ABSTRACT
The MIP gene, the founder of the MIP family of channel proteins, is specifically expressed in fiber cells of the ocular lens and expression is regulated temporally and spatially during development. We previously found that a DNA fragment containing 253 bp of 5'-flanking sequence and 42 bp of exon 1 of the human MIP gene contains regulatory elements responsible for lens-specific expression of the MIP gene. In this report we have analyzed the function of overlapping Sp1 and AP2 binding sites present in the MIP promoter. Using DNase I footprinting analysis we found that purified Sp1 and AP2 transcription factors interact with several domains of the human MIP promoter sequence -253/+42. Furthermore, addition of purified Sp1 to Drosophila nuclear extracts activates in vitro transcription from the MIP promoter -253/+42. This promoter activity is competed by oligonucleotides containing domains footprinted with Sp1. Using promoter-reporter gene (CAT) constructs we found that the sequence -39/-70 contains a cis regulatory element essential for promoter activity in transient assays in lens cells. EMSA analysis showed that lens nuclear extracts contain factors that bind to the MIP 5'-flanking sequence containing overlapping Sp1 and AP2 binding domains at positions -37/-65. Supershift experiments with lens nuclear extracts indicated that Sp3 is also able to interact with this regulatory element, suggesting that Sp1 and Sp3 may be involved in regulation of transcription of the MIP gene in the lens.
Major intrinsic protein (MIP) is the most abundant protein of the ocular lens fiber membrane and is a member of an ancient family of membrane channel proteins (1). The MIP gene is specifically expressed in lens fiber cells, which arise by differentiation of the lens epithelium (2). MIP may play an important role in maintaining lens transparency by reducing the interfiber space, as suggested by its ability to function as a weak water channel (3,4) and possibly as an adhesion molecule (5). MIP undergoes selective proteolysis during cataractogenesis and aging, which may modify gating of the MIP channels (6). Moreover, mutations in the MIP gene have been linked to the mouse genetic cataract Fraser mutation (CatFr) and lens opacity mutation (lop) (7). Expression of the MIP gene in the lens is tightly regulated temporally and spatially during embryogenesis and starts in the primary lens fibers.
Several transcription factors regulating crystallin gene expression in the lens have been characterized in the past few years. Pax-6, a paired domain and homeodomain-containing protein, which is encoded by a master gene in eye formation and belongs to the PAX family of transcription factors, interacts with control elements present in the 5'-flanking region of the [alpha]A-, [delta]1- and [zeta]-crystallin genes (8,9,10). The Sox-2 transcription factor, which belongs to a protein family containing the HMG domain, functions as an activator of chicken [delta]1- and mouse [gamma]F-crystallin gene expression in the lens (11). Neither Pax-6 nor Sox-2 are lens specific. The transcription factor [delta]EF-1, a repressor of [delta]1-crystallin gene expression, interacts with an element of the lens-specific enhancer located in the third intron of the chicken [delta]1-crystallin gene (12). [delta]EF-1 is a zinc finger protein and besides being expressed in lens, is also expressed in mesodermal tissues and the nervous system during embryogenesis (13). The [alpha]ACRYBP-1 transcription factor, which interacts with an NF[kappa]B consensus site in the mouse [alpha]A-crystallin 5'-flanking sequence regulatory region DE-1, is a zinc finger protein expressed in various tissues (14). Three novel zinc finger proteins, produced by alternative splicing from the [gamma]FBP gene, interact with a [gamma]F-crystallin enhancer element, which is similar to the [alpha]CE2 element of the chicken [alpha]A-crystallin promoter. [gamma]FBP functions as a repressor of [gamma]F-crystallin gene expression in lens cells and is expressed in presomitic mesoderm and epithelial somites during embryogenesis (15).
