Characterization of 5′ End of Human Thromboxane Receptor Gene
Organizational Analysis and Mapping of Protein Kinase C– Responsive Elements Regulating Expression in Platelets
Abstract Platelet thromboxane receptors are acutely and reversibly upregulated after acute myocardial infarction. To determine if platelet thromboxane receptors are under transcriptional control, we isolated and characterized human genomic DNA clones containing the 5′ flanking region of the thromboxane receptor gene. The exon-intron structure of the 5′ portion of the thromboxane receptor gene was determined initially by comparing the nucleotide sequence of the 5′ flanking genomic clone with that of a novel human uterine thromboxane receptor cDNA that extended the mRNA 141 bp further upstream than the previously identified human placental cDNA. A major transcription initiation site was located in three human tissues ≈560 bp upstream from the translation initiation codon and 380 bp upstream from any previously identified transcription initiation site. The thromboxane receptor gene has neither a TATA nor a CAAT consensus site. Promoter function of the 5′ flanking region of the thromboxane receptor gene was evaluated by transfection of thromboxane receptor gene promoter/chloramphenicol acetyltransferase (CAT) chimera plasmids into plateletlike K562 cells. Thromboxane receptor promoter activity, as assessed by CAT expression, was relatively weak but was significantly enhanced by phorbol ester treatment. Functional analysis of 5′ deletion constructs in transfected K562 cells and gel mobility shift localized the major phorbol ester–responsive motifs in the thromboxane receptor gene promoter to a cluster of activator protein-2 (AP-2) binding consensus sites located ≈1.8 kb 5′ from the transcription initiation site. These studies are the first to determine the structure and organization of the 5′ end of the thromboxane receptor gene and demonstrate that thromboxane receptor gene expression can be regulated by activation of protein kinase C via induction of an AP-2–like nuclear factor binding to upstream promoter elements. These findings strongly suggest that the mechanism for previously described upregulation of platelet thromboxane receptors after acute myocardial infarction is increased thromboxane receptor gene transcription in platelet-progenitor cells.
Membrane receptors coupled to G proteins transduce the effects of circulating humoral factors in virtually every cell type. Detailed mechanistic information is available regarding G protein–coupled receptor desensitization. As initially demonstrated in adrenergic receptors,1 cellular responses can be diminished by phosphorylation-dependent uncoupling of receptors from effectors, by receptor sequestration, and by receptor downregulation. In contrast to these desensitization studies, relatively little information is available regarding the mechanisms regulating cell hypersensitivity to G protein–coupled receptor agonists.
Our laboratory has had a long-standing interest in defining mechanisms regulating platelet sensitivity to thromboxane A2. Thromboxane is one of the most potent platelet aggregating and vasoconstricting substances known and is crucial for the maintenance of normal hemostasis.2 Thromboxane receptors are typical G protein–coupled receptors in that they possess seven transmembrane-spanning domains.3 In platelets, thromboxane receptors are coupled to phospholipase C via interactions with G proteins of the Gq/G12 class.4 5 6 Cloning of thromboxane receptor cDNAs from several human tissues has shown that different receptor proteins result from alternate splicing of a single gene product,3 7 8 which indicates that regulation of a single thromboxane receptor gene modulates thromboxane receptor expression in various organs. We recently reported that cultured human megakaryoblast cells express increased levels of thromboxane receptor steady state mRNA when stimulated to differentiate into (platelet precursor) megakaryocytes by activation of protein kinase C with PMA.9 This suggests that regulation of thromboxane receptor gene expression by activated protein kinase C in platelet progenitor cells could be responsible for the previously observed upregulation of platelet thromboxane receptors after acute myocardial infarction.10 However, characterizing the precise molecular events regulating thromboxane receptor gene expression requires isolation and analysis of the gene’s promoter.
Recently, Nusing et al11 reported the structure and chromosomal localization of the human thromboxane receptor gene. Based on 5′ RACE, it was concluded that transcription initiation within the gene occurs at multiple loci within the first two exons. Consistent with this organizational paradigm, two putative promoter regions were identified; they are designated promoters 1a and 1b, one upstream of each of the first two exons. However, as yet, there is no functional evidence demonstrating that promoter elements regulate thromboxane receptor gene expression in platelets or any other cell type.
In the present studies, we have addressed misperceptions regarding the human thromboxane receptor gene and demonstrate for the first time regulation of reporter gene expression in plateletlike cells by functional elements in the 5′ flanking region of the thromboxane receptor gene. Our results indicate that increased levels of thromboxane receptor mRNA transcript observed after phorbol ester–stimulated megakaryocytic differentiation is most likely the result of induction of an AP-2–like transcription factor and activation of enhancer elements in the thromboxane receptor gene promoter. Demonstration of regulated thromboxane receptor gene expression and the initial analysis of this gene’s promoter indicates that transcriptional regulation of receptor expression may be an important mechanism regulating specific target cell responses to thromboxane and possibly to G protein–coupled receptors in general.
