Regulation of Corneal Keratin-12 Gene Expression by the Human ...

2 downloads 0 Views 655KB Size Report
KLF6 mRNA and protein in five human total corneas. The same experimental design was used to explore human corneal epi- thelial (HCE) cells in 20 patients ...
Regulation of Corneal Keratin-12 Gene Expression by the Human Kru ¨ ppel-like Transcription Factor 6 Fre´de´ric Chiambaretta,1,2 Loı¨c Blanchon,2 Be´ne´dicte Rabier,2 Winston W.-Y. Kao,3 Janice J. Liu,4 Bernard Dastugue,2 Danie`le Rigal,1 and Vincent Sapin2 PURPOSE. The keratin-12 (K12) protein is essential for the integrity of the corneal epithelium. This study was conducted to investigate the possible involvement of Kru ¨ ppel-like factor 6 (KLF6) in the corneal regulation of K12 gene expression, in view of the presence of one KLF6 potential binding site in the human K12 promoter and the known role of KLF6 in regulating keratin gene expression. METHODS. RT-PCR, Western blot analysis, and immunolocalization experiments were used to investigate the expression of KLF6 mRNA and protein in five human total corneas. The same experimental design was used to explore human corneal epithelial (HCE) cells in 20 patients and a HCE cell line. The ability of the KLF6 protein to modulate K12 promoter activity was studied in the HCE cell line, by transient transfections with a KLF6 expression plasmid and promoter–reporter gene assays. Gel-shift assays were performed to confirm the interactions between the KLF6 protein and specific sequences of the K12 promoter. RESULTS. The presence of KLF6 transcripts and proteins in human total corneal extracts was demonstrated. Immunohistofluorescence experiments showed positive staining specifically present in the corneal epithelial layer. KLF6 transcripts and proteins were also present in corneal epithelial cells in 20 patients and the HCE cell line. Transient transfections of KLF6 showed statistical transactivation of the K12 promoter in HCE cells. The gel-shift assay showed a physical interaction between KLF6 and the K12 promoter. CONCLUSIONS. The expression of KLF6 in HCE cells and its role in the regulation of K12 gene expression were demonstrated. (Invest Ophthalmol Vis Sci. 2002;43:3422–3429)

T

he integrity of corneal epithelial cells is vital for the maintenance of a normal human ocular surface and adequate vision. As in other epithelia, this corneal integrity depends on the harmonious expression of a cornea-specific keratin pair: keratin-12 (K12) and keratin-3 (K3). For example, mutations in the human K12 gene have been linked to Meesmann’s dystrophy, characterized by the fragility of the anterior corneal epithelium and intracorneal microcysts.1 In addition, homozygous

From the 1Department of Ophthalmology, University of ClermontFerrand, France; 2National Institute of Health and Medical Research (INSERM) Unit 384, Faculty of Medicine, Clermont-Ferrand, France; the 3 Department of Ophthalmology, University of Cincinnati, Cincinnati, Ohio; and the 4Department of Ophthalmology, University of Washington School of Medicine, Seattle, Washington. Submitted for publication April 4, 2002; revised June 17, 2002; accepted July 8, 2002. Commercial relationships policy: N. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact. Corresponding author: Vincent Sapin, INSERM U.384, Laboratoire de Biochimie, Faculte´ de Me´decine, 28, place Henri-Dunant, 63001 Clermont-Ferrand, France; [email protected].

3422

knockout mice without the K12 gene display mild corneal epithelial erosion.2 Keratins are a group of water-insoluble proteins that contribute to the formation of cytoskeletal networks by intermediate filaments in epithelial cells.3,4 Their relative charges permit a division into acidic and basic neutral subfamilies. Usually, a basic keratin is coexpressed and paired with an acidic keratin to form a functional pair. The expression of keratin pairs is tissue specific, differentiation dependent, and developmentally regulated. Expression of the K12-K3 pair has been found in human, bovine, guinea pig, rabbit, and chicken corneas.3,5,6 The expression of K12 is restricted to the corneal epithelium, although K3 may be undetectable in mouse cornea.6,7 Despite the physiological importance of K12 in corneal epithelium, the molecular mechanisms that govern cornea epithelial cell–specific expression of K12 remain poorly understood. Vitamin A and several growth factors8 –10 are known to be extracellular signals that control differentiation and proliferation of corneal epithelial cells. However, little is known about intracellular transcription factors that play a pivotal role in expression of the K12 gene. To date, it has been demonstrated only that the paired box homeotic gene 6 (PAX6) and the Ets family transcription factor ESE-1 may activate the expression of the K12 gene.11,12 Recently, Kru ¨ ppel-like factor 6 (KLF6/Zf9/CPBP) has been shown to regulate the human keratin-4 gene in esophageal squamous epithelium.13 KLF6 is a member of the family of Kru ¨ ppel-like factors (KLFs), composed of 15 nuclear transcription factors sharing a highly conserved C-terminal DNA-binding domain containing three zinc fingers, primarily described in Drosophila protein.14 KLF6 contains a proline- and serine-rich amino terminal activation domain, and like other KLFs, three carboxyl-terminal C2H2 zinc fingers that interact directly with the promoter of target genes through a GC box element. By binding to a promoter region that possesses CACCC homology or is rich in CG content, KLFs are known to play a critical role in the regulation of genes involved in tissue development, differentiation, angiogenesis, hematopoiesis, cellular cycle control, proliferation, and differentiation.15 Cloned originally from cDNA libraries of placenta, human KLF6 mRNA is ubiquitously expressed, with a high level of expression in lung, intestine, prostate, and placenta,16,17 but its expression has never been explored in the eye. Given the presence of one potential KLF6 binding site (GC-rich region) in the human K12 promoter, we investigated the presence of KLF6 in the cornea and its ability to regulate K12 gene expression. Our purpose was to determine whether KLF6 mRNAs and proteins are present in corneal tissue and, more specifically, in corneal epithelium. Further experiments were performed to study the colocalization of KLF6 and K12 in corneal epithelium. We also demonstrated that KLF6 proteins activate the K12 promoter, and we determined the region responsible for the binding of KLF6 in the K12 promoter. Our results indicate that KLF6 is an important transcription factor for cornea-specific K12 expression. Investigative Ophthalmology & Visual Science, November 2002, Vol. 43, No. 11 Copyright © Association for Research in Vision and Ophthalmology

