Molecular and Cellular Biochemistry 188: 169–176, 1998. © 1998 Kluwer Academic Publishers. Printed in the Netherlands. 169 An A/G-rich motif in the rat fibroblast growth factor-2 gene confers enhancer activity on a heterologous promoter in neonatal rat cardiac myocytes Karen A. Detillieux, Adrienne F.A. Meyers, Johanna T.A. Meij and Peter A. Cattini Gene Technology Group and Departments of Physiology and Pharmacology and Therapeutics, University of Manitoba, Winnipeg, Manitoba, Canada Abstract We have cloned the rat fibroblast growth factor-2 (FGF-2) promoter region including 1058 base pairs (bp) of 5′-flanking DNA. Complete sequencing of this promoter region revealed a 74 bp domain between nucleotides –793 and –720 that was greater than 97% A/G-rich. A repeat of the sequence 5′-AGGGAGGG-3′ separated by 11 bp was located at the core of this domain. A 37 bp A/G-rich oligonucleotide containing these AGGG-repeat sequences was synthesised, and tested for function on a minimal herpes simplex virus thymidine kinase (TK) promoter, fused to the firefly luciferase gene (TKp.luc), in transiently transfected neonatal rat cardiac myocytes. Promoter activity was stimulated ~3 fold in the presence of AGGG-repeat sequences. This effect was neither tissue or species-specific since TK promoter activity was increased ~11 fold in both rat and human glial tumor cells. Four specific complexes (C1–4) were detected between neonatal rat heart nuclear proteins and the 37 bp A/G-rich oligonucleotide by gel mobility shift assay. Competition with excess unlabelled 37 bp A/G-rich oligonucleotide revealed that two complexes represented very high affinity/specificity interactions (C2 > C4) while C1 and C3 were of lower affinity. As a result, competition with up to a 25 fold molar excess of 37 bp A/G-rich oligonucleotide led to the loss of C2 and C4, and a corresponding and transient increase in the levels of C1 and C3, which themselves were reduced with more competitor oligonucleotide. The AGGG-repeat resembles the 5′-gGGGAGGG-3′ sequence previously implicated in the response of the atrial natriuretic factor promoter to the α-adrenergic agonist, phenylephrine. Although an additional 1.5 fold increase in TK promoter activity was detected in the presence of the 37 bp A/G-rich oligonucleotide with phenylephrine treatment of transfected myocytes, this effect was not statistically significant. Furthermore, there was no difference in the gel mobility shift (C1–4) pattern obtained with the 37 bp A/G-rich oligonucleotide and nuclear protein isolated from neonatal rat cardiac myocytes grown in the presence or absence of norepinephrine. These data suggest that the A/G rich sequences in the rat FGF-2 gene 5′-flanking DNA, including the AGGGrepeat, are able to confer stimulatory activity on a promoter in a tissue- and species-independent manner, but alone are not able to induce a significant phenylephrine response in neonatal rat cardiac myocytes. (Mol Cell Biochem 188: 169–176, 1998) Key words: FGF-2, transcription, gene transfer, HSV-thymidine kinase promoter Introduction Fibroblast growth factor-2 (FGF-2) or basic fibroblast growth factor (bFGF) and its high affinity receptor FGFR-1, a member of the tyrosine kinase family, are essential for the normal growth and development of the myocardium [1, 2]. With regard to the post natal heart, there is increasing evidence that FGF-2, which is released with contraction [3, 4], is Address for offprints: P.A. Cattini, Gene Technology Group and Department of Physiology, University of Manitoba, 730 William Avenue, Winnipeg, Manitoba, Canada R3E 3J7 170 involved in the maintenance of a healthy myocardium through its angiogenic [5–7] as well as cardioprotective properties [8, 9]. Many of the studies regarding FGF-2 have made use of exogenous addition of this factor, and focussed on its interaction with its high affinity cell surface receptor, which signals its mitogenic and angiogenic responses. In contrast, there have been relatively few studies on the control of FGF-2 synthesis, particularly at the transcriptional