Materials and Methods
The cell lines K562 (chronic myelogenous leukemia) and HEK-293 (human embryonal kidney fibroblast) were obtained from American Type Culture Collection. Human cDNA and genomic libraries were from Clontech. Hybond N+ was purchased from Amersham. The Fast CAT chloramphenicol acetyltransferase kits were from Molecular Probes. Plasmid pBLCAT612 was a gift from Muthu Periasamy, University of Cincinnati. pcDNA3CAT was from Invitrogen, and pBluescript SK+ was from Stratagene. Unless otherwise indicated, other reagents were of the highest quality available from Sigma Chemical Co or GIBCO/BRL.
Thromboxane A2 Receptor cDNA Probe Construction
Two thromboxane receptor cDNA fragments were generated by PCR on the basis of the reported sequence of the placental thromboxane receptor cDNA.3 A 464-bp fragment spanning nucleotides 739 to 1203* (TXR 739 to 1203, primer sequences 5′-CTCCTTCCTCACCTTCCTCTGC-3′ [sense] and 5′-GTGTTCAGCAGGAAGGACAGC-3′ [antisense]) was amplified from CHRF 288-11 megakaryoblast cell cDNA.9 13 14 15 A 373-bp fragment was amplified from baboon lung cDNA spanning nucleotides 1228 to 1601 (TXR 1228 to 1601, primer sequences 5′-CGTCTACCACGGGCAGGAGGCGGCC [sense] and 5′-CTCTGTCCACTTCCTACTGCA-3′ [antisense]). PCR products were subcloned into HincII-cut pBluescript for verification of nucleotide sequence. Inserts were released from the plasmid and gel-purified before 32P labeling by random priming. A third probe was generated by Pst I digestion of a K562 thromboxane receptor cDNA,8 resulting in a 1124-bp fragment containing the complete coding region spanning nucleotides 508 to 1632 (TXR 508 to 1632).
Genomic and cDNA Library Screening
Thromboxane receptor genomic clones were isolated from a human genomic DNA EMBL3 SP6/T7 library (Clontech) by screening with TXR 1228 to 1601 and TXR 739 to 1203 separately. Human uterine thromboxane receptor cDNAs were isolated from a λgt11 library (Clontech) by screening with TXR 508 to 1632. All library screenings were performed essentially as previously described8 by hybridization to phage plaques (≈106 recombinants per screen). Hybridizing phages were purified to homogeneity through subsequent rounds of plating and screening. Phage DNA was isolated from large-scale liquid preparations by using the Lambda Maxi Kit (Quiagen). cDNA inserts were released with EcoRI, and genomic inserts were released with Xho 1. Both were subcloned into pBluescript for sequencing.
RNA was extracted from K562 cells by using the method of Chirgwin et al16 and components from Pharmacia. Poly A+ RNA was enriched by binding to oligo dT-cellulose, and the mRNA was stored in ethanol at −70°C. Northern analysis of thromboxane receptor mRNA was performed as previously described.8 9
Transcription Initiation Site Mapping
Primer extension was performed by using an AMV reverse-transcriptase primer extension system from Promega. Briefly, antisense primer R3 (5′-GGCCCTGGCTGACTTGGAGTGCAT-3′, base pairs 49 to 72) was 32P-labeled at its 5′ end by using T4 polynucleotide kinase, and 10 fmol was annealed to 5 μg of poly A+–enriched unstimulated K562 RNA at 58°C for 1 hour. The primer was extended using AMV reverse transcriptase, size-separated on an 8% acrylamide/7 mol/L urea sequencing gel with HinfI-digested φX174 as size markers, and visualized by autoradiography for 3 days at −80°C.
