HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 28, 25329-25336, July 12, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/28/25329    most recent M203688200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Boffa, M. B. Articles by Koschinsky, M. L. Search for Related Content PubMed PubMed Citation Articles by Boffa, M. B. Articles by Koschinsky, M. L. Social Bookmarking                      What's this?

A Role for CCAAT/Enhancer-binding Protein in Hepatic Expression of Thrombin-activable Fibrinolysis Inhibitor^*

Michael B. Boffa, Jeffrey D. Hamill, Nazareth Bastajian, Rebecca Dillon, Michael E. Nesheim^§, and Marlys L. Koschinsky^¶

From the Departments of  Biochemistry and ^§ Medicine, Queen's University, Kingston, Ontario K7L 3N6, Canada

Received for publication, April 16, 2002, and in revised form, May 8, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Thrombin-activable fibrinolysis inhibitor (TAFI) is a procarboxypeptidase B-like zymogen that upon activation by thrombin, thrombin-thrombomodulin, or plasmin attenuates fibrin clot lysis by inhibiting positive feedback in the fibrinolytic cascade. The concentration of TAFI in plasma varies in the human population and thus may constitute a risk factor for thrombotic disorders. In addition, TAFI has been reported to be a positive acute phase reactant in mice. We have initiated molecular analysis of the human TAFI promoter to understand the mechanisms underlying regulation of TAFI gene expression. We identified a putative C/EBP-binding site between 53 and 40 of the promoter. Mutations in this site that abolish C/EBP binding decrease TAFI promoter activity in human hepatoma (HepG2) cells by ~80%. Gel mobility shift analyses indicated that C/EBP- present in HepG2 nuclear extracts and C/EBP- and - present in adult rat liver nuclear extracts bind to the C/EBP site. C/EBP-, -, and - isoforms are all capable of binding to the C/EBP site and activating the TAFI promoter. The identification of a functional C/EBP-binding site in the human TAFI promoter may have important implications for the regulation of expression of this gene during development and in response to inflammatory stimuli.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The balance between the activities of the blood coagulation and fibrinolytic cascades is crucial for ensuring normal hemostasis at sites of tissue injury while preventing inappropriate coagulation of the blood at sites remote from the injury. Imbalances lead to a tendency to bleed, as in hemophilia, or to thrombose, as in myocardial infarction, stroke, and deep vein thrombosis. Both the coagulation and fibrinolytic cascades feature inherent regulatory mechanisms that allow for localization, amplification, and subsequent attenuation of the respective activities of the cascades. In addition, regulatory pathways that communicate between the cascades influence the balance between their activities.

A novel example of the latter regulatory mechanism is provided by the recent identification of thrombin-activable fibrinolysis inhibitor (TAFI)¹ (1). TAFI, which is also known as plasma procarboxypeptidase B (2) or procarboxypeptidase R (3) or U (4), is a plasma zymogen that is activated by thrombin, the terminal enzyme of the coagulation cascade. Activation of TAFI by thrombin is accelerated over 1000-fold in the presence of the endothelial cell membrane protein thrombomodulin (5). Plasmin, the terminal enzyme of the fibrinolytic cascade, has also been identified as an activator of TAFI (6). Activated TAFI (TAFIa) is a basic carboxypeptidase that inhibits fibrinolysis by removing carboxyl-terminal lysine and arginine residues from partially degraded fibrin thereby inhibiting the development of positive feedback in the fibrinolytic cascade (7, 8). Additional substrates for TAFIa have been identified, such as the anaphylatoxins and bradykinin (9-12), thus suggesting additional roles for the TAFI pathway beyond regulation of hemostasis.

It has been determined that the concentration of TAFI antigen in human plasma varies considerably (up to 10-fold) in the population (reviewed in Ref. 13), largely as a result of genetic factors (14). Indeed, many sequence polymorphisms have been identified throughout the TAFI gene, including in the 5'-flanking region and regions encoding the protein sequence and 3'-untranslated region (15-18); many of these polymorphisms have been shown to be associated with variation in plasma TAFI concentrations (15, 16, 18), although a direct functional role for the polymorphisms remains to be demonstrated. Because the plasma concentration of TAFI is likely to impact directly the rate of TAFIa generation in response to activation of the coagulation cascade (3, 19), variation in plasma concentrations of TAFI may constitute a risk factor for the development of thrombotic disorders (15, 20). In addition, studies in mice have identified TAFI as a positive acute phase reactant (21); indeed, studies in humans have demonstrated associations between plasma TAFI concentrations and markers of inflammation such as C-reactive protein (22).² However, the molecular bases for these observations remain to be elucidated because no information currently exists concerning the mechanisms by which expression of the gene encoding TAFI may be regulated. Accordingly, following from our characterization of the human gene encoding TAFI (23), we have initiated investigations into the molecular architecture of the TAFI promoter. We report here the identification of a functional C/EBP-binding site in the TAFI promoter, which represents the first description of a cis-acting sequence in this promoter and which may have important implications for regulation of TAFI gene expression.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Restriction and modification enzymes were from New England Biolabs, Invitrogen, Promega, and Stratagene. [-³²P]ATP and [-³²P]dATP and fast protein liquid chromatography-pure Escherichia coli DNase I (7500 units/ml) were purchased from Amersham Biosciences. The Geneclean III kit was from Bio-101, Inc. Minimum essential medium, Dulbecco's modified Eagle's medium/Nutrient Mixture F-12, and penicillin/streptomycin/fungizone (PSF) were obtained from Invitrogen. Fetal calf serum was purchased from ICN. Synthetic oligonucleotides were purchased from Cortec DNA Service Laboratories, Inc. (Kingston, Ontario, Canada). Protease inhibitor mixture, containing 4-(2-aminoethyl)benzenesulfonyl fluoride, pepstatin A, trans-epoxysuccinyl-L-leucyl-amido(4-guanidino)butane, bestatin, leupeptin, and aprotinin, was from Sigma. Adult rat liver nuclear extracts were purchased from Geneka Biotechnology, Inc. (Montreal, Quebec, Canada). Polyclonal antibodies directed against human C/EBP-, -, and - were from Santa Cruz Biotechnology, Inc.

