- Email: [email protected]
USF binding
Biochemical and Biophysical Research Communications 361 (2007) 775–781 www.elsevier.com/locate/ybbrc
First functional polymorphism in CFTR promoter that results in decreased transcriptional activity and Sp1/USF binding M. Taulan
a,b,c,*
, E. Lopez a,b,1, C. Guittard a,c,1, C. Rene´ a,b,c, D. Baux a,c, J.P. Altieri M. DesGeorges a,c, M. Claustres a,b,c, M.C. Romey a,b
a,c
,
a
c
INSERM U827, Laboratoire de Ge´ne´tique de Maladies Rares, Montpellier, France b Universite´ Montpellier1, UFR de Me´decine, Montpellier, France CHU Montpellier, Hoˆpital Arnaud de Villeneuve, Laboratoire de Ge´ne´tique Mole´culaire, Montpellier F-34000, France Received 14 July 2007 Available online 25 July 2007
Abstract Growing evidences show that functionally relevant polymorphisms in various promoters alter both transcriptional activity and affinities of existing protein–DNA interactions, and thus influence disease progression in humans. We previously reported the 94G>T CFTR promoter variant in a female CF patient in whom any known disease-causing mutation has been detected. To investigate whether the 94G>T could be a regulatory variant, we have proceeded to in silico analyses and functional studies including EMSA and reporter gene assays. Our data indicate that the promoter variant decreases basal CFTR transcriptional activity in different epithelial cells and alters binding affinities of both Sp1 and USF nuclear proteins to the CFTR promoter. The present report provides evidence for the first functional polymorphism that negatively affects the CFTR transcriptional activity and demonstrates a cooperative role of Sp1 and USF transcription factors in transactivation of the CFTR gene promoter. 2007 Elsevier Inc. All rights reserved. Keywords: CFTR promoter variant; Functional analysis; Sp1; USF transcription factor
Cystic fibrosis (CF), the most common life-limiting recessively transmitted genetic disease in Caucasians, is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene. While the most common mutation is the Phe508del, other putative diseasecausing mutations identified in the CFTR gene, both in coding and non-coding regions, are rare or private [1,2]. In the promoter region, there is a relative paucity of sequence variations and their relationship to CF remains yet uncertain (see www.genet.sicckids.on.ca). Nevertheless, several studies have showed that functionally relevant polymorphisms in various promoters including CFTR alter *
Corresponding author. Address: Inserm U827, UFR de Me´decine, IURC 641, Avenue du Doyen Gaston Giraud, 34093 Montpellier Cedex 5, France. Fax: +33 467415365. E-mail address: [email protected] (M. Taulan). 1 These authors contributed equally to this work. 0006-291X/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2007.07.091
both transcriptional activity and affinities of existing protein–DNA interactions, and thus influence disease progression in humans [3–5]. While a putative deleterious mutation (741T>G) within the transcriptional regulatory region has been partially characterized by EMSA [6], only one study has shown the impact of promoter variant on transcriptional activity [3]. Indeed, it has previously demonstrated that sequence variant in the human minimal promoter, associated with an attenuated CF phenotype [7] increase CFTR expression levels [3]. Moreover, despite the recent identification of functional antagonism between the SRF and YY1 transcription factors through competition for the CFTR-CArG-like binding [8], no clear mechanism responsible for CFTR transcriptional regulation has yet been reported. In the current study, we have focused on a female patient harbouring any putative disease-causing mutation except to the 94G>T promoter nucleotide change [2].
776
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
Table 1 Sequences of oligonucleotides for functional assays Site-directed mutagenesis experiment 5 0 -CAAATTTGGGGCCTGACCAGGCAGCACTCG-3 0 EMSA probes and competitors 94G (WT) 94T (M) USF Sp1a Pax-5 mutantb (IC)
5 0 -CCGCTAGAGCAAATTTGGGGCCGGACCAGGCAGCACTCGGC-3 0 5 0 -CCGCTAGAGCAAATTTGGGGCCTGACCAGGCAGCACTCGGC-3 0 5 0 -CACCCGGTCACGTGGCCTACACC-3 0 5 0 -ATTCGATCGGGGCGGGGCGAGC-3 0 5 0 -GAATCTGGAACAGAGTCGCTCCCACCG-3 0
Bold and italicized type indicates mutated nucleotides. The underlined nucleotides indicate the core site. a Oligonucleotide probe purchased from Promega, Corporation (E323A). b Oligonucleotide probe purchased from Santa Cruz (sc-2590).
The currently most-favored hypothesis advocates the 94G>T promoter variant could be viewed as a modifier polymorphism. Therefore, we tested the functional significance of this variant by using in silico analysis, reporter gene and band shift assays. We identified putative regulatory composite cis-acting element for CFTR promoter activity involving GC-rich and E-box motifs. We explored the functional relevance of Sp1 and USF transcription factors in the transcriptional control of the CFTR gene. Materials and methods Plasmids constructs. The original luciferase expression vector pGL3Basic containing the wild-type human CFTR minimal promoter was previously described [3]. The nucleotidique change 94T was introduced into the WT-pGL3 plasmid using an oligonucleotide directed mutagenesis system (QuickChangeSite-Directed Mutagenesis Kit, Stratagene) according to the manufacturer’s instructions. The mutagenesis primers were listed in Table 1. The presence of mutations and sequence fidelity were verified by direct sequencing. Cell culture and transient transfections. Hela, Beas2B and Caco-2 human epithelial cell lines were maintained as previously reported [3]. Transient transfections assays were performed with PolyFect transfection reagent (Qiagen, France) as previously described [8], except some minor modifications. All cells were seeded at a density of 40,000 cells/400 ll of medium and plated in 24-well dishes (NUNClone, Merck-Eurolab, Inc.). We used 0.36 lg of plasmids reporter (WT-94 GpGL3 or M-94T-pGL3) and 40 ng of internal control pRL-SV40 containing Renilla luciferase (Promega, France). All luciferase activities represent at least three independent experiments with each construct tested in triplicate per experiment. To minimize the possibility of errors in DNA amplification, at least two independently constructed clones were tested. For cotransfection assays, 0.36 lg of luciferase reporter and 100 ng of each expression vector were used. Expression vectors encoding either pCR3-USF2a or pN3-Sp1 were generously provided by B. Viollet and G Suske, respectively. Electrophoretic mobility shift assay (EMSA). Nuclear extracts and radiolabeled double-stranded oligomeric probes were prepared as previously reported [8]. EMSAs were performed in a total volume of 20 ll containing 2 ng of the labeled probe, 10 ll of binding buffer (2·), 2–5 lg polydIdC as a nonspecific competitor, 10–20 lg of nuclear extracts. For EMSA competition and antibody interference assays, proteins were incubated with either cold specific competitors (WT, M, Sp1, or USF oligonucleotides, see Table 1), irrelevant competitor (Pax5 mutant oligonucleotide) or purified antibodies (anti-Sp1: sc-14027X; anti-USF: sc862X or anti-HA, Santa Cruz, TEBU, France) for 30 min before addition of labeled probes. The complexes were resolved as previously reported [8].
