Tài liệu Báo cáo khoa học: Fra-1 targets the AP-1 site/2G single nucleotide polymorphism (ETS site) in the MMP-1 promoter docx

Fra-1 targets the AP-1 site/2G single nucleotide polymorphism (ETS site) in the MMP-1 promoter Grant B. Tower 1, *, Charles I. Coon 2 , Karine Belguise 3 , Dany Chalbos 3 and Constance E. Brinckerhoff 1,2 Department of 1 Biochemistry and 2 Medicine at Dartmouth Medical School, Hanover, NH, USA; 3 Institut National de la Sante ´ et de la Recherche ´ Me ´ dicale, Endocrinologie Moleculaire et Cellulaire des Cancers, Montpellier, France The matrix metalloproteinase (MMP) family degrades the extracellular matrix. One member of this family, MMP-1, initiates the breakdown of interstitial collagens. The expression of MMP-1 is controlled by the mitogen activated protein kinase (MAPK) pathway(s) via the activity of acti- vator protein-1 (AP-1) and polyoma enhancing activity-3/ E26 virus (PEA3/ETS) transcription factors through con- sensus binding sites present in the promoter. Another ETS site in the MMP-1 promoter is created at )1607 bp by a single nucleotide polymorphism (SNP), which contains two guanines (5¢-GGAT-3¢; Ô2G SNPÕ), rather one guanine (5¢-GAT-3¢; Ô1G SNPÕ), adjacent to an AP-1 binding site at )1602 bp. The 2G SNP displays greater transcriptional activity than the 1G SNP, and AP-1 and Ets families of transcription factors cooperate to increase transcription. The 2G SNP has been linked to the incidence and the progression of several cancers and is also associated with non-neoplastic diseases; although the underlying mechanism(s) has yet to be elucidated. In this study we demonstrate that the expression of Fos-like region antigen (Fra-1), an AP-1 transcription factor component that also correlates strongly with neo- plastic disease, is necessary for MMP-1 transcription in A2058 melanoma cells. The inhibition of Fra-1 expression preferentially downregulates transcription from the MMP-1 promoter DNA containing the 2G SNP, compared to DNA containing the 1G SNP. This study provides evidence that, in cooperation with the 2G DNA polymorphism, the AP-1 family member, Fra-1, contributes to the high constitutive expression of MMP-1 in melanoma cells. Keywords: antisense; MAPK; matrix metalloproteinase; melanoma and metastasis; signal transduction. The extracellular matrix (ECM) acts as a structural support network within tissues and as a barrier to cell migration. Additionally, the destruction of the ECM can regulate disease progression and severity in a variety of pathological situations [1–3]. The matrix metalloproteinase (MMP) family is responsible for degradation of the ECM, and the MMP subfamily of collagenases specifically cleaves the interstitial collagens (types I, II and III). MMP-1 is a collagenase that is expressed at very low levels in normal physiological situations, and expression can be increased transiently when remodeling of the ECM is required (e.g. wound healing, uterine resorption and development). However, in many pathological states, MMPs are expressed at high levels. This is due to continuous induction by external stimuli, as seen in rheumatoid arthritis and athlerosclerosis, or to constitutive expression because of an activating mutation in a regulatory pathway [4–7]. High constitutive levels of MMP-1 correlate with a poor prog- nosis in many types of cancer [8–10]. The mitogen activated protein kinase (MAPK) signaling pathway is composed of four main cascades, three of which are known to control the expression of MMP-1. They are the extracellular response kinase (ERK 1/2), the p38 and the Jun N-terminal kinase, and they activate factors that target multiple activator protein-1 (AP-1) and polyoma enhancing activity-3/E26 virus (PEA3/ETS) factor binding sites within the MMP-1 promoter [11–16]. These families of transcrip- tion factors act synergistically when AP-1 and ETS consensus sites are located in proximity to one another [17,18], such as the AP-1/ETS site located at )73/)88 bp in the MMP-1 promoter [19]. We have described a single nucleotide polymorphism (SNP) at )1607 bp that generates an additional ETS site when two guanines (5¢-GGAT-3¢; Ô2G SNPÕ) are present instead of one guanine (5¢-GAT-3¢; Ô1G SNPÕ) (Fig. 1), and this SNP results in increased transcription from the 2G-containing promoter relative to the 1G-containing promoter [20,21]. We have verified this ETS site by demonstrating that it binds recombinant Ets-1 protein, and that the adjacent AP-1 site at )1602 bp binds recom- binant c-Jun. Further, we carried out gel shift analyses with nuclear extracts from A2058 melanoma cells and Correspondence to C. E. Brinckerhoff, Department of Biochemistry, 505 Vail Building, Hanover, NH 03755, USA. Fax: + 1 603 650 1128, Tel.: + 1 603 650 1609, E-mail: brinckerhoff@dartmouth.edu Abbreviations:MMP,matrixmetalloproteinase;MAPK,mitogen activated protein kinase; AP-1, activating protein; PEA3/ETS, poly- oma enhancing activity-3/E26 virus; FGF, fibroblast growth factor; SNP, single nucleotide polymorphism; ERK, extracellular response kinase; Fra, Fos-related antigen; Elf, ETS-like factor; DMEM, Dulbecco’s Modified Eagles Medium; CMV, cytomegalovirus; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. *Present address: Schering-Plough, 2015 Galloping Hill Rd., K-15, D-129C Kenilworth, NJ 07033, USA. (Received 11 June 2003, revised 4 September 2003, accepted 8 September 2003) Eur. J. Biochem. 270, 4216–4225 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03821.x competition experiments with AP-1 and PEA3 oligonucleo- tides to show specificity of binding [20]. Several studies correlate the presence of the 2G SNP with invasive cancers, including melanoma [22,23], and colorectal [24–26], lung [27], endometrial [28] and ovarian cancers [29,30]. The 2G polymorphism is also associated with non-neoplastic diseases, such as premature rupture of fetal membranes [31], increased risk of internal carotid artery stenosis [32], rate of decline in lung function [33], lower bone mineral density [34] and increased severity of chronic periodontitis [35]. These studies support the hypothesis that the 2G polymorphism may increase expression from the MMP-1 promoter and influence the degree of pathology resulting from dysregulated matrix degradation. However, the mechanisms mediating the increase in MMP-1 from the 2G promoter have not been elucidated. We have previously described one mechanism responsible for the increase in expression from the 2G-containing promoter in A2058 melanoma cells, which constitutively express copious amounts of MMP-1 [15]. In this system, we determined by means of transient transfec- tions with MMP-1 reporter constructs, that the MAPK pathway, ERK 1/2, targets the 2G SNP at )1607 bp and theadjacentAP-1siteat)1602 bp. Activation and/or production