Interferon regulatory factor-1 up-regulates angiotensin II type 2 receptor and induces apoptosis.

The expression of the angiotensin II type 2 (AT2) receptor is developmentally and growth regulated. In cultured R3T3 cells, expression of this receptor is markedly induced at the confluent state and with serum deprivation. In this study we demonstrated that the removal of serum from culture media resulted in the induction of apoptosis in these cells and the addition of angiotensin II further enhanced apoptosis. We have previously identified an interferon regulatory factor (IRF) binding motif in the mouse AT2 receptor gene promoter region. In this report, we observed that serum removal increased IRF-1 expression, with a rapid and transient decrease of IRF-2. To prove that the changes in IRFs after serum removal mediated apoptosis and up-regulated AT2 receptor, we transfected antisense oligonucleotides for IRF-1 or IRF-2 into R3T3 cells and observed that IRF-1 antisense oligonucleotide attenuated apoptosis and abolished the up-regulation of AT2 receptor. IRF-2 antisense oligonucleotide pretreatment did not affect the onset of apoptosis after serum removal; instead, it increased AT2 receptor binding and enhanced angiotensin II-mediated apoptosis. Taken together, these results suggest that increased IRF-1 after serum starvation contributes to the induction of apoptosis and that increased IRF-1 up-regulates the AT2 receptor expression after serum starvation, resulting in enhanced angiotensin II-mediated apoptosis.

The expression of the angiotensin II type 2 (AT 2 ) receptor is developmentally and growth regulated. In cultured R3T3 cells, expression of this receptor is markedly induced at the confluent state and with serum deprivation. In this study we demonstrated that the removal of serum from culture media resulted in the induction of apoptosis in these cells and the addition of angiotensin II further enhanced apoptosis. We have previously identified an interferon regulatory factor (IRF) binding motif in the mouse AT 2 receptor gene promoter region. In this report, we observed that serum removal increased IRF-1 expression, with a rapid and transient decrease of IRF-2. To prove that the changes in IRFs after serum removal mediated apoptosis and up-regulated AT 2 receptor, we transfected antisense oligonucleotides for IRF-1 or IRF-2 into R3T3 cells and observed that IRF-1 antisense oligonucleotide attenuated apoptosis and abolished the up-regulation of AT 2 receptor. IRF-2 antisense oligonucleotide pretreatment did not affect the onset of apoptosis after serum removal; instead, it increased AT 2 receptor binding and enhanced angiotensin II-mediated apoptosis. Taken together, these results suggest that increased IRF-1 after serum starvation contributes to the induction of apoptosis and that increased IRF-1 up-regulates the AT 2 receptor expression after serum starvation, resulting in enhanced angiotensin IImediated apoptosis.
Angiotensin II (Ang II) 1 exerts various actions in its diverse target tissues controlling vascular tone, hormone secretion, tissue growth, and neuronal activities. Most of the known effects of Ang II in adult tissues are mediated by the Ang II type 1 (AT 1 ) receptor. Recently, a second receptor subtype known as AT 2 receptor has been cloned by us and others (1,2). The AT 2 receptor is abundantly and widely expressed in fetal tissues, but present only in at limited levels in adult tissues such as adrenal medulla, specific brain regions, uterine myometrium, heart, and atretic ovarian follicles (3)(4)(5)(6)(7)(8). The highly abundant expression of this receptor during embryonic and neonatal growth and quick disappearance after birth has led to the suggestion that this receptor is involved in growth, development, and/or differentiation. Our studies using in vivo gene transfer of the rat AT 2 receptor into an injured rat carotid artery have demonstrated an attenuation of neointimal hyperplasia resulting from AT 2 receptor transgene overexpression (9), and Stoll et al. (10) also demonstrated that the antiproliferative actions of the AT 2 receptor offset the growth promoting effects mediated by the AT 1 receptor.
