Distinct and Common Pathways in the Regulation of Insulin-like Growth Factor-1 Receptor Gene Expression by Angiotensin II and Basic Fibroblast Growth Factor*

Angiotensin II (Ang II) and basic fibroblast growth factor (bFGF) are important modulators of cell growth under physiological and pathophysiological conditions. We and others have previously shown that these growth factors increase insulin-like growth factor-1 receptor (IGF-1R) number and mRNA in vascular smooth muscle cells and that this effect is transcriptionally regulated. To study the mechanisms and the signaling pathways involved, IGF-1R promoter reporter constructs were transiently transfected in CHO-AT1 cells that overexpress angiotensin AT1 receptors. Our findings indicate that Ang II and bFGF significantly increased IGF-1R promoter activity up to 7- and 3-fold, respectively. The effect induced by Ang II was mediated via a tyrosine kinase-dependent mechanism, since tyrphostin A25 largely inhibited the Ang II-induced increase in promoter activity. In addition, co-transfection of dominant negative Ras, Raf, and MEK1 or pretreatment with the MEK inhibitor PD 98059 dose-dependently decreased both the Ang II- and bFGF-induced increase in IGF-1R transcription and protein expression, suggesting that the Ras-Raf-mitogen-activated protein kinase kinase pathway is required for both growth factors. Reactive oxygen species have been shown to act as second messengers in Ang II-induced signaling, and activation of the transcription factor NF-κB is redox-sensitive. While co-transfection of dominant negative IκBα mutant completely inhibited the Ang II-induced increase in transcription, it had no effect on the bFGF signaling. In contrast, co-transfection studies indicated that the transcription factors STAT1, STAT3, and c-Jun and the Janus kinase 2 kinase are required in the signaling pathway of bFGF, whereas only dominant c-Jun inhibited the Ang II-induced effect. In summary, these data demonstrate that Ang II and bFGF increase IGF-1R gene transcription via distinct as well as shared pathways and have important implications for understanding growth-stimulatory effects of these growth factors on vascular cells.

Angiotensin II (Ang II) and basic fibroblast growth factor (bFGF) are important modulators of cell growth under physiological and pathophysiological conditions. We and others have previously shown that these growth factors increase insulin-like growth factor-1 receptor (IGF-1R) number and mRNA in vascular smooth muscle cells and that this effect is transcriptionally regulated. To study the mechanisms and the signaling pathways involved, IGF-1R promoter reporter constructs were transiently transfected in CHO-AT 1 cells that overexpress angiotensin AT 1 receptors. Our findings indicate that Ang II and bFGF significantly increased IGF-1R promoter activity up to 7-and 3-fold, respectively. The effect induced by Ang II was mediated via a tyrosine kinase-dependent mechanism, since tyrphostin A25 largely inhibited the Ang II-induced increase in promoter activity. In addition, co-transfection of dominant negative Ras, Raf, and MEK1 or pretreatment with the MEK inhibitor PD 98059 dose-dependently decreased both the Ang II-and bFGF-induced increase in IGF-1R transcription and protein expression, suggesting that the Ras-Raf-mitogen-activated protein kinase kinase pathway is required for both growth factors. Reactive oxygen species have been shown to act as second messengers in Ang II-induced signaling, and activation of the transcription factor NF-B is redox-sensitive. While co-transfection of dominant negative IB␣ mutant completely inhibited the Ang II-induced increase in transcription, it had no effect on the bFGF signaling. In contrast, co-transfection studies indicated that the transcription factors STAT1, STAT3, and c-Jun and the Janus kinase 2 kinase are required in the signaling pathway of bFGF, whereas only dominant c-Jun inhibited the Ang II-induced effect. In summary, these data demonstrate that Ang II and bFGF increase IGF-1R gene transcription via distinct as well as shared pathways and have important implications for understanding growth-stimulatory effects of these growth factors on vascular cells.
The vascular response to injury requires a coordinated inter-action between hemostatic and inflammatory systems and is regulated by cytokines and growth factors that act locally to regulate cellular proliferation and tissue repair. Among the many growth factors that have been shown to be implicated in the response to vascular injury, angiotensin II (Ang II) 1 is of particular interest. It stimulates a variety of physiological responses related to regulation of blood pressure, salt, and fluid homeostasis (1). However, Ang II has also been shown to function, either directly or indirectly, as a growth factor for vascular smooth muscle cells, cardiac fibroblasts, and cardiac myocytes (2)(3)(4)(5)(6)(7). The array of genes induced by Ang II includes protooncogenes such as c-fos, c-jun, c-myc, and egr-1 (5,(7)(8)(9), genes encoding extracellular matrix proteins such as collagen, fibronectin, and tenascin (10 -12), and genes for growth factors like transforming growth factor ␤ 1 , platelet-derived growth factor (PDGF), insulin-like growth factor-1 (IGF-1), and its receptor (5,7,(12)(13)(14). Similarly, basic fibroblast growth factor (bFGF) has been implicated in the vascular injury response. In particular, bFGF increases endothelial cell migration and proliferation and also stimulates angiogenesis in vitro and in vivo (15,16). The role of bFGF in vessel injury and repair is further supported by evidence that bFGF is released from vessel wall cells after injury (17) and that bFGF mRNA is up-regulated in atherosclerotic lesions (18).