Little is known about regulation of membrane gene expression in the lens. We are studying regulation of the MIP gene, a non-crystallin gene which exhibits stringent lens specificity, to elucidate the mechanisms involved in regulation of expression of lens-specific genes. To this end we have cloned the human MIP gene and identified two negative regulatory elements in its 5'-flanking sequence (16,17). We found that the human MIP gene sequence -253/+42 contains an active promoter in cultured lens cells but is inactive in non-lens cells (17). In the present study we have characterized a cis regulatory element required for MIP gene promoter activity in lens cells, which contains overlapping Sp1 and AP2 binding sites. Sp3, another member of the Sp family of transcription factors (18,19), is present in lens nuclear extracts and also interacts with this Sp1 binding site. Even though Sp1 is a ubiquitous factor, it regulates tissue-specific expression of a variety of genes (20-25). AP2 is expressed in selected tissues and is involved in epidermis-specific gene expression (26-28). Both the Sp1 and AP2 transcription factors are developmentally regulated (26-31) and essential for embryonic mouse development (32-34). Interaction of Sp1, Sp3 and AP2 with the MIP promoter suggest the involvement of these transcription factors in regulating expression of the MIP gene in the lens.
Oligonucleotides were synthesized in an Applied Biosystems synthesizer and purified either on G-25 Sephadex columns for construction and sequencing or by urea-acrylamide gel electrophoresis for EMSA.
Antibodies to Sp1, Sp2, Sp3 and Sp4 were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Sp1 was a monoclonal antibody corresponding to amino acids CKDSEGRGSG DPGKKKQHI (19), Sp2 is a polyclonal antibody to amino acids KGTRSNANIQYQAVPQIQAS (18,19), Sp3 is a polyclonal antibody to amino acids DILTNTEIPLQLVTVSGNET (18,19) and Sp4 is a polyclonal antibody to amino acids VTVAAISQDSNPATPNVSTN (18).
pEN57I. The pHMIP3 (16) EcoNI-Eco57I restriction fragment containing the human MIP gene sequence -311/+71 was purified from a 0.7% agarose gel using a Gene Clean kit (Bio101). Then the fragment was subcloned into the EcoRI and HindIII sites of pBluescriptSK+ using synthetic linkers/adaptors for EcoRI-Eco57I (5'-AATTCGATATCCTAATGTTGA-3' and 5'-AACATTAGGATATCG-3') and EcoNI-HindIII (5'-TGGAGGGCTTAATTAATTAAGCA-3' and 5'-AGCTTGCTTAATTAATTAAGCCC TCC-3'). The resultant plasmid contains -321 to +71 of the human MIP gene. Deletion mutant constructs were made by PCR using pHMIP253CAT (17) as template.
pHMIP100CAT. The following primers for PCR were synthesized: 5'-primer, 5'-AGATCTAAGCTTCTCGAGTCGACCAGCTGTGAAGGGGTTAAGAG-3', containing HindIII, XhoI and SalI sites and the sequence from -100 to -85 of the MIP 5'-flanking region; 3'-primer, 5'-GGATTTGTCCTACTCAAGCTTCAGG-3', containing the junction between the hMIP and CAT coding sequences of pHMIP253CAT, which contains a HindIII site. The 186 bp PCR product was digested with SalI and HindIII and ligated into the SalI and HindIII sites of vector pSVOATCAT. Other constructs were made in the same way except for the 5'-primer sequence for PCR. The 5'-primers for -90, -70, -48, -33 and -20 constructs were 5'-AGATCTAAGCTTCTCGAGTCGACGGGGTTAAGAGGCTGGGCCTGC-3', 5'-AGATCTAAGCTTCTCGAGTCGACGCTACCTCAGCCTGCCCCTCCC-3', 5'-AGATCTAAGCTTCTCGAGTCGACAGGGATTAGGAGTCCTCTATAA-3', 5'-AGATCTAAGCTTCTCGAGTCGACTCTATAAAGGGGACTGTCCAC-3' and 5'-AGATCTAAGCTTCTCGAGTCGACCTGTCCACCCAGACAAGGCCA-3' respectively. The sizes of the PCR products were 176, 156, 134, 119 and 106 bp respectively.
Mutations in the -39/-70 human MIP sequence. Substitution mutations in the -39/-70 region of pHMIP70CAT were also introduced by PCR. Six double-stranded 55mer oligonucleotides containing a SalI site and the human MIP 5'-flanking sequence -70/-30 with seven substitutions as shown in Figure 2B were used as 5'-primer. A double-stranded oligonucleotide containing the CAT sequence and a HindIII site was used as a 3'-primer. After PCR was performed with each pair of primers PCR products were digested with SalI and HindIII and ligated into SalI/HindIII-digested pSVOATCAT. The sequence of the PCR-generated insert in all constructs was confirmed by DNA sequencing.