Greater than 99.9% of the ovarian follicles present at birth undergo atresia, an event dependent on apoptosis or "programmed cell death." Interestingly, AT 2 receptor expression is tightly associated with ovarian follicular atresia (3,8). AT 2 receptor is also abundantly expressed in immature brain and some specific regions of the brain (5,6). Approximately half of the neurons produced during embryogenesis die by apoptosis before adulthood. These results have led us to examine the role of AT 2 receptor in apoptosis. We recently demonstrated that the activation of the AT 2 receptor induces apoptosis in a rat pheochromocytoma cell line (PC12W) and confluent mouse fibroblast cell line (R3T3 cell) (11). These cells have been shown to express abundant AT 2 receptor (12)(13)(14). We further observed that higher AT 2 receptor expression is especially associated with apoptosis (11).
In R3T3 cells, AT 2 receptor expression is regulated by the growth state of the cells. AT 2 receptor binding density is very low in actively growing cells, but increases markedly in the confluent, quiescent state (13,14). We recently reported that the growth-regulated AT 2 receptor expression in R3T3 cells is transcriptionally controlled by the competitive binding of interferon regulatory factor-1 (IRF-1) and IRF-2 for an IRF binding sequence in the AT 2 receptor promoter region (15). IRF-1 and IRF-2 are structurally related, particularly in their aminoterminal regions that confer DNA binding specificity. They recognize the same DNA sequence elements (IRF-Es) (16,17). IRF-1 functions as a transcriptional activator, whereas IRF-2 represses IRF-1 function by competing for the same cis-element (16 -19). It has been shown that the growth state of certain cells depends on a balance between IRF-1 and IRF-2, i.e. a change in the IRF-1/IRF-2 ratio perturbs cell growth control (20 -23).
In this study, we examined the role of AT 2 receptor expression in IRF-regulated apoptosis. We hypothesized that changes in the ratio of IRFs under specific conditions regulate AT 2 receptor expression which subsequently influences apoptosis. We studied the effect of serum starvation on R3T3 cells and observed that an increase in IRF-1 with transient decrease in IRF-2 stimulated AT 2 receptor expression and enhanced AT 2 receptor-mediated apoptosis.

EXPERIMENTAL PROCEDURES
Cell Culture-R3T3 cells were cultured as described (13) using Dulbecco's modified essential medium supplemented with 10% fetal bovine * This work was supported in part by National Institutes of Health Grants HL46631, HL35252, HL35610, HL48638, and HL07708; by the American Heart Association Bugher Foundation Center for Molecular Biology in the Cardiovascular System; and by a grant from Ciba-Geigy, Basel, Switzerland. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
serum. Cells were seeded at 7 ϫ 10 4 cells/T-75 flask (Falcon) for transfection experiments for internucleosomal DNA fragmentation (DNA laddering), and cells were seeded on a 24-well plate (surface area, 2 cm 2 ) (Falcon) at 2.5 ϫ 10 4 cells/well for receptor binding studies. Two days after reaching confluent state, the cells were used for experiments. Cell number was counted by Coulter counter.
Transfection of Oligonucleotides to R3T3 Cells-On the day of transfection, the medium was changed to fresh medium and incubated for 1 h at 37°C. The transfection was performed using LipofectAMINE Reagent (Life Technologies, Inc.), 3:1 (w/w) liposome formulation of the polycationic lipid 2,3-dioleyloxy-N-[2(sperminecarboxamide)ethyl]-N,N-dimethyl-1-propanaminium trifluoroacetate (DOSPA) and the neutral lipid dioleoyl phosphatidylethanolamine (DOPE), according to the manufacturer's instructions. Cells were incubated with the oligonucleotides (0.5 M) and LipofectAMINE (1:3, w/w) for 3 h, and the transfection reagent was replaced with culture medium as described in each figure legend.
