Transcription Factor NF-κB Is Necessary for Up-regulation of Type 1 Angiotensin II Receptor mRNA in Rat Cardiac Fibroblasts Treated with Tumor Necrosis Factor-α or Interleukin-1β

Tumor necrosis factor-α (TNF-α) and interleukin-1β up-regulate type 1 angiotensin II receptor (AT1) mRNA and protein in cultured neonatal rat cardiac fibroblasts. The use of pharmacologic inhibitors and a degradation-resistant mutant IκB-α demonstrated that the transcription factor nuclear factor-κB (NF-κB) is necessary for cytokine-induced AT1 up-regulation. The increase in AT1 mRNA with TNF-α treatment is slow, reaching significance by 6–12 h and peaking by 24–48 h. Electrophoretic mobility shift assays revealed that NF-κB nuclear translocation was maintained for ≥24 h with a single dose of TNF-α. Since prolonged NF-κB activation appeared necessary to maximize AT1up-regulation, the mechanism of persistent NF-κB activation was studied further. Stimulation with TNF-α induced a >10× increase in IκB kinase (IKK) activity that quickly diminished by 20 min. IκB-α and IκB-β proteins were degraded during this time, and IκB-α was resynthesized subsequently by NF-κB-dependent transcription. However, IκB isoforms and IKK activity did not return completely to unstimulated values during a 12-h time course. These results suggest that low but persistent IKK activity and IκB degradation lead to prolonged NF-κB nuclear translocation and maximal AT1 up-regulation in the continued presence of TNF-α.

extracellular matrix proteins (4 -6). Administration of angiotensin-converting enzyme inhibitors or AT 1 blockers inhibits extracellular matrix remodeling and reduces mortality in experimental animal models of MI (7)(8)(9)(10)(11)(12). Angiotensin-converting enzyme inhibitors also increase survival of humans after a MI (13,14). Thus, effects mediated by AT 1 appear to be critical determinants of both extracellular matrix remodeling and the clinical course of patients post-MI.
Increases in AT 1 receptors have been demonstrated in the peri-infarction zone and in remote noninfarcted segments of myocardium after a MI (15)(16)(17). However, the mechanisms responsible for this up-regulation are not understood. Only a few agents such as dexamethasone and growth hormone have been shown to increase AT 1 levels transcriptionally (18,19). Furthermore, most studies of AT 1 regulation have been performed using vascular smooth muscle cells or kidney cells, and little is known about AT 1 regulation in cardiac fibroblasts. We have demonstrated previously that AT 1 mRNA and protein is up-regulated in cultured neonatal rat cardiac fibroblasts by TNF-␣ and IL-1␤ but not by other endogenous mediators found in the post-MI heart (20). These studies showed that AT 1 mRNA increases were sustained for at least 24 -48 h following a single application of TNF-␣ and that increased gene transcription was likely involved. Proinflammatory cytokines such as TNF-␣ and IL-1␤ are increased in the rat heart following experimental MI (21)(22)(23). Since cytokines could be responsible for post-MI increases in AT 1 receptor density, we sought to determine the mechanisms and signaling pathways involved in AT 1 up-regulation by TNF-␣ and IL-1␤.
The transcription factor NF-B is a known regulator of genes that control apoptosis, inflammation, and cell division. It is composed of homodimers and heterodimers of the NF-B/Rel family of proteins but often consists of a heterodimer of p65 and p50. In unstimulated cells, NF-B is retained in the cytoplasm bound to a family of inhibitory IB proteins. Of these, IB-␣ and IB-␤ predominate in a variety of cell types. Stimulation of cells with cytokines such as TNF-␣ or IL-1␤ leads to rapid activation of the IB kinases (IKK1 and IKK2 of which the latter seems to be indispensable) that then phosphorylate IB on serine residues (Ser 32 and Ser 36 on human IB-␣). Phosphorylation leads to ubiquitination of IB on lysine residues (Lys 21 and Lys 22 on human IB-␣) and degradation by the 26 S proteasome. After removal of IB, free NF-B can translocate to the nucleus and affect transcription by binding consensus DNA sequences (for reviews, see Refs. 24 -26). We now provide evidence that NF-B activation is required for increased expression of the AT 1 gene by the cytokines TNF-␣ and IL-1␤ in cultured neonatal rat cardiac fibroblasts. In addition, a single