Reporter Plasmids-- A series of point mutations were introduced into the TAFI[1128]-luc luciferase reporter plasmid (23), as diagrammed in Fig. 1. Numbering is per Boffa et al. (23) and refers to the number of nucleotides upstream of the first nucleotide of the TAFI cDNA reported by Eaton et al. (2). With the exception of TAFI[1128/siteII]-luc, mutations were introduced by PCR using the method of Nelson and Long (24). PCR products encompassing the mutations spanned the XbaI and HindIII sites in the TAFI 5'-flanking region. PCR products were cloned and sequenced, after which the 5'-flanking region up to position 1128 (i.e. at the SacI restriction site) was reconstructed in the context of pBluescript II SK+ (Stratagene). The resultant mutant 5'-flanking region fragments were excised using SacI and HindIII and inserted into the pGL3 Basic luciferase reporter vector (Promega) digested with these enzymes. To construct TAFI[1128/siteII]-luc, overlapping PCR products were generated that together spanned the XbaI and HindIII restriction sites; mutations introduced into the overlapping region introduced a Bsp68I restriction site. The PCR products were cloned and sequenced, digested with XbaI and Bsp68I or Bsp68I and HindIII, respectively, and inserted into pBluescript II SK+ digested with XbaI and HindIII. The 5'-flanking region fragment spanning the SacI and HindIII sites was reconstructed and inserted into pGL3 Basic as described above.

Expression Plasmids-- The C/EBP expression plasmids pMSV-C/EBP, pMEX-C/EBP, and pMEX-C/EBP were the kind gift of Dr. David P. Lillicrap (Department of Pathology, Queen's University). Each plasmid contains the cDNA encoding respective human C/EBP isoforms under the control of the mouse sarcoma virus promoter.

Reporter Gene Assays-- HepG2 cells (human hepatocellular carcinoma) were grown in minimum essential medium containing 10% fetal calf serum and 1% PSF. Baby hamster kidney cells (BHK) (the gift of Dr. Ross McGillivray, Department of Biochemistry, University of British Columbia) were grown in Dulbecco's modified Eagle's Medium/Nutrient Mixture F-12 containing 5% fetal calf serum and 1% PSF. Cells were maintained in a humidified 37 °C incubator under a 95% air, 5% CO₂ atmosphere. For luciferase reporter gene assays, cells were grown in 6-well plates (Corning Glass) and transfected by the method of calcium phosphate co-precipitation (25). Typically, cells received ~1.3 µg of luciferase reporter plasmid and 0.3 µg of -galactosidase internal control plasmid (RSV-gal; Ref. 26) (to control for transfection and harvesting efficiency). In some experiments, cells also received 0.6 µg of C/EBP expression plasmid or the corresponding empty expression vector. After a 6-h exposure to the precipitate, the cells were washed three times in phosphate-buffered saline (PBS) and given fresh medium. After a further 42 h of incubation, the cells were harvested for preparation of cytoplasmic extracts for luciferase and -galactosidase assays as described previously (27). For each sample, the relative luciferase activity was calculated to be the luciferase activity per unit of -galactosidase activity per unit volume of cell extract.

Preparation of Nuclear Extracts from Cultured Cells-- Source material for nuclear extracts isolated from cultured cells was either HepG2 cells or BHK cells transiently transfected with C/EBP expression plasmids. Five to ten subconfluent plates (100 mm in diameter) of cells were used for extract preparation. BHK cells were transiently transfected with C/EBP expression plasmids (10 µg/plate) by the method of calcium phosphate co-precipitation (25). After a 6-h exposure to the precipitate, the cells were washed three times in PBS and given fresh medium. After a further 42 h of incubation, the cells were harvested for nuclear extract preparation. Cells were washed twice with ice-cold PBS and then scraped into a conical centrifuge tube in ice-cold PBS. The cells were pelleted by centrifugation at 400 × g for 5 min at 4 °C. The cells were resuspended in 10 ml of Buffer A (10 mM HEPES, pH 7.6, 15 mM KCl, 2 mM EDTA, 0.5 mM spermidine, 0.15 mM spermine, 0.5% (v/v) Nonidet P-40, 1% (w/v) dry nonfat milk, 5% (v/v) protease inhibitor mixture, 1 mM DTT) and incubated for 5 min on ice. The suspension was underlaid with 2 ml of sucrose cushion (Buffer A containing 0.88 M sucrose, omitting the dry nonfat milk) and centrifuged at 800 × g for 10 min at 4 °C. After aspiration of the supernatant, the nuclear pellet was resuspended in Buffer B (10 mM HEPES, pH 7.6, 100 mM KCl, 0.1 mM EDTA, 3 mM MgCl₂, 10% (v/v) glycerol, 5% (v/v) protease inhibitor mixture, 1 mM DTT, 1 mM benzamidine), and the nuclei were lysed by the dropwise addition of 3 M KCl to a final concentration of 0.55 M. The mixture was incubated for 30 min on ice, with occasional mixing, and then centrifuged at 15,000 × g for 20 min at 4 °C. The supernatant containing the nuclear extract was immediately removed and stored in small aliquots at 70 °C. The protein concentration in the extracts was measured using the BCA Protein Assay (Pierce), with bovine serum albumin as the standard.