Receptor Binding Assay-Cells on the wells were washed twice with 500 l of phosphate-buffered saline containing 0.1% bovine serum albumin. Reactions were performed with 0.2 nM 125 I-CGP42112A (2,176 Ci mmol Ϫ1 , Peptide Radioiodination Center, Washington State University) at 37°C for 1 h. After incubation, cells were washed with ice-cold phosphate-buffered saline twice and solubilized with 200 l of 0.5 N NaOH and solubilized material was counted for radioactivity. Nonspecific binding was defined in the presence of 1 M unlabeled CGP42112A.
Preparation of cDNAs for IRF-1 and IRF-2 by Reverse Transcription-PCR-Total RNA was prepared from cultured R3T3 cells using RNAzol (Tel-Test), reverse transcribed using reverse transcriptase and random hexamers (Perkin-Elmer), and applied to PCR. PCR primers for IRF-1 used were as follows: 5Ј-CCTGATGACCACAGCAGTTAC-3Ј and 5Ј-CTTCATCTCCGTGAAGACATG-3Ј. For IRF-2, PCR primers were 5Ј-GCGGTCCTGACTTCAGCTATA-3Ј and 5Ј-CTTCTTGATGACACTG-GCCGG-3Ј. PCR reaction was carried out with 30 cycles of 1 min f denaturation at 94°C, 1 min of annealing at 55°C, and 3 min of extension at 72°C, followed by 15 min of final extension step. PCRamplified DNAs were subcloned into pCR™II Vector (Invitrogen). cDNA probes for Northern blotting were prepared from these plasmid vector by restriction enzyme cut by EcoRI.
Northern Blot Analysis-Total RNA was prepared from cultured R3T3 cells using RNAzol (Tel-Test) 24 h after oligonucleotide treatment. RNA (20 g/lane) was separated by electrophoresis and transferred onto a nylon membrane (Amersham), and hybridization was carried out using a 32 P-labeled IRF-1 and IRF-2 cDNA, a 32 P-labeled HindIII-NsiI fragment of mouse AT 2 receptor cDNA (25), or 0.78-kb PstI-XbaI fragment of a human glyceraldehyde-3-phosphate dehydrogenase clone. Densitometric analysis of autoradiograms was performed by scanning densitometer (GS300, Hoeffer) and NIH Image software.
Internucleosomal DNA Fragmentation (DNA Ladder)-DNA extraction, subsequent 3Ј end-labeling of DNA, gel electrophoresis, and quantitation of DNA fragmentation were performed as described previously (26). Briefly, 500 ng of DNA prepared from each treated cells was FIG. 1. IRF-1 and IRF-2 mRNA expression after serum starvation. Total RNA was prepared from cultured confluent R3T3 cells 3, 6, 12, 24, and 48 h after serum removal. RNA (20 g/lane) was separated by electrophoresis and hybridized sequentially with a probe for IRF-1, IRF-2, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as internal control to standardize the amount of total RNA actually blotted onto the membrane. Panel A is a representative of data obtained in four different experiments (IRF-1, 1-day autoradiographic exposure; IRF-2, 2-day autoradiographic exposure). In panel B, the signal density of each RNA sample hybridized to IRF-1 and IRF-2 was divided by that hybridized glyceraldehyde-3-phosphate dehydrogenase. The corrected density for each time point is represented as percent of the value obtained before serum depletion (time 0). The values were expressed as mean Ϯ S.D. obtained from four separate experiments. * shows p Ͻ 0.01 compared with time 0.
end-labeled with [␣-32 P]ddATP (Amersham) and terminal transferase (Boehringer Mannheim) for 60 min at 37°C. Labeled DNA was loaded onto a 2% agarose gel, separated by electrophoresis, and autoradiographed. The amount of radiolabeled ddATP incorporated into low (Ͻ20 kb) molecular weight DNA fractions were quantitated by cutting the respective fraction of DNA from the dried gel and counting in a ␤-counter. The results were expressed as percentage of the counts in control samples as described in each figure legend.