IOVS, November 2002, Vol. 43, No. 11

Corneal Regulation of K12 by hKLF

3423

FIGURE 1. KLF6 transcripts expression in human epithelial cornea. cDNAs from total corneal extract (lanes 1, 2, 3), epithelium corneal cells (lanes 4, 5), and HCE cell line (lanes 6, 7) were amplified by PCR with KLF6 specific primers used to generate 323-bp products (lanes 3, 5, 7). Lanes 2, 4 and 6: cDNAs were amplified by PCR with K12-specific primers (398 bp). cDNAs of total corneal extract were amplified by PCR with GAPDH-specific primers (434 bp). Lane 8 has no DNA template (negative control).

MATERIALS

AND

METHODS

Human Cornea and Human Corneal Epithelial Cells In compliance with the Declaration of Helsinki on research involving human subjects, human corneal epithelial (HCE) cells were obtained from 20 eyes of 20 patients (with astigmatism or myopia) who were undergoing epithelial ablation for excimer laser photorefractive keratectomy. Informed consent was obtained from the subjects after the nature and possible consequences of the study had been explained to them. The research was approved by the institutional human experimentation committee. Just before keratectomy, an area 8 mm in diameter encompassing the whole corneal epithelial layer was scraped mechanically, immediately frozen in liquid nitrogen, and stored at ⫺80°C until use. Five human corneas (not retained by the regional

FIGURE 2. Expression of KLF6 protein in human corneal epithelium. The Western blot of total corneal extract (lane 1) and of protein extracts from epithelium corneal cells of patients (lane 2, 3) and the HCE cell line (lanes 4) were probed with anti hKLF6. Lane 5: total corneal extract was incubated without primary antibody. Equal loading of protein in each lane was affirmed by quantitative measurements, and confirmed with ponceau staining.

cornea graft bank) were rinsed using sterile cold phosphate-buffered saline (PBS) and embedded with optimal cutting temperature compound (Tissue Tec; Sakura, Zoeterwounde, The Netherlands) and stored at ⫺80°C until used. For immunohistologic experiments, cryosections (5 ␮m thick) were cut and mounted on slides (Super Frost; Fischer Scientific, Pittsburgh, PA).

Cell Cultures A human corneal epithelium (HCE) cell line transformed with simian virus (SV40; CRL11135American Type Culture Collection [ATTC], Philadelphia, PA) was cultured under standard conditions (5% CO2, 95% humidified air, 37°C) in DMEM-F12 supplemented with 5% fetal calf serum, 5 ␮g/mL insulin, 0.1 ␮g/mL cholera toxin, 50 mg/mL streptomycin, 50 IU/mL penicillin, 0.5 mg/mL epithelial growth factor and

3424

Chiambaretta et al.

IOVS, November 2002, Vol. 43, No. 11

FIGURE 3. Colocalization of KLF6 and K12 in the human corneal epithelial layer. KLF6 and K12 protein expression were examined in human cornea (A– H; adjacent sections) by immunohistologic experiments using anti-KLF6 (E–H) and anti-K12 antibody (A, B). (C, D) Negative data obtained using 4⬘,6⬘-diamino-2-phenylindole (DAPI) staining (blue) without primary antibody incubations. (G, H) Respectively, the same as (E) and (F) without DAPI staining, to illustrate the cytoplasmic and nuclear immunostaining of KLF6 in the epithelial layer. Secondary antibodies FITC conjugated for KLF6 and rhodamine conjugated for K12 were used. Micrographs were acquired with a standard fluorescence microscope. Magnifications: (A, C, E, G) ⫻20; (B, D, F, H) ⫻100. 0.5% dimethyl sulfoxide (DMSO). The COS-7 cell line was cultured in DMEM supplemented with 5% fetal calf serum, 50 mg/mL streptomycin, and 50 IU/mL penicillin. All the media and supplements were obtained from NEN Life Science (Paris, France). Cells were plated in 25- or 75-cm2 flasks (Falcon Labware; BD Biosciences, Plymouth, UK). For immunohistologic experiments, cells were seeded into eight-well chamber slides (Laboratory-Tek; Merck Eurolab, Strasbourg, France) at a density of 2 ⫻ 104 cells/well.