In silico tools. The nucleotide change within the CFTR promoter was screened by TFSEARCH v1.3 (www.cbrc.jp/research/db/TFSEARCH. html), and AliBaba v2.1 (www.gene-regulation.com/pub/programs/alibaba/index.html) and compared to the wild-type sequence to identify any loss or gain of transcription factor binding affinity. Statistical analyses. Transfection data are expressed as the mean ± SE Paired comparisons were made using student’s t-test. Data were considered statistically significant at p < 0.05. All graphical data and statistical analyses were generated with GraphPAD Prism software (Version 3.0).
Results Characterization of deleterious effect of CFTR promoter nucleotide change To explore specifically the effects of the 94G>T sequence variation on CFTR expression, we used two luciferase reporter constructs driven by the minimal CFTR promoter, which differed only by the presence of either a guanine or a thymine at the 94 position. Multiple plasmid preparations were made from each reporter construct and transiently transfected into both Beas2B and Caco-2 cells, which express endogenous CFTR [8], and also into Hela cells, which do not express endogenous CFTR. The relative luciferase activity of the mutant construct (M-94T-pGL3) was significantly decreased by 25%, 26%, or 38% in Beas2B, Caco-2, or Hela cells, respectively, compared with the activity of the wild-type construct (Fig. 1A). These results suggest that the 94T variant is functionally significant. In order to examine whether the CFTR promoter sequence encompassing the 94G>T nucleotide variant interacts with epithelial cell nuclear proteins, EMSAs were performed. G and T variant binding region oligonucleotide probes were incubated in the presence of Beas2B, Caco-2 or Hela nuclear extracts and then migrated through a non-denaturing acrylamide gel. As shown in Fig. 1B, four major nucleoprotein complexes with similar electrophoretic mobility (noted I, II, III, and IV with a cell type-dependent upper-case before) were detected with both probes in all cell types examined. However, a more accurate examination of the band shift patterns revealed that the complexes formed with the 94T probe were consistently less intense
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
777
Fig. 1. Analysis of allele-specific transcriptional effects of the 94G>T CFTR promoter variant. (A) Panel A shows transcriptional activities of the 94G and 94T alleles in Beas2B, Caco-2 and Hela epithelial cell lines. Bars represent Firefly/Renilla luciferase ratios for the different constructs. Results from pGL3-Basic plasmid (grey bars) are given as a negative control. Thick solid bars show luciferase activities from cell lines transfected with the WT-94GpGL3 construct containing the wild-type allele. Open bars represent luciferase activities from cell lines transfected with the M-94T-pGL3 construct containing the mutant allele. Luciferase activity obtained from the wild-type allele 94G was defined as 100%, and relative luciferase activity from mutant allele 94T, indicated above the Open bars, was expressed from this value. The (*) indicates that the value is statistically significant (p < 0.05) compared to the wild-type construct. (B) Panel B shows allele-specific binding of transcription factors. 32P-labeled G or T allele-specific probes (noted WT and M, respectively) were incubated with nuclear proteins from different epithelial cell lines, followed by gel electrophoresis and autoradiography. Four major nuclear protein complexes were detected and designated by a cell type-dependent upper-case to the left of autoradiograms. Specificity was assessed by competition with a 100-fold molar excess of unlabeled oligonucleotide encompassing either each allele (noted 94G or 94T) or an irrelevant sequence (noted IC).
than the wild-type ones and resulted in a pattern slightly different for the complexes II and III. These first observations suggested that the 94T allele affects the affinities of existing protein–DNA interactions. Competition assays with both unlabeled specific oligonucleotides and an irrele-
vant competitor evidenced that the formation of three DNA-protein complexes II, III, and IV is specific for both WT and M probes tested (Fig. 1B). Complex I is nonspecific because it is not inhibited by unlabeled homologous oligonucleotides. Taken together with the reduced
778
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
transcriptional activity of the 94T allele relative to the 94G allele, these results suggest that the binding affinity of factors is seriously affected by the promoter variant. Specific binding of both Sp1 and USF transcription factors to the CFTR promoter region encompassing the 94G>T nucleotide change To assess the effect of the 94G>T promoter variation on putative transcription-factor binding sites, we performed an analysis with both TFSEARCH and AliBaba algorithms. It was observed that the G>T sequence variation at position 94 is located within overlapping non-consensus Sp1 and USF binding sequences (data not shown). The analysis also revealed that the matrix match score of the 94G variant (wild-type sequence) is slightly higher than that of the 94T variant (mutated sequence), indicating that both Sp1 and USF transcription factors have a higher affinity to the 94G variant than to the 94T variant. The consideration of these parameters would tend to reinforce the likehood that the 94T allele is deleterious. To further characterize the transcription factors binding to the studied CFTR promoter region, additional competition experiments were performed with various oligonucleotides corresponding to Sp1 and USF2 factors. As shown in Fig. 2, effective competition was obtained with specific
competitors (sequences listed in Table 1). Nuclear protein complexes BII, BIII, CII, and CIII, were significantly decreased by the addition of cold competitor containing a core Sp1 element (Fig. 2). The addition of cold competitor containing a core USF element resulted in decrease of complexes BII, BIV, CII, and CIV (Fig. 2). These observations suggest that both Sp1 and USF2 proteins contained into the complexes BII and CII, could simultaneously occupy the region encompassing the 94G>T sequence variation. In contrast, an irrelevant oligonucleotide (sequence listed in Table 1) did not compete for both 94G and 94T allele-specific binding protein complexes under the same conditions. To get further insights into the nature of nuclear proteins binding antibody interference assays were performed with specific antibodies. Although the incubation with both anti-Sp1 and anti-USF2 antibodies apparently failed to give any discrete supershift from Beas2B nuclear proteins, a significant decrease or total disappearance of BII complex was observed and subsequently confirmed the presence of these two transcription factors within this complex (Fig. 2). The effect seemed to be specific because none of nuclear protein-complexes was supershifted or abolished when a non-specific antibody was used. In addition, neither the retarded nor supershifted complexes were found when antibodies were incubated with the labeled probes in the absence of nuclear extracts
Fig. 2. Characterization of transcription factors that bind the sequence encompassing the 94G>T variant. EMSAs were performed with nuclear extracts from both Beas2B and Caco-2 epithelial cell lines and the 32P-labeled G and T allele-specific probes (noted WT and M, respectively). Four major DNA– protein complexes were detected and designated by a cell type-dependent upper-case to the left of autoradiograms. DNA competition experiments were carried out with a 100-fold molar excess of the indicated cold competitors. IC indicates irrelevant competitor. SC indicates specific competitor. Immunologic assays were performed with either specific antibodies (noted anti-Sp1 and anti-USF) or anti-HA as nonspecific antiserum (noted NS). The arrow SS indicates the supershifted complex.