level, and specifically in the heart. The human and rat FGF2 promoter sequences have been reported [10, 11], and although they do not share extensive structural similarity, short (~20 base pair) stretches of highly conserved sequences were identified in the proximal promoter region [11]. Also, both promoters lacked a TATA box but contained G/C rich regions and Spl sites [10, 11]. The human FGF-2 promoter was shown to be regulated by mutant and wild type p53 in a positive and negative manner, respectively, in reporter gene/transfection studies using glioblastoma and hepatocellular carcinoma cells [12]. In addition, neuropeptides were shown to influence the human FGF-2 promoter via the zinc finger transcription factor Egr-1 in glial cells [13]. A consensus Egr-1 site is also contained in the rat FGF-2 promoter region [11]. Based on an alignment of reported human and rat FGF-2 genomic sequences [10, 11], the rat fragment appears to extend further upstream [11]. Here we report the detection of an A/G rich domain in this extended DNA region containing a repeat of the sequence 5′-AGGGAGGG-3′. This sequence shows some similarity to a previously reported phenylephrine responsive element (5′-gGGGAGGG-3′) in the rat atrial natriuretic factor promoter, which is functional in transfected neonatal rat cardiac myocytes [14]. We have assessed these sequences for activity using the firefly luciferase reporter gene assay and transient gene transfer into neonatal cardiac myocytes and non cardiac cells, as well as their ability to bind nuclear protein. Our data indicate that these A/G rich sequences can confer stimulatory activity on a promoter, without tissue- or species-specificity, and make high affinity/ specificity interactions with nuclear protein. These sequences did not, however, show a significant response to phenylephrine in neonatal rat cardiac myocytes after gene transfer. Materials and methods Cell culture Neonatal rat cardiac myocyte cultures were prepared essentially as previously described [15]. Briefly, ventricles from Sprague-Dawley rat pups (at 1–2 days after birth) were dissected and the cells dissociated in a spinner flask using a combination of trypsin and DNase I. Myocytes were separated from non muscle cells on a discontinuous Percoll gradient and plated on collagen coated plates at a density of 1 × 106 cells per 35 mm dish. Cells were initially plated in Ham’s F10 medium containing 10% fetal bovine serum (FBS), 10% horse serum, antibiotic (1000 units/ml penicillin, 1 mg/ml streptomycin) and calcium chloride supplemented to 1.05 mM. Rat glioma C6 and human astrocytoma U87MG cells were obtained from the American Type Culture Collection and grown in monolayer culture in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) FBS with antibiotic at 37°C in the presence of 5% CO2. Plating densities for C6 and U87-MG cells were 0.5 × 106 and 1.0 × 106 per 100 mm dish, respectively. All cultureware was purchased from Corning (Fisher Scientific, Nepean, ON, Canada), and all media and reagents from Gibco-BRL (Life Technologies, Burlington, ON, Canada). Nuclear protein preparations Heart tissue was dissected aseptically from neonatal (1–2 days) Sprague Dawley rats, and nuclear protein was prepared essentially as described previously 116], with final dialysis against 20 mM Hepes pH 7.9, 20% v/v glycerol, 0.1 M KCl, 0.2 mM EDTA, 0.5 mM DTT and 1 mM PMSF. The nuclear extract was mixed with heparin-agarose and washed with 0.1M KCl before elution of protein with 0.6 M KCl. Following determination of protein concentration using the Bradford Assay (Bio-Rad Laboratories, Mississauga, ON, Canada), nuclear extracts were aliquoted and maintained at –70 °C. A modified, rapid extraction protocol [17] was used for isolation of nuclear protein from cultured neonatal rat cardiac ventricular myocytes grown in the presence or absence of 0.01 mM norepinephrine for 6 hours under serumfree conditions. Hybrid gene constructions The plasmid TKp.luc (pT81.luc, 118]) contains a portion of the herpes simplex virus thymidine kinase (TK) promoter (positions –81 to +52) fused to the firefly luciferase gene. A 37 bp A/G rich (double-stranded) oligonucleotide, corresponding to nucleotide positions –785/–749 [11], and containing the putative AGGG-repeat sequences, 5′GGGAAAGGGAGGGGGAAGGAAAGGAGGGAGGGAAGGA3′, was synthesised and inserted upstream of the TK promoter in TKp.luc to generate A/G-TKp.luc. The 5 bp of sequence upstream and downstream of the AGGG-repeat was included in an attempt to limit any end effect that might interfere with specific protein binding. The promoterless firefly luciferase gene (–p.luc, contained in the vector pXPl) was described previously [11]. The pRL-CMV vector, containing the Renilla luciferase gene under the control of the cytomegalovirus 171 promoter was obtained from Promega Corporation (Fisher Scientific). Gene transfer Cardiac myocytes and glial tumor cells were transfected by the calcium phosphate-DNA precipitation method. Briefly, 30 µg of test plasmid (hybrid firefly luciferase gene) and 3 µg of control plasmid (pRL-CMV) was made up to a volume of 0.5 ml in 252 mM CaCl2 and added gradually to an equal volume of aerated 2 × HEBS buffer (280 mM NaCl, 50 mM HEPES-KOH, pH 7.1, and 1.5 mM Na2PO4). Precipitate was allowed to form at room temperature for 30 min, and 310 µl was added to each of 3 culture dishes of cardiac myocytes (35 mm) or glial tumor cells (100 mm) containing 4 ml or 8 ml of DMEM/10% FBS, respectively. Cells were transfected for 16 h and then washed thoroughly with calcium- and magnesium-free phosphate-buffered saline. Cells were refed with DMEM/10% FBS and maintained for 48 h before processing. For phenylephrine treatment, transfected cardiac myocytes were refed with DMEM-F12 containing l × Redu-Ser II (Upstate Biotechnology, Lake Placid, NY, U.S.A), 0.02 mg/ml ascorbic acid and antibiotics with or without 0.1 mM phenylephrine, and maintained for 48 h before processing. Cells were also transfected with –p.luc as a control for random transcription initiation. Cotransfection with pRL-CMV was used as a control for DNA uptake and values were used subsequently to normalize the firefly luciferase activity (firefly luciferase/Renilla luciferase) from the ‘test’ genes. For experiments with phenylephrine treatment, firefly luciferase activity was normalized against lysate protein content (luciferase activity/ng protein) which was determined using the Bradford Assay (Bio-Rad Laboratories). was 32P-end-labelled, in the presence of 2 µg of poly dI-dC. Incubation of the reaction (in 10 mM Hepes-NaOH pH 7.9, 50 mM KCl, 0.5 mM EDTA, 10% glycerol, 1 mM DTT with 5 mM MgCl2) at room temperature for 30 min was followed by electrophoresis in non-denaturing 4% polyacrylamide gels. For competition, specific unlabelled 37 bp A/G-rich oligonucleotide or non specific RF-1 element [21] (at a 1– 100 fold molar excess over radiolabelled oligonucleotide) were added to nuclear extracts and incubated at room temperature for 10 min prior to the addition of the radiolabelled DNA and incubation at room temperature for a further 20 min. Statistical analysis Data presented in the text and figures represent the mean ± S.E. from at least 2 independent experiments each done in triplicate. Statistical analysis of the results was done using the Mann-Whitney nonparametric test. Results were considered significant if p was determined to be < 0.05. Results Sequence analysis reveals an A/G-rich domain in the rat FGF-2 gene We have reported the cloning of a fragment of the rat FGF2 gene, including a promoter region of 1058 bp based on a primary transcription start site (+1) determined using brain RNA [11]. Inspection of these sequences reveals a 74 bp region which is 97% adenine/guanine (A/G-rich, located between nucleotide positions –793 and –720 (Fig. 1). Indeed, a 65 bp stretch within this region (–793/–729) contains only A or G residues. Located at the core of this region are two Reporter gene assays Both firefly and Renilla luciferase activities were measured, following ‘active lysis of cells by scraping’, using the ‘DualLuciferase™ Reporter Assay System’ (Promega) and a luminometer (ILA900 Luminometer, Tropix Inc., Bedford, MA, USA) according to the manufacturer’s instructions. The protein content of cell Iysates generated in this manner was determined by the bicinchoninic acid protein assay [19]. Gel mobility shift assay The gel mobility shift assay was performed essentially as described by Baldwin [20]. Nuclear extract (5 µg) was incubated with the 37 bp A/G-rich oligonucleotide which Fig. 1. Region containing A/G-rich sequences in the rat FGF-2 5-flanking DNA. The A/G-rich region corresponds to nucleotides –793/–720, and its position is based on the major transcription initiation site detected in rat brain RNA [11]. The two copies of the AGGG-repeat sequence, 5′AGGGAGGG-3′ are boxed. 172 copies of the an AGGG-repeat sequence, 5′-AGGGAGGG3′. This sequence shows a high degree of similarity with a sequence, 5′-gGGGAGGG-3′, found in the atrial natriuretic factor (ANF) promoter and implicated in its response to α1specific adrenergic activation by phenylephrine [14]. A 37 bp double-stranded A/G-rich oligonucleotide, corresponding to nucleotides –785/–749 of the FGF-2 promoter and containing both copies of 5′-AGGGAGGG-3′, was synthesized (Fig. 1, boxed sequence). An A/G-rich region containing an AGGG-repeat stimulates heterologous promoter activity in a tissue- and species-independent manner To assess any effect of the AGGG-repeat sequences on promoter activity, the 37 bp A/G-rich oligonucleotide was inserted upstream of a minimal viral TK gene promoter, which was fused to the firefly luciferase gene (TKp.luc) to generate A/G-TKp.luc. Both TKp.luc and A/G-TKp.luc. were co-transfected with the hybrid Renilla luciferase gene (pRL CMV) into neonatal rat cardiac myocytes as well as rat C6 and human U87 glial tumor cells, and then tested for activity after 48 h. The results (firefly luciferase/Renilla luciferase activity) are shown in Fig. 2A. A significant increase in TK promoter activity was observed in the presence of the A/Grich sequences in neonatal rat cardiac myocytes (p < 0.005, n = 6), rat C6 glioma cells (p < 0.01, n = 5) and human U87MG astrocytoma cells (p < 0.05, n = 5). However, the level of stimulation was greater in rat or human glial tumor cells Fig. 2. (A) Effect of the 37 bp A/G-rich oligonucleotide on TK promoter activity (firefly luciferase/Renilla luciferase) in neonatal rat cardiac myocytes (CM) as well as rat C6 and human U87 glial tumor cells after transient gene transfer. Results are expressed as the mean from at least two independent experiments. The bars represent the standard error of the mean. (B) The results from (A) are presented to show fold effect of the A/G-rich sequences on TK promoter activity (A/G-TKp.luc/TKp.luc) in the various cell types. (~11 fold) versus primary cardiac myocytes (~3 fold; Fig. 2B). Neonatal rat heart nuclear proteins make high affinity/ specificity interactions with the A/G-rich sequences The gel mobility shift assay was used to investigate the presence of specific neonatal rat heart nuclear protein interactions with the 37 bp A/G-rich oligonucleotide. The radiolabelled DNA (0.5 ng) was incubated with 5 µg of neonatal rat heart nuclear protein in the absence or presence of a 25, 50 or 100 fold molar excess of unlabelled A/G-rich oligonucleotide. As a further control, an RF-1 DNA element, containing an unrelated sequence, was also used at a 25, 50 or 100 fold molar excess as a non specific oligonucleotide competitor. Four specific complexes (C1–4) were identified (Fig. 3). Both C2 and C4 were competed completely with a 25 fold molar excess of specific (A/G-rich oligonucleotide) but not non specific (RF-1 element) competitor. A slight increase in the amount of C1 and C3 complexes was detected corresponding to the complete competition of C2 and C4 with a 25 (and to a lesser extent with a 50) fold molar excess of specific competitor. The C1 and C3 complexes required a 100 fold molar excess of specific A/G-rich oligonucleotide to be competed efficiently (Fig. 3). Fig. 3. Gel mobility