To assess human lung, human uterus, and K562 cells for the presence of transcripts corresponding to the initiation site identified by primer extension, PCR was performed with pfu polymerase (98°C for 45 seconds, 45°C for 1.5 minutes, and 72°C for 2 minutes for 30 cycles) with primers 1 (5′-TCGAATTCGCCATTGCATCCCTGCCACCGGT-3′) (sense), 2 (5′-CCGTCCCAGCTCGGCTT-3′) (sense), or 3 (5′-CACGCCCTCCATCTGTGTGG-3′) (sense) and primer 6 (5′-ACTGGTTCAGGCACACC-3′) (antisense) by using 3 μL of human K562 cell λgt11, human lung λgt10, or human uterus λgt11 cDNA libraries (Clontech) as templates. A second “nested” PCR was performed with primers 1, 2, or 3 and primer 5 (5′-AGAGACCTCATCTGCGGG-3′) (antisense) with 1 μL of the products from the initial PCR or 1 ng genomic clone 21-2 DNA (positive control). The products from the second PCR were electrophoresed through 1.2% agarose, visualized with ethidium bromide, and transferred to Hybond N+ nylon membrane. Membranes were hybridized (6× SSC, 2× Denhardt’s solution, 0.25% SDS, and 100 μg/mL salmon sperm DNA) with 32P-labeled oligonucleotide 4 (5′-ACTGAGTCAGTCTGGCTGTGACC-3′) (antisense) for 12 hours at 70°C and washed with 2× SSC for 30 minutes before autoradiography for 72 hours at room temperature.
Promoter activity in genomic fragments was characterized by using chimeric thromboxane receptor gene/CAT constructs in the promoterless expression vector pBLCAT6.12 The chimeras were transfected into K562 cells as follows: 10 μg DNA (total) plus 50 μL Lipofectamine (GIBCO/BRL) in 100 μL Optimem I (GIBCO/BRL) was vortexed and incubated at room temperature for 45 minutes. K562 cells grown to a density of 105 cells per milliliter in RPMI containing 10% FCS were washed twice and resuspended in serum-free RPMI at a density of 3×106 cells per milliliter. The DNA/lipofectamine mixture was mixed with 800 μL of cells in one well of a six-well culture dish and placed in a 37°C/5% CO2 tissue culture incubator. After 5 hours, 4 mL of RPMI/10% FCS was added. Transfection efficiency was monitored by cotransfecting pCMVβ (1 μg) (Clontech) in which the β-galactosidase gene is driven by the CMV promoter. The variance in transfection efficiency was 12% (standard error/mean, n=18). Unless otherwise indicated, PMA or vehicle was added to cells in RPMI 24 hours after transfection.
Forty eight hours after transfection, cells were pelleted by centrifugation and hypotonically lysed in 0.25 mol/L Tris-HCl (pH 7.4) and three cycles of freeze-thawing followed by microcentrifugation to remove particulate matter. The supernatant was heated to 65°C for 7 minutes to destroy endogenous acetyltransferase activity and was routinely stored at −70°C for up to 3 days before assay. CAT activity was measured with the Fast CAT chloramphenicol acetyltransferase assay kit by using the manufacturer’s recommended protocol. Fluorescent acetylated products were resolved by thin-layer chromatography on silica gels, acetylated and nonacetylated products were separately pooled, and fluorescence was measured on a Photon Technologies spectrofluorometer. In all studies, transfection of promoterless pBLCAT6 served to measure background activity, and pcDNA3CAT (driven by the CMV promoter) served as an index of maximal CAT expression by K562 cells.
Extract for β-galactosidase activity was prepared as described for CAT assay, but the 65°C incubation was omitted. Assay was performed by incubating the cell extract with 0.26 mg O-nitrophenyl-β-d-galactoside, 0.1 mol/L MgCl2, and 5 mol/L β-mercaptoethanol at 37°C for 30 minutes. The reaction was stopped with 1 mol/L Na2CO3, and β-galactosidase activity was quantified by its absorbance at 410 nm.
Preparation of K562 Nuclear Extract
Nuclear extract from vehicle or PMA-treated K562 cells was prepared as previously described.17 Briefly, K562 cells were pelleted and washed in hypotonic buffer, followed by dounce homogenization. Nuclei were pelleted at 3300g, extracted with 300 mmol/L KCl for 30 minutes, and then dialyzed against 100 mmol/L KCl, followed by centrifugation at 25 000g. Aliquots of nuclear extract were stored at −70°C before use.
Gel Mobility Shift Assays
Gel mobility shift assays were performed essentially as previously described.18 19 Briefly, 10 000 cpm of 32P-labeled fragment of the thromboxane receptor gene promoter or double-stranded oligonucleotide encoding AP-2 binding sites (5′-GATCGAACTGACCGCCCGCGGCCCGT-3′) (Promega) was incubated with either purified AP-2 (1 footprinting unit) or 2.5 μg K562 cell nuclear extract with or without a 50-fold molar excess of a competing double-stranded oligonucleotide encoding either an AP-1 or an AP-2 binding site (Promega) for 1 hour. Resultant complexes were electrophoresed through 5% polyacrylamide, followed by autoradiography at −70°C with intensifying screen for 16 hours.