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
(data not shown). The same trend was also observed with Caco-2 nuclear extracts cells with a few variations (Fig. 2). In EMSA carried out with the labeled WT probe, the addition of the anti-Sp1 antibody resulted in discrete supershift (Fig. 2). The absence of any supershift with other nuclear extracts (Beas2) may result from shielding of the epitope for antibody recognition by binding of another factor to Sp1 and/or USF in BII higher molecular complexes (Fig. 2). Relevant experiments with other nuclear extracts (Hela) were done and the results were similar to those presented in this report (data not shown). Together, these data indicate that the 94T allele in the human CFTR promoter decreases the binding affinity of both Sp1 and USF2 transcription factors to their respective overlapping like-elements, leading to decreased transcriptional activity. Transactivation of the CFTR promoter by Sp1 and USF2 in epithelial cells To assess whether Sp1 and/or USF proteins affect CFTR transcriptional regulation, we transiently co-transfected exogenous full-length Sp1 and/or USF with each reporter construct. Beas2B cells were co-transfected with either the WT (WT-94G-pGL3b) or mutant (M-94TpGL3b) reporter constructs and either pN3-Sp1 or pCR3-USF2, or both expression plasmids simultaneously in cells (Fig. 3). Compared to cells co-transfected with the reporter constructs alone, Sp1 transactivates both the WT-94G-pGL3b and M-94T-pGL3b plasmids 1,9-fold and 1,4-fold, respectively. Similarly, USF2 up-regulates luciferase activity 2,1-fold with both reporter constructs. In addition, overexpression of both Sp1 and USF proteins with WT-94G-pGL3b plasmid resulted in ca 66% luciferase
779
activity increase (Fig. 3). Unexpectedly, we also noted a transactivation when both Sp1 and USF were co-transfected with the M-94T-pGL3b plasmid, suggesting the implication of other factors. Discussion Analysis of DNA from a female patient diagnosed CF on the basis of two positive sweat tests and severe chronic lung disease revealed only one sequence variation so far. Indeed, despite an exhaustive direct sequencing and a gross genomic rearrangements search, we failed to identify any known pathogenic mutation except the previously reported 94G>T CFTR promoter variant [2]. Hence, to determine if this promoter nucleotide change was neutral or of clinical significance, we proceeded to both in silico analyses and functional assays. The results of gene reporter assays indicated that the 94T variant produced significant decrease of the basal CFTR transcriptional activity in the three epithelial cell lines tested, suggesting that this promoter variant could affect existing protein-DNA interactions. To test this hypothesis, in silico analysis was performed and revealed that the G to T substitution at position 94 is located within overlapping Sp1 and USF binding motifs. Sp1, an ubiquitous zinc finger transcription factor, which binds to GC rich sequences known as « GC boxes » [9], has been identified in 5 0 regions of many genes including the CFTR gene [10]. Sp1 sites are reported to be implicated in initiation of transcription in TATA-less promoters [11] and tissue specific gene expression [12]. USFs proteins have been postulated to regulate the expression of housekeeping genes [13] and tissue-specific genes expression [14]. EMSA analyses demonstrated that the 94G>T substitution
Fig. 3. Combinatorial activation by Sp1 and USF in Beas2B cells. Induction of wild-type or mutant reporter gene expression by co-transfection with Sp1 and/or USF expression vectors. Luciferase activity obtained from either WT-94G-pGL3 construct containing the wild-type allele (left panel) or M-94TpGL3 construct containing the mutant allele construct (right panel) was defined as 100% (grey bars). Open bars show luciferase activities from Beas2B cells co-transfected with either pN3-Sp1 or pCR3-USF2 expression vectors. Thick solid bars show luciferase activity from Beas2B cells co-transfected with both Sp1 and USF expression vectors and wild-type (left panel) or mutant (right panel) constructs. The (*) indicates that the value is statistically significant (p < 0.05).
780
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
decreased the affinity of both Sp1 and USF nuclear proteins to the binding site of CFTR covering the 94 position. Although, at this stage we can not exclude that binding of other transcription factors are affected by the 94G>T promoter variant, our data suggest in line with the reduced DNA-binding, the polymorphic change at position -94 has a negative effect on basal CFTR transcriptional activation. These results are of interest since at the best of knowledge, this study reports the first functional evidence for CFTR promoter variant associated with severe pulmonary outcome. Thereby, this 94G>T promoter variant might account for the phenotype CF, possibly acting in concert with not yet identified mutations in the unexplored regions of the CFTR gene (deep intronic sequences and 5 0 and 3 0 regulatory regions) or mutations in modifying genes and reinforces the importance of functional studies on 5 0 UTR regions. Using reporter gene assays, we demonstrated that USF and Sp1 factors cooperate to transactivate the CFTR promoter activity, as previously documented for other genes [15,16]. However and unexpectedly, despite we demonstrated that the promoter variant decreases basal CFTR transcriptional activity and affinities of both Sp1 and USF proteins to their CFTR promoter binding motifs, we observed that these factors also cooperate to transactivate the M-94T-pGL3b promoter activity. A possible explanation is that it might exist a competition between exogenous USF/Sp1 proteins and other TRANS-regulators that present a higher affinity than USF and Sp1 on the mutated promoter. Moreover, the transcriptional activity decrease induced by the promoter mutant might be due simply to the binding of repressors to the polymorphic sequence at position 94 or with a more complex view to functional interactions between USF/Sp1 proteins and others factors or co-factors. It is well defined that genes expression is regulated by an exquisite balance between various activators and/or repressors and co-regulators in a coordinated fashion. An attractive biological feature of USF and Sp1 proteins is their abilities to mediate a wide range of transcriptional activities via interactions with additional factors or cofactors [17,18]. While USFs and Sp1 proteins have been traditionally described as transcriptional activators, studies have suggested that in specific instances [19,20], they may act as a negative regulators by interacting with transcription factors which bound simultaneously to the same site or at other sites, suggesting a dual mode of regulation of genes by these factors. USF functionality including its tissue-specificity may also be regulated by an interplay between its various isoforms [21–23]. North et al. have reported the possibility that USF2b homodimer may be a transrepressor while USF1/USF2b heterodimer could activate the same gene [24]. Moreover, it has been shown that binding of USF or Sp1 to the E-box or GC-rich motifs, respectively, induces an acute bend in the DNA [25]. So, such bending could probably facilitate interactions with other proteins like those already involved in the CFTR