of transcription factors binding to these sites are responsible for the elevated expression of MMP-1 from the 2G allele compared to that from the 1G allele [15]. However, it is likely that other mechanisms exist, which may also influence this response. Our previous study showed that the chemical inhibitors PD098059 and U0126 blocked the phosphorylation of ERK1/2 and reduced expression of MMP-1 mRNA [15]. We also determined that continuous synthesis of new proteins is necessary for MMP-1 expression [15] and that the levels of these proteins are decreased when ERK 1/2 signaling is disrupted by PD098059. Since the expression and the mRNA stability of the AP-1 family member Fra-1 is dependent on ERK activity [36], we hypothesized that Fra-1 may be produced by A2058 cells, that it may be necessary for MMP-1 gene expression, and that it may be reduced by PD098059. Fra-1 has been implicated in the regulation of MMP-1 [37–40], MMP-13 [41,42] and the urokinase plasminogen activator receptor [43], a receptor in the fibrinolytic cascade that activates MMPs. Fra-1 is necessary for the induction of MMP-1 by mitogens [44,45] and it binds to the AP-1 responsive element in the proximal promoter [39,46]. As the ERK pathway often regulates induction of the mitogen, fibroblast growth factor (FGF), which induces expression of MMP-1, Fra-1 may contribute to the transcriptional regulation of both FGF and MMP-1. Indeed, FGF responsiveness in the MMP-1 promoter requires both the AP-1 and ETS sites within a 63-bp region, from )123 bp to )61 bp, to which Fra-1 binds [39]. These studies have suggested that Fra-1 interacts with Ets transcription factors to induce MMP-1 gene expression. The present study extends these findings to include a role for Fra-1 in augmenting MMP-1 transcription from the AP-1 site at )1602 bp and the 2G SNP (ETS site) at )1607 bp. Materials and methods Cell culture Stock cultures of A2058 melanoma cells were grown in 150-mm culture plates in Dulbecco’s Modified Eagles Medium (DMEM) containing 10% fetal bovine serum, L -glutamine, and penicillin/streptomycin (37 °C, 5% CO 2 ), and passaged at confluency [20]. For most experiments, confluent cultures of cells were washed twice with Hanks balanced salt solution to remove traces of serum and cultured in serum-free DMEM and 0.2% lactalbumin hydrolysate. Cells were treated with increasing concentra- tions of PD098059 (Calbiochem), an inhibitor of the ERK 1/2 pathway, solubilized in dimethyl sulfoxide. Electrophoresis mobility shift assay (EMSA) Nuclear extracts were prepared from untreated A2058 cells, or from cells treated with PD098059 [20]. Oligonucleotide probes containing 35 or 36 bp from regions of the MMP-l promoter encompassing the lG/2G site at )1607 bp and the AP-1 site at )1602 bp, from )1621 bp to )1586 bp [20], were end-labeled with [ 32 P]dATP[cP] by T4 Kinase (Gibco/ BRL) in 1 · polynucleotide kinase buffer (Roche) and 2 · 10 6 c.p.m. were incubated with 5 lg of nuclear extract for 20 min at 4 °C. This amount was needed for probe excess with the 2G oligonucleotide. Lesser amounts showed depletion of free probe at the bottom of the gel. Binding reactions were separated on 7% Tris/glycine polyacrylamide gels and visualized by autoradiography. Transient transfection and luciferase assay One microgram of 4372 bp of the MMP-l promoter DNA containing either 1G or 2Gs at )1607 bp, linked to the luciferase reporter, was transiently transfected into cells using Geneporter (Gene Therapy Systems). To further elucidate the role of AP-1 sites in the MMP-1 promoter, reporter constructs containing mutations in selected AP-1 sites were used [15,20]. To alter the expression levels of Fra-1, expression constructs containing the Fra-1 cDNA in the sense and antisense orientation driven by the Fig. 1. Diagram of the MMP-1 promoter region spanning the region from )1585 bp to )1620 bp. The 1G SNP DNA contains an AP-1 consensus at )1602 bp, and the sequence 5¢-GAA-3¢ at )1607 bp. The 2G SNP DNA has the AP-1 site at )1602 bp and the sequence 5¢-GGAA-3¢ at )1607 bp, conferring a consensus ETS site. Ó FEBS 2003 Fra-1 and the 2G SNP (Eur. J. Biochem. 270) 4217 cytomegalovirus (CMV) minimal promoter in a pCl vector [47] as well as empty vector to keep the DNA concentration constant, were used. Twenty-four hours post-transfection, cells were washed and incubated with serum free medium, with or without 5 l M PD098059 for 24 h and cell lysates were then assayed for luciferase activity in a luminometer. All transfections were carried out in triplicate. Disparities in relative light unit (RLU) values among different experi- ments are due to the use of different luminometers. Northern blot analyses Total RNA (10 lg) was harvested from cells with Trizol (Gibco) and separated on a 5% formaldehyde, 1% agarose gel, transferred to a GeneScreen membrane (NEN Life Sciences) and probed with a 32 P-labelled cDNA for MMP-1 [47], c-Fos (a generous gift from I. Verma, the Salk Institute, La Jolla, CA, USA) and Fra-1. The Fra-1 probe was created by PCR, using primers to amplify a region unique to the transcript. The forward primer for Fra-1 was 5¢-TCTGGGCTGCAGCGAGAGATTGAGGAG-3¢ and the reverse primer was 5¢-GGAGGAGACATTGGC TAGGGTGGCATC-3¢, giving a product of 435 bp. PCR reactions were separated on a 1.5% agarose gel by electrophoresis; the band of the correct size was excised and purified. The purified PCR fragment was labeled with [ 32 P]dCTP[aP] for probing. Levels of glyceraldehyde- 3-phosphate dehydrogenase (GAPDH) mRNA probed with a 32 P-labelled cDNA controlled for loading. Western blot analysis Whole cell lysates were harvested by mechanical scraping and collected with 2 · SDS buffer [15]. Proteins were separated on 10% Tris/HCl Ready Gels (Bio-Rad) and transferred to Immobilon TM -P poly(vinylidene difluoride) membranes (Millipore). All membranes were blocked in 5% dry milk (Carnation) at room temperature for 1 h. Blots were probed with antibodies against Fra-1, JunD, c-Fos, Ets-1, Ets-2, and ETS-like factor (Elf-1; Santa Cruz, Inc., Santa Cruz, CA, USA) overnight at 4 °C, as suggested by manufacturer. Blots were then incubated with an anti-rabbit horseradish peroxidase-linked secondary antibody (Cell Signaling, Beverly, MA, USA) at a dilution of 1 : 2000 at room temperature for 1 h. Bands were visualized by electro- chemiluminesence. The membranes were stripped as des- cribed in [48] and reprobed with an antibody against actin (Oncogene, Boston, MA, USA) for 1 h at room tempera- ture. Blots were subsequently probed with an anti-goat IgM secondary antibody (Oncogene) at a dilution of 1 : 2000 for 1 h at room temperature. Bands were visualized by electro- chemiluminesence. Results Increasing concentrations of the ERK specific inhibitor, PD098059, lead to a decrease in protein