RNA Extraction and RT-PCR Experiments mRNA was extracted from human total cornea, HCE cell line and HCE cells with a mRNA purification kit (Quickprep Micro; Amersham Pharmabiotech, Les Ulis, France), according to the recommendations of the manufacturer. cDNA for RT-PCR was generated with a synthesis system (Superscript First-Strand Synthesis System; Gibco-BRL, CergyPontoise, France). The specific oligonucleotide primers used for the

PCR reaction were originally generated using the Web program “Primer3” based on the published full-length human mRNA sequences of each specific gene: KLF6 sense (S) 5⬘-ACCCGGCCCGACATGGACG TG-3⬘, KLF6 antisense (AS) 5⬘-CAGGCTGTTGTTCTCTAAAG TT-3⬘, K12 S 5⬘-TTGTGACAGACTCCAAATCA-3⬘ and K12 AS 5⬘-TACTCCAGTTGTCCAGAAGG-3⬘. PCR amplification was performed on 2 ␮L cDNA according to the following program: initial denaturing at 95°C for 10 minutes, followed by denaturing at 95°C for 45 seconds, annealing at 55°C for 45 seconds, and extension at 72°C for 1 minute, followed by a final extension of 72°C for 7 minutes (Mastercycler; Eppendorf, Fremont, CA). The PCR products were electrophoresed on a 2% agarose gel. To confirm the KLF6 or K12 identity of the PCR products, the generated bands were sequenced on both strands, with the same primers described earlier used in the amplification and the DNA dye terminator cycle sequencing kit (Applied Biosystems, Courtaboeuf, France). Sequence analysis was performed with an automated DNZ sequencer (model 377; Applied Biosystems).

IOVS, November 2002, Vol. 43, No. 11

Corneal Regulation of K12 by hKLF

3425

FIGURE 4. Stimulation of K12 promoter activity in HCE cells by KLF6. (A) An increasing amount of human KLF6 expression vector (pEGFP-C1hCPBP) was cotransfected with 0.5 ␮g pTK/CAT and 10 ␮g K12-1, K12-2, and K12-3 ␤-galactosidase reporter plasmid into 50% confluent HCE cells, with 50 ␮g transfection reagent. ␤-Galactosidase activity was normalized according to CAT activity. Experiments were repeated six times. Error bars represent the SD of nine experiments. The same superscripts (a–f) were used to designate incubation conditions that do not differ statistically (P ⬍ 0.05). (B) Normal (pEGFP-C1-hCPBP) and mutated KLF6 (X137) expression vectors were cotransfected with 0.5 ␮g pTK/CAT, and 10 ␮g K12-1, K12-2, K12-3, mK12-1, mK12-2, and mK12-3 ␤-galactosidase reporter plasmids into 50% confluent HCE cells using 50 ␮g transfection reagent. ␤-Galactosidase activity was normalized according to CAT activity. Experiments were repeated six times. Error bars represent the SD of nine experiments. The same superscripts (a– c) were used to designate incubation conditions that do not differ significantly (P ⬍ 0.05).

Immunohistologic Experiments Cryosections of total cornea and cells grown in the 8-well chamber slides, were fixed in 4% paraformaldehyde in PBS at 4°C for 1 hour, rinsed with PBS three times and incubated in 5% bovine serum albumin (Sigma Aldrich, St. Quentin Fallavier, France) at 25°C for 30 minutes. Immunohistologic staining was performed with the epitope-specific polyclonal antibody anti-K12 (1:300)18 and anti-KLF6 (1:300; Tebu, Le Perray-en-Yvelines, France) overnight at 4°C, followed by incubation with anti rabbit IgG FITC-conjugated or rhodamine-conjugated secondary antibody for 1 hour at room temperature. Histologic examination was performed by microscope (Axioscope; Carl Zeiss, Oberkochen, Germany) after 4⬘,6⬘-diamino-2-phenylindole (DAPI; nuclear staining; 1 minute, 1:500 dilution in PBS).

