type I. Jpn J Ophthalmol 2003; 47:13-7.

26. Kuchle M, Green WR, Volcker HE, Barraquer J. Reevaluation of corneal dystrophies of Bowman's layer and the anterior stroma (Reis-Bucklers and Thiel-Behnke types): a light and electron microscopic study of eight corneas and a review of the literature. Cornea 1995; 14:333-54.

27. Okada M, Yamamoto S, Tsujikawa M, Watanabe H, Inoue Y, Maeda N, Shimomura Y, Nishida K, Quantock AJ, Kinoshita S, Tano Y. Two distinct kerato-epithelin mutations in Reis-Bucklers corneal dystrophy. Am J Ophthalmol 1998; 126:535-42.

28. Korvatska E, Henry H, Mashima Y, Yamada M, Bachmann C, Munier FL, Schorderet DF. Amyloid and non-amyloid forms of 5q31-linked corneal dystrophy resulting from kerato-epithelin mutations at Arg-124 are associated with abnormal turnover of the protein. J Biol Chem 2000; 275:11465-9.

29. Streeten BW, Qi Y, Klintworth GK, Eagle RC Jr, Strauss JA, Bennett K. Immunolocalization of beta ig-h3 protein in 5q31-linked corneal dystrophies and normal corneas. Arch Ophthalmol 1999; 117:67-75.

30. Kim JE, Park RW, Choi JY, Bae YC, Kim KS, Joo CK, Kim IS. Molecular properties of wild-type and mutant betaIG-H3 proteins. Invest Ophthalmol Vis Sci 2002; 43:656-61.

31. Schmitt-Bernard CF, Chavanieu A, Derancourt J, Arnaud B, Demaille JG, Calas B, Argiles A. In vitro creation of amyloid fibrils from native and Arg124Cys mutated betaIGH3((110-131)) peptides, and its relevance for lattice corneal amyloid dystrophy type I. Biochem Biophys Res Commun 2000; 273:649-53.

32. Jankovic I, Egeter O, Bruckner R. Analysis of catabolite control protein A-dependent repression in Staphylococcus xylosus by a genomic reporter gene system. J Bacteriol 2001; 183:580-6.

33. Taniyama Y, Sato K, Sugawara A, Uruno A, Ikeda Y, Kudo M, Ito S, Takeuchi K. Renal tubule-specific transcription and chromosomal localization of rat thiazide-sensitive Na-Cl cotransporter gene. J Biol Chem 2001; 276:26260-8.

34. Fiegler H, Bassias J, Jankovic I, Bruckner R. Identification of a gene in Staphylococcus xylosus encoding a novel glucose uptake protein. J Bacteriol 1999; 181:4929-36.

35. Maruyama K, Sugano S. Oligo-capping: a simple method to replace the cap structure of eukaryotic mRNAs with oligoribonucleotides. Gene 1994; 138:171-4.

36. Takakura M, Kyo S, Kanaya T, Hirano H, Takeda J, Yutsudo M, Inoue M. Cloning of human telomerase catalytic subunit (hTERT) gene promoter and identification of proximal core promoter sequences essential for transcriptional activation in immortalized and cancer cells. Cancer Res 1999; 59:551-7.

37. Wang Y, Newton DC, Robb GB, Kau CL, Miller TL, Cheung AH, Hall AV, VanDamme S, Wilcox JN, Marsden PA. RNA diversity has profound effects on the translation of neuronal nitric oxide synthase. Proc Natl Acad Sci U S A 1999; 96:12150-5.

38. Ahn AH, Kunkel LM. The structural and functional diversity of dystrophin. Nat Genet 1993; 3:283-91.

39. Newton DC, Bevan SC, Choi S, Robb GB, Millar A, Wang Y, Marsden PA. Translational regulation of human neuronal nitric-oxide synthase by an alternatively spliced 5'-untranslated region leader exon. J Biol Chem 2003; 278:636-44.

40. Hromas R, Collins SJ, Hickstein D, Raskind W, Deaven LL, O'Hara P, Hagen FS, Kaushansky K. A retinoic acid-responsive human zinc finger gene, MZF-1, preferentially expressed in myeloid cells. J Biol Chem 1991; 266:14183-7.

41. Georgopoulos K, Bigby M, Wang JH, Molnar A, Wu P, Winandy S, Sharpe A. The Ikaros gene is required for the development of all lymphoid lineages. Cell 1994; 79:143-56.

42. Leonard M, Brice M, Engel JD, Papayannopoulou T. Dynamics of GATA transcription factor expression during erythroid differentiation. Blood 1993; 82:1071-9.

43. Goltry KL, Dobry CJ, Jensen TC, Smith AK. Identification of a putative novel spliced variant of a TGF-beta induced gene, BIGH3, expressed in bone marrow: potential role in regulating hematopoietic cell growth. Blood 1999; 94:39a.

44. O'Leary KA, McQuiddy P, Kasper CB. Transcriptional regulation of the TATA-less NADPH cytochrome P-450 oxidoreductase gene. Arch Biochem Biophys 1996; 330:271-80.

45. Emami KH, Burke TW, Smale ST. Sp1 activation of a TATA-less promoter requires a species-specific interaction involving transcription factor IID. Nucleic Acids Res 1998; 26:839-46.

46. Buckhaults P, Rago C, St Croix B, Romans KE, Saha S, Zhang L, Vogelstein B, Kinzler KW. Secreted and cell surface genes expressed in benign and malignant colorectal tumors. Cancer Res 2001; 61:6996-7001.

47. Zhao YL, Piao CQ, Hei TK. Overexpression of Betaig-h3 gene downregulates integrin alpha5beta1 and suppresses tumorigenicity in radiation-induced tumorigenic human bronchial epithelial cells. Br J Cancer 2002; 86:1923-8.

48. Zhao YL, Piao CQ, Hei TK. Downregulation of Betaig-h3 gene is causally linked to tumorigenic phenotype in asbestos treated immortalized human bronchial epithelial cells. Oncogene 2002; 21:7471-7.

49. Tsujimoto H, Nishizuka S, Redpath JL, Stanbridge EJ. Differential gene expression in tumorigenic and nontumorigenic HeLa x normal human fibroblast hybrid cells. Mol Carcinog 1999; 26:298-304.

50. Massague J, Wotton D. Transcriptional control by the TGF-beta/Smad signaling system. EMBO J 2000; 19:1745-54.

51. Poncelet AC, Schnaper HW. Sp1 and Smad proteins cooperate to mediate transforming growth factor-beta 1-induced alpha 2(I) collagen expression in human glomerular mesangial cells. J Biol Chem 2001; 276:6983-92. Erratum in: J Biol Chem 2001; 276:47746.

52. Inagaki Y, Nemoto T, Nakao A, ten Dijke P, Kobayashi K, Takehara K, Greenwel P. Interaction between GC box binding factors and Smad proteins modulates cell lineage-specific alpha 2(I) collagen gene transcription. J Biol Chem 2001; 276:16573-9.