shift assay of neonatal rat heart nuclear proteins and the 37 bp A/G-rich oligonucleotide. Specificity was determined by competition with ‘specific’ unlabelled A/G-rich or ‘non specific’ RF-1 element oligonucleotide competitors. The radiolabelled A/G-rich fragment (free probe, FP) was incubated in the (a) absence or (b-h) presence of (5 µg) nuclear protein. For competition, (b) none or (c) 25, (d) 50, or (e)100 fold molar excess of specific competitor, or (f) 25, (g) 50, or (h)100 fold molar excess of non specific competitor was used. The positions/mobilities of four specific complexes (C1-C4) are indicated. 173 Fig. 4. Detection of very high affinity interactions between neonatal rat heart nuclear protein and the 37 bp A/G-rich DNA fragment. Affinity was determined by gel mobility shift assay and competition with low amounts of ‘specific’ unlabelled A/G-rich oligonucleotide competitor. The radiolabelled A/G-rich fragment (FP) was incubated in the (a) absence or (b-g) presence of (5 µg) nuclear protein, (b) without or with a (c) 1, (d) 2, (e) 5, (f) 10, or (g) 15 fold molar excess of specific competitor. The positions/ mobilities of the four specific complexes (C1-C4) are indicated. Fig. 5. Effect of 0.1 mM phenylephrine treatment for 48 h on TK promoter activity in the presence (A/G-TKp.luc) or absence (TKp.luc) of the 37 bp A/G-rich oligonucleotide in transiently transfected neonatal rat cardiac myocytes. Results are expressed as mean promoter activity (firefly luciferase/ng protein). Basal levels for –p.luc in the presence and absence of phenylephrine were 0.078 ± 0.007 and 0.030 ± 0.004, respectively. Bars represent standard error of the mean. To assess the relative affinity of nuclear protein for DNA in the C2 vs. C4 complexes, gel mobility shift assays were done using lesser amounts (1, 2, 5, 10, and 15 fold molar excess) of 37 bp A/G-rich oligonucleotide for competition (Fig. 4). Complex C2 represents a very high affinity/ specificity interaction since it was competed efficiently with only a 2 fold molar excess of specific A/G-rich oligonucleotide. C4 was competed completely with a 10 fold molar excess of specific competitor. The transient increase in the amount of C1 and C3 with competition of C2 and C4 was also apparent. sequences, respectively. However, although there was an additional ~1.5 fold increase in activity observed in the presence of the 5′-AGGGAGGG-3′ sequence with phenylephrine treatment, this was not considered statistically significant (p = 0. 19). To complement this study, we compared the gel mobility shift assay patterns obtained using the 37 bp A/G-rich oligonucleotide, containing two copies of the 5′-AGGGAGGG3′ sequence, with nuclear protein isolated from neonatal rat cardiac myocytes grown in the absence versus presence of 0.01 mM norepinephrine. Four complexes (C1–4) were observed (Fig. 6). This pattern was not altered by norepinephrine stimulation. There was also no difference in the degree of competition with a 50 or 100 fold molar excess of specific A/G-rich oligonucleotide, suggesting no change in affinity of these complexes with adrenergic stimulation (Fig. 6). Effect of phenylephrine treatment on the pattern of interaction between cardiac myocyte nuclear protein and DNA containing the 5′-AGGGAGGG-3′ sequence To assess the effect of phenylephrine on TK promoter activity in the absence or presence of the 5′-AGGGAGGG3′ sequences, neonatal rat cardiac myocytes were transfected with TKp.luc or A/G-TKp.luc. Transfected cells were treated without or with 0.1 mM phenylephrine for 48 h, harvested and firefly luciferase activity per ng protein was determined (Fig. 5). Phenylephrine treatment increased TK promoter activity 4.6 ± 0.8 (p < 0.0001, n = 17) and 6.7 ± 1.1 fold (p < 0.008, n = 5) in the absence and presence of the A/G-rich Discussion We have used both functional and structural approaches in the form of transient gene transfer and gel mobility shift assays to characterise the repeat elements 5′-AGGGAGGG3′ contained in a 37 bp DNA fragment comprised totally of adenosine (A) or guanosine (G) monophosphates. These 174 Fig. 6. Comparison of gel mobility shift patterns seen with the 37 bp A/ G-rich oligonucleotide and nuclear protein from isolated cardiac myocytes grown in the absence (b-f) or presence (e-k) of 0.01 mM norepinephrine for 6 h. Affinity/specificity was assessed by competition with ‘specific’ unlabelled A/G-rich or ‘non specific’ RF-1 oligonucleodde competitors. The radiolabelled A/G-rich fragment (FP) was incubated in the (a) absence or (b, g) presence of 2.5 ±g or (c-f, h-k) 5 µg of nuclear protein, (c, h) without or (d, i) with a 50, or (e, j) 100 fold molar excess of specific competitor, or (f, k) 100 fold molar excess of non specific competitor. The positions of four specific complexes (C 1-C4) are indicated. sequences represent the core of a 74 bp region located at nucleotide position –793/–720 that is 97% A/G residues, and of which a 65 bp stretch contains only A or G nucleotides (Fig. 1). A/G-rich regions have been linked to gene regulation [14] as well as alterations in DNA structure, specifically with regard to the human β-globin locus control region [22]. The ability of the A/G-rich oligonucleotide to increase the activity of a minimal 81 bp viral thymidine kinase promoter in neonatal rat cardiac myocytes as well as rat C6 and human U87 glial tumor cells (Fig. 2) suggests that these sequences, and more specifically the A/G-rich elements, can regulate gene expression. Furthermore, the stimulation of promoter activity in cells of different tissues (cardiac myocytes vs. brain glial cells) or species (rat versus human glial cells) origin correlates with the ubiquitous pattern of FGF-2 synthesis. FGF-2 has been found in every tissue examined so far [23, 24] and its structure is highly conserved between species, such that mouse and human FGF-2 share 94% sequence similarity at the amino acid level [25]. However, although the A/G oligonucleotide was able to stimulate promoter activity in the context of a heterologous promoter and reporter gene (Fig. 2), a deletion analysis of the rat FGF2 5′-flanking DNA does not support a major role for the AGGG-repeats, alone, in the regulation of the FGF-2 promoter in cardiac myocytes or glial cells [Detillieux, Meyers and Cattini, unpublished observations]. Of course, this does not rule out the possibility that additional elements and/or factors participate in common complexes with those associated with the A/G-rich sequences, thereby modifying their action and, thus, relative importance. The data obtained from gel mobility shift assays indicate the presence of multiple protein/DNA interactions. Four complexes (C1–4) between neonatal rat heart nuclear protein and the A/G-rich oligonucleotide were identified (Figs 3 and 4). The same pattern of complexes was also suggested when nuclear protein isolated from neonatal rat cardiac myocytes was used (Fig 5B), indicating that these interactions are not only the product of non muscle cell proteins present in the heart. All four complexes were specific as they were competed by increasing amounts (25–100 fold molar excess) of unlabelled A/G-rich oligonucleotide, but not by equivalent doses of the RF-1 DNA element [21] containing unrelated sequences (Fig. 3). The complete competition of C2 and C4 with a 25 fold molar excess indicates that these complexes possess a higher affinity/specificity than C1 and C3, which required 100 fold molar excess of specific oligonucleotide to see efficient competition. Interestingly, an increase in C1 and C3 was suggested with the complete competition of C2 and C4 in the presence of a 25 fold molar excess of A/G-rich oligonucleotide (Fig. 3). The experiment was repeated with a reduced level of specific competitor DNA fragment (1–15 fold molar excess) to further dissect this observation as well as determine the relative affinity of the C2 and C4 complexes (Fig. 4). C2 and C4 were efficiently competed with a 2 and 10 fold molar excess, respectively, confirming that these are both very high affinity/ specificity interactions (C2 > C4). These results