Western Blot Analysis Isolation of KLF6 proteins from cells and tissues was achieved as previously described.13 Protein concentrations of the homogenates were determined by the biuret method on a clinical chemistry system19 (Roche/Hitachi 912; produced by Roche, Mannheim, Germany, in collaboration with Hitachi Ltd., Tokyo, Japan). Total protein (10 ␮g per sample) was boiled for 10 minutes and then separated on a 10% sodium

dodecyl sulfate (SDS) polyacrylamide gel. The proteins were transferred to a polyvinylidene difluoride (PVDF) membrane (Immobilon-P; Millipore Corp., Bedford, MA). The membrane was blocked in 5% bovine serum albumin in 10 mM Tris-HCl (pH 8.0), 150 mM NaCl, and 0.05% Tween-20 (TTBS) for 1 hour at room temperature. The membrane was then incubated with primary polyclonal KLF6 antibody (1:500 in TTBS) overnight at 4°C, followed by incubation with the anti-rabbit secondary antibody (1:3000 in TTBS) conjugated with horseradish peroxidase (Sigma Aldrich) for 1 hour at room temperature. The peroxidase activity was developed with 3,3⬘-diaminobenzidine (DAB)-H2O2 solution (Vector Laboratories Inc., Burlingame, CA).

Transfection of Cultured HCE Cells Three K12 promoter sequences (K12-1, 1.03 kb, ⫺989 to ⫹40; K12-2, 0.71 kb, ⫺669 to ⫹40; K12-3, 0.29 kb, ⫺251 to ⫹40; see Fig. 6) were generated after PCR with primers designed to introduce appropriate restriction sites and subcloned into a p␤-gal basic ␤-galactosidase reporter vector.20 Three other promoter constructs containing a mutated KLF6-binding site in a K12 promoter (mK12-1, mK12-2, and mK12-3) were generated using a kit (Quick Change XL Site Directed Mutagenesis Kit; Stratagene, Amsterdam, The Netherlands). The empty

3426

Chiambaretta et al.

IOVS, November 2002, Vol. 43, No. 11 DNA binding site previously described16 and called RS in our work; wild-type potential K12 promoter KLF6 DNA binding site called TS1 (CTCCACCCA possessing 88.89% homology with RS), and two TS1 sequence mutants called TS2 and TS3 (with the respective sequence CTCCGTACA and CTCCACTTG). Double-stranded oligonucleotides were radiolabeled with [␣-32P]. Radiolabeling was performed with a T4 polynucleotide kinase kit (Life Technologies). Electrophoretic mobility shift assays (EMSAs) were performed by incubating 10 ␮g nuclear protein extract with 0.1 pmol of the ␣-32P-labeled oligonucleotide DNA probes (25,000 cpm) in a 20-␮L binding reaction containing 10 mM HEPES (pH 7.9), 80 mM KCl, 2.5 mM MgCl2, 1 mM dithiothreitol (DTT), 10% glycerol, 0.1 ␮g of poly(dI-dC). After incubation on ice for 10 minutes, the samples were loaded on a 5% PAGE and separated by electrophoresis at 10 V/cm for 90 minutes. The gels were dried at ⫺80°C for 40 minutes and detected by autoradiography (exposed for 12–24 hours at ⫺80°C). For competition experiments, the protein fractions were preincubated (20 minutes on ice) with a 10⫻ and a 20⫻ excess of unlabeled double-stranded oligonucleotides.

RESULTS FIGURE 5. Binding of KLF6 on K12 promoter. Electrophoretic mobility shift assays with nuclear extracts from COS cells and radiolabeled (*) oligonucleotides corresponding to the consensus binding site of KLF6 previously described16 and called RS* (positive control for KLF6 binding) and putative binding motif of KLF6 (TS1*, TS2*, and TS3*) in K12 promoter. Lane 1: extract from normal COS cells and RS*. Lane 2: extract from COS cells transiently transfected with empty expression plasmid and RS*. Lanes 3, 4, 5, and 6: extracts from COS cells transiently transfected with KLF6 expression vector, respectively, with RS*, TS1*, TS2*, and TS3*. Lanes 7 and 8: extracts from COS cells transiently transfected with KLF6 expression vector with TS1*, respectively, in competition with nonradiolabeled TS1 at 10⫻ and 20⫻ concentrations.

p␤-gal basic reporter was transfected as a baseline control. To normalize transfection efficiency an internal control vector containing the chloramphenicol acetyl transferase (CAT) gene (driven by the cytomegalovirus [CMV] promoter) was cotransfected with reporter and/or expression plasmids (pTK/CAT; Invitrogen, San Diego, CA). The production of KLF6 protein was achieved using a plasmid expression vector containing the KLF6 gene (pEGFP-C1-hCPBP) or the mutated KLF6, unable to bind DNA and called X137 (a generous gift from Scott L. Friedman). HCE cells were trypsinized 16 hours before transfection. Cells were grown to 50% to 80% confluence in 60-mm culture dishes. Transient transfection of all DNA constructs was performed by a liposome-mediated method using transfection reagents (Lipofectamine and Plus; Life Technologies, Cergy-Pontoise, France) recommended protocol, with 0.5 ␮g of pTK/CAT, 10 ␮g of different K12 promoter constructs, and an increasing amount of pEGFP-C1-hCPBP or X137 expression vector plasmid. After 2 days, the cells were washed twice with PBS, and treated with 700 ␮L cell lysis buffer for 1 hour at 4°C (Roche). The lysed cells were centrifuged at 950 g for 5 minutes at 4°C. The determination of ␤-galactosidase and CAT production was performed by an imunoenzyme assay on 100 ␮L of supernatants (Roche). Results refer to mean ⫾ SEM and are averages of six values per experiment. Comparison of means was done by analysis of variance (ANOVA) and Fisher’s t-test on computer (Statview II, ver. 1.03; Abacus Concepts, Inc., Berkeley, CA). Throughout, values were considered significantly different when P ⬍ 0.05.