Gel Mobility Shift Assays-- Complementary sets of oligonucleotides encompassing site II were synthesized: sense 5'-AGAAGGCTGTTATGCAATCAATGATC-3' and antisense 5'-GATCATTGATTGCATAACAGCCTTCT-3'. Mutant oligonucleotides encompassing the same range, corresponding to the site II mutations shown in Fig. 1, were also synthesized. For radiolabeled binding site probes for gel mobility shift assays, 5 pmol of sense strand oligonucleotide was end-labeled using [-³²P]ATP and T4 polynucleotide kinase. Unincorporated label was removed using a NAP-5 column (Amersham Biosciences). The labeled oligonucleotide was combined with a 5-fold molar excess of cold antisense oligonucleotide, and the two were annealed by placing in boiling water and allowing to cool slowly at room temperature. Unlabeled competitor binding site probes were made by annealing equimolar amounts of sense and antisense oligonucleotides.

Binding reactions were performed in binding buffer (10 mM HEPES, pH 7.8, 40 mM KCl, 3 mM MgCl₂, 4% (w/v) Ficoll, 0.5 mM DTT) and contained 10 µg of nuclear extract, 2 µg of poly(dI·dC), and 10 fmol of radiolabeled probe (~20,000 cpm). Binding reactions were incubated for 30 min on ice. In some binding reactions, an excess of unlabeled binding site competitor and/or antibodies specific for C/EBP isoforms (1 µl) were included. Reactions were loaded on a 5% polyacrylamide gel in 0.5× Tris borate/EDTA, 5% glycerol that had been pre-electrophoresed at 300 V for 20 min at 4 °C. Electrophoresis was continued for a further 1.5 h, at which time the gel was fixed, dried, and exposed to film (Kodak X-OMAT AR).

DNase I Footprinting Analysis-- To prepare the probe, the luciferase reporter plasmid TAFI[417]-luc (23) was digested with SacI and HindIII; the former restriction site is in the multiple cloning site upstream of the 5'-most nucleotide (417) of the TAFI 5'-flanking region in this construct, and the latter site is immediately downstream of the initiator methionine codon. The digestion products were then incubated with the Klenow fragment of E. coli polymerase I, [-³²P]dATP, and unlabeled dTTP, dCTP, and dGTP to label the TAFI promoter fragment specifically at the 3'-(HindIII) end. The labeled fragment was purified by agarose gel electrophoresis followed by isolation of the DNA using the Geneclean III kit. Binding reactions were performed in 25 mM HEPES, pH 7.6, 60 mM KCl, 7.5% (v/v) glycerol, 0.1 mM EDTA, 0.75 mM DTT, and 5 mM MgCl₂ and contained ~25 µg of nuclear extract (isolated either from mock-transfected BHK cells or BHK cells transiently transfected with pMSV-C/EBP-) and 2 µg poly(dI·dC). Control reactions lacked nuclear extract, and some reactions also contained different amounts (0.25, 1.25, or 6.25 pmol) of unlabeled double-stranded oligonucleotides corresponding to wild-type site II or the  site II mutation (see above). After a 15-min incubation on ice, 0.13 pmol (~100,000 cpm) of radiolabeled probe was added to each reaction. After a further 15 min on ice, 11.25 units of DNase I was added to each reaction, and digestion was performed on ice for 2 (in the absence of nuclear extract) or 10 min (in the presence of nuclear extract) before addition of 2.5 volumes of stop buffer (400 mM sodium acetate, 0.2% (w/v) SDS, 10 mM EDTA, 50 µg/ml yeast tRNA, 10 µg/ml proteinase K). Reactions were incubated for 10 min at 55 °C, extracted with phenol/chloroform, and precipitated with ethanol. The digestion products were dissolved in 95% (v/v) formamide, 20 mM EDTA, 0.05% (w/v) bromphenol blue, 0.05% (w/v) xylene cyanol FF, heated at 95 °C for 5 min, and loaded onto a 6% polyacrylamide sequencing gel containing 7 M urea. The gel was fixed, dried, and exposed to film (Kodak X-Omat AR). Footprints were mapped by electrophoresing Maxam-Gilbert sequencing reactions (performed using the same asymmetrically end-labeled probe as the footprinting reactions) alongside the footprinting reactions.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Mutational Analysis of Two Putative Transcription Factor-binding Sites-- As part of our previous characterization of the human gene encoding TAFI, we determined the nucleotide sequence of the 5'-flanking region of this gene, mapped the transcription start sites, and performed deletion analysis in order to identify sequences in the 5'-flanking region required for promoter activity in hepatic cells (23). We found that the TAFI promoter is transcribed from multiple (~9) major transcription start sites and lacks a consensus TATA box. Deletion of sequences between 140 and 73 in the 5'-flanking region results in loss of promoter activity in human hepatocellular carcinoma (HepG2) cells. By using a computer program (Matinspector version 2.2 (28)) to search the TAFI 5'-flanking region sequence for consensus transcription factor-binding sites, we identified two regions (site I and site II) each containing two overlapping consensus sequences (Fig. 1). Based on comparison of the respective sequence matrices for the consensus sites, contained in the Transfac version 3.4 data base, nucleotides were identified that are absolutely required for binding to either or both of the overlapping sites (Fig. 1). Accordingly, a mutagenesis strategy was devised such that either or both overlapping sites were selectively abolished in the context of the TAFI[1128]-luc luciferase reporter plasmid (23) (Fig. 1).