expression regulation such as NF-jB [26], C/EBP [27] or the repressor YY1 [8]. As genes are typically regulated by multiple transcription factors, each with its associated set of coactivators and repressors, which respond to different signaling cascades, it is evident that major advances in the characterization of transcription factors that influence genes transcription will enlighten on the mechanisms involved in the CFTR gene expression control. Acknowledgments This work was supported in part by the Centre Hospitalier Universitaire of Montpellier and the association Vaincre La Mucoviscidose (VLM, Grant FC0527). References [1] S. Carles, M. Desgeorges, A. Goldman, R. Thiart, C. Guittard, C.A. Kitazos, T.J. de Ravel, A.T. Westwood, M. Claustres, M. Ramsay, First report of CFTR mutations in black cystic fibrosis patients of southern African origin, J. Med. Genet. 33 (1996) 802–804. [2] M.C. Romey, C. Guittard, S. Carles, J. Demaille, M. Ramsay, M. Claustres, First putative sequence alterations in the minimal CFTR promoter region, J. Med. Genet. 36 (1999) 263–264. [3] M.C. Romey, N. Pallares-Ruiz, A. Mange, C. Mettling, R. Peytavi, J. Demaille, M. Claustres, A naturally occurring sequence variation that creates a YY1 element is associated with increased cystic fibrosis transmembrane conductance regulator gene expression, J. Biol. Chem. 275 (2000) 3561–3567. [4] R. Rowntree, A. Harris, DNA polymorphisms in potential regulatory elements of the CFTR gene alter transcription factor binding, Hum. Genet. 111 (2002) 66–74. [5] J. Wittwer, J. Marti-Jaun, M. Hersberger, Functional polymorphism in ALOX15 results in increased allele-specific transcription in macrophages through binding of the transcription factor SPI1, Hum. Mutat. 27 (2006) 78–87. [6] T. Bienvenu, V. Lacronique, M. Raymondjean, C. Cazeneuve, D. Hubert, J.C. Kaplan, C. Beldjord, Three novel sequence variations in the 5 0 upstream region of the cystic fibrosis transmembrane conductance regulator (CFTR) gene: two polymorphisms and one putative molecular defect, Hum. Genet. 95 (1995) 698–702. [7] M.C. Romey, C. Guittard, J.P. Chazalette, P. Frossard, K.P. Dawson, M.A. Patton, T. Casals, T. Bazarbachi, E. Girodon, G. Rault, D. Bozon, F. Seguret, J. Demaille, M. Claustres, Complex allele [-102T>A+S549R(T>G)] is associated with milder forms of cystic fibrosis than allele S549R(T>G) alone, Hum. Genet. 105 (1999) 145–150. [8] C. Rene, M. Taulan, F. Iral, J. Doudement, A. L’Honore, C. Gerbon, J. Demaille, M. Claustres, M.C. Romey, Binding of serum response factor to cystic fibrosis transmembrane conductance regulator CArGlike elements, as a new potential CFTR transcriptional regulation pathway, Nucleic Acids Res. 33 (2005) 5271–5290. [9] J.T. Kadonaga, K.R. Carner, F.R. Masiarz, R. Tjian, Isolation of cDNA encoding transcription factor Sp1 and functional analysis of the DNA binding domain, Cell 51 (1987) 1079–1090. [10] K. Yoshimura, H. Nakamura, B.C. Trapnell, W. Dalemans, A. Pavirani, J.P. Lecocq, R.G. Crystal, The cystic fibrosis gene has a ‘‘housekeeping’’-type promoter and is expressed at low levels in cells of epithelial origin, J. Biol. Chem. 266 (1991) 9140–9144. [11] R. Kollmar, K.A. Sukow, S.K. Sponagle, P.J. Farnham, Start site selection at the TATA-less carbamoyl-phosphate synthase (glutamine-hydrolyzing)/aspartate carbamoyltransferase/dihydroorotase promoter, J. Biol. Chem. 269 (1994) 2252–2257. [12] D.E. Zhang, C.J. Hetherington, S. Tan, S.E. Dziennis, D.A. Gonzalez, H.M. Chen, D.G. Tenen, Sp1 is a critical factor for the
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
[13]
[14]
[15]
[16]
[17]
[18]
monocytic specific expression of human CD14, J. Biol. Chem. 269 (15) (1994) 11425–11434. A.L. Roy, M. Meisterernst, P. Pognonec, R.G. Roeder, Cooperative interaction of an initiator-binding transcription initiation factor and the helix-loop-helix activator USF, Nature 354 (1991) 245–248. J. Liu, M. Antaya, D. Boerboom, J.G. Lussier, D.W. Silversides, J. Sirois, The delayed activation of the prostaglandin G/H synthase-2 promoter in bovine granulosa cells is associated with down-regulation of truncated upstream stimulatory factor-2, J. Biol. Chem 274 (1999) 35037–35045. Y. Ge, T.L. Jensen, L.H. Matherly, J.W. Taub, Physical and functional interactions between USF and Sp1 proteins regulate human deoxycytidine kinase promoter activity, J. Biol. Chem. 278 (2003) 49901–49910. E. Meccia, L. Bottero, F. Felicetti, C. Peschle, M.P. Colombo, A. Care, HOXB7 expression is regulated by the transcription factors NF-Y, YY1, Sp1 and USF-1, Biochim. Biophys. Acta 1626 (2003) 1–9. M.A. Renault, S. Jalvy, M. Potier, I. Belloc, E. Genot, L.V. Dekker, C. Desgranges, A.P. Gadeau, UTP induces osteopontin expression through a coordinate action of NFkappaB, activator protein-1, and upstream stimulatory factor in arterial smooth muscle cells, J. Biol. Chem. 280 (2005) 2708–2713. M.K. Bennett, T.T. Ngo, J.N. Athanikar, J.M. Rosenfeld, T.F. Osborne, Co-stimulation of promoter for low density lipoprotein receptor gene by sterol regulatory element-binding protein and Sp1 is specifically disrupted by the yin yang 1 protein, J. Biol. Chem. 274 (1999) 13025–13032.
781
[19] A.K. Ghosh, P.K. Datta, S.T. Jacob, The dual role of helix-loop– helix-zipper protein USF in ribosomal RNA gene transcription in vivo, Oncogene 14 (1997) 589–594. [20] K.M. Kelley, H. Wang, M. Ratnam, Dual regulation of ets-activated gene expression by SP1, Gene 307 (2003) 87–97. [21] M.L. Moore, E.A. Park, J.B. McMillin, Upstream stimulatory factor represses the induction of carnitine palmitoyltransferase-Ibeta expression by PGC-1, J. Biol. Chem. 278 (2003) 17263–17268. [22] B. Viollet, A.M. Lefrancois-Martinez, A. Henrion, A. Kahn, M. Raymondjean, A. Martinez, Immunochemical characterization and transacting properties of upstream stimulatory factor isoforms, J. Biol. Chem. 271 (1996) 1405–1415. [23] S. Yan, B.F. Sloane, Isolation of a novel USF2 isoform: repressor of cathepsin B expression, Gene 337 (2004) 199–206. [24] S. North, X. Espanel, F. Bantignies, B. Viollet, V. Vallet, P. Jalinot, G. Brun, G. Gillet, Regulation of cdc2 gene expression by the upstream stimulatory factors (USFs), Oncogene 18 (1999) 1945–1955. [25] D. Kumari, A. Gabrielian, D. Wheeler, K. Usdin, The roles of Sp1, Sp3, USF1/USF2 and NRF-1 in the regulation and three-dimensional structure of the Fragile X mental retardation gene promoter, Biochem. J. 386 (2005) 297–303. [26] F. Brouillard, M. Bouthier, T. Leclerc, A. Clement, M. BaudouinLegros, A. Edelman, NF-kappa B mediates up-regulation of CFTR gene expression in Calu-3 cells by interleukin-1beta, J. Biol. Chem. 276 (2001) 9486–9491. [27] N. Pittman, G. Shue, N.S. LeLeiko, M.J. Walsh, Transcription of cystic fibrosis transmembrane conductance regulator requires a CCAAT-like element for both basal and cAMP-mediated regulation, J. Biol. Chem. 270 (1995) 28848–28857.