Isolation of Cardiac Nonmyocytes (Fibroblasts) from Neonatal Rats
Neonatal rat cardiac fibroblasts were isolated from the hearts of 1-2-day-old Sprague-Dawley rats as described previously (27). Second passage cells were used for all experiments. Fibroblasts were grown on tissue culture plates in medium (Dulbecco's modified Eagle's medium high glucose, Life Technologies, Inc.) supplemented with 10% fetal bovine serum and penicillin/streptomycin/Fungizone (Life Technologies, Inc.) in a humidified incubator at 37°C and 10% CO 2 . Prior to initiating experiments (except those involving adenovirus), cells were grown to confluence then equilibrated in Dulbecco's modified Eagle's medium ϩ 0.5% fetal bovine serum for 24 h.

RNA Slot-blotting
Total RNA was isolated from cultured cells using the RNeasy Mini kit (Qiagen, Valencia, CA). RNA concentration was determined by absorbance at 260 nm. Six micrograms of RNA was diluted into water, denatured by heating at 70°C for 10 min, and then chilled on ice. An equal volume of 20ϫ SSC (3 M NaCl, 0.3 M sodium citrate (pH 7.0)) was then added. Samples were slotted onto a nylon membrane (Magna-Graph, Micron Separations, Westborough, MA) using the Bio-Dot SF Microfiltration Apparatus (Bio-Rad Laboratories). Slots were washed once with 10ϫ SSC. The membrane was removed from the apparatus, washed in 10ϫ SSC, and allowed to air-dry. RNA was fixed to the membrane by heating at 80°C for 2 h in a desiccating oven. Membranes were incubated at 42°C in prehybridization solution (50% formamide, 5ϫ saline/sodium phosphate/EDTA, 5ϫ Denhardt's solution, 0.5% SDS, 15 g/ml sheared, heat-denatured salmon DNA) for several hours in roller bottles. Heat-denatured 32 P-labeled probe (radiolabeled using the Multiprime DNA labeling system, Amersham Biosciences, Inc.) was added, and the hybridization was allowed to proceed overnight. Membranes were washed thrice at low stringency (2ϫ SSC, 0.1% SDS; 42°C; 20 min per wash) and thrice at high stringency (0.2ϫ SSC, 0.1% SDS; 65°C; 20 min per wash). Damp membranes were exposed to a Storage Phosphor Screen and then visualized on the Storm 860 fluorescent scanner (Molecular Dynamics, Sunnyvale, CA). Signal intensities were quantified using the software supplied with the scanner (ImageQuant). The probe for AT 1 mRNA was the ϳ1.4 kbp EcoRI fragment of plasmid ATIIR/pG4z (provided by Dr. David Pribnow, Oregon Health Sciences University). To standardize loading between slots, membranes were stripped and reprobed with a cDNA representing nucleotides 2837-4436 of human 28 S rRNA.

Immunoblotting
Cells were washed with phosphate-buffered saline, placed on ice, lysed with cytoplasmic extraction buffer (10 mM Tris-HCl (pH 7.9), 60 mM KCl, 1 mM EDTA, 0.4% Igepal CA-630, 1 mM DL-dithiothreitol, 10 g/ml leupeptin, 100 kallikrein inhibitory units/ml aprotinin, 0.1 mg/ml phenylmethylsulfonyl fluoride) for 5 min and scraped from the plate. The mixture was microcentrifuged at 2500 rpm for 3 min at 4°C, and the supernatant (cytoplasmic extract) was collected. The pellet was washed by gentle resuspension in cytoplasmic extraction buffer and centrifuged again. Nuclear extraction buffer (50 mM Tris-HCl (pH 8.0), 1.5 mM MgCl 2 , 420 mM NaCl, 25% glycerol, 10 g/ml leupeptin, 100 kallikrein inhibitory units/ml) was added to the pellet and vortexed for 1 min. The mixture was allowed to stand on ice for 10 min and then microcentrifuged at maximum speed for 10 min at 4°C. The supernatant (nuclear extract) was collected. If phosphoproteins were to be analyzed, extraction buffers also contained 0.5 mM sodium orthovanadate and 5 mM ␤-glycerophosphate to inhibit phosphatases.
Protein concentrations of extracts were quantified with a Bio-Rad protein assay (catalog no. 500-0006). Equal amounts of protein were electrophoresed on a Bio-Rad Mini-PROTEAN II apparatus using the discontinuous SDS-polyacrylamide gel electrophoresis system of Laemmli (28). Resolved proteins were transferred electrophoretically to polyvinylidene difluoride membrane (Immobilon-P, Millipore) in 192 mM glycine, 25 mM Tris, 10% methanol. The membrane was allowed to air-dry, then wetted in methanol, and floated in distilled water for 1-2 min. At this point, protein bands could be seen as "shiny" areas on a dull membrane background to verify equal loading between lanes. The rewetted membrane was immunoblotted using ECF Western blotting Reagent Packs (Amersham Biosciences, Inc.) following the instructions of the manufacturer. Fluorescent bands were visualized on the Storm 860 scanner in blue fluorescence mode with a photomultiplier tube voltage of 600 -700. Band intensities were quantified using the software supplied with the scanner.