binding to the 2G SNP. Our earlier studies demonstrated that the 2G SNP functioned as a bone fide ETS site, as measured by competition experiments and by the specific binding of recombinant Ets-1 protein [20]. Further, treatment of A2058 melanoma cells with the chemical inhibitor PD098059 blocked the phosphorylation of ERK 1/2 and reduced MMP-1 mRNA [15]. This reduction in MMP-1 mRNA occurred at the level of transcription, was greater in the 2G SNP-containing promoter than in the 1G SNP-containing promoter, and required the AP-1 site at )1602 bp [15]. Therefore, we hypothesized that this inhibition may be a consequence of reduced binding of nuclear proteins to the AP-1 site and SNP at )1602 bp and )1607 bp, respectively. We tested this hypothesis with an EMSA, using a 35/36 bp nucleotide probe containing either 1G or 2Gs and nuclear extracts from untreated cells or from cells treated with PD098059 for 24 h. Figure 2 illustrates that extracts from untreated cells bind to both the 1G and 2G probes. In agreement with previous data [20,49], the binding to the 2G probe is greater than to the 1G probe, with a 2G specific band, indicated by a star. Treating cells with increasing concentrations of PD098059 results in a decrease in the intensity and/or a complete disappearance of some bands, indicated by arrows. In contrast, bands that are more intense in the lanes containing the 1G probe appear unchanged in the presence of the inhibitor (–). Thus, a decrease in the expression of MMP-1 in the presence of PD098059 may be due, at least in part, to a decrease in the binding of transcription factors to the MMP-1 promoter Fig. 2. Effects of the MEK/ERK inhibitor PD098059 on transcription factor binding. EMSA of transcription factors binding to [ 32 P]dATP[cP] end-labeled 35- and 36-mer, 1G and 2G (respectively) oligonucleotides in the presence of increasing concentrations of PD098059 for 24 h. Lanes 1 and 2, free probe alone; lanes 3 and 4, extract from cells incubated in serum-free media alone; lanes 5–8, extract from cells incubated in the presence of 5 l M (5 and 6) and 10 l M (7 and 8) PD098059. *, 2G specific band; fi , inhibitable activities that bind 2G to a greater extent; –, bands with greater affinity to 1G and not inhibited by PD098059. EMSA was performed four times for reproducibility. 4218 G. B. Tower et al. (Eur. J. Biochem. 270) Ó FEBS 2003 region containing the AP-1 site at )1602 bp and the 2G SNP at )1607 bp. Fra-1 protein is decreased in response to PD098059 Previously, we demonstrated that recovery of MMP-1 mRNA expression after inhibition by PD098059 in A2058 melanoma cells required de novo protein synthesis [15]. This finding, together with the data in Fig. 2, led us to hypothesize that inhibition of the MEK/ERK pathway by PD098059 may decrease the level of a transcription factor(s) required for MMP-1 expression and may explain the decrease in protein binding to the AP-1/2G site. Therefore, to determine which factors in the AP-1 or Ets family might be downregulated, we compared protein levels of AP-1/Ets family members in cells left untreated or treated for 24 h with increasing concentrations of PD098059. Western blot analysis showed that in A2058 cells, c-Fos, Fra-1, and JunD of the AP-1 family and Ets-1, Ets-2 and Elf-1 of the Ets family are expressed constitutively (Fig. 3A, and data not shown). We could not detect Fra-2. Fra-1 levels were downregulated in response to PD098059 (Fig. 3A). The two bands in Fig. 3A represent the phosphorylated protein (upper band) and the nonphosphorylated form (lower band) of Fra-1. Both forms of the protein are decreased by PD098059 treatment, suggesting that levels of Fra-1 protein and its subsequent phosphorylation are mediated by the ERK pathway. We found that c-Fos was also downregulated by PD098059, but its level of expres- sion was very low (data not shown). Thus, Fra-1 is the principal Fos family member expressed in these cells. Further, while c-Fos expression can be regulated by many pathways [50,51], Fra-1 mRNA synthesis is solely regu- lated by the ERK 1/2 pathway [36,52–55]. Since MMP-1 expression in these cells is unaffected by other signal transduction inhibitors [15], it is likely that Fra-1 is the dominant Fos family member involved in constitutive MMP-1 expression in these cells. Therefore, we have focused on the possible role of Fra-1 in the transcription of MMP-1. Fra-1 mRNA is reduced in response to PD098059 prior to inhibition of MMP-1 gene expression To ascertain whether there is a correlation between Fra-1 and MMP-1 expression, dose–response and time-courses comparing the inhibition of Fra-1 and MMP-1 mRNAs were performed. Fra-1 has two mRNA transcripts, with the 1.7-kb transcript being dominantly expressed compared to the 3.3-kb transcript, which results from an extended 3¢ untranslated region [56], presumably due to an alternative polyadenylation site. Since both transcripts are presumed to encode the Fra-1 protein [56], whether the reduction of one transcript precedes the other or is more completely down- regulated may be unimportant. The dose–response for inhibiting Fra-1 and MMP-1 gene expression was similar, with complete inhibition of MMP-1 mRNA and the 3.3-kb transcript of Fra-1 at 5–10 l M PD098059 and substantial inhibition of the 1.7-kb Fra-1 transcript with 10 l M PD098059 (Fig. 3B). For the time-course experiment, A2058 cells were either left untreated or treated with PD098059 (5 l M ). RNA was harvested immediately after addition of PD098059 (time zero), and at subsequent time points for measurement of Fra-1 and MMP-1 mRNA. The level of the 3.3-kb transcript of Fra-1 began to decrease within 1 h of treatment, and was undetectable by 6 h. Reduction of the 1.7-kb transcript lagged behind the 3.3-kb transcript, beginning at 2 h and reaching maximal inhibition by 4 h. This reduction of the Fra-1 transcript preceded that of MMP-1 mRNA, which began at 4 h and was nearly complete by 10 h (Fig. 4). The increase in MMP-1 mRNA levels in untreated cultures beginning at 4 h has been observed previously in response to removal of serum- containing medium [57]. This increase in MMP-1 mRNA may be a stress response or a consequence of removing transforming growth factor-b, which is present in serum and can repress the expression of MMP-1 [58]. These findings indicate that Fra-1 is downregulated in response to PD098059 prior to MMP-1, and suggest that Fra-1 may be involved in regulating MMP-1 expression in these cells. Fig. 3. Inhibition of Fra-1 protein and mRNA levels by PD098059. (A) Fra-1 protein levels from confluent cultures of A2058 cells incubated in serum-free media with or without increasing concentrations of PD098059. Whole cell extracts were separated by SDS/PAGE, trans- ferredtopoly(vinylidenedifluoride)membranesandprobedwithFra-1 antibody. The blot was stripped as described [48] and reprobed with an antibody against a-actin for 1 h. (B) Dose–response of PD098059 on Fra-1 and MMP-1 mRNA levels in A2058 cells. GAPDH was used as a loading control. Dose–response experiments were performed at least three times for reproducibility. Ó FEBS 2003 Fra-1 and the 2G SNP (Eur. J. Biochem. 