Ang II exerts its effects through specific G-protein-coupled receptors, predominantly through the AT 1 receptor subtype. These receptors induce intracellular calcium mobilization; activation of tyrosine kinases such as p125 FAK , p46 SHC , and p54 SHC ; induction of serine/threonine kinases, including protein kinase C and mitogen-activated protein kinases (MAPKs) (13, 30 -35); and stimulation of the Janus kinase (Jak)/signal transducer and activator of transcription (STAT) pathway (36). The bFGF receptor, however, belongs to the receptor tyrosine kinase family (37)(38)(39) and couples to a variety of signaling pathways, including phospholipase C-␥ and phospholipase A 2 activation (40,41), activation of the MAPK pathway (42)(43)(44), and activation of the Jak/STAT cascade (45,46). Similarly to Ang II, bFGF has been shown to induce the expression of the early response gene c-fos (47). However, the signaling pathways by which Ang II or bFGF increase transcriptional activity of the IGF-1R gene are unknown, with the exception that the bFGF, but not the Ang II effect, is protein kinase C-dependent (48).
The purpose of the present studies was to localize the Ang IIand the bFGF-responsive elements in the IGF-1R promoter and define the signaling cascades whereby these two growth factors increase IGF-1R gene transcription or protein expression. We show that Ang II and bFGF positively regulate transcriptional activity of the IGF-1R gene and that they increase IGF-1R gene expression via common as well as distinct signaling pathways.

EXPERIMENTAL PROCEDURES
Materials-Cell culture media and LipofectAMINE were purchased from Life Technologies (Basel, Switzerland), and human Ang II was from Sigma. Human recombinant bFGF, PD 98059, genistein, tyrphostin A25, eicosatetrayonic acid (ETYA), SB 203580, and BAPTA/AM were from Calbiochem. The Dual-Luciferase TM Reporter Assay was from Promega (Madison, WI), and salmon sperm DNA was purchased from Stratagene (La Jolla, CA). The enhanced chemiluminescence reagents and the horseradish peroxidase-conjugated anti-rabbit immunoglobulin were from Amersham (Pharmacia Biotech). The polivinylidene fluoride blotting membranes were from Millipore Corp. (Bedford, MA), and the anti-MAPK and phosphospecific anti-MAPK (p42/p44) antibodies were purchased from New England Biolabs (Beverly, MA). The antibody against the ␤-subunit of the IGF-1 receptor was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
Cell Culture-CHO-AT 1 cells (kindly provided by Dr. E. Clauser, INSERM, Paris) stably overexpressing the Ang II AT 1A receptor (49) were grown in Ham's F-12 medium supplemented with 10% heatinactivated fetal bovine serum, 2 mM glutamine, 100 units/ml penicillin, 100 g/ml streptomycin, and 0.75 mg/ml G418 and incubated at 37°C in a humidified 5% CO 2 atmosphere. The parental cell line CHO-K1 (ATCC, Rockville, MD) was cultured under the same conditions but without the addition of G418.