EMSA
Nuclear extracts were prepared from cultured fibroblasts as described for Western blotting. Gel-shift assays for NF-B were performed as described by Jobin et al. (29). Briefly, a double-stranded synthetic oligodeoxyribonucleotide corresponding to the NF-B consensus sequence in the B-cell light chain enhancer region (5Ј-AGTTGAGGG-GACTTTCCCAGGC-3Ј) (30) was radiolabeled with T4 polynucleotide kinase and [␥-32 P]ATP and purified with a QIAquick Nucleotide Removal kit (Qiagen). Nuclear extract (2 g of protein) was incubated with the radiolabeled probe. Bound probe and free probe were separated by electrophoresis on a nondenaturing polyacrylamide gel. The gel was dried and exposed either to film (Kodak BioMax MS) or to a Storage Phosphor Screen.

Packaging of Replication-deficient Adenovirus-5
Ad-IB␣M-Adenovirus expressing a degradation-resistant mutant of IB-␣ (Ad-IB␣M) was a generous gift from Dr. Inder Verma of the Salk Institute (La Jolla, CA). Construction of the mutant has been described elsewhere (31).
Ad-Less-Luc (Promoterless Luciferase)-An adenovirus containing a promoterless luciferase gene was used to determine the effect of adenoviral infection alone on cardiac fibroblasts. For this purpose, a promoterless luciferase gene was excised from the plasmid pGL3-Basic (Promega, Madison, WI) by digestion with BglII and BamHI and ligated into the NotI sites of the adenoviral shuttle vector pACCMV⅐pLpASR(Ϫ) using synthetic linker-adapters. A clone that had the luciferase gene inserted in a positive orientation with respect to the adenoviral genome was selected. Adenovirus was produced by recombination between this shuttle vector and the full-length adenovirus-5 vector pJM17 as described by Gómez-Foix et al. (32).

Adenoviral Infection of Cardiac Fibroblasts
Fibroblasts were grown to confluence in 6-cm tissue culture plates. Two extra plates were seeded to determine cell number via hemacytometer counting. Fibroblasts were infected with adenovirus (at a multiplicity of infection of 5 plaque-forming units/fibroblast) in 3 ml of Dulbecco's modified Eagle's medium ϩ 2% fetal bovine serum (heatinactivated at 65°C for 15 min) for 18 h. The medium was replaced with Dulbecco's modified Eagle's medium ϩ 0.5% fetal bovine serum, and the cells were allowed to recover for 6 h. Cells were then treated and processed according to the protocol of each experiment.

IKK Kinase Assay
Neonatal rat fibroblasts were assayed for IKK kinase activity as described previously (33,34) with minor modifications. Immunoprecipitation was performed using a rabbit polyclonal antibody against IKK␣/␤ (Santa Cruz Biotechnology, Inc., catalog no. sc-7607) on 200 g of total extracted protein (determined using a Bio-Rad protein assay, catalog no. 500-0006). Kinase assays contained 2.5 Ci of [␥-32 P]ATP, 2.5 M cold ATP, and 1 g GST-IB-␣-(1-54) (34). Polyacrylamide gels were fixed in 50% methanol, 10% acetic acid for 2 h and then soaked overnight in several changes of water. Dried gels were exposed to a Storage Phosphor Screen and visualized on the Storm 860 fluorescent scanner. Signal intensities were quantified using the software supplied with the scanner.