Cloning and Characterization of a Human Uterine Thromboxane Receptor cDNA
The initial description of a human thromboxane receptor cDNA from placenta3 described a large 5′ untranslated region, 703 bases of which were later attributed to a cloning artifact.11 We extended the 5′ region of the human thromboxane receptor cDNA by obtaining additional cDNA clones from cDNA libraries of several tissues with high-level thromboxane receptor gene expression. Isolates from lung and plateletlike K562 cell libraries did not extend the 5′ sequence and are reported elsewhere.8 Three independent clones were isolated from a human uterine cDNA library. Restriction mapping and sequence analysis demonstrated that the nucleotide sequence within the translated region was identical to that previously reported for the human placental thromboxane receptor,3 except for a single nucleotide substitution (from T1484 to C), which did not alter the amino acid sequence. One of the uterine clones, designated HU4-2, extended 141 bp further in the 5′ region than the placental thromboxane receptor cDNA (Fig 1⇓). Comparison with the previously reported human thromboxane receptor gene11 shows that transcription of the uterine cDNA begins at least 254 bp upstream from the most 5′ previously identified putative transcription initiation site.
Cloning and Characterization of the Human Thromboxane Receptor Gene
To further define the structure and organization of the human thromboxane receptor gene, a human genomic library was screened with two thromboxane receptor PCR fragments spanning nucleotides 739 to 1601.3 In three separate rounds of screening, three distinct genomic clones containing portions of the thromboxane receptor gene were isolated and designated λ11, λ10-1, and λ21-2. The relation of the genomic clones to the structure of the thromboxane receptor gene is illustrated in Fig 2⇓, top. λ11 contained only the fourth exon. λ10-1 contained exons two and three but not the first and fourth exons. λ21-2 contained the first three exons and 5′ flanking sequence and was chosen for detailed analysis.
A 2.45-kb BamHI–Xho I fragment of λ21-2 encoding a portion of the first untranslated exon of the human thromboxane receptor gene and ≈2.1 kb of 5′ flanking DNA was sequenced over both strands (Fig 2⇑, bottom). A major transcription initiation site was localized within this fragment by primer extension of K562 cell mRNA. A primer antisense to genomic DNA sequence 50 bp 5′ of the HU4-2 overlap region (see Fig 2⇑, bottom) produced a single-labeled fragment migrating at a size of ≈75 bp (Fig 3⇓, left). In separate but identical experiments, comparison of the primer extension product to known sequence in pBluescript showed the fragment to be 72 bp in length, corresponding to nucleotide G in the sequence AAGGCG of the gene (see Fig 3⇓, left). Because there is no TATA or CAAT consensus sequence within this genomic fragment, the thromboxane receptor gene belongs to an atypical class of genes lacking these elements.
Primer extension located a major transcription initiation site in K562 cells that is ≈380 bp 5′ of the previously identified putative transcription initiation sites for this gene.11 To confirm that transcription of the thromboxane receptor gene could start in this region in other human tissues, we performed nested PCR analysis of cDNA from human lung, human uterus, and K562 cells. An antisense primer was designed in the third exon (which is not alternatively spliced) and was used to amplify against three 5′ primers, of which two were downstream from the transcription initiation site and one was upstream (Fig 3⇑, right). To increase specificity of the PCR, the products of the first reactions were reamplified by using a nested antisense primer within the first exon. A thromboxane receptor genomic fragment was used as a positive control for the second PCR. After size separation on agarose and blotting, fragments of amplified thromboxane receptor cDNA were identified by Southern analysis using as a probe an oligonucleotide internal to the smallest PCR product. Each PCR reaction produced the expected size fragment from genomic DNA. However, only the two oligonucleotides downstream from the start site amplified thromboxane receptor from K562 cells, lung, and uterus. These results indicate that transcription of the human thromboxane receptor gene can be initiated in a region ≈470 bp 5′ of the first exon-intron splice site in at least these three human tissues.
Mapping of the Promoter Region of the Human Thromboxane Receptor Gene
Previous studies have suggested that platelet thromboxane receptor expression could be under transcriptional control and modulated by activation of protein kinase C.9 We confirmed that plateletlike K562 cells respond to phorbol ester treatment (which induces megakaryocytic differentiation) by increasing steady state levels of thromboxane receptor mRNA (Fig 4⇓). To further characterize and localize the functional elements within the thromboxane receptor gene promoter, we assayed the expression of chimeric thromboxane receptor promoter/CAT reporter gene constructs transfected into plateletlike K562 cells.