270) 4219 Fra-1 expression drives transcription from the MMP-1 promoter in a dose-responsive manner Evidence thus far has indirectly implicated Fra-1 as one factor contributing to constitutive expression of MMP-1 in A2058 melanoma cells: Fra-1 is expressed in these cells, and its pattern of repression in response to PD098059 is coordinate with that of MMP-1. To more directly link Fra-1 to MMP-1 expression, we used a fra-1 sense expression construct under control of the CMV promoter, which has been described previously [47]. Cells were transiently cotransfected with MMP-1 promoter DNA containing either the 1G or 2G SNP, driving luciferase, along with increasing amounts of the Fra-1 expression construct. Expression from the 2G SNP was initially 2.7-fold higher than the 1G SNP, and increasing quantities of Fra-1 preferentially augmented transcription from this promoter (Fig. 5A). With increasing amounts of Fra-1, expression of the 1G promoter construct increased 2.9-, 5.2- and 8.8-fold over the 1G control construct, compared to increases of 4.8-, 10.4- and 12.9-fold for the 2G construct. For each concentration of Fra-1, the increase in expression from the 2G SNP was approximately twofold greater than the increase in expression from the 1G SNP. Therefore, Fra-1 can drive transcription from both the 1G and 2G-containing promoters but has the more dramatic effect on the 2G allele. Fra-1 antisense reduces MMP-1 transcription in A2058 cells To further address the role of Fra-1 in MMP-1 expres- sion, we used an antisense construct. A2058 cells were cotransfected with the 1G or 2G MMP-1 promoter constructs and increasing doses of a construct expressing Fra-1 antisense cDNA, which has been shown to lower AP-1 activity in a dose-dependent manner in MCF-7 and MDA-231 breast cancer cells [47]. We reasoned therefore that it should antagonize production of endogenous Fra-1. Unfortunately, however, because the transfection effi- ciency was too low (< 10%), we could not measure changes in the level of Fra-1 protein. Nonetheless, Fig. 5B shows that expression of both the 1G and 2G-containing constructs was inhibited by fra-1 antisense. Repression of luciferase activity from the 1G SNP was  50% at all doses of Fra-1 antisense. In contrast, reporter expression from the 2G SNP promoter was inhibited in a dose- dependent fashion (Fig. 5B), with maximal inhibition at  75%. These results are similar to those seen with PD098059, where the 1G allele was inhibited by 50% and the 2G allele by  50–70% [15]. Thus, the effects of decreasing Fra-1 expression mimic the effects of inhibition by PD098059, and the reduction in Fra-1 expression by the addition of 3 lg of antisense construct abolished the difference in expression between the 1G and 2G alleles. Therefore, in these melanoma cells, Fra-1 expression is required for the increase in transcription from the 2G SNP promoter over the 1G SNP promoter. The AP-1 site at )1602 bp is necessary for modulation of MMP-1 expression from the 2G SNP promoter To identify the sequences in the MMP-1 promoter that are targeted by Fra-1, 1G/2G MMP-1 constructs with either wild-type or mutated AP-1 sites at )73 bp or )1602 bp were Fig. 4. Downregulation of Fra-1 and MMP-1 mRNA levels prior to MMP-1 by PD098059. A2058 cells were grown until confluent, washed with Hanks balanced salt solution and transferred to serum-free media with or without 5 l M PD098959.RNAwasharvestedat0,0.5,1,2,4,6,8,10and 12 h. RNA blots were probed with Fra-1 and MMP-1 and subsequently, with GAPDH probes as a loading control, as described in Materials and methods. 4220 G. B. Tower et al. (Eur. J. Biochem. 270) Ó FEBS 2003 transfected into A2058 cells. The AP-1 site at )73 bp is important for basal and growth factor induced expression (via ERK activation) [19]. The AP-1 site at )1602 bp is adjacent to the 2G SNP, which is targeted by the ERK 1/2 pathway, and it may also be responsible for growth factor induced expression of MMP-1 [15]. As seen in Fig. 6, expression of luciferase from the MMP-1 reporter constructs with a wild-type AP-1 site at )1602 bp was equivalent to previous reports [15], in that there is greater expression from the wild-type 2G construct than from the 1G. However, as noted previously [15], this difference in expression was absent in the constructs containing a mutated AP-1 site at )1602 bp (Fig. 6). In this experiment, some cells were also transfected with the Fra-1 antisense construct in order to determine whether the absence of Fra-1 can lower gene expression when the AP-1 sites are mutated. We found that expression of Fra-1 antisense abolished the higher level of expression from the 2G MMP-1 reporter construct so that all constructs, wild-type or with a mutated AP-1 site at )1602 bp, show equivalent activity. These findings corroborate our previous results, which show that when A2058 cells were transfected with the same wild-type and mutant constructs and subsequently treated with PD098059, reporter levels were nearly identical [15], thereby suggesting that the ERK inhibitor and Fra-1 antisense may be working in a similar manner. We next examined the effect of mutating the AP-1 site at )73 bp in the MMP-1 promoter (Fig. 6, inset). Com- pared to constructs containing a wild-type site at )73 bp, mutating this site substantially decreased expression ( 90%) from both the 1G and 2G SNP promoters (compare 1G vs. 2G wild-type in Fig. 6 with 1G vs. 2G inset). While the absolute levels of expression from both alleles were reduced when the AP-1 site at )73 bp is mutated, the difference between expression of the 1G vs. 2G alleles remained intact, further demonstrating that the proximal AP-1 site is not involved in the augmented expression from the 2G allele. Co-transfection of con- structs with the mutated AP-1 site at )73 bp and Fra-1 antisense inhibited expression from both the 1G and 2G constructs but downregulated the 2G construct to a greater extent (49% repression vs. 77% repression, respectively). The fact that Fra-1 antisense can inhibit transcription from the 1G SNP construct with the mutant AP-1 site at )73 bp indicates that Fra-1 can induce MMP- 1 through AP-1 sites other than that at )73 bp. We conclude that the )73 bp site, in contrast to the AP-1 site at )1602 bp, is not involved in the difference between the 1G and 2G SNP. Fra-1 rescues the 1G/2G response in cells treated with PD098059 To confirm that the attenuation of Fra-1 by PD098059 