also show that there is a transient increase in C1 and C3 binding which corresponds to the competition of C2 and C4. This suggests the possibility that the protein/DNA interactions reflected by C2/C4 and C1/C3 are mutually exclusive with a preference, under the experimental conditions used, for C2/C4 to form. When the higher affinity C2/C4 events are made invisible by competition (with a low molar excess of A/G-rich oligonucleotide), there is an increased opportunity for the proteins associated with C1 and C3 to bind. The ability to detect all four complexes (C14) simultaneously suggests that the proteins associated with lower affinity C1 and C3 are present in excess in the neonatal rat heart. This scenario offers a possible mechanism for regulation of nuclear protein binding and hence function. The AGGG-repeat sequence 5′-AGGGAGGG-3′ present in the FGF-2 gene is related structurally to the sequence 5′gGGGAGGG-3′, which was reported previously to be the phenylephrine responsive element in the ANF promoter [14]. Indeed, the sequence 5′AGGGAGGG-3′ exists in the 175 promoter of human α-skeletal actin, and was suggested to play a role in its phenylephrine responsiveness [14]. A subsequent investigation demonstrated the requirement of at least one additional element, resembling a serum response element, to produce an efficient response of the ANF promoter to phenylephrine treatment [26]. Although we detected an additional ~1.5 fold increase in viral thymidine kinase promoter activity that could be attributed to the presence of AGGG-repeat sequences after phenylephrine treatment of transfected neonatal rat cardiac myocytes, this effect was not statistically significant (Fig. 5). Furthermore, there was no change in the pattern of DNA interaction with nuclear proteins (C1–4) isolated from neonatal rat cardiac myocytes grown in the absence or presence of norepinephrine (Fig. 6). Thus, it is possible that the duplicated 5′-AGGGAGGG-3′ sequences present in the FGF-2 gene constitute elements that contribute to phenylephrine regulation of the FGF-2 promoter, but are themselves insufficient for a complete response. In summary, examination of sequences upstream of the rat FGF-2 coding region revealed a 65 bp domain containing only A or G nucleotides. Located at the core of this region are two copies of the repeat sequence 5′-AGGGAGGG-3′, which shows a strong similarity to a phenylephrine response element. These A/G-rich sequences bind cardiac nuclear protein with high affinity and are able to confer stimulatory activity on a viral promoter in a tissue- and species-independent manner, but do not respond significantly to phenylephrine in transfected neonatal rat cardiac myocytes. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. Acknowledgements 14. 15. The authors would like to thank Rama Mohan Surabhi for his careful review of the manuscript. This work was supported by a grant from the Medical Research Council of Canada (MT-13398). KAD is the recipient of a Heart and Stroke Foundation Studentship and AFAM is the recipient of a Medical Research Council Studentship Award. References 16. 17. 18. 19. 1. 2. 3. Sugi Y, Sasse J, Lough J: Inhibition of precardiac mesoderm cell proliferation by antisense oligodeoxynucleotide complementary to fibroblast growth factor-2 (FGF-2). Dev Biol 157: 28–37, 1993 Mima T, Ueno H, Fischman DA, Williams LT, Mikawa T: Fibroblast growth factor receptor is required for in vivo cardiac myocyte proliferation at early embryonic stages of heart development. Proc Natl Acad Sci USA 92: 467–471, 1995 Clarke MS, Caldwell RW, Chiao H, Miyake K, McNeil PL: Contractioninduced cell wounding and release of basic fibroblast growth factor. Circ Res 76: 927–934, 1995 20. 21. 