First functional polymorphism in CFTR promoter that results in decreased transcriptional activity and Sp1/USF binding M. Taulan
a,b,c,*
, E. Lopez a,b,1, C. Guittard a,c,1, C. Rene´ a,b,c, D. Baux a,c, J.P. Altieri M. DesGeorges a,c, M. Claustres a,b,c, M.C. Romey a,b
a,c
,
a
c
INSERM U827, Laboratoire de Ge´ne´tique de Maladies Rares, Montpellier, France b Universite´ Montpellier1, UFR de Me´decine, Montpellier, France CHU Montpellier, Hoˆpital Arnaud de Villeneuve, Laboratoire de Ge´ne´tique Mole´culaire, Montpellier F-34000, France Received 14 July 2007 Available online 25 July 2007
Abstract Growing evidences show that functionally relevant polymorphisms in various promoters alter both transcriptional activity and affinities of existing protein–DNA interactions, and thus influence disease progression in humans. We previously reported the 94G>T CFTR promoter variant in a female CF patient in whom any known disease-causing mutation has been detected. To investigate whether the 94G>T could be a regulatory variant, we have proceeded to in silico analyses and functional studies including EMSA and reporter gene assays. Our data indicate that the promoter variant decreases basal CFTR transcriptional activity in different epithelial cells and alters binding affinities of both Sp1 and USF nuclear proteins to the CFTR promoter. The present report provides evidence for the first functional polymorphism that negatively affects the CFTR transcriptional activity and demonstrates a cooperative role of Sp1 and USF transcription factors in transactivation of the CFTR gene promoter. 2007 Elsevier Inc. All rights reserved. Keywords: CFTR promoter variant; Functional analysis; Sp1; USF transcription factor
Cystic fibrosis (CF), the most common life-limiting recessively transmitted genetic disease in Caucasians, is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene. While the most common mutation is the Phe508del, other putative diseasecausing mutations identified in the CFTR gene, both in coding and non-coding regions, are rare or private [1,2]. In the promoter region, there is a relative paucity of sequence variations and their relationship to CF remains yet uncertain (see www.genet.sicckids.on.ca). Nevertheless, several studies have showed that functionally relevant polymorphisms in various promoters including CFTR alter *
Corresponding author. Address: Inserm U827, UFR de Me´decine, IURC 641, Avenue du Doyen Gaston Giraud, 34093 Montpellier Cedex 5, France. Fax: +33 467415365. E-mail address: [email protected] (M. Taulan). 1 These authors contributed equally to this work. 0006-291X/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2007.07.091
both transcriptional activity and affinities of existing protein–DNA interactions, and thus influence disease progression in humans [3–5]. While a putative deleterious mutation (741T>G) within the transcriptional regulatory region has been partially characterized by EMSA [6], only one study has shown the impact of promoter variant on transcriptional activity [3]. Indeed, it has previously demonstrated that sequence variant in the human minimal promoter, associated with an attenuated CF phenotype [7] increase CFTR expression levels [3]. Moreover, despite the recent identification of functional antagonism between the SRF and YY1 transcription factors through competition for the CFTR-CArG-like binding [8], no clear mechanism responsible for CFTR transcriptional regulation has yet been reported. In the current study, we have focused on a female patient harbouring any putative disease-causing mutation except to the 94G>T promoter nucleotide change [2].
776
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
Table 1 Sequences of oligonucleotides for functional assays Site-directed mutagenesis experiment 5 0 -CAAATTTGGGGCCTGACCAGGCAGCACTCG-3 0 EMSA probes and competitors 94G (WT) 94T (M) USF Sp1a Pax-5 mutantb (IC)
5 0 -CCGCTAGAGCAAATTTGGGGCCGGACCAGGCAGCACTCGGC-3 0 5 0 -CCGCTAGAGCAAATTTGGGGCCTGACCAGGCAGCACTCGGC-3 0 5 0 -CACCCGGTCACGTGGCCTACACC-3 0 5 0 -ATTCGATCGGGGCGGGGCGAGC-3 0 5 0 -GAATCTGGAACAGAGTCGCTCCCACCG-3 0
Bold and italicized type indicates mutated nucleotides. The underlined nucleotides indicate the core site. a Oligonucleotide probe purchased from Promega, Corporation (E323A). b Oligonucleotide probe purchased from Santa Cruz (sc-2590).
The currently most-favored hypothesis advocates the 94G>T promoter variant could be viewed as a modifier polymorphism. Therefore, we tested the functional significance of this variant by using in silico analysis, reporter gene and band shift assays. We identified putative regulatory composite cis-acting element for CFTR promoter activity involving GC-rich and E-box motifs. We explored the functional relevance of Sp1 and USF transcription factors in the transcriptional control of the CFTR gene. Materials and methods Plasmids constructs. The original luciferase expression vector pGL3Basic containing the wild-type human CFTR minimal promoter was previously described [3]. The nucleotidique change 94T was introduced into the WT-pGL3 plasmid using an oligonucleotide directed mutagenesis system (QuickChangeSite-Directed Mutagenesis Kit, Stratagene) according to the manufacturer’s instructions. The mutagenesis primers were listed in Table 1. The presence of mutations and sequence fidelity were verified by direct sequencing. Cell culture and transient transfections. Hela, Beas2B and Caco-2 human epithelial cell lines were maintained as previously reported [3]. Transient transfections assays were performed with PolyFect transfection reagent (Qiagen, France) as previously described [8], except some minor modifications. All cells were seeded at a density of 40,000 cells/400 ll of medium and plated in 24-well dishes (NUNClone, Merck-Eurolab, Inc.). We used 0.36 lg of plasmids reporter (WT-94 GpGL3 or M-94T-pGL3) and 40 ng of internal control pRL-SV40 containing Renilla luciferase (Promega, France). All luciferase activities represent at least three independent experiments with each construct tested in triplicate per experiment. To minimize the possibility of errors in DNA amplification, at least two independently constructed clones were tested. For cotransfection assays, 0.36 lg of luciferase reporter and 100 ng of each expression vector were used. Expression vectors encoding either pCR3-USF2a or pN3-Sp1 were generously provided by B. Viollet and G Suske, respectively. Electrophoretic mobility shift assay (EMSA). Nuclear extracts and radiolabeled double-stranded oligomeric probes were prepared as previously reported [8]. EMSAs were performed in a total volume of 20 ll containing 2 ng of the labeled probe, 10 ll of binding buffer (2·), 2–5 lg polydIdC as a nonspecific competitor, 10–20 lg of nuclear extracts. For EMSA competition and antibody interference assays, proteins were incubated with either cold specific competitors (WT, M, Sp1, or USF oligonucleotides, see Table 1), irrelevant competitor (Pax5 mutant oligonucleotide) or purified antibodies (anti-Sp1: sc-14027X; anti-USF: sc862X or anti-HA, Santa Cruz, TEBU, France) for 30 min before addition of labeled probes. The complexes were resolved as previously reported [8].