results in reduced MMP-1 expression, we attempted to reverse this repression by ectopic expression of Fra-1. A2058 cells were treated with PD098059 and cotransfected with the 1G or the 2G MMP-1 promoter constructs along with increasing amounts of the Fra-1 expression vector. Exo- genous Fra-1, expressed under control of the CMV promoter, would not be subject to the same regulation as endogenous Fra-1 gene, and the high level of expression should allow the exogenous protein to be phosphorylated even in the presence of 5 l M PD098059. Although higher concentrations of the drug might block the phosphorylation of all exogenous Fra-1, they may also introduce toxicity and nonspecific effects. Thus, as expected, in cells transfected with the empty vector, there is about a threefold increase in expression of the 2G allele compared to the 1G allele, and treatment with 5 l M PD098059 abolished this difference (Fig. 7). We reasoned that any increase in expression from the 2G allele in these treated cells cotransfected with Fra-1 would be due to rescue by Fra-1. Indeed, compared to the empty vector, we found that cotransfection of increasing amounts of the Fra-1 expression plasmid augments the level Fig. 5. Effect of Fra-1 expression on MMP-1 transcription. (A) Fra-1 sense expression. A2058 cells were transfected with 1 lgof1G/2G MMP-1 promoter constructs driving luciferase and the pCl empty vector or with increasing concentrations of the Fra-1 overexpressing vector, as described in Materials and methods. For the empty vector, the difference in expression between the 1G and 2G constructs is indicated above the line spanning both the 1G and 2G bars (mean ± SD). With increasing amounts of Fra-1 expression, (*) indicates that the value is statistically significant (P £ 0.02), compared to the 1G or 2G value in cells transfected with the empty vector. (B) Fra-1 antisense expression. A2058 cells were transfected with 1 lgof 1G/2GMMP-1promoterconstructsdrivingaluciferasereporterand the pCl empty vector, or with increasing quantities of pCl fra-1 anti- sense (AS) vector. The percent inhibition compared to the empty vector control is indicated above the bar. The (*) indicates that the value is statistically significant (P £ 0.05) compared to the previous respective 1G or 2G data point. Transfection experiments were per- formed three times for reproducibility. Ó FEBS 2003 Fra-1 and the 2G SNP (Eur. J. Biochem. 270) 4221 of transcription from both the 1G and 2G promoter constructs. The increase is greater in cells transfected with the 2G construct compared to the 1G construct, increasing from 1.06-fold for the empty vector to 2.1-fold for cells transfected with 3.0 lg Fra-1. Increasing the amount of the Fra-1 not only restored the difference between the 1G and 2G alleles, but also partially abrogated the repression of the 2G allele by PD098059. Therefore, expression of Fra-1 is reduced by PD098059 and this protein is required for increased expression from the 2G allele over the 1G allele. Discussion The SNP in the MMP-1 promoter at )1607 bp results in an ETS binding site (the 2G SNP) and is correlated with increased transcription [20]. Subsequent investigations that demonstrated an association between the 2G allele and several cancers have generated interest as to the identity of factors that activate this site. In this study, we identify Fra-1 as one of the potential transcription factors driving MMP-1 transcription and enhancing expression through the AP-1 site at )1602 bp, but only in combination with the 2G SNP (ETS site) at )1607 bp. In the presence of increasing concentrations of the ERK 1/2 inhibitor, PD098059, the dose–responses of Fra-1 and MMP-1 expression are similar (Fig. 3B). The time-course of inhibi- tion of Fra-1 and MMP-1 mRNAs in response to PD098059 treatment (Fig. 4) further supports a probable role of Fra-1 in MMP-1 transcription. These correlations between Fra-1 and MMP-1 expression prompted us to find direct evidence implicating Fra-1 in constitutive MMP-1 production. AP-1 factors bind DNA as Jun/Jun homodimers or Jun/ Fos heterodimers, which have affinity for AP-1 responsive elements. A2058 melanoma cells express both JunD (data not shown) and Fra-1 constitutively and these proteins may form a transactivating complex with other transcription factor families, such as members of the Ets family [18]. Furthermore, Fra-1 is a potent transactivator in the presence of JunD, where either factor may be inhibitory without the other [42,59]. In addition to the reduction in phosphorylation and expression of Fra-1, inhibition of the ERK pathway also decreases the stability of Fra-1 protein and its ability to bind to DNA [36]. With the exception of Fra-2, this contrasts with other AP-1 members, which bind Fig. 6. Mutational analysis of AP-1 sites at )1602 bp and )73 bp. A2058 cells were cotransfected with 1G/2G MMP-1 reporter constructs containing a wild-type AP-1 site or a mutated AP-1 site at either )1602 bp or (inset) )73 bp. Additionally all cells were transfected with either a Fra- 1 expression, Fra-1 antisense (AS) or control construct. Data points are the mean ± SD and the statistical significance of the 1G compared to the 2G is **P ¼ 0.002. The transfection experiment was performed three times for consistency. Fig. 7. Partial rescue of MMP-1 transcription in cells treated with PD098059. A2058 cells were transfected with 1 lg of 1G or 2G MMP-1 promoter constructs driving luciferase and the pCl vector or with increasing quantities of pCl Fra-1 overexpression vector. After 24 h cells were incubated in serum-free media with or without 5 l M PD098059 for an additional 24 h. The fold increase of 2G expression over 1G expression from the 2G allele treated with 5 l M PD098059 compared to the expression from the untreated 1G allele are charted below the graph. Data points represent the mean ± SD. **P £ 0.05 is the statistical significance of expression of the 1G allele compared to the 2G allele. 4222 G. B. Tower et al. (Eur. J. Biochem. 