Electrophoretic Mobility Shift Assays Nuclear extracts were performed on normal COS-7 cells previously and transiently transfected with KLF6. The sequence of tested oligonucleotides was as follows: consensus sequence (CCCCACCCA) of the KLF6

Expression of KLF6 in HCE Cells and the HCE Cell Line To test for the expression of human KLF6 mRNA in the cornea, we performed RT-PCR experiments with total mRNA isolated from five human corneas with hKLF6-specific primers. A specific 323-bp PCR product was generated by RT-PCR (and checked for KLF6 sequence identity) in the five corneas tested (Fig 1, lane 3 and data not shown). We then assayed hKLF6 protein in total human extract by Western blot analysis. As expected, we detected a 37-kDa band (Fig. 2, lane 1), as previously described in other human tissues.16 These data helped us determine the precise cellular localization of hKLF6 protein in corneal tissue by using an immunohistologic approach with a specific KLF6 antibody. As shown in Figures 3E to 3H, positive (cytoplasmic and nuclear) KLF6 immunostaining was strongly present in the epithelial layer. Therefore, RT-PCR and Western blot analysis in corneal epithelial cells (of patients) and the HCE cell line confirmed that KLF6 transcripts (Fig. 1, lanes 5, 7) and proteins (Fig. 2, lanes 2– 4) were essentially present in the epithelial layer of human cornea.

Colocalization of hKLF6 and K12 in Human Cornea Epithelium To be able to modulate the transcription of K12, the intracellular transcription factor hKLF6 had to be colocalized with K12, which is detected at the transcriptional level in the extracts of cornea, epithelial corneal cells and HCE cell line (Fig. 1, lanes 2, 4, 6). As shown in Figures 3A and 3B, K12 was colocalized in corneal epithelial cells with hKLF6. Collectively, the results for hKLF6 expression and its colocalization with K12 mean that the transcription factor hKLF6 may play a physiological role in the corneal epithelium, prompting us to explore its involvement in K12 transcriptional regulation.

hKLF6 Transactivates K12 Promoter in HCE Cells It has been shown that KLF6 can transactivate the GC region (similar to our potential binding site of KLF6) promoters of target genes such as collagen ␣1 and TGF-␤ type I and II receptors.14,15 We therefore explored the ability of KLF6 to influence K12 promoter activity in the HCE cell line, using three reporter gene constructs containing the K12 5⬘ flanking genomic DNA region (see Fig. 6). Transient transfections of 1 ␮g KLF6 plasmid DNA showed significant transactivation of the

IOVS, November 2002, Vol. 43, No. 11

Corneal Regulation of K12 by hKLF

3427

FIGURE 6. Sequence of the K12 promoter. DNA sequence and elements of the 1-kb fragment of the K12 gene promoter. Nucleotide numbers are related to the transcriptional initiation residue A (⫹1). TATA boxes are underscored. AP1, activator protein 1 (AP1) binding sites; CREB/ATF, cyclic AMP response element binding protein or activating transcription factor; CTF/NFI, CCAAT-binding transcription factor or nuclear factor 1; ISRE, interferon (IFN)-stimulated response element; KFK6, the KLF6 binding site; PAX6, the paired box homeotic gene 6 binding site.

K1-1, K12-2, and K12-3 promoter sequence compared with the empty expression vector (p␤-gal control) used as control basal activity (Fig. 4A). hKLF6 stimulated the activity of the three constructs dose dependently (compare ␤-galactosidase inductions between 1 and 3 ␮g) reaching a steady state at approximately 3 ␮g (no statistical differences in ␤-galactosidase inductions between 3 and 4 ␮g). As shown in Figure 4A, the induction of K12-3 promoter was higher than that observed for K12-1 and K12-2 for hKLF6 expression plasmid concentrations between 1 and 4 ␮g. No induction of K12 reporter–promoter constructs could be observed in corneal keratinocytes, as already described for ESE-1,20 suggesting that this DNA region drives the transcription of K12 in a corneal epithelium–specific manner.20 Mutated KLF6 (expression plasmid containing the X137 stop mutant, which does not have the DNA binding domain and should not bind the DNA promoter) failed to transactivate the K12 promoter–reporter constructs (Fig. 4B). In addition, mutation of the potential KLF6-binding site abolished induction of all the mutated K12 promoter–reporter constructs (Fig. 4B). All these results taken together indicate that hKLF6 can transactivate the K12 promoter in a region present in the K12-3 construct (⫺251 to ⫹40), using a single KLF6 binding site.