Chromatin Binding Dye Staining-R3T3 cells seeded onto six-well plates (Falcon) were used for this experiment. Chromatin binding dyes (Hoechst 33342 and propidium iodide (Molecular Probes)) were added to the serum-free medium at a concentration of 5 or 10 M to examine the morphological changes of nuclei. After incubation at 37°C for 1 h, cells were collected. After centrifugation, the pellet was resuspended in phosphate-buffered saline and cells were viewed under UV microscopy.
Statistical Analysis-All values are expressed as mean Ϯ S.D. Statistical significance was assessed by analysis of variance (ANOVA) followed by Scheffe's test. p Ͻ 0.01 was considered to be significant.

Role of IRF-1 and IRF-2 in AT 2 Receptor: Up-regulation after
Serum Growth Factor Removal-We first examined the changes of IRF-1 and IRF-2 gene expressions after serum removal which induces apoptosis in this cell line (11). As shown in Fig. 1, expression of IRF-1 increased in a time-dependent manner and showed a maximum at 24 h after serum removal, whereas IRF-2 mRNA transiently decreased 3 h after serum removal and returned to basal level at 24 h. To document that these change in IRF-1 and IRF-2 contributed to the increased AT 2 receptor expression, we treated confluent R3T3 cells with double-stranded oligonucleotides spanning the IRF-Es in the mouse AT 2 receptor promoter region (15) as transcriptional factor decoy, which could bound with IRF-1 and IRF-2 in R3T3 cells (15). This strategy is based on the competition for binding to a specific transcription factor between a cis-element present in an endogenous target gene and an exogenously added oligonucleotide corresponding to that cis-sequence. One day after transfection of the oligonucleotides, the cells were placed in serum-free medium, and AT 2 receptor binding was determined 3 days later by radioligand receptor binding assay. As shown in Fig. 2A, serum depletion enhanced AT 2 receptor binding. Pretreatment with IRF-binding decoy oligonucleotide attenuated the increase of AT 2 receptor in response to serum removal. IRF-binding decoy oligonucleotide treatment did not change the AT 2 receptor density in serum-fed R3T3 cells. Mutant IRF-FIG. 2. Effects of oligonucleotides spanning IRF binding sequence in mouse AT 2 receptor promoter region and antisense oligonucleotides for IRF-1 and IRF-2 on AT 2 receptor expression after serum starvation. Two double-stranded oligonucleotides (IRF and mutant IRF-binding decoy oligonucleotides) (see "Experimental Procedures") were transfected into confluent R3T3 cells, and 1 day after oligonucleotides transfection, serum was removed from the medium. The AT 2 receptor binding was determined 3 days after serum removal by radioligand binding assay. A, the values were expressed as mean Ϯ S.D. obtained from five different culture wells. Total RNA was prepared from R3T3 cells 24 h after serum removal or serum-fed cells. RNA (20 g/lane) was separated by electrophoresis and hybridization was carried with a probe for the AT 2 receptor or for glyceraldehyde-3phosphate dehydrogenase (GAPDH) as a control. Similar results were obtained in three different experiments (B). Antisense oligonucleotides for IRF-1 and IRF-2 were transfected into R3T3 cells. One day after   FIG. 3. The effect of antisense oligonucleotides for IRF-1 and IRF-2 on IRF expression after serum removal. Antisense oligonucleotides for IRF-1 and IRF-2 were transfected into R3T3 cells in confluent state, and then serum was removed from the medium 1 day after oligonucleotide transfection. The cell were harvested 12 h after serum depletion, and the cell lysates were prepared. Cell lysates were resolved by 12% SDS-polyacrylamide gel electrophoresis, electroblotted onto nitrocellulose membrane, and immunoblotted with IRF-1 or IRF-2 antibody. Antibodies were detected by horseradish peroxidase-linked secondary antibody using ECL system. S, sense; AS, antisense. oligonucleotide transfection, serum was removed from the medium. The AT 2 receptor binding was determined 3 days after serum removal by radioligand binding assay (C). The values were expressed as mean Ϯ S.D. obtained from five different cell culture wells. * shows p Ͻ 0.01 compared with LipofectAMINE transfection alone. binding decoy oligonucleotide treatment did not affect AT 2 receptor density. We also measured AT 2 receptor mRNA levels at 24 h after serum removal and observed that pretreatment with IRF-binding decoy oligonucleotide inhibited the AT 2 receptor mRNA increase (Fig. 2B), supporting the contention that the change in the ratio of IRF-1/IRF-2 after serum depletion is important for the transcriptional up-regulation of AT 2 receptor expression.