The plasmid pEN57I AccI-EcoRI restriction fragment, spanning positions -321 to +71 of the MIP promoter region, was used to label the upper strand 3'-end. After purification of the DNA fragment from 0.7% agarose gel the AccI site (+71) was labeled with [[alpha]-32P]dCTP and Klenow DNA polymerase (Boehringer Mannheim). After incubation the labeled fragment was extracted with phenol-chloroform, ethanol precipitated and dissolved in TE buffer, pH 7.5. The pHMIP253CAT (17) SalI-HindIII restriction fragment, corresponding to positions -253/+42 was labeled at the SalI site lower strand (3'-end) with [[alpha]-32P]dTTP and Klenow DNA polymerase. DNase I footprinting reactions were performed with Core Footprinting System (Promega). One to eight footprinting units of purified Sp1 and AP2 proteins or AP2 extract (Promega) were incubated with labeled probe in 50 µl reaction buffer (25 mM Tris-HCl, pH 8.0, 50 mM KCl, 6.25 mM MgCl2, 0.5 mM EDTA, 10% glycerol, 0.5 mM DTT) for 10 min on ice. After adding 50 µl 5 mM CaCl2, 10 mM MgCl2 the mixture was digested with appropriately diluted DNase I for 1 min at room temperature. DNase I digestions were terminated by addition of 90 µl buffer containing 200 mM NaCl, 30 mM EDTA, 1% SDS and 100 µg/ml yeast tRNA. The DNA was extracted with phenol/chloroform, precipitated with ethanol and analyzed on 6% polyacrylamide-8 M urea sequencing gels alongside a Maxam-Gilbert G+A reaction. The gels were autoradiographed on Kodak XAR5 film at -80°C with intensifying screens.
The Drosophila embryo nuclear extract (35) and primer extension kit were from Promega Corp. (Madison, WI). Plasmid pHMIP253CAT (17) was incubated with Drosophila embryo nuclear extract with or without purified Sp1 (Promega) in a transcription buffer containing 6.25 mM HEPES, pH 7.6, 50 mM potassium glutamate, 3 mM MgCl2, 0.025 mM EDTA, 1.25 mM DTT, 2.5% glycerol and 0.3 mM each four rNTPs in a 30 µl final volume. The mixture was incubated for 60 min at 21°C, the reaction stopped with 100 µl 20 mM EDTA, pH 8.0, 0.2 M NaCl, 1% SDS, 0.25 mg/ml glycogen. Nucleic acids in the mixture were extracted once with phenol/chlorophorm and ethanol precipitated. Primer extensions were performed with [[gamma]-32P]ATP 5'-end-labeled CAT oligo 5'-CAACGGTGGTATATCCAGTG-3', corresponding to the CAT sequence coding region 96 bp from the MIP promoter junction in pHMIP253CAT (17). The expected size of the extended product with this primer is 138 bp. The primer extension reaction was performed as indicated by the manufacturer and the extended products were analyzed on 8% polyacrylamide-8 M urea sequencing gels. pBR322 MspI DNA fragments labeled with [[alpha]-32P]dCTP and Klenow DNA polymerase were used as molecular weight markers. The gels were autoradiographed on Kodak XAR5 film at -80°C with an intensifying screen. The Sp1 oligonucleotide (double-stranded 5'-ATTCGATCGGGGCGGGGCGAGC-3') used as competitor was from Promega Corp. The double-stranded oligonucleotides corresponding to human MIP gene 5'-flanking sequences -73/-36 and -160/-129, used as competitors, were added at 15 pM (100-fold molar excess).