hKLF6 Interactions with DNA Sequence of the K12 Promoter The 291-bp sequence of the K12-3 promoter fragment (⫺251 to ⫹40) was found to contain one hKLF6-binding site. Using a genomic database and the Blast program (provided in the public domain by the National Center for Biotechnology Information and available at www.ncbi.nlm.nih.gov/BLAST), this putative binding motif was found in K12-3: CTCCACCCA (TS1) with 88.89% similarity with the consensus sequence of the hKLF6 binding motif already described in other target genes, CCCCACCCA.16 To check the ability of this sequence to interact physically with hKLF6, we performed EMSAs using COS-7 cells transiently transfected with hKLF6 expression vector. As shown in Figure 5 (lane 4), hKLF6 bound the TS1 motif demonstrating a physical interaction of hKLF6 and K12 promoter. This specific interaction was confirmed by the absence of complex detections when mutant oligonucleotide sequences derived from TS1 (Fig. 5, see TS2 and TS3 in lanes 5, 6) or mutated KLF6 (expression plasmid containing the X137 stop mutant that does not possess the DNA-binding domain and that should not bind the DNA promoter) were used (data not shown). In addition, competition with 10⫻ and 20⫻ excess of unlabeled double-stranded oligonucleotide TS1 blocked the

3428

Chiambaretta et al.

formation of the radiolabeled complex (Fig. 5, lanes 7, 8), demonstrating the specificity of this interaction.

DISCUSSION In this work, we demonstrate for the first time that hKLF6/ Zf9/CPBP is expressed in HCE cells. We also demonstrate that hKLF6 is able to regulate the transcription of the K12 gene positively, by binding directly to a CTCCACCCA sequence located between bases ⫺221 and ⫺213 (Fig. 6). We found a new target gene of KLF6, adding to those already described for this transcription factor: collagen ␣1,21 keratin-4,13 placental glycoprotein PSG5,16 urokinase type plasminogen activator,22 the human immunodeficiency virus long-terminal repeat23 and TGF-␤ type I and II receptors.24 Regulation of K12 by hKLF6 is a new finding that attests links between keratin and the KLF family. Gut-KLF/GKLF/KLF4 is known to regulate the transcription of the keratin-19 and keratin-4 gene promoter.13,25 The localization of high promoter activity for the short 291-bp fragment of the K12 promoter is not surprising, because high in vitro activity was demonstrated for short fragments of other keratin promoters: human keratin-5, human keratin-6b, or rabbit K3, respectively, 90, 300, and 315 bp in length.26 –28 Our results agree with previous observations of Shiraishi et al.20 suggesting that the promoter-positive elements responsible for the corneal specificity of K12 gene expression may lie within the region 5⬘ to the first 250 bp of this promoter. In addition, K12-1 (1.03 kb) and K12-2 (0.71 kb) constructs have a tendency toward slightly lower promoter activity than the K12-3 (0.29 kb) pointing to the possible existence of a later silencer element between 0.3 and 1.03 kb, as also suggested elsewhere.20 Further studies are necessary to identify this region and the negative transcriptional regulators. Here, hKLF6 is the third intracellular transcription factor (with ESE-1 and PAX-6) to be experimentally described as a positive regulator of K12.12,20 ESE-1 and PAX-6 are also known to be expressed in adult and embryonic cornea and to regulate the expression of K12.11,12 In our complementary study of the developmental implications of KLF6 in K12 gene expression, preliminary results showed that KLF6 is also expressed during embryonic corneal development, similar to Sp1, another KLFrelated gene.29 Because of the presence of the same target gene for these three transcription regulators, the molecular cross-talks among KLF6, PAX-6, and ESE-1 had to be studied in human corneal development and physiology, focusing on the regulation of K12. Expression of K12 by epithelial cornea cells could also be presented as a biochemical marker of differentiation of these epithelial cells, an important process in the maintenance of a normal ocular surface.12 It is well established that KLF family members are deeply involved in cell growth, proliferation, and differentiation.30 A direct positive effect of intestine-KLF/IKLF/ KLF5 on cell growth and cellular proliferation has recently been demonstrated.31 In contrast to IKLF, GKLF/KLF4 has been implicated in the p53-transactivating effect of p21WAF1/ Cip1 promoter induction, cyclin D1 promoter activity repressor, and inhibition of cell proliferation, indicating that this protein may play a direct role in negative growth control.32–34 It has recently been demonstrated that KLF6 reduces cell proliferation. Induction of KLF6 in the NIH 3T3 cell line resulted in a reduction in proliferating cell nuclear antigen expression and in an increase in the expression of p21(WAF1/ Cip1), an inhibitor of several cyclin-dependent kinases and a key regulator of the G1-to-S transition.35 This induction of p21 by KLF6 is mediated by its binding to the two GC boxes present in the p21 promoter. These results collectively suggest