In silico tools. The nucleotide change within the CFTR promoter was screened by TFSEARCH v1.3 (www.cbrc.jp/research/db/TFSEARCH. html), and AliBaba v2.1 (www.gene-regulation.com/pub/programs/alibaba/index.html) and compared to the wild-type sequence to identify any loss or gain of transcription factor binding affinity. Statistical analyses. Transfection data are expressed as the mean ± SE Paired comparisons were made using student’s t-test. Data were considered statistically significant at p < 0.05. All graphical data and statistical analyses were generated with GraphPAD Prism software (Version 3.0).
Results Characterization of deleterious effect of CFTR promoter nucleotide change To explore specifically the effects of the 94G>T sequence variation on CFTR expression, we used two luciferase reporter constructs driven by the minimal CFTR promoter, which differed only by the presence of either a guanine or a thymine at the 94 position. Multiple plasmid preparations were made from each reporter construct and transiently transfected into both Beas2B and Caco-2 cells, which express endogenous CFTR [8], and also into Hela cells, which do not express endogenous CFTR. The relative luciferase activity of the mutant construct (M-94T-pGL3) was significantly decreased by 25%, 26%, or 38% in Beas2B, Caco-2, or Hela cells, respectively, compared with the activity of the wild-type construct (Fig. 1A). These results suggest that the 94T variant is functionally significant. In order to examine whether the CFTR promoter sequence encompassing the 94G>T nucleotide variant interacts with epithelial cell nuclear proteins, EMSAs were performed. G and T variant binding region oligonucleotide probes were incubated in the presence of Beas2B, Caco-2 or Hela nuclear extracts and then migrated through a non-denaturing acrylamide gel. As shown in Fig. 1B, four major nucleoprotein complexes with similar electrophoretic mobility (noted I, II, III, and IV with a cell type-dependent upper-case before) were detected with both probes in all cell types examined. However, a more accurate examination of the band shift patterns revealed that the complexes formed with the 94T probe were consistently less intense
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
777
Fig. 1. Analysis of allele-specific transcriptional effects of the 94G>T CFTR promoter variant. (A) Panel A shows transcriptional activities of the 94G and 94T alleles in Beas2B, Caco-2 and Hela epithelial cell lines. Bars represent Firefly/Renilla luciferase ratios for the different constructs. Results from pGL3-Basic plasmid (grey bars) are given as a negative control. Thick solid bars show luciferase activities from cell lines transfected with the WT-94GpGL3 construct containing the wild-type allele. Open bars represent luciferase activities from cell lines transfected with the M-94T-pGL3 construct containing the mutant allele. Luciferase activity obtained from the wild-type allele 94G was defined as 100%, and relative luciferase activity from mutant allele 94T, indicated above the Open bars, was expressed from this value. The (*) indicates that the value is statistically significant (p < 0.05) compared to the wild-type construct. (B) Panel B shows allele-specific binding of transcription factors. 32P-labeled G or T allele-specific probes (noted WT and M, respectively) were incubated with nuclear proteins from different epithelial cell lines, followed by gel electrophoresis and autoradiography. Four major nuclear protein complexes were detected and designated by a cell type-dependent upper-case to the left of autoradiograms. Specificity was assessed by competition with a 100-fold molar excess of unlabeled oligonucleotide encompassing either each allele (noted 94G or 94T) or an irrelevant sequence (noted IC).
than the wild-type ones and resulted in a pattern slightly different for the complexes II and III. These first observations suggested that the 94T allele affects the affinities of existing protein–DNA interactions. Competition assays with both unlabeled specific oligonucleotides and an irrele-
vant competitor evidenced that the formation of three DNA-protein complexes II, III, and IV is specific for both WT and M probes tested (Fig. 1B). Complex I is nonspecific because it is not inhibited by unlabeled homologous oligonucleotides. Taken together with the reduced
778
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
transcriptional activity of the 94T allele relative to the 94G allele, these results suggest that the binding affinity of factors is seriously affected by the promoter variant. Specific binding of both Sp1 and USF transcription factors to the CFTR promoter region encompassing the 94G>T nucleotide change To assess the effect of the 94G>T promoter variation on putative transcription-factor binding sites, we performed an analysis with both TFSEARCH and AliBaba algorithms. It was observed that the G>T sequence variation at position 94 is located within overlapping non-consensus Sp1 and USF binding sequences (data not shown). The analysis also revealed that the matrix match score of the 94G variant (wild-type sequence) is slightly higher than that of the 94T variant (mutated sequence), indicating that both Sp1 and USF transcription factors have a higher affinity to the 94G variant than to the 94T variant. The consideration of these parameters would tend to reinforce the likehood that the 94T allele is deleterious. To further characterize the transcription factors binding to the studied CFTR promoter region, additional competition experiments were performed with various oligonucleotides corresponding to Sp1 and USF2 factors. As shown in Fig. 2, effective competition was obtained with specific
competitors (sequences listed in Table 1). Nuclear protein complexes BII, BIII, CII, and CIII, were significantly decreased by the addition of cold competitor containing a core Sp1 element (Fig. 2). The addition of cold competitor containing a core USF element resulted in decrease of complexes BII, BIV, CII, and CIV (Fig. 2). These observations suggest that both Sp1 and USF2 proteins contained into the complexes BII and CII, could simultaneously occupy the region encompassing the 94G>T sequence variation. In contrast, an irrelevant oligonucleotide (sequence listed in Table 1) did not compete for both 94G and 94T allele-specific binding protein complexes under the same conditions. To get further insights into the nature of nuclear proteins binding antibody interference assays were performed with specific antibodies. Although the incubation with both anti-Sp1 and anti-USF2 antibodies apparently failed to give any discrete supershift from Beas2B nuclear proteins, a significant decrease or total disappearance of BII complex was observed and subsequently confirmed the presence of these two transcription factors within this complex (Fig. 2). The effect seemed to be specific because none of nuclear protein-complexes was supershifted or abolished when a non-specific antibody was used. In addition, neither the retarded nor supershifted complexes were found when antibodies were incubated with the labeled probes in the absence of nuclear extracts
Fig. 2. Characterization of transcription factors that bind the sequence encompassing the 94G>T variant. EMSAs were performed with nuclear extracts from both Beas2B and Caco-2 epithelial cell lines and the 32P-labeled G and T allele-specific probes (noted WT and M, respectively). Four major DNA– protein complexes were detected and designated by a cell type-dependent upper-case to the left of autoradiograms. DNA competition experiments were carried out with a 100-fold molar excess of the indicated cold competitors. IC indicates irrelevant competitor. SC indicates specific competitor. Immunologic assays were performed with either specific antibodies (noted anti-Sp1 and anti-USF) or anti-HA as nonspecific antiserum (noted NS). The arrow SS indicates the supershifted complex.