The oligonucleotides corresponding to human MIP sequences -67/-38 (wild-type and mutants) and -160/-129 plus strand were 5'-end-labeled with [[gamma]-32P]ATP and T4 polynucleotide kinase (Promega Corp., Madison, WI), purified on Sephadex G-25 spin columns (5 Prime -3 Prime Inc.) and annealed to cold oligonucleotide corresponding to the complementary strand. Chicken lens nuclear extracts were prepared from 14 day chicken embryo lenses as described by Dignam et al. (36). Mouse lens nuclear extract was prepared from newborn mice by the method described by Schreiber et al. (37). Aliquots of 3 µg chicken or mouse lens nuclear extract were incubated with a gel shift assay kit (Promega) containing 4 µl 5* binding buffer [250 mM NaCl, 50 mM Tris-HCl, pH 7.5, 2.5 mM EDTA, 2.5 mM DTT, 5 mM MgCl2, 20% glycerol, 0.5 µg poly(dI·dC)·poly(dI·dC)] and 1 µl 32P-labeled double-stranded oligonucleotide (0.15 pmol, ~10 000-20 000 c.p.m.) at 30°C for 30 min. For competition experiments 100-fold molar excess of cold double-stranded oligonucleotides were added to the incubation mixture. In the experiments with Sp1, Sp2, Sp3 and Sp4 antibodies the indicated antibody was preincubated with lens nuclear extract at a final concentration of 1 µg/µl for 20 min followed by a 20 min incubation with the 32P-labeled probe at room temperature. The binding reaction mixtures were electrophoresed in a 5% polyacrylamide gel in 0.5* TBE (45 mM Tris, 45 mM sodium borate, 1 mM EDTA) at 150 V at 4°C for 2 h, dried and subjected to autoradiography.
Embryonic chicken lens epithelia primary cultures (PLE) and transfections were performed as previously described (17,38). Lenses from 14-day-old chicken embryos were co-transfected 48 h after establishing the culture using the calcium phosphate precipitation method with 10 µg CAT plasmid and 1.5 µg pSV-[beta]gal (Promega). Cells were harvested 48 h after transfection and homogenized in 100 µl 0.1 M Tris-HCl, pH 7.9. The CAT assays were performed by the biphase method as indicated before (17,38). [beta]-Galactosidase assays were performed as described (38) and used as a control to normalize transfection efficiency. CAT activity was expressed relative to the activity of the pSVOATCAT promoterless vector.
We have previously found that the human MIP gene 5'-flanking sequence -253/+42 contains a functional promoter in lens cells but is inactive in non-lens cells (17). To further characterize the cis elements responsible for activating the MIP promoter in lens cells several deletion mutants fused to the CAT reporter gene were constructed and transfected into chicken lens cell primary cultures. As shown in Figure 1, -253, -100, -90 and -70 constructs are able to activate CAT gene expression. In contrast, when the MIP 5'-flanking sequence -70 to -49 is deleted CAT gene expression is completely lost (-48, -33 and -20 constructs). These results revealed that the sequence between -70 and -49 is critical for activity of the MIP gene promoter in lens cells.
As -70/+42 was the minimal deletion containing an active promoter in lens cells, we designed six mutant constructs to further characterize the elements in the -70/-39 domain that are important for promoter activity. In each mutant the -70/+42 MIP promoter fragment fused to the CAT gene contains a 7 bp substitution between nt -70 and -39 (M1-M6, as shown in Fig. 2B). These constructions were transfected into chicken lens cell primary cultures and assayed for CAT activity. Compared with the wild-type (pHMIP70CAT), the ability to activate CAT gene expression was decreased by ~50% in the M1 mutant, which contains mutations between -64 and -70. When mutations were introduced at nt -59/-65, -54/-60, -49/-55, -44/-50 and -39/-45 (mutants M2-M6 respectively) a drastic decrease in CAT activity compared with the wild-type was observed. The CAT activity levels were similar to the activity observed with the promoterless control vector pSV0ATCAT (see Fig. 2A). Therefore, the domain -65 to -39, proximal to the TATA box, contains an element required for activity of the MIP promoter. Analysis of this sequence shows the presence of a CT box at positions -49/-56, which has been shown to bind the Sp1 transcription factor in the CYP2D5 and the CYP11A genes (22,24). A sequence with similarity to the AP2 consensus binding site (26) is observed at positions -45/-51. As these two motifs overlap with each other and are affected in mutants M3-M6, we studied the possible interaction of these transcription factors with the MIP promoter.