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Mutagenesis of the human TAFI promoter. A, topology of the luciferase (Luc) reporter plasmid used as the basis for mutagenesis. A SacI/HindIII fragment of the human TAFI gene, encompassing the 5'-flanking region up to 1128 bp upstream of the +1 nucleotide (corresponding to one of the transcription start sites (23)) and including two potential sites for transcription factor binding (sites I and II), all possible transcription start sites (bent arrow) and the entire 5'-untranslated region, was inserted into the pGL3 Basic luciferase reporter vector. The initiator methionine codon immediately upstream of the HindIII site was changed to TTG (23). All mutants in sites I and II were introduced in the context of this parental construct (TAFI[1128]-luc). B, mutagenesis of site I. Point mutations were introduced that would prevent binding of either c-Ets-1, RFX1, or both, based on the consensus transcription factor-binding site matrices in the Transfac version 3.4 data base. C, mutagenesis of site II. Point mutations were introduced that would prevent binding of either HLF, C/EBP, or both, based on the consensus transcription factor-binding site matrices in the Transfac version 3.4 data base. An additional mutant was constructed (TAFI[1128/ site II]-luc) that contains five nucleotide substitutions in site II and results in the introduction of a Bsp68I restriction site.

The respective luciferase reporter plasmids were transiently transfected into HepG2 cells in order to assess the impact of the mutations on TAFI promoter activity. The mutations in site I either resulted in a small decrease (~20%) or a moderate increase (~50%) in TAFI promoter activity (Fig. 2), indicating that neither c-Ets-1 nor RFX1 are likely to play a role in hepatic expression of the TAFI gene. A mutation in site II (T49G) that would be expected to abolish binding of the liver-enriched transcription factor hepatic leukemia factor (HLF) increased TAFI promoter activity ~50%, suggesting that this factor also does not play a role in TAFI gene transcription in the liver. However, both point mutations that would be expected to abolish C/EBP binding (G46A and A43C; Fig. 1) markedly decreased TAFI promoter activity (70-80%) (Fig. 2). A reporter plasmid containing a more extensive series of mutations in site II possessed a similarly decreased promoter activity ( site II; Fig. 2), providing further evidence that the role of site II in TAFI promoter activity can be accounted for by C/EBP binding alone. Importantly, reporter plasmids containing C/EBP-binding site mutations retained some promoter activity, relative to the empty luciferase reporter vector pGL3 Basic (Fig. 2), suggesting that C/EBP binding to site II was not absolutely required for TAFI promoter activity.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of mutations in site I and II on TAFI promoter activity. Luciferase reporter plasmids containing the wild-type human TAFI 5'-flanking region (TAFI[1128]-luc) or the indicated mutations or the empty luciferase reporter plasmid (pGL3 Basic) were transiently transfected into HepG2 cells along with the internal control plasmid RSV-gal. Forty eight hours after transfection, cytoplasmic extracts were prepared for measurement of luciferase and -galactosidase activities. Corrected luciferase activities for the respective mutants are shown relative to the wild-type TAFI[1128]-luc construct, the activity of which was designated as 100%. The data are the mean of three independent experiments performed in duplicate, and the error bars represent the S.E. of the mean.

Binding of C/EBP Isoforms to Site II-- In order to demonstrate explicitly that C/EBP is able to bind to site II, gel mobility shift assays were performed using a radiolabeled double-stranded oligonucleotide probe spanning site II and nuclear extracts isolated from HepG2 cells (Fig. 3). Two prominent complexes of reduced mobility were observed upon incubation of the site II probe with the nuclear extract (lane 2); the upper complex (bound) was completely abolished when a 50-fold molar excess of unlabeled wild-type site II competitor oligonucleotide was included in the binding reaction (lane 3), indicating that this complex is specific. By contrast, the lower complex is likely nonspecific (NS), as its intensity was not diminished in the presence of the competitor binding site. A 50-fold molar excess of an unlabeled site II competitor oligonucleotide containing the A43C substitution was a poor competitor for the specific complex (lane 4), in keeping with the observation that this mutation decreased TAFI promoter activity. Additionally, when a radiolabeled probe containing the A43C mutation was utilized, no specific complex was observed (data not shown)

View larger version (68K): [in this window] [in a new window]   Fig. 3.   Binding of HepG2 nuclear proteins to site II. An end-labeled, double-stranded oligonucleotide probe corresponding to site II was incubated with nuclear extracts prepared from HepG2 cells. Nuclear extract was omitted from the sample in lane 1. Some binding reactions contained a 50-fold molar excess of unlabeled double-stranded oligonucleotides corresponding to wild-type (wt) site II or site II containing the A43C substitution, as indicated. Some binding reactions also contained antibodies specific for C/EBP- (lanes 5-7), - (lanes 8-10), or - (lanes 11-13). Binding reactions were electrophoresed on a non-denaturing 5% polyacrylamide gel; the gel was fixed, dried, and exposed to x-ray film. The positions of the unbound labeled probe (free) as well as specific (bound), nonspecific (NS), and supershifted, specific (supershift) complexes between the probe and nuclear proteins are indicated to the right of the autoradiogram.