To prove that the changes of IRF-1 and IRF-2 expression after serum depletion regulates the AT 2 receptor expression, we transfected antisense oligonucleotides for IRF-1 or IRF-2 into R3T3 cells 24 h before serum removal. To confirm whether these antisense oligonucleotides inhibit IRFs expression, we examined the IRFs protein expression 12 h after serum depletion. As shown in Fig. 3, consistent with the mRNA expression of IRF-1 (Fig. 1A), we observed that IRF-1 protein was upregulated after serum depletion and IRF-1 antisense pretreatment inhibited increased IRF-1 expression, whereas IRF-1 sense and IRF-2 sense and antisense oligonucleotides did not influence the IRF-1 expression. On the other hand, IRF-2 protein level was not changed 12 h after serum depletion, although IRF-2 mRNA transiently decreased 3 h after serum removal and returned to basal level rapidly. IRF-2 antisense oligonucleotide treatment decreased the expression of IRF-2. Next we examined the effects of IRFs antisense oligonucleotides on AT 2 receptor expression and observed that IRF-2 antisense enhanced AT 2 receptor binding and IRF-1 antisense oligonucleotide attenuated AT 2 receptor binding in confluent serum-fed R3T3 cells (Fig. 2C). We demonstrated further that IRF-1 antisense oligonucleotide pretreatment abolished the up-regulation of AT 2 receptor after serum removal (Fig. 2C), demonstrating that the increase in IRF-1 mediated the up-regulation of AT 2 receptor expression. In the serum-depleted state, IRF-2 antisense treatment increased the AT 2 receptor density. These data suggest that increased IRF-1 and decreased IRF-2 after serum removal may exert synergistic effects on AT 2 receptor up-regulation.
Induction of Apoptosis by IRF-1-To test the possibility that the increased expression of IRF-1 in R3T3 cells after serum growth factor removal contributes to the induction of apoptosis, we treated the R3T3 cells with antisense oligonucleotides for IRF-1 or IRF-2 1 day before serum starvation. As shown in Fig.  4, serum depletion induced the internucleosomal cleavage of DNA, which resulted in the generation of DNA fragments of multiples of ϳ180 base pairs (the size of a nucleosome): a hallmark of apoptosis. IRF-1 antisense oligonucleotide treatment inhibited DNA fragmentation in serum-starved R3T3 cells, whereas IRF-1 antisense did not affect DNA fragmentation in serum-fed R3T3 cells, suggesting that the increase of IRF-1 after serum depletion plays a critical role in the induction of apoptosis. On the other hand, IRF-2 antisense oligonucleotide did not influence the development of apoptosis. However, we cannot exclude the possibility that the transient decrease of IRF-2 mRNA after serum starvation as shown in Fig. 1 enhanced the IRF-1 effect and consequently contributed to the process of apoptosis as well. In addition to the DNA fragmentation, we also examined the morphological changes of apoptosis. As shown in Fig. 5, serum depletion caused apoptotic morphological changes in R3T3 cells. IRF-1 antisense oligonucleotide pretreatment attenuated the induction of apoptosis after serum growth factor removal, consistent with the results of DNA laddering.