To investigate the possible involvement of Sp1 and AP2 transcription factors in MIP gene expression in the lens we performed DNase I in vitro footprinting using either Sp1 or AP2 purified proteins. We found multiple Sp1 footprinted regions in the MIP sequence -253 to +71, S-1-S-7 (Fig. 3A and B). As indicated in Figure 4, the purified Sp1 binding domain S-2 contains the classical consensus Sp1 binding site, a GC box, GGCGGG (39,40). The Sp1 footprinted domains S-3 and S-5 contain a GA (CT) box in opposite orientations. This box, which binds Sp1 with lower affinity than the GC box (39), has been characterized as an Sp1 binding site in several genes (21,22,24,41). The Sp1 footprinted regions S-1 and S-6 contain a GT (CA) box. This motif, which binds Sp1 with lower affinity than the consensus sequence (39), has also been characterized in several genes as an Sp1 binding site (23,30,40,42). The Sp1 footprinted regions S-4 and S-7 contain sequence similaritites to GT and GA boxes respectively, which may be responsible for Sp1 footprinting in these two regions. The motif GGCTGG, present in the S-4 domain, has also been characterized as an Sp1 binding site in monocyte-specific expression of the CD14 gene (25).
To investigate whether the Sp1 binding sites characterized by footprinting analysis are functional in regulating transcription of the MIP gene we used Drosophila embryo nuclear extracts as an in vitro transcription system. Both Drosophila and mammalian genes have been shown to be transcribed in this in vitro system (35). As Drosophila nuclear extracts lack Sp1, it provides a convenient in vitro transcription system to assay for activity of this transcription factor. Without addition of Sp1 weak basal transcription from the MIP promoter (Fig. 5A, lane 2, and B, lane 1), initiating at the same site as the MIP gene in vivo (16), was observed. When purified human Sp1 was added to the Drosophila nuclear extract the transcription levels were increased in a dose-dependent manner (see Fig. 5A, lanes 2-5). This activation by Sp1 was inhibited when an oligonucleotide containing the Sp1 consensus binding site was used as competitor (Fig. 5, lane 3). Moreover, synthetic oligonucleotides corresponding to the MIP 5'-flanking sequences -73/-36 and -160/-129, which contain Sp1 footprinted regions S-5 and S-2 respectively, also abolished activation of transcription by Sp1 (Fig. 5B, lanes 4 and 5). These results suggest that transcription factor Sp1 is able to activate transcription of the MIP promoter in vitro by interacting with the Sp1 binding sites footprinted with purified Sp1 in the human MIP promoter region -253/+42.
As the overlapping Sp1 and AP2 footprinting region A-1/S-5 is important for promoter activity in transient assays in chicken lens cells (see Fig. 2) we performed EMSAs with chicken lens nuclear extracts, to test whether this domain interacts with lens nuclear factors. When a double-stranded oligonucleotide corresponding to sequence -67/-38, containing the overlapping footprinting regions A-1/S-5, was incubated with lens extracts a retarded band was observed (Fig. 6A, lanes 2, 8 and 14). Mutations were introduced in the MIP gene sequence -67/-38 to determine which nucleotides are responsible for interaction with the nuclear factors (see Fig. 6B). Substitutions in nt -46 to -48 of mutants M7 and M10 do not affect either interaction with lens nuclear factors (Fig. 6A, lanes 4 and 12) or competition with wild-type oligonucleotide (Fig. 6A, lane 17 for mutant M10). However, when substitutions were introduced at nt -49 to -51 (mutant M8) and -53 to -55 (mutant M9) the ability of the oligonucleotide corresponding to the sequence -65/-40 to interact with nuclear factors is lost (Fig. 6A, lanes 6 and 10) and does not compete with wild-type oligonucleotide (Fig. 6A, lane 16 for mutant M9). These results suggest that mutations which affect the CT box in mutants M8 and M9 affect binding to factors present in chicken lens nuclear extract. However, mutations affecting the putative AP2 binding site in mutants M7 and M10 do not affect interaction with factors present in chicken lens nuclear extracts. The apparent increase in binding affinity of mutant M7 may be due to transversions introduced in the AP2 site; this mutation creates a longer stretch of C residues in the CT box. AP2 is expressed in lens epithelia but not in lens fibers (43). As the lens nuclear extract is derived from whole lens, both lens epithelia and fibers are present. Therefore, the AP2 levels in this nuclear extract may not be high enough to be able to observe interaction with regulatory elements.