IOVS, November 2002, Vol. 43, No. 11 that KLF6 may play a key role in the regulation of corneal epithelial proliferation and differentiation. Regulated and strong expression of K12 is needed for corneal integrity. This has been clearly demonstrated by the physiological consequences of a lower expression of the K12 gene and more particularly the fragility of epithelia, which fail to adhere firmly to the corneal surface, as has been described in the homozygous knockout mice without K12.2 By regulating expression of K12, KLF6 may be strongly implicated in the integrity of corneal surface. Two other KLF-related transcription factors, Sp1 and Sp3, have also been implicated in the wound healing of corneal epithelial cells by positively regulating the expression of the gene encoding integrin subunit ␣5.36 Note that this recent result confirms previous molecular implications of Sp1 known to regulate corneal genes such as ␣1proteinase inhibitor.37 In conclusion, we demonstrate for the first time that KLF6, a member of a family of transcription factors, is strongly implicated in the regulation of corneal specific K12 expression. Recent study has implicated another member of the KLF family, KLF15, in retinal physiology,38 suggesting that KLF members may play key roles in eye physiology.

References 1. Nishida K, Honma Y, Dota A, et al. Isolation and chromosomal localization of a cornea-specific human keratin 12 gene and detection of four mutations in Meesmann corneal epithelial dystrophy. Am J Hum Genet. 1997;61:1268 –1275. 2. Kao WW, Liu CY, Converse RL, et al. Keratin 12-deficient mice have fragile corneal epithelia. Invest Ophthalmol Vis Sci. 1996;37: 2572–2584. 3. Moll R, Franke WW, Schiller DL, Geiger B, Krepler R. The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell. 1982;31:11–24. 4. Cooper D, Schermer A, Sun TT. Classification of human epithelia and their neoplasm using monoclonal antibodies to keratin: strategies, applications and limitations. Lab Invest. 1996;52:243–256. 5. Chaloin-Dufau C, Sun TT, Dhouailly D. Appearance of the keratin pair K3/K12 during embryonic and adult corneal epithelial differentiation in the chick and in the rabbit. Cell Diff Dev. 1990;32:97– 108. 6. Chaloin-Dufau C, Pavitt I, Delorme P, Dhouailly D. Identification of keratins 3 and 12 in corneal epithelium of vertebrates. Epithelial Cell Biol. 1993;2:120 –125. 7. Kinoshita S, Adachi W, Sotozono C, et al. Characteristics of the human ocular surface epithelium. Prog Retinal Eye Res. 2001;20: 639 – 673. 8. Wilson SE, He YG, Weng J, Zieske JD, Jester JV, Schultz GS. Effect of epidermal growth factor, hepatocyte growth factor, and keratinocyte growth factor, on proliferation, motility and differentiation of human corneal epithelial cells. Exp Eye Res. 1994;5:665– 678. 9. Nishimura T, Toda S, Mitsumoto T, Oono S, Sugihara H. Effects of hepatocyte growth factor, transforming growth factor-beta1 and epidermal growth factor on bovine corneal epithelial cells under epithelial-keratocyte interaction in reconstruction culture. Exp Eye Res. 1998;66:105–116. 10. Kruse FE, Tseng SC. Retinoic acid regulates clonal growth and differentiation of cultured limbal and peripheral corneal epithelium. Invest Ophthalmol Vis Sci. 1994;35:2405–2420. 11. Liu J, Kao WY, Wilson S. Corneal epithelium-specific mouse keratin K12 promoter. Exp Eye Res. 1999;68:295–301. 12. Yoshida N, Yoshida S, Araie M, Handa H, Nabeshima Y. Ets family transcription factor ESE-1 is expressed in corneal epithelial cells and is involved in their differentiation. Mech Dev. 2000;1–2:27–34. 13. Okano J, Opitz OG, Nakagawa H, Jenkins TD, Friedman SL, Rustgi AK. The Kruppel-like transcriptional factors Zf9 and GKLF coactivate the human keratin 4 promoter and physically interact. FEBS Lett. 2000;473:95–100.