Our initial experiments tested whether a 2.13-kb sequence upstream from the start site was indeed a promoter and, if so, its relative ability to direct transcription in K562 cells. The 2.45-kb BamHI–Xho I fragment of λ21-2 encoding ≈325 bp of exon 1 plus 2.13 kb of the 5′ flanking sequence was directionally cloned into the promoterless CAT reporter gene plasmid pBLCAT6 (−2.13 CAT) (Fig 5A⇓). Two smaller restriction fragments (1.32-kb HindIII–Xho I and 0.49-kb Pst I–Xho I) were also cloned into pBLCAT6 (−1.32 CAT and −0.49 CAT, respectively) to generate a total of three thromboxane receptor gene promoter/CAT reporter constructs with serial 5′ deletions. In addition, a series of 3′ deletion constructs was generated by using the BamHI–Pst I (−2.13 to −0.49 CAT), BamHI-HindIII (−2.13 to −1.32 CAT), and HindIII–Pst I (−1.32 to 0.49 CAT) fragments of the 5′ flanking region. Preliminary studies indicated that none of the three 3′ deletion constructs, which do not contain the major transcription initiation site, drove reporter gene expression in K562 cells (not shown). In contrast, K562 cells transfected with all three of the 5′ deletion constructs expressed CAT at low levels, although the longest construct (−2.13 CAT) drove CAT expression at higher levels than the two shorter constructs, suggesting the presence of upstream enhancer elements (Fig 5B⇓). Furthermore, only −2.13 CAT exhibited an enhanced expression of reporter gene after phorbol ester stimulation (see below).
Tissue specificity of the thromboxane receptor gene promoter was assayed by comparing CAT reporter gene activity of the longest and shortest constructs in K562 cells, which express the receptor,20 and in human HEK-293 fibroblasts, which do not express thromboxane receptors.8 As shown in Fig 5B⇑, CAT expression was measurable in K562 cells but not in fibroblasts transfected with thromboxane receptor gene promoter constructs.
Since it was previously reported that the intronic region between the first two exons of the thromboxane receptor gene was one of two putative promoters for this gene,11 we constructed pBLCAT6 chimeras containing a 2.9-kb Pst I genomic fragment that included all of the second exon, 2.4 kb of the upstream intronic sequence, and 0.3 kb of the downstream intronic sequence. This construct did not exhibit any CAT activity above background in transfected K562 cells (data not shown).
Identification of Protein Kinase C–Responsive Elements in the Human Thromboxane Receptor Gene Promoter
Preliminary studies in which K562 cells transfected with −2.13 CAT, −1.32 CAT, or −0.49 CAT were treated with vehicle or 100 nmol/L PMA for 24 hours localized phorbol ester responsiveness to the ≈800-bp region between 1.3 and 2.1 kb 5′ of the major transcription initiation site. We localized cis elements conferring phorbol ester responsiveness by using a combination of DNA-protein binding assays and functional assays. Computerized analysis of the nucleotide sequence of the phorbol ester–responsive region revealed no consensus sequence for AP-1 binding but a number of putative AP-2 binding sites (see Figs 2, bottom, and 6, left). To investigate the role of AP-2 protein as an enhancer of thromboxane receptor gene transcription, we examined the ability of the AP-2 motifs to bind AP-2 protein in the context of the phorbol ester–responsive fragment of the thromboxane receptor gene promoter. DNA gel mobility shift assays were performed by using radiolabeled PCR fragments of the gene promoter as probes and assessing binding to purified recombinant human AP-2. As shown in Fig 6⇓, left, only the two most 5′ promoter fragments bound AP-2. Competition experiments were performed to show specificity of AP-2 binding. Binding was competed by a 50-molar excess of unlabeled double-stranded oligonucleotide encoding AP-2 binding sites, but no competition was observed with DNA encoding AP-1 binding sites (Fig 6⇓, top right).
The presence of an AP-2–like DNA binding factor in K562 cells was confirmed by gel mobility shift assays (Fig 6⇑, bottom right). K562 cell nuclear extract shifted a double-stranded oligonucleotide encoding AP-2 DNA binding sites, and the shift was enhanced by phorbol ester treatment of the K562 cells (the presence of double bands on the nuclear extract gel shift may indicate the presence of multiple binding factors with AP-2–like DNA binding properties). These results indicate that AP-2 and/or an AP-2–like factor in K562 cell nuclear extract is upregulated after PMA treatment.