Statistical Analyses
Data were collected from at least three independent experiments. Quantitative data are expressed as the mean Ϯ S.E. Statistical significance was determined by one-way analysis of variance followed by multiple comparison testing using GraphPad Prism software (version 3.00, GraphPad Software, San Diego, CA).

RESULTS
We have demonstrated previously that TNF-␣ increases the steady state levels of AT 1 mRNA in cultured neonatal rat cardiac fibroblasts. AT 1 mRNA up-regulation peaked at 24 -48 h with an EC 50 of 4.6 ng/ml (0.26 nM) TNF-␣. The stability of AT 1 mRNA was unchanged by treatment with TNF-␣, which suggested that increased transcription was responsible for the up-regulation (20). IL-1␤ also up-regulated AT 1 mRNA, an effect that peaks at ϳ12 h with an EC 50 of 0.32 ng/ml (17 pM). 2 TNF-␣ and IL-1␤ are known to affect numerous signaling pathways including those that activate NF-B or mitogen-activated protein kinase and those that generate ceramide or nitric oxide (35,36). To gain insight into the mechanism responsible for AT 1 mRNA up-regulation pharmacologic compounds that are known to inhibit such signaling pathways were tested (Fig. 1). TNF-␣ and IL-1␤ produced similar profiles of pharmacologic inhibition. Pyrrolidine dithiocarbamate, an antioxidant/prooxidant, MG-132, a proteasome inhibitor, SB203580 and SB202190, both p38 mitogen-activated protein kinase inhibitors, D609, a phosphatidylcholine-specific phospholipase C inhibitor, and Ro-31-8220, a protein kinase C inhibitor, significantly reduced cytokine-induced AT 1 mRNA up-regulation (Fig. 1). SB202474, an inactive pyridinyl imidazole compound, N G -monomethyl-L-arginine, a nitric-oxide synthase inhibitor, indomethacin, a cyclo-oxygenase inhibitor, fumonisin B 1 , a ceramide synthase inhibitor, and Z-Val-Ala-Asp(OMe)-CH 2 F, a caspase inhibitor, had no significant effect on AT 1 up-regulation (Fig. 1). When considered collectively, all of the compounds that inhibited AT 1 up-regulation have been shown to inhibit NF-B activity by inhibiting nuclear translocation, DNA binding, or transcriptional competency (37)(38)(39)(40). In addition, the mitogen-activated protein kinase kinase inhibitor PD98059 increased AT 1 mRNA levels significantly with TNF-␣ treatment (Fig. 1). Funakoshi et al. (41) have demonstrated that cytokineinduced NF-B activation can be augmented by PD98059 in T98G cells. Therefore, the screen of pharmacologic compounds suggested the involvement of NF-B, but additional evidence using a more selective inhibitor was needed.
To confirm the role of NF-B, cultured cardiac fibroblasts were infected with an adenovirus-5 expressing a mutant (Ser 32,36 3 Ala 32,36 ) IB-␣ (Ad-IB␣M) that cannot be phosphorylated and degraded and thus sequesters NF-B in the cytoplasm (31). Ad-IB␣M completely blocked NF-B nuclear translocation induced by IL-1␤, while a negative control adenovirus (Ad-Less-Luc) had no effect ( Fig. 2A, upper panel). In contrast, Ad-IB␣M did not affect phosphorylation of the transcription factors c-Jun and ATF-2, which is an indirect measure of mitogen-activated protein kinase activation ( Fig. 2A, middle  and lower panels). Thus, Ad-IB␣M appeared to specifically inhibit NF-B activation in the fibroblasts. When tested on AT 1 expression, Ad-IB␣M completely blocked IL-1␤-induced AT 1 mRNA up-regulation, while Ad-Less-Luc had no significant effect (Fig. 2B). Therefore, the use of the signaling mutant and pharmacologic inhibitors provides strong evidence that the transcription factor NF-B is necessary for cytokine-induced up-regulation of AT 1 mRNA.