In order to substantiate that the specific bound complex contains C/EBP and to identify the C/EBP isoform(s) present in the complex, binding reactions were performed in the presence of antibodies specific for the  (lanes 5-7),  (lanes 8-10), and  (lanes 11-13) isoforms of C/EBP (Fig. 3). Only in the presence of C/EBP--specific antibodies was a "supershifted" complex observed (its reduced mobility a function of the increased size of the complex as a result of the bound antibody), suggesting that in HepG2 cells site II binds C/EBP- homodimers. Note that the supershifted complex is abolished in the presence of a 50-fold molar excess of unlabeled wild-type competitor oligonucleotide but is only slightly diminished in the presence of a 50-fold molar excess of unlabeled competitor oligonucleotide containing the A43C substitution.

Similar gel mobility shift experiments were performed by using nuclear extracts isolated from adult rat liver (Fig. 4). By using this material, two complexes of reduced mobility were observed (lane 2), both of which were completely abolished by a 50-fold molar excess of unlabeled wild-type competitor oligonucleotides (lane 3) but which were only moderately competed by a 50-fold molar excess of unlabeled competitor oligonucleotides containing the A43C substitution (lane 4). Binding reactions performed in the presence of antibodies specific for C/EBP isoforms resulted in supershifted complexes in the case of C/EBP-- and C/EBP--specific antibodies (lanes 5-10) but not C/EBP--specific antibodies (Fig. 4). These findings indicate that the complexes observed between site II and rat liver nuclear proteins contain both C/EBP- and C/EBP-, as the respective homodimeric species and/or as heterodimers.

View larger version (65K): [in this window] [in a new window]   Fig. 4.   Binding of rat liver nuclear proteins to site II. An end-labeled, double-stranded oligonucleotide probe corresponding to site II was incubated with nuclear extracts prepared from livers of adult rats. Nuclear extract was omitted from the sample in lane 1. Some binding reactions contained a 50-fold molar excess of unlabeled double-stranded oligonucleotides corresponding to wild-type (wt) site II or site II containing the A43C substitution, as indicated. Some binding reactions also contained antibodies specific for C/EBP- (lanes 5-7), - (lanes 8-10), or (lanes 11-13). Binding reactions were electrophoresed on a non-denaturing 5% polyacrylamide gel; the gel was fixed, dried, and exposed to x-ray film. The positions of the unbound labeled probe (free) as well as specific (bound) and supershifted, specific (supershift) complexes between the probe and nuclear proteins are indicated to the right of the autoradiogram.

To examine more systematically the ability of C/EBP isoforms to bind to site II, gel mobility shift experiments were performed using radiolabeled double-stranded oligonucleotides corresponding to site II and nuclear extracts prepared from BHK cells transiently transfected with C/EBP expression vectors (Fig. 5). Binding reactions were also performed in the presence of unlabeled competitor oligonucleotides corresponding to the respective mutations in site II outlined in Fig. 1. By using nuclear extracts prepared from BHK cells that had been transfected with C/EBP-, -, or - expression plasmids, intense complexes of lower mobility were observed (Fig. 5, lanes 8, 15, and 21) that, by comparison with the results observed using nuclear extracts prepared from untransfected BHK cells (lane 2), likely correspond to the overexpressed C/EBP isoforms. These complexes were competed efficiently with unlabeled oligonucleotides corresponding to the wild-type sequence and the T49G mutations (Fig. 5, lanes 3 and 4) but to a lesser extent with unlabeled oligonucleotides corresponding to the A43C mutation and not at all with unlabeled oligonucleotides corresponding to the G46A and  site II mutations. Interestingly, the pattern of the effects of the mutations on the ability of the oligonucleotides to compete parallels their effects on TAFI promoter activity; the mutations that decreased TAFI promoter activity resulted in less efficient competition in gel mobility shift assays.

View larger version (59K): [in this window] [in a new window]   Fig. 5.   Binding of C/EBP isoforms expressed in BHK cells to site II. An end-labeled, double-stranded oligonucleotide probe corresponding to site II was incubated with nuclear extracts prepared from BHK cells mock-transfected (lanes 2-7) or transiently transfected with expression plasmids for C/EBP- (lanes 8-13), - (lanes 15-20), or - (lanes 21-26). Nuclear extract was omitted from the samples in lanes 1 and 14. Some binding reactions contained a 50-fold molar excess of unlabeled double-stranded oligonucleotides corresponding to wild-type (wt) site II or the site II mutations shown in Fig. 1C, as indicated. Binding reactions were electrophoresed on a non-denaturing 5% polyacrylamide gel; the gel was fixed, dried, and exposed to x-ray film. The positions of the unbound labeled probe (free) as well as specific (bound) complexes between the probe and nuclear proteins are indicated to the right of the autoradiogram.

The less intense complexes observed using nuclear extracts prepared from untransfected BHK cells show a similar pattern of susceptibility to competition, suggesting that these complexes arise from C/EBP isoforms expressed endogenously in BHK cells.