Finally, to assess the functional significance of AP-2–like binding, an extended series of 5′ deletion mutants was constructed and transfected into K562 cells. The cells were exposed in a paired fashion to vehicle or PMA, and CAT activity was assayed and corrected for expression of β-galactosidase. As depicted in Fig 7⇓, the two shortest constructs, −0.49 CAT and −1.3 CAT, were unresponsive to phorbol ester treatment. Addition of the 5′ sequence that did not bind AP-2 (−1.55 CAT) conferred modest phorbol ester responsiveness (approximately twofold over vehicle treated) to the promoter. This degree of responsiveness was not increased by the addition of 300 bp of the gene that contained AP-2 binding sites (−1.84 CAT). However, the three constructs (−2.13 CAT, −2.04 CAT, and −1.95 CAT) that contained a cluster of closely situated functional AP-2 binding sites (as demonstrated by gel-shift analysis) all exhibited greatly augmented transcription after exposure to PMA (averaging fivefold over vehicle treated, Fig 7⇓). An in vivo competition experiment in which 5 μg of −2.13 CAT and 5 μg pBluescript DNA containing six AP-2 binding sites were cotransfected into K562 cells resulted in a 19±5-fold (n=3, P<.001) reduction in PMA-stimulated promoter activity (data not shown). Thus, the majority of PMA/protein kinase C responsiveness of the human thromboxane receptor gene can be attributed to AP-2–like binding located at sites between 1.85 kb and 1.95 kb upstream from the major transcription initiation site.
In the present study, we report the cloning of a thromboxane receptor cDNA from human uterus and the cloning, structural characterization, and analysis of promoter function of the human thromboxane receptor gene, and we map protein kinase C–responsive elements, which mediate increased thromboxane receptor expression during megakaryocytic differentiation of plateletlike K562 cells.
Full-length human thromboxane receptor cDNAs have been cloned from placenta,3 endothelium,7 K562 cells,8 and uterus (present study). The predicted receptor protein was found to be nearly identical in each of the tissues except endothelium. In the case of endothelium, the receptor protein is identical except for the extreme carboxyl terminus, where the amino acid sequence diverges because of alternate splicing within the fourth exon.7 Nusing et al11 have reported that alternate splicing of the second exon can alter the 5′ untranslated region of thromboxane receptors. Although the physiological significance of these alternate splicing events has yet to be determined, all evidence to date points to the existence of a single gene encoding thromboxane receptors in various human tissues.
Primer extension and PCR indicated that a major transcription initiation site of the thromboxane receptor gene is located ≈560 bp upstream from the translation-initiator AUG in K562 cells, human lung, and human uterus. These results differ substantially from those of Nusing et al,11 who interpreted the results of 5′-RACE experiments as identifying multiple transcription initiation sites, the furthest 5′ being ≈90 nt short of the 5′ end of the previously described placental cDNA.3 The issue of where transcription of this gene can start becomes more complex when our results are taken into consideration. Our uterine cDNA extends an additional ≈130 bp 5′ of the placental cDNA, indicating that it is at least possible for transcription to start over 200 bp 5′ of any site proposed by Nusing et al. Furthermore, all of the promoter/reporter gene constructs used in the present study used a 3′ Xho restriction site, which eliminated all of the “start sites” (Figs 2, top, and 3, right) described by Nusing et al, yet we clearly get transcription with these promoter fragments. Finally, we found no functional evidence that transcription can be initiated anywhere within the first intron, as previously proposed.11 The differences between the results of Nusing et al and the present study may arise because of the use of 5′ RACE, which (although being highly sensitive) is extremely susceptible to premature termination by reverse transcriptase and can therefore result in incomplete 5′ ends. Our approach using PCR methodology in concert with traditional primer extension has localized what we believe is a major transcription initiation site for this gene in multiple human tissues.
Interestingly, the thromboxane receptor gene lacks TATA and CAAT boxes. Transcription initiation sites lacking a TATA box have been described in constitutively active genes with GC–rich promoters, multiple transcription initiation sites, and several binding sites for the transcription factor Sp-1.21 The thromboxane receptor gene promoter region is ≈70% GC and contains two potential SP-1 binding sites in the basal promoter (see Fig 2⇑, bottom). However, the thromboxane receptor gene is clearly not constitutively active in K562 cells. Another class of TATA-less genes tends to be regulated during differentiation and to have only one or several tightly clustered transcription initiation sites. Examples of this class of gene include the T-lymphocyte β-chain V-region genes,22 which lack consensus TATA elements but possess a conserved decanucleotide motif 40 to 75 nt upstream from the transcription initiation site. The β-chain gene decanucleotide shares limited homology with the AP-1 and cAMP-responsive elements and has been postulated to mediate tissue-specific activation of the gene. Interestingly, a similar sequence (TGAATAA) is located 140 nt upstream from the start of transcription in the thromboxane receptor gene (see Fig 2⇑, bottom). Whether this sequence has a regulatory function similar to the lymphocyte gene sequence remains to be determined.