Plasmids and Transfections-The full length promoter of the IGF-1R (Ϫ2350/ϩ640-Luc) and a shorter promoter construct (Ϫ476/ϩ640-Luc) were a generous gift from Dr. H. Werner (National Institutes of Health, Bethesda, MD). Deletion fragments were made from the full-length promoter construct and subcloned upstream of the firefly luciferase cDNA, resulting in fragments extending from nucleotides Ϫ476 to ϩ21, Ϫ416 to ϩ21, Ϫ330 to ϩ21, Ϫ270 to ϩ21, and Ϫ135 to ϩ21. The following constructs have previously been described: dominant negative mutant p21 ras (N17) (50), dominant negative Raf (301) (51), dominant negative MAPKK 1 (MEK1) mutant A221 (52), dominant negative Jun kinase (stress-activated extracellular signal-regulated kinase (SEK1)) (53), dominant negative c-Jun (Tam67) (54), dominant negative STAT1 Tyr 701 (55), dominant negative STAT3 Tyr 705 (56), kinase-deficient Jak2 kinase (57), and dominant negative IB␣ K21/22R (58). To control for transfection efficiency and interwell variation, cells were co-transfected with the internal control vector pRL-TK according to the manufacturer (Herpes simplex virus thymidine kinase promoter region driving Renilla luciferase expression). Cells were plated in 24-well plates and transfected with 1 g of reporter plasmid and 5 ng of pRL-TK/well with LipofectAMINE reagent. In co-transfection experiments with dominant negative Ras, Raf, MEK1, STAT1, STAT3, Jak2, c-Jun, SEK1, and the IB␣ mutant, cells were transfected as described above with the addition of increasing amounts of the above mentioned plasmids. The total amount of DNA transfected was kept constant using salmon sperm DNA. Twenty hours after transfection, the DNA-containing medium was changed to Ham's F-12, and the cells were treated with or without Ang II (100 nM) or bFGF (10 ng/ml). To determine the signaling pathways, transfected cells were incubated with inhibitors for 1 h prior to the addition of Ang II or bFGF. After 24 h, cells were washed and lysed according to the manufacturer's instruction. Firefly and Renilla luciferase activities were measured using an EG & G Berthold luminometer (Bad Wildbad, Germany). Firefly luciferase activity was normalized to the internal control Renilla luciferase (Luc/Ren).
Western Blot Analysis-Cultured CHO-AT 1 or CHO-K1 cells were serum-starved for 24 h prior to the addition of Ang II (100 nM) or bFGF (10 ng/ml) for the times indicated. In some experiments, cells were incubated with the MAPKK inhibitor PD 98059 (10 M), the tyrosine kinase inhibitor tyrphostin (10 M), the p38 MAPK inhibitor SB 203580 (10 M), or ETYA (10 M) 1 h prior to the addition of Ang II or bFGF. Cells were washed in ice-cold phosphate-buffered saline and lysed in lysis buffer containing 150 mM NaCl, 10 mM Tris-Cl, pH 7.4, 5 mM EDTA, 1% Triton X-100, 5 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, 5 mM ⑀-aminocaproic acid, 1 mM sodium orthovanadate, 0.1 M okadaic acid, 0.1 M aprotinin, 10 g/ml leupeptin, and 10 mM NaF. Lysates were subjected to SDS-PAGE on 7.5 or 12% gels, and separated proteins were transferred to polivinylidene fluoride membranes. Blots were blocked with 5% dry milk; incubated with polyclonal anti-MAPK, anti-phosphospecific MAPK (p44, p42), or anti-IGF-1R␤ antibodies; and then incubated with peroxidase-conjugated donkey anti-rabbit antibody. Immunopositive bands were visualized by enhanced chemiluminescence. Purified MAPK protein was included as positive control; nonimmune rabbit IgG was used as negative control.

Effect of Ang II on IGF-1R Promoter Activity and Localization of the Ang II-responsive Element-To measure the effect of
Ang II on IGF-1R gene expression and to localize the Ang II-responsive region, CHO-AT 1 cells were transiently transfected with IGF-1R promoter constructs containing a luciferase reporter gene under the control of the proximal promoter region of the IGF-1R gene together with the Renilla luciformis thymidine kinase expression vector (pRL-TK) as internal control. Ang II (100 nM) significantly increased IGF-1R promoter activity between 2.3-and 7-fold depending on the promoter constructs (Fig. 1B). No effect was observed when the promoterless pOLUC was used (data not shown). The biggest effect induced by Ang II was seen with the promoter construct containing 476 base pairs of the 5Ј-flanking region and 640 base pairs of the 5Ј-untranslated region (UTR) (p(Ϫ476/ϩ640-Luc)), and a lesser response occurred with the construct containing a shorter 5Ј-UTR. It appeared that there were some negative regulatory elements between nucleotides Ϫ2350 and Ϫ476, because the longest reporter construct responded less to Ang II stimulation, compared with (Ϫ476/ϩ640-Luc). When the sequence between nucleotides Ϫ270 and Ϫ135 was deleted, the stimulatory effect of Ang II was greatly diminished, although not completely, suggesting that the major Ang II-responsive element may be located between nucleotides Ϫ270 and Ϫ135 of the 5Ј-flanking region.