We also performed DNase I footprint analysis using a fragment of the TAFI 5'-flanking region spanning from nucleotide 417 to the HindIII restriction site and using nuclear extracts isolated from mock-transfected BHK cells or BHK cells transfected with a C/EBP- expression plasmid. By using the nuclear extracts containing ectopically expressed C/EBP-, we observed a footprint in the region corresponding to site II, and a similar footprint was absent using mock-transfected BHK cell extracts (Fig. 6). No additional footprints were observed using either extract. The site II footprint could be abolished by including a 50-fold molar excess of unlabeled double-stranded oligonucleotides corresponding to wild-type site II (Fig. 6), whereas corresponding oligonucleotides containing the  site II mutations were much less efficient competitors. Of note, we also observed a footprint corresponding to site II using adult rat liver nuclear extracts, along with numerous other footprints.³

View larger version (83K): [in this window] [in a new window]   Fig. 6.   Binding of C/EBP- to site II as revealed by DNase I footprint analysis. An asymmetrically end-labeled fragment of the TAFI 5'-flanking region spanning from nucleotide 417 to the HindIII site located immediately downstream of the initiator methionine codon was incubated with nuclear extracts harvested from BHK cells (BHK) or BHK cells that had been transiently transfected with a C/EBP- expression plasmid (BHK-C/EBP-); some binding reactions contained a 2-10- or 50-fold molar excess of unlabeled double-stranded oligonucleotides corresponding to the wild-type site II sequence (wt) or to the  site II mutations. The reaction mixtures were treated with limiting quantities of DNase I, and the digestion products were resolved on a polyacrylamide/urea sequencing gel. Shown to the left of the autoradiogram is the sequence of the region protected from DNase I digestion (as determined by Maxam-Gilbert sequencing of the probe fragment), which encompasses the putative C/EBP-binding site.

trans-Activation of the TAFI Promoter by C/EBP Isoforms-- Our previous studies (23) revealed that the TAFI promoter is transcriptionally silent in BHK cells, in keeping with the lack of expression of this gene in human kidney. Therefore, we investigated whether ectopic expression of C/EBP isoforms in BHK cells would activate the TAFI promoter. These experiments also afforded the opportunity to assess the relative potency with which the respective isoforms could trans-activate this promoter. BHK cells were transiently transfected with the luciferase reporter plasmids (both wild-type and containing site II mutations) together with expression plasmids for C/EBP-, -, or - or the respective empty expression plasmids (Fig. 7). The data show that all three C/EBP isoforms can activate the wild-type TAFI promoter in BHK cells, with C/EBP- activating the most strongly (~20-fold enhancement relative to the empty expression plasmid) and C/EBP- activating the least strongly (~11-fold enhancement). In agreement with the results from the gel mobility shift analyses (Fig. 5), the T49G mutation had little or no effect on the ability of C/EBP to trans-activate the TAFI promoter, whereas the other mutations in site II all markedly decreased (but did not eliminate) activation by C/EBP.

View larger version (29K): [in this window] [in a new window]   Fig. 7.   Effect of ectopic expression of C/EBP isoforms on TAFI promoter activity in BHK cells. Luciferase reporter plasmids containing the wild-type 5'-flanking region (TAFI[1128]-luc) or the indicated mutations in site II (see Fig. 1C) were transiently transfected into BHK cells along with the internal control plasmid RSV-gal. Transfections also included expression plasmids for C/EBP-, -, or , or combinations thereof, or the respective empty expression vectors as indicated. The data shown are the mean of duplicate transfections with the error bars representing the range of the data; similar results were obtained in two independent experiments.

The effect of ectopic expression of combinations of the respective C/EBP expression plasmids was also assessed in order to model the presence of more than one C/EBP isoform within the cell (Fig. 7). It would be expected that homodimers of the respective isoforms as well as heterodimers would exist within the nucleus of transfected cells. Co-expression of C/EBP- and - (as is observed in the adult rat liver; Fig. 4) resulted in a pattern of TAFI promoter activation similar to C/EBP- alone, whereas co-expression of C/EBP- and - (which would ensue after exposure of liver cells to inflammatory cytokines) resulted in modest increases in promoter activation relative to C/EBP- or C/EBP- alone.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

It has been demonstrated that plasma concentrations of TAFI vary considerably in the human population (reviewed in Ref. 13). The largest contributor to this variation is genetic factors (16); in this context, many polymorphisms in the human TAFI gene have been identified that are strongly associated with plasma TAFI concentrations (15, 16, 18), including polymorphisms in the 5'-flanking region that may alter TAFI promoter activity. It remains to be determined what, if any, direct effect the known polymorphisms have on TAFI gene expression or if as yet undiscovered mutations play a role. It is likely that TAFI gene expression can also be altered by a variety of physiological stimuli to effect an alteration in the balance between coagulation and fibrinolysis. TAFI has been demonstrated to be a positive acute phase reactant in mice (21), and elevated plasma TAFI concentrations in humans may be associated with elevations in the inflammatory marker C-reactive protein² (22) indicating that TAFI gene expression may be under the control of inflammatory stimuli such as cytokines and glucocorticoid hormones. In addition, plasma TAFI concentrations in women rise as a function of age (14, 20, 29, 30) and may be influenced by oral contraceptive or climacteric hormone use (14, 20, 29) as well as pregnancy (31), suggesting a role for sex hormones in regulation of TAFI gene expression. Collectively, these emerging data imply a role for control of TAFI gene expression, and particularly TAFI transcription, in mediating the balance between coagulation and fibrinolysis and the interplay between coagulation and inflammation. Our functional analysis of the human TAFI promoter has resulted in the identification of a functional C/EBP-binding site, which may have important implications for the control of TAFI gene expression.