Sp1 is able to interact with the CT box of several promoters (21,22,24,41). However, as other members of the Sp family also interact with this motif (18,19) it was important to determine which member was involved in interaction with the CT box present at positions -55/-49 of the functional MIP promoter. When the oligonucleotide -67/-38 was incubated with mouse lens nuclear extract two retarded bands, C1 and C2, were observed (Fig. 7, lane 9). Complex C1 was not observed when antibody to Sp3 was added to the nuclear extract (Fig. 7, lane 13), suggesting interference with formation of this DNA-protein complex. The presence of Sp1, Sp2 or Sp4 antibodies did not affect formation of the retarded bands (Fig. 7, lanes 11, 12 and 14). Similar retarded bands were observed when the oligonucleotide -160/-129, which contains the classical Sp1 binding site (GC box) in the Sp1 footprinted region S-2, was incubated with mouse lens nuclear extract (Fig. 7, lane 2). However, in this case C1 was supershifted with Sp1 antibody (Fig. 7, lane 4, band SS); no effect was observed with antibodies to Sp2, Sp3 or Sp4 (Fig. 7, lanes 5-7). These results suggest that Sp1 and Sp3 are present in mouse lens nuclear extracts and under these conditions Sp3 instead of Sp1 interacts with the CT box. Another factor(s) unrelated to the Sp family may also interact with this motif, as the C2 complex was not affected by any of the antibodies to the four members of the family.
Precise regulation of tissue-specific gene expression is critical for cell differentiation during embryogenesis and post-natal development. Synergism between different transcription factors, some of them ubiquitous and some of them cell type specific, is required to achieve this precise regulation. In fact, non-lens-specific transcription factors are involved in lens-specific expression of several crystallin genes (8-11,14). Here we report that Sp1 and AP2, two transcription factors that are not lens specific, interact with the MIP gene promoter, which is activated only in the lens.
We found two AP2 binding sites in the hMIP gene promoter. One of them overlaps with the initiation site of transcription; the other one, located in a domain proximal to the TATA box, overlaps with a Sp1 binding site and is required for promoter activity in lens cells. The transcriptional activator AP2 plays a critical role in regulation of epidermis-specific keratin gene expression (26,28). However, AP2 functions as a repressor of acetylcholinesterase gene transcription (44). AP2 represses transcriptional activation by Myc of the prothymosin-[alpha] and ornithine decarboxylase gene promoters, by competing for their overlapping DNA binding sites or by AP2-Myc interaction (45). The K3 keratin gene, which contains overlapping AP2 and Sp1 binding sites, is activated in differentiated corneal epithelial cells by a drastic decrease in the AP2:Sp1 ratio (46). AP2 may also function as a repressor due to the absence of either the activation domain (47) or the DNA binding/dimerization domain (48) produced by alternative splicing. We found AP2 gene expression in lens epithelium but not in lens fibers (43). As the MIP gene is specifically expressed in lens fibers, AP2 may function as a negative regulator in lens epithelium by interfering with interaction of transcription factor(s) with their DNA binding site, resulting in repression of MIP gene expression in lens epithelium.
We found seven Sp1 binding sites in the -253/+42 region of the MIP gene promoter and one of them, at positions -65/-37, is a critical element for promoter activity in lens cells. Sp1 has been considered a ubiquitous transcription factor required for transcriptional activation of many housekeeping genes. However, the Sp1 gene is differentially expressed in mouse tissues and its ability to function as an activator of transcription changes during liver and erythroid cell differentiation (29-31). Therefore, Sp1 is likely to play an important regulatory role in cell differentiation during development. In fact, Sp1 is involved in tissue-specific expression of a variety of genes and functions synergistically with other transcription factors (20-25,30,41,42). Genes encoding Sp1-related proteins, Sp2, Sp3 and Sp4, have been cloned, indicating the existence of a Sp family of transcription factors (18,19). A three zinc finger domain is highly conserved in members of this family, resulting in similar binding affinities to GC, GT and GA boxes. Sp1, Sp2 and Sp4 activate gene transcription, whereas Sp3 may function either as an activator or a repressor (49-55).