Possible Role of Up-regulated AT 2 Receptor in the Induction of Apoptosis by IRFs-As shown in Figs. 6 and 7, Ang II (10 Ϫ7 M) treatment further enhanced DNA fragmentation after serum removal, and this effect was blocked by the specific AT 2 receptor antagonist, PD123319 (Fig. 6, A and B). The effect of Ang II on the induction of apoptosis was observed both in serum-fed and serum-starved cells; however, the effect of Ang II was significantly greater in serum-starved cells. Next we examined the effects of IRFs on AT 2 receptor-mediated DNA fragmentation after serum removal (Fig. 7, A and B). Antisense oligonucleotides for IRF-1 and IRF-2 were transfected into R3T3 cells; serum was removed from the culture medium 1 day after antisense transfection. The cells were then stimulated with Ang II (10 Ϫ7 M). Three days after Ang II stimulation, DNA fragmentation was measured. IRF-1 antisense pretreatment inhibited Ang II-mediated DNA fragmentation, whereas IRF-2 antisense pretreatment further enhanced the effect of Ang II as a result of an increased expression of AT 2 receptor as shown in Fig. 2C. Taken together these results suggest that the upregulated AT 2 receptor density, as a result of the changes in IRF-1 and IRF-2, enhances the AT 2 receptor-mediated apoptosis after serum depletion.

DISCUSSION
The antigrowth effects of AT 2 receptor on vascular smooth muscle cells (9) and endothelial cells (10) have been reported recently. The highly abundant expression of AT 2 receptor during embryonic and neonatal growth, the rapid disappearance after birth (4,6), and the up-regulation of AT 2 receptor in myocardial infarction (27), cardiac hypertrophy (28), and skin wound (29) suggest that this receptor is closely involved with growth, development, and/or differentiation. Dudley et al. (14) reported that the addition of growth factors, such as basic fibroblast growth factor or serum, to quiescent R3T3 cells caused a rapid decrease in the number of surface AT 2 receptor sites. Leung et al. (30) reported that nerve growth factor treatment decreased AT 2 receptor expression in PC12W cells. Kambayashi et al. (31) also reported that the expression of AT 2 receptor in cultured vascular smooth muscle cells was downregulated by the addition of growth factors such as plateletderived growth factor-BB, epidermal growth factor, and endothelin-1. Recently we have reported that this receptor induces apoptosis in PC12W cells by antagonizing the effect of nerve growth factor, and that apoptotic PC12W cells exhibited upregulated AT 2 receptor density (11). Moreover, we demonstrated that Ang II enhanced apoptosis in serum-starved R3T3 cells, along with the increased concentration of the AT 2 receptor. These studies suggest that the expression of the AT 2 receptor is inversely related to cell growth and is positively correlated with apoptosis. Consistent with our previous reports (11), we also demonstrated in this paper that AT 2 receptor induces apoptosis in R3T3 cells; however, the "genetic evidence" that AT 2 receptor induces apoptosis still remains to be proved.
To examine the molecular mechanism of AT 2 receptor upregulation in apoptosis, we studied R3T3 cells that express very low level of AT 2 receptor in the growing state and abundant AT 2 receptor after reaching the confluent state (14). To One day after oligonucleotide transfection, serum was removed and cells were incubated with or without Ang II (10 Ϫ7 M) for 3 days. Quantitative analysis was performed by calculating the radioactivity incorporated into the low molecular weight fraction (Ͻ20 kb) divided by the radioactivity of serum-fed, non-treated cells (n ϭ 4 each condition, mean Ϯ S.D.). * shows p Ͻ 0.01 versus serum-starved, non-Ang IItreated cells. † shows p Ͻ 0.01 versus serum-starved, Ang II-treated cells.
investigate the growth-regulated AT 2 receptor expression, we first cloned mouse AT 2 receptor genomic DNA and studied its promoter function in R3T3 cells (15). We identified IRF binding motif (IRF-Es) in the promoter region of AT 2 receptor gene and demonstrated that IRF-2 attenuated the AT 2 receptor expression in both growing and confluent R3T3 cells, whereas IRF-1 enhanced AT 2 receptor expression in the confluent cells only. Consistent with this result, gel mobility shift assay demonstrated that growing R3T3 cells exhibited only IRF-2 binding, whereas confluent cells exhibited both IRF-1 and IRF-2 bindings. In this paper, we demonstrated further that, in confluent R3T3 cells, the removal of serum increased IRF-1 expression but caused a rapid and transient decrease in IRF-2 mRNA expression. These changes of IRFs were associated with an up-regulation of AT 2 receptor and the enhancement of AT 2 receptor-mediated apoptosis. These results suggest that the IRFs play an important role in the up-regulation of this receptor associated with apoptosis.