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
(data not shown). The same trend was also observed with Caco-2 nuclear extracts cells with a few variations (Fig. 2). In EMSA carried out with the labeled WT probe, the addition of the anti-Sp1 antibody resulted in discrete supershift (Fig. 2). The absence of any supershift with other nuclear extracts (Beas2) may result from shielding of the epitope for antibody recognition by binding of another factor to Sp1 and/or USF in BII higher molecular complexes (Fig. 2). Relevant experiments with other nuclear extracts (Hela) were done and the results were similar to those presented in this report (data not shown). Together, these data indicate that the 94T allele in the human CFTR promoter decreases the binding affinity of both Sp1 and USF2 transcription factors to their respective overlapping like-elements, leading to decreased transcriptional activity. Transactivation of the CFTR promoter by Sp1 and USF2 in epithelial cells To assess whether Sp1 and/or USF proteins affect CFTR transcriptional regulation, we transiently co-transfected exogenous full-length Sp1 and/or USF with each reporter construct. Beas2B cells were co-transfected with either the WT (WT-94G-pGL3b) or mutant (M-94TpGL3b) reporter constructs and either pN3-Sp1 or pCR3-USF2, or both expression plasmids simultaneously in cells (Fig. 3). Compared to cells co-transfected with the reporter constructs alone, Sp1 transactivates both the WT-94G-pGL3b and M-94T-pGL3b plasmids 1,9-fold and 1,4-fold, respectively. Similarly, USF2 up-regulates luciferase activity 2,1-fold with both reporter constructs. In addition, overexpression of both Sp1 and USF proteins with WT-94G-pGL3b plasmid resulted in ca 66% luciferase
779
activity increase (Fig. 3). Unexpectedly, we also noted a transactivation when both Sp1 and USF were co-transfected with the M-94T-pGL3b plasmid, suggesting the implication of other factors. Discussion Analysis of DNA from a female patient diagnosed CF on the basis of two positive sweat tests and severe chronic lung disease revealed only one sequence variation so far. Indeed, despite an exhaustive direct sequencing and a gross genomic rearrangements search, we failed to identify any known pathogenic mutation except the previously reported 94G>T CFTR promoter variant [2]. Hence, to determine if this promoter nucleotide change was neutral or of clinical significance, we proceeded to both in silico analyses and functional assays. The results of gene reporter assays indicated that the 94T variant produced significant decrease of the basal CFTR transcriptional activity in the three epithelial cell lines tested, suggesting that this promoter variant could affect existing protein-DNA interactions. To test this hypothesis, in silico analysis was performed and revealed that the G to T substitution at position 94 is located within overlapping Sp1 and USF binding motifs. Sp1, an ubiquitous zinc finger transcription factor, which binds to GC rich sequences known as « GC boxes » [9], has been identified in 5 0 regions of many genes including the CFTR gene [10]. Sp1 sites are reported to be implicated in initiation of transcription in TATA-less promoters [11] and tissue specific gene expression [12]. USFs proteins have been postulated to regulate the expression of housekeeping genes [13] and tissue-specific genes expression [14]. EMSA analyses demonstrated that the 94G>T substitution
Fig. 3. Combinatorial activation by Sp1 and USF in Beas2B cells. Induction of wild-type or mutant reporter gene expression by co-transfection with Sp1 and/or USF expression vectors. Luciferase activity obtained from either WT-94G-pGL3 construct containing the wild-type allele (left panel) or M-94TpGL3 construct containing the mutant allele construct (right panel) was defined as 100% (grey bars). Open bars show luciferase activities from Beas2B cells co-transfected with either pN3-Sp1 or pCR3-USF2 expression vectors. Thick solid bars show luciferase activity from Beas2B cells co-transfected with both Sp1 and USF expression vectors and wild-type (left panel) or mutant (right panel) constructs. The (*) indicates that the value is statistically significant (p < 0.05).
780
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
decreased the affinity of both Sp1 and USF nuclear proteins to the binding site of CFTR covering the 94 position. Although, at this stage we can not exclude that binding of other transcription factors are affected by the 94G>T promoter variant, our data suggest in line with the reduced DNA-binding, the polymorphic change at position -94 has a negative effect on basal CFTR transcriptional activation. These results are of interest since at the best of knowledge, this study reports the first functional evidence for CFTR promoter variant associated with severe pulmonary outcome. Thereby, this 94G>T promoter variant might account for the phenotype CF, possibly acting in concert with not yet identified mutations in the unexplored regions of the CFTR gene (deep intronic sequences and 5 0 and 3 0 regulatory regions) or mutations in modifying genes and reinforces the importance of functional studies on 5 0 UTR regions. Using reporter gene assays, we demonstrated that USF and Sp1 factors cooperate to transactivate the CFTR promoter activity, as previously documented for other genes [15,16]. However and unexpectedly, despite we demonstrated that the promoter variant decreases basal CFTR transcriptional activity and affinities of both Sp1 and USF proteins to their CFTR promoter binding motifs, we observed that these factors also cooperate to transactivate the M-94T-pGL3b promoter activity. A possible explanation is that it might exist a competition between exogenous USF/Sp1 proteins and other TRANS-regulators that present a higher affinity than USF and Sp1 on the mutated promoter. Moreover, the transcriptional activity decrease induced by the promoter mutant might be due simply to the binding of repressors to the polymorphic sequence at position 94 or with a more complex view to functional interactions between USF/Sp1 proteins and others factors or co-factors. It is well defined that genes expression is regulated by an exquisite balance between various activators and/or repressors and co-regulators in a coordinated fashion. An attractive biological feature of USF and Sp1 proteins is their abilities to mediate a wide range of transcriptional activities via interactions with additional factors or cofactors [17,18]. While USFs and Sp1 proteins have been traditionally described as transcriptional activators, studies have suggested that in specific instances [19,20], they may act as a negative regulators by interacting with transcription factors which bound simultaneously to the same site or at other sites, suggesting a dual mode of regulation of genes by these factors. USF functionality including its tissue-specificity may also be regulated by an interplay between its various isoforms [21–23]. North et al. have reported the possibility that USF2b homodimer may be a transrepressor while USF1/USF2b heterodimer could activate the same gene [24]. Moreover, it has been shown that binding of USF or Sp1 to the E-box or GC-rich motifs, respectively, induces an acute bend in the DNA [25]. So, such bending could probably facilitate interactions with other proteins like those already involved in the CFTR