22. Kaye D, Pimental D, Prasad S, Maki S, Berger HJ, McNeil PL, Smith TW, Kelly RA: Role of transiently altered sarcolemmal membrane permeability and basic fibroblast growth factor release in the hypertrophic response of adult rat ventricular myocytes to increased mechanical activity in vitro. J Clin Invest 97: 281–291, 1996 Unger EF, Banai S, Shou M, Lazarous DF, Jaklitsch MT, Scheinowitz M, Correa R, Ningbeil C, Epstein SE: Basic fibroblast growth factor enhances myocardial collateral flow in a canine model. Am J Physiol 266 (Heart Circ Physiol 35): H158–H1595, 1994 Lazarous DF, Scheinowitz M, Shou M, Hodge E, Majanayagam MAS, Hunsberger S, Robinson Jr WG, Stiber JA, Correa R, Epstein SE, Unger EF: Effects of chronic systemic administration of basic fibroblast growth factor on collateral development in the canine heart. Circulation 91: 145–153, 1995 Fox JC, Shanley JR: Antisense inhibition of basic fibroblast growth factor induces apoptosis in vascular smooth muscle cells. J Biol Chem 271: 12578–12584, 1996 Padua RR, Sethi R, Dhalla NS, Kardami E: Basic fibroblast growth factor is cardioprotective in ischemia-reperfusion injury. Mol Cell Biochem 143: 129–135, 1995 Padua RR, Sethi R, Davey-Forgie SE, Lui L, Dhalla N, Kardami E: Cardioprotection and basic fibroblast growth factor. In: N.S. Dhalla, P.K. Singal, R.E. Beamish (eds). Heart Hypertrophy and Failure. Kluwer Academic Publishers, Boston, 1996, pp 501–518. Shibata F, Baird A, Florkiewicz RZ: Functional characterization of the human basic fibroblast growth factor promoter. Growth Factors 4: 277–287, 1991 Pasumarthi KBS, Jin Y, Cattini PA: Cloning of the rat fibroblast growth factor-2 promoter region and its response to mitogenic stimuli in glioma C6 cells. J Neurochem 68: 898–908, 1997 Ueba T, Nosaka T, Takahashi JA, Shibata F, Florkiewicz RZ, Vogelstein B, Oda Y, Kikuchi H, Hatanaka M: Transcriptional regulation of basic fibroblast growth factor gene by p53 in human glioblastma and hepatocellular carcinoma cells. Proc Natl Acad Sci USA 91: 9009– 9013, 1994 Biesiada E, Razandi M, Levin ER: Egr-1 activates basic fibroblast growth factor transcription. J Biol Chem 271: 18576–18581, 1996 Ardati A, Nemer M A nuclear pathway for α1-adrenergic receptor signaling in cardiac cells. EMBO J 12: 5131–5139, 1993 Pasumarthi KBS, Kardami E, Cattini PA: High and low molecular weight fibroblast growth factor-2 increase proliferation of neonatal rat cardiac myocytes but have differential effects on binucleation and nuclear morphology: Evidence for both paracrine and intracrine actions of fibroblast growth factor-2. Circ Res 78: 126–136, 1996 Dignam JD, Lebovitz RM, Roeder RG: Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucl Acids Res 11: 1475–1489, 1983 Andrews NC, Faller DV: A rapid micropreparation technique for extraction of DNAbinding proteins from limiting numbers of mammalian cells. Nucl Acids Res 19: 2499, 1991 Nordeen SK: Luciferase reporter gene vectors for analysis of promoters and enhancers. Biotechniques 6: 454–458, 1988 Smith PK, Krohn RI, Hermansen GT, Mallia AK, Gartner FH, Fugimoto EK, Goeke NM, Olson BJ, Klenk DC: Measurement of protein using bicinchoninic acid. Anal Biochem 150: 76–85, 1985 Baldwin AS: Analysis of sequence-specific DNA-binding proteins by the gel mobility shift assay. DNA Protein Eng Tech 2: 73–76, 1990 Lytras A, Cattini PA: Human chorionic somatomammotropin gene enhancer activity is dependent on the blockade of a repressor mechanism. Mol Endocrinol 8: 478–489, 1994 Boulikas T: Homeodomain protein binding sites, inverted repeats, and nuclear matrix attachment regions along the human β-globin gene complex. J Cell Biochem 52: 23–36, 1993 176 23. Kardami E, Liu L, Padua RR, Fandrich RR, Pasumarthi SKB, Cattini PA: Regulation of basic fibroblast growth factor (bFGF) and FGF receptors in the heart. Ann NY Acad Sci 752: 353–369, 1995 24. Bikfalvi A, Klein S, Pintucci G, Rifkin DB: Biological roles of fibroblast growth factor-2. Endocrine Rev 18: 26–45, 1997 25. Ornitz DM, Xu J, Colvin JS, McEwen DG, MacArthur CA, Coulier F, Gao G, Goldfarb M: Receptor specificity of the fibroblast growth factor family. J Biol Chem 271: 15292–15297, 1996 26. Sprenkle AB, Murray SF, Glembotski CC: Involvement of multiple cis-elements in basal and α-adrenergic agonist-inducible ANF transcription. Roles for serum response elements and an Sp-l-like element. Circ Res 77: 1060–1069, 1995