IRF-1 and IRF-2 are structurally related, and they recognize the same DNA sequence elements AAGTGA motif or G(A)AAAG/CT/CGAAAG/CT/C (16,17). Gene transfection studies have demonstrated that these two factors are mutually antagonistic; IRF-1 activates transcription, whereas IRF-2 inhibits it (16,19). Harada et al. (23) have shown that the IRF-1/IRF-2 expression ratio oscillates throughout the cell cycle of NIH 3T3 cells, with IRF-1 expression at its highest when cells are growth-arrested owing to serum starvation and with IRF-2 being more abundant in growing cells. IRF-1 manifests antiproliferative properties (20,22), whereas overexpression of the repressor IRF-2 leads to cell transformation and increased tumorigenicity, and this phenotype can be reverted by the concomitant overexpression of IRF-1 (23). Accordingly, we speculate further that the changes of IRF-1 and IRF-2 after serum removal are closely linked to the induction of apoptosis in R3T3 cells. Indeed our data demonstrated an increased IRF-1 expression and transiently decreased IRF-2 expression during apoptosis induced by serum removal.
Several lines of evidence has shown that the expression of oncogenes can sensitize cells to undergo apoptosis, particularly under conditions of low serum concentration or high cell density (32,33). It has been demonstrated that IRF-1 and IRF-2 have antioncogenic and oncogenic activities, respectively. Deletion or inactivation of the IRF-1 gene at one or both alleles have been detected in leukemia and myelodysplastic syndromes (34). Tanaka et al. (35) reported that IRF-1 may be a critical determinant of oncogene-induced cell transformation or apoptosis in mouse embryonic fibroblasts (EFs). They demonstrated that Ras signaling, under conditions of low serum or at high cell density or following treatment by anticancer drugs or ionizing radiation, induces the death of EFs derived from wild type mice and from mice with a null mutation in the IRF-2 gene (IRF-2 Ϫ/Ϫ mice), but not of EFs from IRF-1 Ϫ/Ϫ and double knockout mice. These data demonstrate that the absence of IRF-1 alone is sufficient to prevent Ras-induced apoptosis. Tamura et al. (36) also showed that the mitogen induction of the interleukin-1␤-converting enzyme, mammalian homologue of the Caenorhabditis elegance cell death gene ced-3, is IRF-1dependent. These studies are consistent with our present results and support further that IRF-1 is an unique transcription factor that functions as an apoptotic inducer.
IRF-1 and IRF-2 are originally identified as regulators of type I interferon system. DNA sequences recognized by IRFs have been also found in the regulatory regions in a number of interferon-inducible genes (16 -19, 37-39). IRF binding consensus element was also identified in inducible nitric oxide synthase (iNOS) gene promoter region, and IRF-1 is essential for iNOS activation in murine macrophages (40). Recently we reported that NO donor molecules, S-nitroso-N-acetylpenicillamine or sodium nitroprusside, induced apoptosis in cultured rabbit vascular smooth muscle cells (41). In this study, we observed that an increase in the ratio of IRF-1/IRF-2 after serum starvation mediated the up-regulation of AT 2 receptor, resulting in the enhancement of Ang II-mediated apoptosis. Our data showed that the blockade of IRF-1 expression resulted in the simultaneous reduction in AT 2 receptor expression and in apoptosis, whereas blockade of IRF-2 expression led to an increase in AT 2 receptor expression and enhanced apoptosis. Taken together these data support the notion that AT 2 receptor is one of the target genes of the IRFs system. We propose further that IRF-regulated gene products such AT 2 receptor and iNOS may play important mediating roles in IRF-regulated apoptosis.