Effect of bFGF on IGF-1R Gene Transcription-The stimulatory effect induced by bFGF in CHO-AT 1 cells ranged between 1.7-and 3.2-fold (Fig. 1C), and the same was found in the parental cell line CHO-K1 (data not shown). Previous studies have suggested that the bFGF-response element may be located between nucleotides Ϫ476 and Ϫ188 of the 5Ј-flanking region (59). In our CHO-AT 1 cells, the Ϫ476/ϩ640 construct responded well to bFGF (ϳ3-fold increase in luciferase activity). In contrast to Ang II, there was no evidence of a repressor sequence in the larger construct (Ϫ2350/ϩ640). The removal of the majority of the 5Ј-UTR resulted similarly in a reduction in basal activity, but the bFGF response persisted. Progressive deletion of the cis-acting sequence up to Ϫ135/ϩ21 still yielded a ϳ2-fold increase in luciferase activity after bFGF treatment (Fig. 1C).
Effect of Ang II and bFGF on IGF-1R Protein Levels-To assess whether Ang II or bFGF increased IGF-1R protein levels, cell lysates of cells treated with or without the corresponding growth factor were assayed for IGF-1R protein levels by Western immunoblot. Fifty micrograms were fractionated on a 7.5% reducing SDS-PAGE. Ang II and bFGF increased IGF-1R protein expression, with a maximum seen after 24 h of incubation (Fig. 2). Incubation with Ang II or bFGF for 48 h did not further increase IGF-1R protein levels (not shown).

Common and Distinct Signaling Pathways of Ang II and bFGF on the Stimulation of IGF-1R Gene Expression and on
Protein Level-The Ang II AT 1 receptor belongs to the family of G-protein-coupled seven-transmembrane domain receptors (60), whereas the bFGF receptor is characterized by a single transmembrane domain that has intrinsic tyrosine kinase activity (37)(38)(39). To define signal transduction pathways by which Ang II and bFGF stimulate transcription of the IGF-1R, cells were transfected with the indicated promoter reporter constructs and either pretreated with various inhibitors or co-transfected with increasing doses of dominant negative expression constructs prior to the addition of Ang II or bFGF. The protein-tyrosine kinase inhibitor tyrphostin A25 (10 M) de- creased the Ang II response by 60 -70%, suggesting that protein tyrosine phosphorylation is involved in the pathway of Ang II (Fig. 3A). Similar results were obtained when genistein, another tyrosine kinase inhibitor was used (54 Ϯ 1.1 to 82 Ϯ 5.3% inhibition of the Ang II response, depending on the promoter construct used; mean Ϯ S.E. of three experiments).
We have previously shown that Ang II signals through a lipoxygenase-dependent pathway to increase macrophage-mediated oxidative modification of low density lipoprotein (61). Therefore, we were interested to see if Ang II would also signal through that pathway to increase IGF-1R transcription. Indeed, ETYA almost completely inhibited the stimulatory effect of Ang II on IGF-1R transcription (representative experiments as follows: for Ϫ476/ϩ640, control, 9.5; Ang II, 35.6; and ETYA/ Ang II, 17.9 Luc/Ren, respectively; or for Ϫ416/ϩ21, control, 5.4; Ang II, 16.5; and ETYA/Ang II, 7.2 Luc/Ren), whereas it had no effect on the increase induced by bFGF (Ϫ2350/ϩ640: control, 10.1 Ϯ 3.6; bFGF, 16.8 Ϯ 1.4; ETYA, 14.7 Ϯ 4.8; and ETYA/bFGF, 19.8 Ϯ 5.9, Luc/Ren respectively (mean Ϯ S.D. of two experiments). ETYA alone had no effect on basal luciferase activity (data not shown). In agreement with the above mentioned experiments using genistein or ETYA to block the Ang II stimulation of IGF-1R gene transcription, both blockers also inhibited Ang II-induced MAPK phosphorylation, suggesting that a tyrosine kinase-and a lipoxygenase-dependent step are upstream of MAPK activation (Fig. 3B).