270) Ó FEBS 2003 constitutively to AP-1 consensus sequences in the presence or absence of MAPK signaling [50,60]. Additionally, Fra-1 is the factor that determines the DNA-binding activity of the AP-1 transcription factor complex [61]. Thus, if Fra-1 is binding to the AP-1 site at )1602 bp, PD098059 should reduce binding activity since ERK 1/2 is required for stability/DNA binding [36]. Our EMSA results suggest the presence of a multi-family association of proteins required to enhance synthesis of MMP-1 through the 2G SNP/AP-1 site [20,31,49], of which Fra-1 may be a component. Specific complexes bind to the probe containing the AP-1 site and 2G SNP as seen in Fig. 2, and these complexes are absent in the lanes containing the probe with the 1G SNP. Although increasing concentrations of PD098059 reduce the binding of complexes to the probe containing the 1G SNP, a more pronounced decrease is observed in the binding of com- plexes to the 2G SNP probe. Attempts to supershift complexes with antibodies raised to Fra-1 or other AP-1 factors were unsuccessful (data not shown), potentially due to the inability of the antibodies to access the epitope. Experiments with a Fra-1 antibody from another source (Geneka, Canada) also failed to supershift (data not shown). The downregulation of Fra-1 may decrease the total number of complexes formed, thereby diminishing the cooperative binding to the promoter and, subsequently decreasing MMP-1 production. Transfection with increasing amounts of Fra-1 sense and antisense expression constructs had a greater effect on transcription from the 2G SNP than from the 1G allele (Fig. 5). It has been previously reported that Fra-1 protein binds to the proximal AP-1 site at )73 bp in the MMP-1 promoter [39]. Thus, the increase in expression from the 1G promoter in the presence of increasing amounts of Fra- 1 may be due to the effects of Fra-1 on the proximal AP-1 site. The greater increase in the induction of the 2G promoter may result from Fra-1 acting not only at the proximal site but also at the AP-1 site at )1602 bp adjacent to the 2G SNP. Furthermore, our data implicate Fra-1 in contributing to the increase in MMP-1 expression from the 2G SNP promoter via a cooperative mechanism between theAP-1siteat)1602 bp and the 2G SNP at )1607 bp (Fig. 6). Our previous results show that when A2058 cells were transfected with wild-type and AP-1 mutant at )1602 bp constructs and treated with PD098059, reporter levels were identical from either the 1G or 2G promoter [15]. Thus, expression of Fra-1 antisense and PD098059 [15] have similar effects on transcription of the MMP-1 promoter. The similarities between the wild-type and AP-1 ()1602 bp) mutant constructs in response to Fra-1 anti- sense and ERK 1/2 inhibition provide additional support for the concept that inhibition of Fra-1 is one target of PD098059 [15], with the subsequent inhibition of MMP-1 mRNA. The ectopic expression of Fra-1 was able to moderately rescue transcription from the 2G allele in the presence of PD098059 (Fig. 7), but was unable to fully overcome this repression. This partial rescue is probably because the inhibition of the ERK 1/2 pathway greatly reduces phos- phorylation and stability of downstream targets, including Fra-1 [62] and because Fra-1 needs to be phosphorylated to transactivate transcription. Young et al. [62] determined that the threonine at 231 in Fra-1 is phosphorylated by the ERK pathway and is essential for the transactivation of transcription by Fra-1. We attempted to overcome the need for phosphorylation by the ERK pathway by changing the threonine at 231 to an aspartic acid residue (T231D), which may mimic a phosphothreonine. However, the T231D Fra-1 mutant was unable to transactivate MMP-1 tran- scription to the same levels as wild-type Fra-1 or rescue MMP-1 expression in the presence of PD098059 (data not shown). Two possible reasons why the T231D mutation did not work are that the mutant protein may have been unstable or the aspartic acid residue was unable to mimic the phosphorylation state. Although the A2058 cells express c-Fos, levels of the protein are low (data not shown). However, it is possible that c-Fos may play a minimal role in the residual expression of MMP-1 in the presence of Fra-1 antisense. Fra-1 had previously been implicated in MMP-1 expres- sion through the AP-1/ETS site in the proximal promoter [39,46], but it is the first transcription factor to be identified that leads to the increase in expression from the MMP-1 2G SNP through its cooperation with the adjacent AP-1 site. Fra-1 expression, like the presence of the 2G polymorphism, has been associated with the more aggressive and metastatic forms of breast cancer [63] and melanoma [64], further supporting the possible connection between Fra-1 expres- sion and the 2G polymorphism. Acknowledgements Supported by grants from the NIH, AR-26599, CA-77267, the Department of Defense DAMD17-00-1-0221 (CEB), predoctoral fellowships from the Komen Foundation Austin, Texas (DISS00- 000354), and the Department of Defense DAMD17-01-1-0225 (GBT), Institut National de la Sante ´ et de la recherche Me ´ dicale, the Association pour la Recherche sur le Cancer (grant no. 5825) (DC), the Fondation pour la Recherche Me ´ dicale and the Ligue Nationale Contre le Cancer (KB). References 1. Hashimoto, G., Inoki, I., Fujii, Y., Aoki, T., Ikeda, E. & Okada, Y. (2002) Matrix metalloproteinases cleave connective tissue growth factor and reactivate angiogenic activity of vascular endothelial growth factor 165. J. Biol. Chem. 277, 36288–36295. 2. Bergers, G. & Coussens, L.M. (2000) Extrinsic regulators of epi- thelial tumor progression: metalloproteinases. Curr. Opin. Genet Dev. 10, 120–127. 3. D’Armiento, J., DiColandrea, T., Dalal, S.S., Okada, Y., Huang, M.T., Conney, A.H. & Chada, K. (1995) Collagenase expression in transgenic mouse skin causes hyperkeratosis and acanthosis and increases susceptibility to tumorigenesis, Mol. Cell Biol. 15, 5732– 5739. 4. Borden,P.&Heller,R.A.(1997)Transcriptionalcontrolofmatrix metalloproteinases and the tissue inhibitors of matrix metallo- proteinases. Crit. Rev. Eukaryot. Gene Expr. 7, 159–178. 5. Grant, G.M., Cobb, J.K., Castillo, B. & Klebe, R.J. (1996) Reg- ulation of matrix metalloproteinases following cellular transfor- mation. J. Cell Physiol. 167, 177–183. 6. Mengshol, J.A., Mix, K.S. & Brinckerhoff, C.E. (2002) Matrix metalloproteinases as therapeutic targets in arthritic diseases: bull’s-eye or missing the mark? Arthritis Rheum. 46, 13–20. 7. Vincenti, M.P. & Brinckerhoff, C.E. (2001) The potential of signal transduction inhibitors for the treatment of arthritis: Is it all just JNK? J. Clin. Invest. 108, 181–183. Ó FEBS 2003 Fra-1 and the 2G SNP (Eur. J. Biochem. 270) 4223 8.Murray,G.I.,Duncan,M.E.,O’Neil,P.,Melvin,W.T.& Fothergill, J.E. (1996) Matrix metalloproteinase-1 is associated with poor prognosis in colorectal cancer. Nat. Med. 2, 461–462. 9. Murray, G.I., Duncan, M.E., O’Neil, P., McKay, J.A., Melvin, W.T. & Fothergill, J.E. (1998) Matrix metalloproteinase-1 is associated with poor prognosis in oesophageal cancer. J. Pathol. 185, 256–261. 10. Brinckerhoff, C.E., Rutter, J.L. & Benbow, U. (2000) Interstitial collagenases as markers of tumor progression. Clin. Cancer Res. 6, 4823–4830. 11. Vincenti, M.P., White, L.A., Schroen, D.J., Benbow, U. & Brinckerhoff, C.E. (1996) Regulating expression of the gene for matrix metalloproteinase-1 (collagenase): mechanisms that control enzyme activity, transcription, and mRNA stability. Crit. Rev. Eukaryot. Gene Expr. 6, 391–411. 