We identified a potential C/EBP-binding site between 53 and 40 of the TAFI 5'-flanking region by computer analysis of the sequence of the TAFI gene (23). Mutations in the putative C/EBP site that would be expected to abolish C/EBP binding markedly decreased TAFI promoter activity in HepG2 cells (Fig. 2) and inhibited binding of C/EBP to this site in gel mobility shift assays (Figs. 3-5). We found that C/EBP-, -, and - were all able to bind to and trans-activate through the C/EBP site (Figs. 5-7); C/EBP- present in nuclear extracts prepared from HepG2 cells as well as C/EBP- and - present in nuclear extracts prepared from adult rat liver nuclei were able to bind to the C/EBP-binding site (Figs. 3 and 4). It has been shown that HepG2 cells are deficient in expression of C/EBP- (32). Because the TAFI promoter is active in HepG2 cells, and the endogenous TAFI gene is expressed by these cells,⁴ it is clear that C/EBP- is not absolutely required for TAFI promoter activity.

The TAFI promoter lacks a consensus TATA box, and its transcription is initiated from multiple sites (23). In these respects the TAFI gene is similar to those encoding several of the vitamin K-dependent coagulation factors, including factors VII, IX, X, and XII as well as protein C (see Ref. 23 and references therein). Of these genes, functional C/EBP sites have been identified only in the promoter of the factor IX gene (33), although some evidence exists for a role for C/EBP in the protein C promoter (34). A mutation in a C/EBP site in the factor IX promoter has been implicated in defective C/EBP binding and hence impaired factor IX expression in some patients with hemophilia B Leyden (33), whereas functional cooperation between C/EBP and D-site-binding protein has been linked to the post-pubertal recovery in factor IX expression in hemophilia B Leyden patients (35). C/EBP sites have also been identified in the promoters of the genes encoding factor VIII (36) and the A-, B-, and, possibly, -chains of fibrinogen (37-39); although all of these genes are induced in the acute phase, C/EBP only appears to play a major role in the induction of the factor VIII promoter under these conditions (36, 40-42). TAFI is closely related to the pancreatic procarboxypeptidases A and B and mast cell procarboxypeptidase A, sharing up to 40% amino acid identity as well as common genomic structures (23). Although a role for C/EBP in expression of these respective genes has not been studied, their promoters show no nucleotide sequence homology to the TAFI promoter, are transcribed from a unique transcription start site, and apparently contain TATA boxes (see Ref. 23 for references).

The ability of respective C/EBP isoforms to trans-activate promoters is a complex function of their site of synthesis and relative levels of expression, the expression of activating and inhibitory variants from the same mRNA by a process of leaky ribosome scanning, and the promoter context of the C/EBP-binding site (43). Our findings from ectopic expression of C/EBP isoforms indicate some differences in the ability of C/EBP isoforms to trans-activate the TAFI promoter. It is reasonable to hypothesize that physiological signals and circumstances mediated by changes in the expression and activity of C/EBP isoforms will result in changes in TAFI gene expression.

It has been demonstrated that both C/EBP- and - are first expressed in the fetal liver relatively late in gestation (13-15 days post-coitum (dpc)) (44). Studies of TAFI gene expression during fetal development in mice reveal that TAFI mRNA is detectable as early as 7.5 dpc (45), before the development of the vascular system or liver. However, TAFI mRNA levels remain low to undetectable through 13.5 dpc (45); TAFI mRNA abundance is increased, although perhaps not to adult levels, at 14.5 dpc (45), at which point C/EBP- and - expression should be well established in the fetal liver. Although it is tempting to speculate that the onset of C/EBP expression is required for TAFI expression in liver, more information is required as to the other liver-specific factors that regulate TAFI transcription (see below). It has been found that C/EBP- and - expression in liver transiently peaks in the perinatal period (44), although no information currently exists as to the pattern of TAFI gene expression during this time. Of note, analysis of mice in which the TAFI gene has been knocked out by homologous recombination revealed no defects in fetal and neonatal growth, development, or viability (46).

The potential ability of different isoforms of C/EBP to complement each other in regulating TAFI gene expression may be relevant in the context of liver regeneration and liver disease. It is well known that there is a reciprocal down-regulation of C/EBP- and up-regulation of C/EBP- and - during liver regeneration after partial hepatectomy (47). It is not known if this process results in changes in plasma TAFI concentrations, although we would speculate that the ability of the TAFI promoter to be activated by all three C/EBP isoforms might result in minimal alteration of TAFI gene expression. On the other hand, it has been demonstrated that plasma TAFI levels are greatly decreased in the setting of various forms of advanced liver disease (48, 49), which may be relevant to the bleeding tendency seen in patients with these conditions. The mechanism underlying these observations, including the potential role of altered expression of C/EBP isoforms, remains to be determined.

TAFI expression in adult humans is restricted to the liver (23), and possibly megakaryocytes (50). However, because C/EBP- and - are expressed in other tissues besides the liver in adults (43), the C/EBP site in the TAFI promoter does not alone account for the restricted expression of this gene in liver. Our computer analysis of the TAFI 5'-flanking region sequence also identified an excellent match for the consensus binding site for the liver-enriched transcription factor HLF between 52 and 42. However, a mutation (T49G) that would be expected to abolish HLF binding did not decrease TAFI promoter activity, suggesting that this factor does not play a role in basal expression of TAFI. Our earlier deletion analysis of the human TAFI promoter revealed a key role for sequences between 140 and 73 in hepatic transcription of the TAFI promoter. Within this region, computer analysis revealed the presence of good matches to the consensus binding site sequences for the c-Ets-1 and RFX1 transcription factors. The ubiquitous Ets factor GABP/ has been shown to cooperate with HLF in transcription of the factor IX promoter in HepG2 cells (51), but in accordance with a lack of a role for HLF in TAFI promoter activity, mutations that would be expected to abolish c-Ets-1 binding also did not decrease TAFI promoter activity. RFX1 is also a ubiquitous factor that has been shown to be important for liver-specific activity of the hepatitis virus B enhancer, presumably in concert with liver-specific factors (52). However, our mutational analysis also serves to rule out a role for this factor in TAFI promoter activity in the liver. Clearly, additional work is required to fully elucidate the basis for liver-specific expression of the TAFI gene.