It is documented that Ang II activates the MAPK pathway in vascular smooth muscle cells (62,63) and that this activation is partially dependent on protein kinase C (62) and apparently requires prior activation of a Ca 2ϩ -dependent tyrosine kinase (64). Co-transfection experiments with dominant negative Ras, Raf, and MEK1 suggested that the Ras-Raf-MAPKK pathway is involved in the transcriptional activation of the IGF-1R by Ang II because all inhibited the luciferase response induced by Ang II (Fig. 4), whereas the empty vectors had no effect (data not shown). Similarly, the specific MAPKK inhibitor PD 98059 (100 M) blocked the Ang II response to control levels in all promoter constructs without having any effect on basal luciferase activity (Ϫ2350/ϩ640: control, 3.3 Ϯ 0.8; Ang II, 13.9 Ϯ 2.7; PD 98059/Ang II, 5.1 Ϯ 1.2 Luc/Ren, respectively; Ϫ135/ ϩ21: control, 6.8 Ϯ 1.1; Ang II, 20.7 Ϯ 2.2; and PD 98059/Ang II, 8.6 Ϯ 1.9 Luc/Ren, respectively (mean Ϯ S.E. of five independent experiments)). In contrast, while the response induced by Ang II required the p44/p42 MAPK activation, the p38 MAPK inhibitor SB 203580 had no effect on the Ang II response, suggesting that p38 MAPK was not involved (Ϫ476/ ϩ21: control, 6.5 Ϯ 0.6; SB 203580, 5.7 Ϯ 0.6; Ang II, 28.8 Ϯ 1.8; and SB 203580/Ang II, 24.2 Ϯ 1.7 Luc/Ren, respectively (mean Ϯ S.E. of four experiments)). To confirm the specificity of these findings, cells were treated with or without Ang II for various times, and total proteins were immunoblotted with phosphospecific antibodies against p44 and p42 (extracellular signal-regulated kinases 1 and 2). Ang II induced a rapid phosphorylation of p44/p42 already after 2 min, with a maximum at 5 min. PD 98059 completely inhibited the Ang II-induced phosphorylation of p44/p42 (data not shown). It is known that MAPKs in turn phosphorylate numerous cellular proteins, including c-Jun among many others (65). When dominant negative c-Jun (Tam67) was co-transfected with the p(Ϫ476/ϩ21-Luc), it completely reduced the stimulatory effect of Ang II to control values, whereas the empty vector had no effect (Fig. 5). This inhibitory effect of dominant negative c-Jun could also be observed with the smaller IGF-1R promoter constructs p(Ϫ270/ ϩ21-Luc) and p(Ϫ135/ϩ21-Luc) (data not shown).
We have seen that the Ang II-induced IGF-1R gene expression is calcium-dependent and is mediated via a redox-sensitive pathway. 2 Indeed, intracellular Ca 2ϩ chelation using BAPTA/AM (10 M) decreased the stimulatory effect of Ang II to control levels without having any effect on basal luciferase activity, suggesting that intracellular Ca 2ϩ is required (Ϫ476/ ϩ640: control, 10.8 Ϯ 1.3; BAPTA/AM, 13.4 Ϯ 1.9; Ang II, 31.5 Ϯ 2.5; and BAPTA/AM/Ang II, 8.7 Ϯ 7.0 Luc/Ren, respectively (mean Ϯ S.D. from two experiments)). Also, reactive oxygen species have been shown to act as second messengers in Ang II-induced signaling (66) and activation of the transcription factor NF-B is redox-sensitive (67). Co-transfection of the IGF-1R promoter construct p(Ϫ476/ϩ640-Luc) with K21/22R, the IB␣ mutant that shows a defect in degradation and in ubiquitin conjugation and therefore inhibits translocation of NF-B to the nucleus (58), completely inhibited the Ang IIinduced increase in IGF-1R transcription (Fig. 6A). This was also true when the shorter construct p(Ϫ270/ϩ21-Luc) was used (data not shown). Interestingly, the IB␣ mutant had no effect on the Ang II response when the short IGF-1R promoter construct p(Ϫ135/ϩ21-Luc) was used, suggesting that a puta-

FIG. 3. Effect of protein tyrosine kinase inhibition and lipoxygenase inhibition on the Ang II-induced increase in IGF-1R transcription and on AngII-induced MAPK phosphorylation. A,
CHO-AT 1 cells were transfected with the p(Ϫ476/ϩ640-Luc) and the p(Ϫ270/ϩ21-Luc) construct and the pRL-TK plasmid to correct for interwell variation. Twenty hours after transfection, cells were incubated with or without tyrphostin A25 (10 M) for 1 h prior to the addition of Ang II (100 nM). Twenty four hours later, cells were lysed, and luciferases were measured. Data are presented as Luc/Ren and are mean Ϯ S.E. of three separate experiments. Tyrphostin A25 alone had no effect on basal luciferase activity. Tyrphostin A25 had the same inhibitory effect on all other IGF-1R promoter constructs (data not shown). B, cells were treated with or without genistein (60 nM) or ETYA (10 M) prior to the addition of Ang II for 24 h. Ten micrograms of total protein were subjected to SDS-PAGE on a 12% reducing gel and probed with anti-phosphospecific p44/p42 antibody. tive NF-B site is located 5Ј of the nucleotide Ϫ135 (Fig. 6B). Quite in contrast to Ang II, however, co-transfection with IB␣ lysine mutant did not decrease the bFGF-induced activation of IGF-1R expression (data not shown).