12. Westermarck, J. & Kahari, V.M. (1999) Regulation of matrix metalloproteinase expression in tumor invasion. FASEB J. 13, 781–792. 13. White, L.A. & Brinckerhoff, C.E. (1995) Two activator protein-1 elements in the matrix metalloproteinase-1 promoter have differ- enteffectsontranscriptionandbindJunD,c-Fos,andFra-2. Matrix Biol. 14, 715–725. 14. White, L.A., Maute, C. & Brinckerhoff, C.E. (1997) ETS sites in the promoters of the matrix metalloproteinases collagenase (MMP-1) and stromelysin (MMP-3) are auxiliary elements that regulate basal and phorbol-induced transcription. Connect. Tissue Res. 36, 321–335. 15. Tower, G.B., Coon, C.C., Benbow, U., Vincenti, M.P. & Brinckerhoff, C.E. (2002) Erk 1/2 differentially regulates the expression from the 1G/2G single nucleotide polymorphism in the MMP-1promoterinmelanomacells.Biochim. Biophys. Acta 1586, 265–274. 16. Mengshol, J.A., Vincenti, M.P., Coon, C.I., Barchowsky, A. & Brinckerhoff, C.E. (2000) Interleukin-1 induction of collagenase 3 (matrix metalloproteinase 13) gene expression in chondrocytes requires p38, c-Jun N-terminal kinase, and nuclear factor kappaB: differential regulation of collagenase 1 and collagenase 3. Arthritis Rheum. 43, 801–811. 17. Buttice, G., Duterque-Coquillaud, M., Basuyaux, J.P., Carrere, S., Kurkinen, M. & Stehelin, D. (1996) Erg, an Ets-family member, differentially regulates human collagenase1 (MMP1) and strome- lysin1 (MMP3) gene expression by physically interacting with the Fos/Jun complex. Oncogene 13, 2297–2306. 18. Wasylyk, B., Wasylyk, C., Flores, P., Begue, A., Leprince, D. & Stehelin, D. (1990) The c-ets proto-oncogenes encode transcrip- tion factors that cooperate with c-Fos and c-Jun for transcrip- tional activation. Nature 346, 191–193. 19. Aho, S., Rouda, S., Kennedy, S.H., Qin, H. & Tan, E.M. (1997) Regulation of human interstitial collagenase (matrix metallopro- teinase-1) promoter activity by fibroblast growth factor. Eur. J. Biochem. 247, 503–510. 20. Rutter, J.L., Mitchell, T.I., Buttice, G., Meyers, J., Gusella, J.F., Ozelius, L.J. & Brinckerhoff, C.E. (1998) A single nucleotide polymorphism in the matrix metalloproteinase-1 promoter creates an Ets binding site and augments transcription. Cancer Res. 58, 5321–5325. 21. Wyatt, C.A., Coon, C.I., Gibson, J.J. & Brinckerhoff, C.E. (2002) potential for the 2G single nucleotide polymorphism in the pro- moter of matrix metalloproteinase to enhance gene expression in normal stromal cells. Cancer Res. 62, 7200–7202. 22. Ye, S., Dhillon, S., Turner, S.J., Bateman, A.C., Theaker, J.M., Pickering, R.M., Day, I. & Howell, W.M. (2001) Invasiveness of cutaneous malignant melanoma is influenced by matrix metallo- proteinase 1 gene polymorphism. Cancer Res. 61, 1296–1298. 23. Noll, W.W., Belloni, D.R., Rutter, J.L., Storm, C.A., Schned, A.R.,Titus-Ernstoff,L.,Ernstoff,M.S.&Brinckerhoff,C.E. (2001) Loss of heterozygosity on chromosome 11q22–23 in mel- anoma is associated with retention of the insertion polymorphism in the matrix metalloproteinase-1 promoter. Am.J.Pathol.158, 691–697. 24. Hinoda, Y., Okayama, N., Takano, N., Fujimura, K., Suehiro, Y., Hamanaka, Y., Hazama, S., Kitamura, Y., Kamatani, N. & Oka, M. (2002) Association of functional polymorphisms of matrix metalloproteinase (MMP) -1 and MMP-3 genes with colo- rectal cancer. Int. J. Cancer. 102, 526–529. 25. Biondi, M.L., Turri, O., Leviti, S., Seminati, R., Cecchini, F., Bernini, M., Ghilardi, G. & Guagnellini, E. (2000) MMP1 and MMP3 polymorphisms in promoter regions and cancer. Clin. Chem. 46, 2023–2024. 26. Ghilardi, G., Biondi, M.L., Mangoni, J., Leviti, S., DeMonti, M., Guagnellini, E. & Scorza, R. (2001) Matrix metalloproteinase-1 promoter polymorphism 1G/2G is correlated with colorectal cancer invasiveness. Clin. Cancer Res. 7, 2344–2346. 27. Zhu, Y., Spitz, M.R., Lei, L., Mills, G.B. & Wu, X. (2001) A single nucleotide polymorphism in the matrix metalloproteinase-1 pro- moter enhances lung cancer susceptibility. Cancer Res. 61, 7825– 7829. 28. Nishioka, Y., Kobayashi, K., Sagae, S., Ishioka, S., Nishikawa, A., Matsushima, M., Kanamori, Y., Minaguchi, T., Nakamura, Y., Tokino, T. & Kudo, R. (2000) A single nucleotide poly- morphism in the matrix metalloproteinase-1 promoter in endometrial carcinomas. Jpn J. Cancer Res. 91, 612–615. 29. Ye, S. (2000) Polymorphism in matrix metalloproteinase gene promoters: implication in regulation of gene expression and susceptibility of various diseases. Matrix Biol. 19, 623– 629. 30. Kanamori, Y., Matsushima, M., Minaguchi, T., Kobayashi, K., Sagae, S., Kudo, R., Terakawa, N. & Nakamura, Y. (1999) Correlation between expression of the matrix metalloproteinase-1 gene in ovarian cancers and an insertion/deletion polymorphism in its promoter region. Cancer Res. 59, 4225–4227. 31. Fujimoto, T., Parry, S., Urbanek, M., Sammel, M., Macones, G., Kuivaniemi, H., Romero, R. & Strauss, J.F. III. (2002) A single nucleotide polymorphism in the matrix metalloproteinase-1 (MMP-1) promoter influences amnion cell MMP-1 expression and risk for preterm premature rupture of the fetal membranes. J. Biol. Chem. 277, 6296–6302. 32. Ghilardi, G., Biondi, M.L., DeMonti, M., Turri, O., Guagnellini, E. & Scorza, R. (2002) Matrix metalloproteinase-1 and matrix metalloproteinase-3 gene promoter polymorphisms are associated with carotid artery stenosis. Stroke 33, 2408–2412. 33. Joos, L., He, J.Q., Shepherdson, M.B., Connett, J.E., Anthonisen, N.R., Pare, P.D. & Sandford, A.J. (2002) The role of matrix metalloproteinase polymorphisms in the rate of decline in lung function. Hum. Mol. Genet. 11, 569–576. 34. Yamada, Y., Ando, F., Niino, N. & Shimokata, H. (2002) Association of a polymorphism of the matrix metalloproteinase-1 gene with bone mineral density. Matrix Biol. 21, 389–392. 35. De Souza, A.P., Trevilatto, P.C., Scarel-Caminaga, R.M., Brito, R.B. & Line, S.R. (2003) MMP-1 promoter polymorphism: association with chronic periodontitis severity in a Brazilian population. J. Clin. Periodontol. 30, 154–158. 36. Gruda, M.C., Kovary, K., Metz, R. & Bravo, R. (1994) Regulation of Fra-1 and Fra-2 phosphorylation differs during the cell cycle of fibroblasts and phosphorylation in vitro by MAP kinase affects DNA binding activity. Oncogene 9, 2537– 2547. 37. Robinson, C.M., Prime, S.S., Huntley, S., Stone, A.M., Davies, M., Eveson, J.W. & Paterson, I.C. (2001) Overexpression of JunB in undifferentiated malignant rat oral keratinocytes enhances the malignant phenotype in vitro without altering cellular differ- entiation. Int. J. Cancer 91, 625–630. 4224 G. B. Tower et al. (Eur. J. Biochem. 270) Ó FEBS 2003 38. Tsuji, M., Hirakawa, K., Kato, A. & Fujii, K. (2000) The possible role of c-fos expression in rheumatoid cartilage destruction. J. Rheumatol. 27, 1606–1621. 