The human MIP gene promoter domain at positions -70/-39, just upstream of the TATA box, is required for promoter activity in lens cells and contains a CT box. Purified Sp1 interacts with the CT box present at positions -49/-56 of the MIP gene promoter. However, Sp1 is not able to interact with this promoter element in the presence of Sp3 contained in lens nuclear extracts. Sp3 has been shown to repress transcriptional activation by Sp1 of several promoters (49-52). However, Sp3 is also involved in induction of the p21 gene promoter during keratinocyte differentiation (53) and in activation of the integrin CD11c and b genes in myelomonocytic cells (54). Therefore, Sp3 may play a role in modulating MIP gene expression in lens fibers or in repressing its expression in lens epithelium. It is interesting to note that the retinoblastoma gene product Rb interacts with Sp1 and Sp3 and synergistically activates Sp1- or Sp3-mediated transcription (56). As Rb plays a role in withdrawal from the cell cycle in differentiating lens fiber cells and MIP expression is markedly decreased in Rb-/- mice (57), it is tempting to speculate that the Sp1-Rb and/or Sp3-Rb interaction may provide a mechanism for activating MIP gene expression in the lens fibers.
Several genes containing CT boxes in their promoter domain proximal to the TATA box interact with tissue-specific factors, like the JC virus early promoter and the acetylcholine receptor [alpha]3 gene promoter, which interact with a glial-specific factor (21) and a neural cell-specific factor (40) respectively. Other transcription factors, like the zinc finger protein MAZ or the single-strand CT binding factors hnRNP K and CNBP, also interact with CT boxes (58,59). Interestingly, the domain required for activation of the MIP promoter contains [alpha]CE1 and [alpha]CE2 motifs adjacent to each other. Cooperative interaction between these two motifs confers lens-specific expression to the [alpha]A-crystallin gene promoter (60). The [alpha]CE1 motif, which overlaps with the Sp1/Sp3 binding site, is at positions -59/-47 (5'-CTGCCCCTCCCAG-3') and the [alpha]CE2 motif is at positions -71/-60 (5'-TGCTACCTCAGC-3'). Mutations in the [alpha]CE2 motif, which has been shown to interact with a member of the Maf family of transcription factors (61), also affected MIP promoter activity. Therefore, various transcription factors may regulate transcription of the MIP gene. Further investigation of the involvement of Sp1, Sp3, AP2and other transcription factors in regulating MIP gene expression will help us to understand the functional roles and synergisms of general or tissue-enriched factors in lens-specific gene expression.
We thank Martin Breitman for the pSVOATCAT vector and Graeme Wistow for critical reading of the manuscript.
Nucleic Acids Research
Pages
Introduction
Materials And Methods
Oligonucleotides
Antibodies
Plasmid constructs
DNase I footprinting
In vitro transcription
Electrophoretic mobility shift assay (EMSA)
CAT assay in transfected primary lens cell culture
Results
Deletion analysis of the MIP promoter in chicken lens cell primary cultures
Mutational analysis of the MIP promoter -70/-39 domain
Purified Sp1 and AP2 transcription factors interact with the human MIP gene promoter
Purified Sp1 activates in vitro transcription directed by the MIP gene promoter
The MIP promoter element footprinted by Sp1 interacts with factors present in chicken lens nuclear extract
Differential effect of Sp1 and Sp3 antibodies to MIP gene Sp1 footprinted domains
Discussion
Acknowledgements
References
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 6 Jan 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. J. DePianto, J. F. Hess, T. N. Blankenship, and P. G. FitzGerald Isolation and Characterization of the Human CP49 Gene Promoter Invest. Ophthalmol. Vis. Sci., January 1, 2003; 44(1): 235 - 243. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Kondrashov and S. A. Shabalina Classification of common conserved sequences in mammalian intergenic regions Hum. Mol. Genet., March 1, 2002; 11(6): 669 - 674. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. YANG, Z.-Z. HUANG, J. WANG, and S. C. LU The role of c-Myb and Sp1 in the up-regulation of methionine adenosyltransferase 2A gene expression in human hepatocellular carcinoma FASEB J, July 1, 2001; 15(9): 1507 - 1516. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Bateman, M. Johannes, P. Flodman, D. D. Geyer, K. P. Clancy, C. Heinzmann, T. Kojis, R. Berry, R. S. Sparkes, and M. A. Spence A New Locus for Autosomal Dominant Cataract on Chromosome 12q13 Invest. Ophthalmol. Vis. Sci., August 1, 2000; 41(9): 2665 - 2670. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||