While both Ang II and bFGF stimulated IGF-1R gene transcription (Fig. 1), the signal pathway by which these two growth factors mediate the increase in IGF-1R expression showed common but also distinct features. Thus, tyrphostin A25 reduced the stimulatory effect of bFGF on IGF-1R transcription by 44.4 Ϯ 3.8% (Ϫ476/ϩ21: control, 3.8 Ϯ 0.10; tyrphostin A25, 3.1 Ϯ 0.08; bFGF, 11.8 Ϯ 0.46; and tyrphostin A25/bFGF, 6.5 Ϯ 0.29, respectively (mean Ϯ S.E. of four experiments)). The Ras-Raf-MAPK pathway seemed also to be required in the transcriptional activation of the IGF-1R by bFGF, as we found with Ang II. Dominant negative Ras, Raf, and MEK1 inhibited the bFGF-induced increase in luciferase activity to a similar degree as seen with Ang II (Fig. 7). Accordingly, the MAPKK inhibitor PD 98059 (100 M) completely abrogated the bFGF-induced stimulation of IGF-1R transcription and inhibited the bFGF-induced phosphorylation of p44/p42 MAPK (data not shown). Furthermore, BAPTA/AM inhibited the bFGF-induced stimulation of IGF-1R expression to control values (representative experiment Ϫ476/ϩ640: control, 9.9; bFGF, 33.3; and BAPTA/bFGF, 13.7 Luc/Ren, respectively); however, the p38 MAPK inhibitor SB 203580 did not inhibit the bFGF response similarly to Ang II-stimulated IGF-1R expression (data not shown). The c-Jun N-terminal kinase/stress-activated protein kinase (JNK/SAPK) has been shown to phosphorylate and regulate the activity of several transcription factors includ-ing c-Jun, ELK-1, and ATF-2 (68 -72). The JNK/SAPK is phosphorylated, resulting in its activation by JNK kinase (JNK kinase/SEK1) (53,(73)(74)(75). Increasing doses of dominant negative SEK1 expression construct produced a dose-dependent decrease in the IGF-1R transcriptional activity induced by bFGF when the p(Ϫ476/ϩ21-Luc) reporter construct was used, suggesting that the SEK1/JNK/SAPK pathway was involved (Fig. 8A). It is of interest that the same dominant negative SEK1 did not have any effect on IGF-1R transcriptional stimulation by Ang II (data not shown), despite the inhibitory effect of dominant negative c-Jun in the response to Ang II (Fig. 5). It is of note that dominant negative c-Jun also dose-dependently reduced the stimulatory effect seen with bFGF (Fig. 8B).
More recently, a novel nuclear signaling pathway has been described that regulates a large family of transcription factors called STATs (76). This pathway, initially described for the interferon receptors, has subsequently been shown to be involved in hormone and growth factor signaling, such as growth hormone (77), Ang II (36), or bFGF (45,46). We have previously shown that Ang II directly stimulates the Jak/STAT pathway in rat aortic smooth muscle cells by phosphorylation of the intracellular Jak2 kinase and its substrates STAT1 and STAT2 (36). We therefore investigated whether kinase-deficient Jak2 or dominant negative STAT1 Tyr 701 and STAT3 Tyr 705 mutants could inhibit the increase in IGF-1R transcriptional activity induced by Ang II or bFGF. While all mutants had no effect on the Ang II response (data not shown), they greatly reduced the effect seen with bFGF ( Fig. 9), suggesting that the Jak/STAT pathway is involved in the response to bFGF but not to Ang II in this cell model.