The promoter region for the human thromboxane receptor gene was identified by assaying reporter gene activity in K562 cells transiently transfected with chimeric CAT plasmids containing fragments of the first exon and 5′ flanking region of the thromboxane receptor gene. K562 cells were chosen for the promoter analyses because we have previously shown that K562 thromboxane receptors are similar to aggregation-coupled platelet thromboxane receptors in studies of ligand binding, cell signaling, and desensitization.20 Furthermore, we have recently cloned, expressed, and performed a detailed pharmacological characterization of a K562 cell thromboxane receptor cDNA8 and have demonstrated upregulation of thromboxane receptor steady state mRNA levels by phorbol ester treatment of K562 cells (present study) and of a similar cultured megakaryoblastic leukemia cell line.9 Thus, K562 cells express the human platelet thromboxane receptor, regulate it in a manner similar to mature platelets and platelet-precursor megakaryocytes, and therefore are an ideal system in which to characterize thromboxane receptor promoter function.
Basal thromboxane receptor gene promoter activity in K562 cells was only three times greater than activity of a promoterless CAT plasmid and was only 1.4% as intense as CAT activity driven by the strong CMV promoter. Thus, the thromboxane receptor gene has a relatively weak promoter, which explains the need to use 10 μg of poly A+ RNA for Northern analysis of K562 thromboxane receptor transcripts and the necessity of prolonged film exposure to visualize good signals. Furthermore, radioligand binding to K562 cells shows only 12 000 to 13 000 thromboxane receptors per cell.8 20
Consistent with results of Northern analyses and binding studies demonstrating upregulation of K562 and CHRF-288 thromboxane receptors and receptor mRNA after treatment with phorbol esters,9 23 promoter activity of the longest construct tested was increased approximately fivefold by treatment with 100 nmol/L PMA. The enhancing effect of PMA was observed in as little as 2 hours after the addition of PMA (data not shown). Interestingly, examination of the nucleotide sequence of the PMA-responsive gene fragment shows no consensus binding sequences for the phorbol ester–responsive AP-1 transcription factor. However, several putative AP-2 binding sites were located, which could confer responsiveness to protein kinase C.19 Therefore, mobility shift and studies of promoter activity in 5′ deletion constructs were performed, demonstrating that AP-2 binding occurred at multiple sites in the promoter between 1.5 and 2 kb upstream from the transcription initiation site. Deletion of a cluster of AP-2 sites between −1.84 and −1.95 resulted in greatly diminished phorbol responsiveness. Recombinant AP-1 failed to shift any DNA fragments from the 800-bp phorbol-responsive promoter fragment (data not shown). Finally, K562 cell nuclear extract contains an AP-2–like binding activity that is upregulated by prior treatment of the cells with PMA. Therefore, our results strongly suggest that activation of protein kinase C by phorbol esters and the induction of AP-2 or an AP-2–like DNA binding factor(s) are responsible for transcriptional upregulation of thromboxane receptors during megakaryocytic differentiation of platelet precursor cells.
The organization of regulatory domains within the human thromboxane receptor gene as described in the present study differs greatly from that recently proposed by Nusing et al,11 who reported multiple transcription initiation sites within the first two exons and who postulated the existence of two promoters, a housekeeping-type promoter upstream from the first exon and a regulated promoter upstream from the second exon. In the present study, we located in multiple tissues a major transcription initiation site in the genomic region flanking the 5′ end of the first exon. We also found a single regulated promoter upstream from this start site. The intronic region upstream from the second exon did not exhibit promoter activity in K562 cells.
Defining the molecular mechanisms regulating thromboxane receptor gene expression has importance because of inferential evidence that transcriptional upregulation of platelet megakaryocyte thromboxane receptors increases platelet aggregability in patients with acute myocardial infarction.10 This increase in platelet thromboxane receptor number and heightened platelet sensitivity to thromboxane may be a key event in perpetuating a vicious cycle of platelet aggregation causing thromboxane release, which, in turn, produces more platelet aggregation. The present results help to define the mechanism by which thromboxane receptor gene expression in plateletlike cells can be increased by activation of protein kinase C. It is reasonable to suggest that circulating humoral factors such as thrombin9 13 are released during acute myocardial infarction and activate protein kinase C in platelet-precursor bone marrow cells, thus stimulating the formation and release of platelets with increased numbers of thromboxane receptors.