It is noteworthy that elimination of the C/EBP-binding site between 53 and 40 does not completely abolish TAFI promoter activity in HepG2 cells (Fig. 2). It is possible that a lower level of promoter activity can occur without the involvement of C/EBP binding to this site or that another, less potent, C/EBP-binding site exists in the TAFI promoter. It is noteworthy that mutations in site II that abolish C/EBP binding decrease, but do not eliminate, the ability of C/EBP to activate the TAFI promoter (Fig. 7). Computer analysis using Matinspector reveals that within the TAFI 5'-flanking region fragment contained within TAFI[1128]-luc, four additional potential C/EBP-binding sites exist (1098 to 1085; 833 to 820; 423 to 410; and 321 to 308).

It has recently been demonstrated that injection of mice with bacterial lipopolysaccharide results in increased hepatic TAFI mRNA abundance and plasma TAFI antigen concentrations, thus identifying TAFI as a positive acute phase reactant (21). It is well established that C/EBP isoforms are critical mediators of immune and inflammatory responses, including the acute phase response (53). Although the potential role for C/EBP in the acute phase response of the TAFI gene remains to be elucidated, numerous mechanisms by which it may occur are possible. The acute phase mediators IL-1 and IL-6 up-regulate C/EBP- and - expression (54), which may in turn give rise to increased transcription of the TAFI gene. IL-6 signaling may also increase phosphorylation of C/EBP- on Thr-235, thus increasing its trans-activating potential (55). Glucocorticoids are required for the maximal stimulation of many acute phase genes; instances of functional cooperation between C/EBP and the glucocorticoid receptor have been described, which either do (56) or do not (57) require DNA binding by the receptor. Interestingly, we have identified a functional glucocorticoid response element in the human TAFI promoter, ~40 bp upstream of the C/EBP-binding site described in this study.⁴ Lipopolysaccharide treatment of mice also alters the relative abundance of the differently sized C/EBP- and - variants that result from leaky ribosome scanning (58). C/EBP has been shown to be able to interact directly with NF-B transcription factors, which are mediators of the effects of IL-1 and tumor necrosis factor- (59, 60). IL-6 can directly stimulate expression of acute phase genes through the STAT family of transcription factors; NF-B, and C/EBP- and - increase IL-6 expression (reviewed in Ref. 53), thus constituting an indirect mechanism by which C/EBP could regulate the TAFI promoter. It has also been demonstrated recently (61) that C/EBP- is required for the acute phase response, likely through facilitation of transcription activation by STAT3. Interestingly, an examination of the TAFI 5'-flanking sequence does not reveal any obvious STAT3 or NF-B-binding sites. We are currently investigating the mechanisms by which transcription of the TAFI gene is activated in the acute phase, a phenomenon in which the role of C/EBP isoforms is likely to be crucial.

	FOOTNOTES

^* This work was supported by Canadian Institutes for Health Research Grant MOP-36491.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. of Biochemistry, Rm. A208 Botterell Hall, Queen's University, Kingston, Ontario, K7L 3N6, Canada. Tel.: 613-533-6586; Fax: 613-533-2987; E-mail: mk11@post.queensu.ca.

Published, JBC Papers in Press, May 8, 2002, DOI 10.1074/jbc.M203688200

² P. Crainich, Z. Tang, E. M. Macy, M. B. Boffa, M. E. Nesheim, M. L. Koschinsky, and R. P. Tracy, unpublished data.

³ N. Bastajian and M. L. Koschinsky, unpublished data.

⁴ M. B. Boffa, J. D. Hamill, D. Brown, M. L. Scott, M. E. Nesheim, and M. L. Koschinsky, unpublished data.

	ABBREVIATIONS

The abbreviations used are: TAFI, thrombin-activable fibrinolysis inhibitor; BHK, baby hamster kidney; C/EBP, CCAAT/enhancer-binding protein; dpc, days post-coitum; HLF, hepatic leukemia factor; IL, interleukin; PBS, phosphate-buffered saline; PSF, penicillin/streptomycin/fungizone; TAFIa, activated TAFI; DTT, dithiothreitol.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				H. Schrem, J. Klempnauer, and J. Borlak Liver-Enriched Transcription Factors in Liver Function and Development. Part II: the C/EBPs and D Site-Binding Protein in Cell Cycle Control, Carcinogenesis, Circadian Gene Regulation, Liver Regeneration, Apoptosis, and Liver-Specific Gene Regulation Pharmacol. Rev., June 1, 2004; 56(2): 291 - 330. [Abstract] [Full Text] [PDF]

				M. B. Boffa, J. D. Hamill, D. Maret, D. Brown, M. L. Scott, M. E. Nesheim, and M. L. Koschinsky Acute Phase Mediators Modulate Thrombin-activable Fibrinolysis Inhibitor (TAFI) Gene Expression in HepG2 Cells J. Biol. Chem., March 7, 2003; 278(11): 9250 - 9257. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/28/25329    most recent M203688200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Boffa, M. B. Articles by Koschinsky, M. L. Search for Related Content PubMed PubMed Citation Articles by Boffa, M. B. Articles by Koschinsky, M. L. Social Bookmarking                      What's this?