Similarly to the transcriptional assays, the MAPKK inhibitor PD 98059 and the protein tyrosine kinase inhibitor tyrphostin A25 decreased the stimulatory effect of Ang II or bFGF on IGF-1R protein expression, whereas the lipoxygenase inhibitor ETYA blocked only the Ang II response. Furthermore, the p38 MAPK inhibitor SB 203580 did not inhibit the Ang II or bFGF effect (Fig. 10), confirming the results observed in the reporter assays. DISCUSSION It has previously been demonstrated that growth factors such as PDGF, thrombin, Ang II, and bFGF increase IGF-1R on vascular smooth muscle cells and that this effect is transcriptionally regulated (23,25,28). Furthermore, the ability of Ang II to up-regulate IGF-1R is a critical determinant of its mitogenic activity on vascular cells, since the Ang II-induced increase in DNA synthesis was inhibited by IGF-1R-specific antisense oligonucleotides (29). The present studies show by which mechanisms and signaling pathways Ang II and bFGF increase IGF-1R gene transcription. By deletional analysis of the IGF-1R promoter region, we determined that the Ang IIresponsive region is located in the proximal promoter, between nucleotides Ϫ270 and Ϫ135 upstream of the transcription start site, as is the bFGF-responsive element. In addition, stimulation of IGF-1R gene promoter activity by Ang II or bFGF in transient transfection experiments correlates well with its effect on endogenous IGF-1R protein levels. Both increased IGF-1R protein expression after 8 -24 h. This is in good agreement with the previous reports of Du et al. (48) and Ververis et al. (25), which showed that Ang II and bFGF caused a significant increase in IGF-1R mRNA peaking at 3 h and 6-9 h, respectively. Of note, Hernandez-Sanchez et al. (59) reported that the bFGF-responsive element was located between nucleotides Ϫ476 and Ϫ188. These findings are somewhat different from ours; however, our studies were performed using different cells, and our deletion constructs contained less 5Ј-UTR sequence. Our data indicate loss of basal activity between nucleotides Ϫ476 and Ϫ135 but conservation of a bFGF-responsive element.
There has been significant interest generated by the observation that growth factors and cytokines, which possess structurally different receptors, with or without intrinsic tyrosine kinase activity, may signal through a common pathway to the nucleus. In order to define the mechanisms and the signaling cascade involved in the Ang II or bFGF regulation of IGF-1R gene expression, we transiently transfected various IGF-1R promoter constructs into CHO-AT 1 or CHO-K1 cells and used different approaches to block the signaling pathways at different levels. Our findings clearly show that Ang II and bFGF share common but also quite distinct pathways. Thus, both Ang II and bFGF increase IGF-1R transcriptional activity via the Ras-Raf-MAPKK-MAPK pathway, since transfection of dominant negative expression constructs for Ras, Raf, or MEK1 dose-dependently reduced the stimulatory effects of these growth factors on IGF-1R promoter activity, whereas they had no effect on IGF-1R promoter activation in the absence of Ang II or bFGF. Further evidence for the involvement of this signaling pathway in the activation of the IGF-1R promoter by Ang II or bFGF was provided by experiments using PD 98059. This compound, which is a specific inhibitor of MAPKK phosphorylation and activation (78,79), completely reversed the stimulatory effect on luciferase activity induced by Ang II or bFGF. Furthermore, analysis at the protein level clearly demonstrated that both Ang II and bFGF induced a rapid phosphorylation of MAPK, which was inhibited by upstream blockade of MAPKK by PD 98059, and inhibition of the MAPKK reduced the stimulatory effect of Ang II and bFGF on IGF-1R protein levels. Thus, the Ras-Raf-MAPK pathway is clearly required for Ang II and bFGF induction of IGF-1R gene and protein expression.
Although the Ang II AT 1 receptor does not possess intrinsic tyrosine kinase activity, its activation leads to intracellular second messenger protein tyrosine phosphorylation by cytosolic tyrosine kinases (80). Thus, our finding that the protein-tyrosine kinase inhibitors genistein and tyrphostin A25 inhibited the Ang II-induced stimulation of IGF-1R gene expression and phosphorylation of MAPK demonstrates a requirement for protein-tyrosine kinase(s) in Ang II-stimulated IGF-1R expres- sion. We have previously shown that lipoxygenases may be involved in the signaling pathway of Ang II (61). Our present study demonstrates that Ang II-induced activation of the IGF-1R promoter requires lipoxygenase activity, since this stimulation was blocked by ETYA, a lipoxygenase inhibitor (81). ETYA and genistein not only reduced the stimulatory effect of Ang II on IGF-1R promoter transcriptional activity to basal levels but also inhibited the Ang II-induced phospho-rylation of MAPK, suggesting that protein tyrosine phosphorylation and lipoxygenase activation is upstream of MAPK activation. Consistent with the results observed in the transcriptional assays, tyrosine kinase and lipoxygenase inhibition abolished the increase in IGF-1R protein level induced by Ang II.