39. Newberry, E.P., Willis, D., Latifi, T., Boudreaux, J.M. & Towler, D.A. (1997) Fibroblast growth factor receptor signaling activates the human interstitial collagenase promoter via the bipartite Ets- AP1 element. Mol. Endocrinol. 11, 1129–1144. 40. Robinson, M.J. & Cobb, M.H. (1997) Mitogen-activated protein kinase pathways. Curr. Opin. Cell Biol. 9, 180–186. 41. Selvamurugan, N. & Partridge, N.C. (2000) Constitutive expres- sion and regulation of collagenase-3 in human breast cancer cells. Mol. Cell Biol. Res. Commun. 3, 218–223. 42. Reboul, P., Pelletier, J.P., Tardif, G., Benderdour, M., Ranger, P., Bottaro, D.P. & Martel-Pelletier, J. (2001) Hepatocyte growth factor induction of collagenase 3 production in human osteoar- thritic cartilage: involvement of the stress-activated protein kinase/ c-Jun N-terminal kinase pathway and a sensitive p38 mitogen- activated protein kinase inhibitor cascade. Arthritis Rheum. 44, 73–84. 43. Bhattacharya, A., Lakka, S.S., Mohanam, S., Boyd, D. & Rao, J.S. (2001) Regulation of the urokinase-type plasminogen acti- vator receptor gene in different grades of human glioma cell lines. Clin. Cancer Res. 7, 267–276. 44. Westermarck, J., Lohi, J., Keski-Oja, J. & Kahari, V.M. (1994) Okadaic acid-elicited transcriptional activation of collagenase gene expression in HT-1080 fibrosarcoma cells is mediated by JunB. Cell Growth Differ. 5, 1205–1213. 45. Wilcox, C.B., Weisberg, E., Dumin, J.A., Wilcox, B.D. & Jeffrey, J.J. (2000) Serotonin-dependent collagenase transcription in myometrial cells requires extended AP-1 site. Mol. Cell. Endocrinol. 170, 41–56. 46. Bakiri, L., Matsuo, K., Wisniewska, M., Wagner, E.F. & Yaniv, M. (2002) Promoter specificity and biological activity of tethered AP-1 dimers. Mol. Cell. Biol. 22, 4952–4964. 47.Benbow,U.,Schoenermark,M.P.,Mitchell,T.I.,Rutter,J.L., Shimokawa, K., Nagase, H. & Brinckerhoff, C.E. (1999) A novel host/tumor cell interaction activates matrix metalloproteinase 1 andmediatesinvasionthroughtypeIcollagen.J. Biol. Chem. 274, 25371–25378. 48. Kaufmann, S.H. & Shaper, J.H. (1993) Erasure of western blots after autoradiography or chemiluminescent detection. Appl. Bio- chem. Biotechnol. 38, 243–255. 49. Benbow,U.,Tower,G.B.,Wyatt,C.A.,Buttice,G.&Brinck- erhoff, C.E. (2002) High levels of MMP-1 expression in the absence of the 2G single nucleotide polymorphism is mediated by p38 and ERK1/2 mitogen-activated protein kinases in VMM5 melanoma cells. J. Cell Biochem. 86, 307–319. 50. Catterall, J.B., Carrere, S., Koshy, P.J., Degnan, B.A., Shingleton, W.D.,Brinckerhoff,C.E.,Rutter,J.,Cawston,T.E.&Rowan, A.D. (2001) Synergistic induction of matrix metalloproteinase 1 by interleukin-1alpha and oncostatin M in human chondrocytes involves signal transducer and activator of transcription and activator protein 1 transcription factors via a novel mechanism. Arthritis Rheum. 44, 2296–2310. 51. Solis-Herruzo, J.A., Rippe, R.A., Schrum, L.W., de La Torre, P., Garcia, I., Jeffrey, J.J., Munoz-Yague, T. & Brenner, D.A. (1999) Interleukin-6 increases rat metalloproteinase-13 gene expression through stimulation of activator protein 1 transcription factor in cultured fibroblasts. J. Biol. Chem. 274, 30919–30926. 52. Ramos-Nino, M.E., Timblin, C.R. & Mossman, B.T. (2002) Mesothelial cell transformation requires increased AP-1 binding activity and ERK-dependent Fra-1 expression. Cancer Res. 62, 6065–6069. 53. Cook, S.J., Aziz, N. & McMahon, M. (1999) The repertoire of fos and jun proteins expressed during the G1 phase of the cell cycle is determined by the duration of mitogen-activated protein kinase activation. Mol. Cell Biol. 19, 330–341. 54. Hurd,T.W.,Culbert,A.A.,Webster,K.J.&Tavare,J.M.(2002) Dual role for mitogen-activated protein kinase (Erk) in insulin- dependent regulation of Fra-1 (fos-related antigen-1) transcription and phosphorylation. Biochem. J. 368, 573–580. 55. Tsuchiya, H., Fujii, M., Niki, T., Tokuhara, M., Matsui, M. & Seiki, M. (1993) Human T-cell leukemia virus type 1 Tax activates transcription of the human fra-1 gene through multiple cis elements responsive to transmembrane signals. J. Virol. 67, 7001–7007. 56. Schreiber, M., Poirier, C., Franchi, A., Kurzbauer, R., Guenet, J.L.,Carle,G.F.&Wagner,E.F.(1997)Structureandchromo- somal assignment of the mouse fra-1 gene, and its exclusion as a candidate gene for oc (osteosclerosis). Oncogene 15, 1171–1178. 57. Nutt, J.E. & Lunec, J. (1996) Induction of metalloproteinase (MMP1) expression by epidermal growth factor (EGF) receptor stimulation and serum deprivation in human breast tumour cells. Eur. J. Cancer 32A, 2127–2135. 58. Uria, J.A., Jimenez, M.G., Balbin, M., Freije, J.M. & Lopez-Otin, C. (1998) Differential effects of transforming growth factor-beta on the expression of collagenase-1 and collagenase-3 in human fibroblasts. J. Biol. Chem. 273, 9769–9777. 59. Hanley, K., Wood, L., Ng, D.C., He, S.S., Lau, P., Moser, A., Elias, P.M., Bikle, D.D., Williams, M.L. & Feingold, K.R. (2001) Cholesterol sulfate stimulates involucrin transcription in kerati- nocytes by increasing Fra-1, Fra-2, and Jun D. J. Lipid Res. 42, 390–398. 60. Mengshol, J.A., Vincenti, M.P. & Brinckerhoff, C.E. (2001) IL-1 induces collagenase-3 (MMP-13) promoter activity in stably transfected chondrocytic cells: requirement for Runx-2 and acti- vation by p38 MAPK and JNK pathways. Nucleic Acids Res. 29, 4361–4372. 61.Cohen,D.R.,Ferreira,P.C.,Gentz,R.,Franza,B.R.Jr & Curran, T. (1989) The product of a fos-related gene, fra-1, binds cooperatively to the AP-1 site with Jun: transcription factor AP-1 is comprised of multiple protein complexes. Genes Dev. 3, 173–184. 62. Young,M.R.,Nair,R.,Bucheimer,N.,Tulsian,P.,Brown,N., Chapp, C., Hsu, T.C. & Colburn, N.H. (2002) Transactivation of Fra-1 and consequent activation of AP-1 occur extracellular signal-regulated kinase dependently. Mol. Cell Biol. 22, 587– 598. 63. Zajchowski, D.A., Bartholdi, M.F., Gong, Y., Webster, L., Liu, H.L., Munishkin, A., Beauheim, C., Harvey, S., Ethier, S.P. & Johnson, P.H. (2001) Identification of gene expression profiles that predict the aggressive behavior of breast cancer cells. Cancer Res. 61, 5168–5178. 64. Voigtlander, C., Rand, A., Liu, S.L., Wilson, T.J., Pittelkow, M.R., Getz, M.J. & Kelm, R.J. Jr (2002) Suppression of tissue factor expression, cofactor activity, and metastatic potential of murine melanoma cells by the N-terminal domain of adenovirus E1A 12S protein. J. Cell Biochem. 85, 54–71. Ó FEBS 2003 Fra-1 and the 2G SNP (Eur. J. Biochem. 270) 4225 . Fra-1 targets the AP-1 site/2G single nucleotide polymorphism (ETS site) in the MMP-1 promoter Grant B. Tower 1, *, Charles I. Coon 2 , Karine Belguise 3 ,. present in the promoter. Another ETS site in the MMP-1 promoter is created at )1607 bp by a single nucleotide polymorphism (SNP), which contains two guanines