expression regulation such as NF-jB [26], C/EBP [27] or the repressor YY1 [8]. As genes are typically regulated by multiple transcription factors, each with its associated set of coactivators and repressors, which respond to different signaling cascades, it is evident that major advances in the characterization of transcription factors that influence genes transcription will enlighten on the mechanisms involved in the CFTR gene expression control. Acknowledgments This work was supported in part by the Centre Hospitalier Universitaire of Montpellier and the association Vaincre La Mucoviscidose (VLM, Grant FC0527). References [1] S. Carles, M. Desgeorges, A. Goldman, R. Thiart, C. Guittard, C.A. Kitazos, T.J. de Ravel, A.T. Westwood, M. Claustres, M. Ramsay, First report of CFTR mutations in black cystic fibrosis patients of southern African origin, J. Med. Genet. 33 (1996) 802–804. [2] M.C. Romey, C. Guittard, S. Carles, J. Demaille, M. Ramsay, M. Claustres, First putative sequence alterations in the minimal CFTR promoter region, J. Med. Genet. 36 (1999) 263–264. [3] M.C. Romey, N. Pallares-Ruiz, A. Mange, C. Mettling, R. Peytavi, J. Demaille, M. Claustres, A naturally occurring sequence variation that creates a YY1 element is associated with increased cystic fibrosis transmembrane conductance regulator gene expression, J. Biol. Chem. 275 (2000) 3561–3567. [4] R. Rowntree, A. Harris, DNA polymorphisms in potential regulatory elements of the CFTR gene alter transcription factor binding, Hum. Genet. 111 (2002) 66–74. [5] J. Wittwer, J. Marti-Jaun, M. Hersberger, Functional polymorphism in ALOX15 results in increased allele-specific transcription in macrophages through binding of the transcription factor SPI1, Hum. Mutat. 27 (2006) 78–87. [6] T. Bienvenu, V. Lacronique, M. Raymondjean, C. Cazeneuve, D. Hubert, J.C. Kaplan, C. Beldjord, Three novel sequence variations in the 5 0 upstream region of the cystic fibrosis transmembrane conductance regulator (CFTR) gene: two polymorphisms and one putative molecular defect, Hum. Genet. 95 (1995) 698–702. [7] M.C. Romey, C. Guittard, J.P. Chazalette, P. Frossard, K.P. Dawson, M.A. Patton, T. Casals, T. Bazarbachi, E. Girodon, G. Rault, D. Bozon, F. Seguret, J. Demaille, M. Claustres, Complex allele [-102T>A+S549R(T>G)] is associated with milder forms of cystic fibrosis than allele S549R(T>G) alone, Hum. Genet. 105 (1999) 145–150. [8] C. Rene, M. Taulan, F. Iral, J. Doudement, A. L’Honore, C. Gerbon, J. Demaille, M. Claustres, M.C. Romey, Binding of serum response factor to cystic fibrosis transmembrane conductance regulator CArGlike elements, as a new potential CFTR transcriptional regulation pathway, Nucleic Acids Res. 33 (2005) 5271–5290. [9] J.T. Kadonaga, K.R. Carner, F.R. Masiarz, R. Tjian, Isolation of cDNA encoding transcription factor Sp1 and functional analysis of the DNA binding domain, Cell 51 (1987) 1079–1090. [10] K. Yoshimura, H. Nakamura, B.C. Trapnell, W. Dalemans, A. Pavirani, J.P. Lecocq, R.G. Crystal, The cystic fibrosis gene has a ‘‘housekeeping’’-type promoter and is expressed at low levels in cells of epithelial origin, J. Biol. Chem. 266 (1991) 9140–9144. [11] R. Kollmar, K.A. Sukow, S.K. Sponagle, P.J. Farnham, Start site selection at the TATA-less carbamoyl-phosphate synthase (glutamine-hydrolyzing)/aspartate carbamoyltransferase/dihydroorotase promoter, J. Biol. Chem. 269 (1994) 2252–2257. [12] D.E. Zhang, C.J. Hetherington, S. Tan, S.E. Dziennis, D.A. Gonzalez, H.M. Chen, D.G. Tenen, Sp1 is a critical factor for the
M. Taulan et al. / Biochemical and Biophysical Research Communications 361 (2007) 775–781
[13]
[14]
[15]
[16]
[17]
[18]
monocytic specific expression of human CD14, J. Biol. Chem. 269 (15) (1994) 11425–11434. A.L. Roy, M. Meisterernst, P. Pognonec, R.G. Roeder, Cooperative interaction of an initiator-binding transcription initiation factor and the helix-loop-helix activator USF, Nature 354 (1991) 245–248. J. Liu, M. Antaya, D. Boerboom, J.G. Lussier, D.W. Silversides, J. Sirois, The delayed activation of the prostaglandin G/H synthase-2 promoter in bovine granulosa cells is associated with down-regulation of truncated upstream stimulatory factor-2, J. Biol. Chem 274 (1999) 35037–35045. Y. Ge, T.L. Jensen, L.H. Matherly, J.W. Taub, Physical and functional interactions between USF and Sp1 proteins regulate human deoxycytidine kinase promoter activity, J. Biol. Chem. 278 (2003) 49901–49910. E. Meccia, L. Bottero, F. Felicetti, C. Peschle, M.P. Colombo, A. Care, HOXB7 expression is regulated by the transcription factors NF-Y, YY1, Sp1 and USF-1, Biochim. Biophys. Acta 1626 (2003) 1–9. M.A. Renault, S. Jalvy, M. Potier, I. Belloc, E. Genot, L.V. Dekker, C. Desgranges, A.P. Gadeau, UTP induces osteopontin expression through a coordinate action of NFkappaB, activator protein-1, and upstream stimulatory factor in arterial smooth muscle cells, J. Biol. Chem. 280 (2005) 2708–2713. M.K. Bennett, T.T. Ngo, J.N. Athanikar, J.M. Rosenfeld, T.F. Osborne, Co-stimulation of promoter for low density lipoprotein receptor gene by sterol regulatory element-binding protein and Sp1 is specifically disrupted by the yin yang 1 protein, J. Biol. Chem. 274 (1999) 13025–13032.
781
[19] A.K. Ghosh, P.K. Datta, S.T. Jacob, The dual role of helix-loop– helix-zipper protein USF in ribosomal RNA gene transcription in vivo, Oncogene 14 (1997) 589–594. [20] K.M. Kelley, H. Wang, M. Ratnam, Dual regulation of ets-activated gene expression by SP1, Gene 307 (2003) 87–97. [21] M.L. Moore, E.A. Park, J.B. McMillin, Upstream stimulatory factor represses the induction of carnitine palmitoyltransferase-Ibeta expression by PGC-1, J. Biol. Chem. 278 (2003) 17263–17268. [22] B. Viollet, A.M. Lefrancois-Martinez, A. Henrion, A. Kahn, M. Raymondjean, A. Martinez, Immunochemical characterization and transacting properties of upstream stimulatory factor isoforms, J. Biol. Chem. 271 (1996) 1405–1415. [23] S. Yan, B.F. Sloane, Isolation of a novel USF2 isoform: repressor of cathepsin B expression, Gene 337 (2004) 199–206. [24] S. North, X. Espanel, F. Bantignies, B. Viollet, V. Vallet, P. Jalinot, G. Brun, G. Gillet, Regulation of cdc2 gene expression by the upstream stimulatory factors (USFs), Oncogene 18 (1999) 1945–1955. [25] D. Kumari, A. Gabrielian, D. Wheeler, K. Usdin, The roles of Sp1, Sp3, USF1/USF2 and NRF-1 in the regulation and three-dimensional structure of the Fragile X mental retardation gene promoter, Biochem. J. 386 (2005) 297–303. [26] F. Brouillard, M. Bouthier, T. Leclerc, A. Clement, M. BaudouinLegros, A. Edelman, NF-kappa B mediates up-regulation of CFTR gene expression in Calu-3 cells by interleukin-1beta, J. Biol. Chem. 276 (2001) 9486–9491. [27] N. Pittman, G. Shue, N.S. LeLeiko, M.J. Walsh, Transcription of cystic fibrosis transmembrane conductance regulator requires a CCAAT-like element for both basal and cAMP-mediated regulation, J. Biol. Chem. 270 (1995) 28848–28857.