The regulation of IGF-1R transcription is not well understood. The IGF-1R gene promoter lacks TATA or CAAT motifs; Increasing amount of dominant negative STAT1 Y701F or STAT3 Y705F and the p(Ϫ270/ϩ21-Luc) as well as kinase-deficient Jak2 together with p(Ϫ476/Ϫ21-Luc) were transfected as described. Twenty hours later, cells were treated with or without bFGF (10 ng/ml). Data are mean Ϯ S.E. of three experiments. thus, transcription starts from a unique initiator sequence (82). The present experiments are therefore of interest in characterizing the signaling pathway of Ang II or bFGF in stimulating the IGF-1R promoter and in determining the interaction between transcription factors and the IGF-1R promoter. The region of the IGF-1R promoter extending from nucleotide Ϫ2350 in the 5Ј-flanking region to nucleotide ϩ640 in the 5Ј-UTR, contains putative consensus sequences for a number of well defined regulatory elements, including Egr-1 (83) and Sp1 (84), as well as a PDGF-responsive element (85) and potential AP-2 (86), AP-1, and NF-B sites. 3 To gain insights into the promoter region of the IGF-1R gene responsive to Ang II or bFGF, cells were transfected with expression vectors encoding dominant negative forms of IB (lysine mutant), JNK kinase, c-Jun, and the transcription factors STAT1 and STAT3. The results of these experiments suggest that one of the other common pathways by which Ang II or bFGF increases IGF-1R gene transcription was the involvement of c-Jun. Thus, our data indicated that dominant negative c-Jun dose-dependently inhibited the Ang II-as well as the bFGF-induced increase in IGF-1R promoter activity. c-Jun is one of the components of the transcription factor AP-1, its best known partner being c-Fos (87). However, the activity known as AP-1 can consist of heterodimers between any of the Jun proteins and any of the Fos proteins (87). The finding that dominant negative c-Jun inhibited the increase in IGF-1R promoter activity by Ang II or bFGF, whereas dominant negative JNK kinase (SEK1), which activates JNK/SAPK and ultimately activates c-Jun, dose-dependently blocked the effect induced by bFGF and not by Ang II can be explained by differential pathways whereby these two growth factors signal. It is possible that although Ang II and bFGF both activate MAP kinases, Ang II may predominantly induce c-Fos through the MAPK pathway, whereas bFGF induces c-Jun through activation of the JNK/SAPK pathway. Obviously, more data are needed to fully comprehend and define this difference.
Another important mitogenic cascade that is activated by cytokines and growth factors involves the Jak family of cytoplasmic tyrosine kinases (76,88). Jak-mediated tyrosine phosphorylation of STATs promotes the translocation of these growth factors to the nucleus, where they bind to specific DNA motifs and induce c-fos gene transcription (76, 88 -90). Marrero et al. (36) have previously demonstrated that Ang II stimulates tyrosine phosphorylation of Jak isoforms, tyrosine kinase activity of Jak2, and tyrosine phosphorylation of STATs in vascular smooth muscle cells. Using a kinase-deficient Jak2 or dominant negative STAT1 or STAT3, we were unable to inhibit the Ang II-induced stimulation of the IGF-1R promoter in transient transfection assays using CHO-AT 1 cells, indicating that the Jak/STAT pathway is not involved in Ang II-induced IGF-1R gene expression. Quite in contrast to Ang II, kinasedeficient Jak2, dominant negative STAT1, and STAT3 completely inhibited the stimulatory effect of bFGF on IGF-1R transcription, whereas neither empty vectors nor both dominant negative expression constructs had any effect on the IGF-1R transcriptional activity in the absence of bFGF. These findings demonstrate that the Jak/STAT pathway and more precisely Jak2, STAT1, and STAT3 are required in the transcriptional activation of the IGF-1R promoter by bFGF but are not involved in the stimulation by Ang II.
The involvement of the transcription factor NF-B seems to be restricted to the stimulatory effect induced by Ang II and not by bFGF, since the IB␣ mutant only inhibited the activation of luciferase activity by Ang II. Furthermore, it was found that by deleting nucleotides between Ϫ270 and Ϫ135, the inhibitory effect of the IB␣ mutant on the Ang II-induced effect was lost, implying that a putative NF-b site is located between nucleotides Ϫ270 and Ϫ135. These results suggest that Ang II and bFGF utilize distinct and specific transcription factors to stimulate the IGF-1R gene promoter.
In conclusion, we have characterized the regulation of IGF-1R gene expression by Ang II and bFGF and the main signaling pathways by which these growth factors increase IGF-1R transcriptional activity and IGF-1R protein expression. Both growth factors increase IGF-1R promoter activity by acting on the proximal promoter region upstream of the transcription start site. Although Ang II and bFGF possess structurally different receptors, they transduce signaling through common pathways, notably the Ras-Raf-MAPKK-MAPK and c-Jun pathways. However, they also use unique signaling pathways, such as lipoxygenase-mediated MAPK activation or the involvement of the transcription factor NF-B in the case of Ang II or the JNK/SAPK cascade and the Jun kinase and Jak/STAT pathway in the bFGF-induced regulation of IGF-1R gene expression. These studies have important implications for understanding growth-stimulatory effects of these growth factors.