IOVS, November 2002, Vol. 43, No. 11 14. Dang DT, Pevsner J, Yang VW. The biology of the mammalian Kruppel-like family of transcription factors. Int J Biochem Cell Biol. 2000;32:1103–1121. 15. Bieker JJ. Kruppel-like factors: three fingers in many pies. J Biol Chem. 2001;276:34355–34358. 16. Koritschoner NP, Bocco JL, Panzetta-Dutari GM, Dumur CI, Flury A, Patrito LC. A novel human zinc finger protein that interacts with the core promoter element of a TATA box-less gene. J Biol Chem. 1997;272:9573–9580. 17. Blanchon L, Bocco JL, Gallot D, et al. Co-localization of KLF6 and KLF4 with pregnancy-specific glycoproteins during human placenta development. Mech Dev. 2001;105:185–189. 18. Moyer PD, Kaufman AH, Zhang Z, Kao CW, Spaulding AG, Kao WW. Conjunctival epithelial cells can resurface denuded cornea, but do not transdifferentiate to express cornea-specific keratin 12 following removal of limbal epithelium in mouse. Differentiation. 1996;60:31–38. 19. Camara PD, Wright C, Dextraze P, Griffiths WC. Comparison of a commercial method for total protein with a candidate reference method. Ann Clin Lab Sci. 1991;21:335–339. 20. Shiraishi A, Converse RL, Liu CY, Zhou F, Kao WC, Kai WW. Identification of the cornea-specific keratin 12 promoter by in vivo particle-mediated gene transfer. Invest Ophthalmol Vis Sci. 1998; 39:2554 –2561. 21. Ratziu V, Lalazar A, Wong L, et al. Zf9, a Kruppel-like transcription factor up-regulated in vivo during early hepatic fibrosis. Proc Natl Acad Sci USA. 1998;95:9500 –9505. 22. Kojima S, Hayashi S, Shimokado K, et al. Transcriptional activation of urokinase by the Kruppel-like factor Zf9/COPEB activates latent TGF-beta1 in vascular endothelial cells. Blood. 2000;95:1309 – 1316. 23. Suzuki T, Yamamoto T, Kurabayashi M, Nagai R, Yazaki Y, Horikoshi M. Isolation and initial characterization of GBF, a novel DNA-binding zinc finger protein that binds to the GC-rich binding sites of the HIV-1 promoter. J Biochem. 1998;124:389 –395. 24. Kim Y, Ratziu V, Choi SG, et al. Transcriptional activation of transforming growth factor beta1 and its receptors by the Kruppellike factor Zf9/core promoter-binding protein and Sp1: potential mechanisms for autocrine fibrogenesis in response to injury. J Biol Chem. 1998;273:33750 –33758. 25. Brembeck FH, Rustgi AK. The tissue-dependent keratin 19 gene transcription is regulated by GKLF/KLF4 and Sp1. J Biol Chem. 2000;275:28230 –28239.

Corneal Regulation of K12 by hKLF

3429

26. Byrne C, Fuchs E. Probing keratinocyte and differentiation specificity of the human K5 promoter in vitro and in transgenic mice. Mol Cell Biol. 1993;13:3176 –3190. 27. Jiang CK, Epstein HS, Tomic M, Freedberg IM, Blumenberg M. Epithelial-specific keratin gene expression: identification of a 300 base-pair controlling segment. Nucleic Acids Res. 1990;18:247– 253. 28. Wu RL, Galvin S, Wu SK, Xu C, Blumenberg M, Sun TT. A 300 bp 5⬘-upstream sequence of a differentiation-dependent rabbit K3 keratin gene can serve as a keratinocyte-specific promoter. J Cell Sci. 1993;105:303–316. 29. Nakamura H, Ueda J, Sugar J, Yue BY. Developmental expression of Sp1 in the mouse cornea. ARVO Abstract 3204, 2002. 30. Black AR, Black JD, Azizkhan-Clifford J. Sp1 and Kru ¨ ppel-like factor family of transcription factors in cell growth regulation and cancer. J Cell Physiol. 2001;188:143–160. 31. Sun R, Chen X, Yang V. Intestinal-enriched Kru ¨ ppel-like factor (Kru ¨ ppel-like factor 5) is a positive regulator of cellular proliferation. J Biol Chem. 2001;276:6897– 6900. 32. Zhang W, Geiman DE, Shields JM, et al. The gut-enriched Kruppellike factor (KLF4) mediates the transactivating effect of p53 on the p21WAF1/Cip1 promoter. J Biol Chem. 2000;275:18391–18398. 33. Shie JL, Chen ZY, Fu M, Pestell RG, Tseng CC. Gut-enriched Kruppel-like factor represses cyclin D1 promoter activity through Sp1 motif. Nucleic Acids Res. 2000;28:2969 –2976. 34. Chen X, Johns DC, Geiman DE, et al. Kruppel-like factor 4, gutenriched Kruppel-like factor, inhibits cell proliferation by blocking G1/S progression of the cell cycle. J Biol Chem. 2001;276:30423– 30428. 35. Narla G, Heath KE, Reeves HL, et al. KLF6, a candidate tumor suppressor gene mutated in prostate cancer. Science. 2001;294: 2563–2566. 36. Gingras ME, Larouche K, Larouche N, et al. Expression of the gene encoding integrin subunit alpha5 is positively regulated by the transcription factors Sp1/Sp3 during wound healing of corneal epithelial cells. ARVO Abstract 3205, 2002. 37. Maruyama Y, Wang X, Li Y, Sugar J, Yue BY. Involvement of Sp1 elements in the promoter activity of genes affected in keratoconus. Invest Ophthalmol Vis Sci. 2001;42:1980 –1985. 38. Otteson DC, Lai H, Zack. KLF15 represses transactivation of the Rhodopsin and IRBP promoters and inhibits their transactivation by Crx in vitro. ARVO Abstract 3634, 2002.