Selected Abbreviations and Acronyms
|AMV||=||avian myeloblastosis virus|
|FCS||=||fetal calf serum|
|PCR||=||polymerase chain reaction|
|PMA||=||phorbol 12-myristate 13-acetate|
|RACE||=||rapid amplification of cDNA ends|
|SSC||=||standard saline citrate|
This study was supported by grant HL-49267 from the National Institutes of Health, a merit review grant from the Veterans Administration, and an American Heart Association, Ohio Affiliate, Inc, Postdoctoral Fellowship. Dr Dorn is an Established Investigator of the American Heart Association, supported with funds contributed in part by its Ohio Affiliate. We gratefully acknowledge the secretarial assistance of Reene Cantwell.
Winner of the American Heart Association’s 1995 Council on Circulation Boots Cardiovascular Research Prize.
1 Unless stated otherwise, all thromboxane receptor gene and cDNA nucleotide numbering is from the start site of transcription, designated +1.
- Received February 22, 1995.
- Accepted June 6, 1995.
- © 1995 American Heart Association, Inc.
Raymond JR, Hnatowich M, Lefkowitz RJ, Caron MG. Adrenergic receptors: models for regulation of signal transduction processes. Hypertension. 1990;15:119-131.
Hamberg M, Svensson J, Samuelsson B. Thromboxanes: a new group of biologically active compounds derived from prostaglandin endoperoxides. Proc Natl Acad Sci U S A. 1975;72:2994-2998.
Shenker A, Goldsmith P, Unson CG, Spiegel AM. The G protein coupled to the thromboxane A2 receptor in human platelets is a member of the novel Gq family. J Biol Chem. 1991;266:9309-9313.
Knezevic I, Borg C, LeBreton GC. Identification of Gq as one of the G-proteins which copurify with human platelet thromboxane A2/prostaglandin H2 receptors. J Biol Chem. 1993;268:26011-26017.
Offermanns S, Laugwitz K-L, Spicher K, Schultz G. G proteins of the G12 family are activated via thromboxane A2 and thrombin receptors in human platelets. Proc Natl Acad Sci U S A. 1994;91:504-508.
Raychowdhury MK, Yukawa M, Collins LJ, McGrail SH, Kent KC, Ware JA. Alternative splicing produces a divergent cytoplasmic tail in the human endothelial thromboxane A2 receptor. J Biol Chem. 1994;269:19256-19261.
D’Angelo DD, Davis MG, Ali S, Dorn GW II. Cloning and pharmacologic characterization of a thromboxane A2 receptor from K562 (human chronic myelogenous leukemia) cells. J Pharmacol Exp Ther. 1994;271:1034-1041.
Dorn GW II, Davis MG, D’Angelo DD. Gene expression during phorbol ester-induced differentiation of cultured human megakaryoblastic cells. Am J Physiol. 1994;266:C1231-C1239.
Dorn GW II, Liel N, Trask JL, Mais DE, Assey ME, Halushka PV. Increased platelet thromboxane A2/prostaglandin H2 receptors in patients with acute myocardial infarction. Circulation. 1990;81:212-218.
Nusing RM, Hirata M, Kakizuka A, Eki T, Ozawa K, Narumiya S. Characterization and chromosomal mapping of the human thromboxane A2 receptor gene. J Biol Chem. 1993;268:25253-25259.
Dorn GW II, Davis MG. Thrombin, but not thromboxane, stimulates megakaryocytic differentiation in human megakaryoblastic leukemia cells. J Pharmacol Exp Ther. 1992;262:1242-1247.
Dorn GW II, Davis MG. Differential megakaryocytic desensitization to platelet agonists. Am J Physiol. 1992;263:C864-C872.
Dorn GW II. Regulation of response to thromboxane A2 in CHRF-288 megakaryocytic cells. Am J Physiol. 1992;262:C991-C999.
Kingston R. Preparation of nuclear and cytoplasmic extracts from mammalian cells. In: Ausubel F, Brent R, Kingston R, Moore D, Seidman JG, Smith J, Strul K, eds. Current Protocols in Molecular Biology. New York, NY: John Wiley & Sons Inc; 1995:12.1.1-12.1.6.
Williams T, Tjian R. Characterization of a dimerization motif in AP-2 and its function in heterologous DNA-binding proteins. Science. 1991;251:1067-1071.
Dorn GW II. Mechanism for homologous downregulation of thromboxane A2 receptors in cultured human chronic myelogenous leukemia (K562) cells. J Pharmacol Exp Ther. 1991;259:228-234.
Sehgal A, Patil N, Chao M. A constitutive promoter directs expression of the nerve growth factor receptor gene. Mol Cell Biol. 1988;8:3160-3167.
Anderson SJ, Chou HS, Loh DY. A conserved sequence in the T-cell receptor β-chain promoter region. Proc Natl Acad Sci U S A. 1988;85:3551-3554.