FIG. 2. Activation of NF-B is necessary for IL-1␤-induced AT 1 mRNA up-regulation.
Confluent fibroblasts were infected with the indicated adenovirus. Cultures also contained 5 g/ml TNFR:Fc (Immunex Corp., Seattle, WA) to neutralize TNF that was secreted by some fibroblast preparations. A, mutant IB-␣ (Ad-IB␣M) blocks NF-B nuclear translocation but not phosphorylation of c-Jun and ATF-2 after IL-1␤ stimulation. Twenty-four hours after infection, fibroblasts were treated with (ϩ) or without (Ϫ) 2 ng/ml IL-␤ for 1 h. Nuclear extracts were prepared and subjected to EMSA for NF-B (top panel), immunoblotting for phospho-c-Jun (p-c-Jun) (middle panel), and immunoblotting for phospho-ATF-2 (p-ATF-2) (bottom panel). Immunoblots contained 2.5 g of total protein/lane. B, mutant IB-␣ (Ad-IB␣M) also blocks IL-1␤-induced AT 1 mRNA up-regulation. Twenty-four hours after infection, fibroblasts were treated without (dark bars) or with (light bars) 2 ng/ml IL-␤ for 12 h. Total RNA was isolated and subjected to slot-blotting for AT 1 mRNA. Bars depict means Ϯ S.E. of six independent experiments. Gurantz et al. (20) demonstrated that AT 1 mRNA was elevated for 24 -48 h in neonatal rat cardiac fibroblasts after a single administration of TNF-␣ to the culture medium. Assuming that the half-life of AT 1 mRNA in these cells is ϳ18 h (42), NF-B would need to be activated for at least 24 h to increase AT 1 mRNA as observed. As shown in Fig. 3A, nuclear translocation of NF-B occurred within 1 h of a single dose of 50 ng/ml TNF-␣ and persisted for at least 24 h as measured by EMSA; NF-B was barely detected in the nuclei of fibroblasts that were not stimulated with cytokine. To verify the identity of the band in Fig. 3A as NF-B, it was competed with excess nonradiolabeled B probe (Fig. 3B, lane 3) and supershifted by an anti-p65 antibody (Fig. 3B, lane 4). Nonspecific rabbit IgG was unable to supershift the complex (Fig. 3B, lane 5).
In the presence of TNF-␣ some cell types transiently activate NF-B, while others maintain a persistent activation (43)(44)(45)(46). Neonatal rat cardiac fibroblasts belong to the latter group according to Fig. 3A. To study the mechanism of persistent NF-B activation in these cells, the protein levels of IB isoforms ␣ and ␤ were analyzed by immunoblotting. During TNF-␣ treatment IB-␣ was degraded and resynthesized rapidly, reaching a minimum by 15 min and then rebounding by 60 min (Fig. 4). In contrast IB-␤ was degraded and resynthesized more slowly, reaching a minimum by 60 min and rising slightly over 12 h (Fig. 4). Despite evidence of resynthesis, both IB isoforms did not return to time 0 values during the 12-h time course (Fig. 4B), an effect that required the continued presence of TNF-␣. For example, after 6 h of treatment with TNF-␣, IB-␣ was 79.5% and IB-␤ was 38.4% of that in untreated cells (Fig. 5A, lanes 3 and 7). However, if TNF-␣ was removed from the medium after 1 h of the 6-h incubation, IB-␣ was 136.4% and IB-␤ was 70.7% (Fig. 5A, lanes 4 and 8). During this period of cytokine removal, p65 protein levels dropped in the nucleus as assessed by immunoblotting (Fig. 5B, compare lanes  11 and 12). These data are consistent with the hypothesis that sustained NF-B activation is caused by a persistent degradation of IB isoforms that depends on the continued presence of TNF-␣. Sustained NF-B activation appears necessary to max-imize AT 1 mRNA increases since removal of TNF-␣ during a 24-h incubation prevents AT 1 mRNA up-regulation (Fig. 5B).
To study further the mechanism of persistent IB degradation, IKK kinase assays were performed. IKK is the proximal kinase that phosphorylates IB, thus regulating IB ubiquitination and degradation (25). Stimulation of neonatal rat car-  FIG. 5. The continued presence of TNF-␣ is required to suppress IB levels, maintain p65 nuclear translocation, and upregulate AT 1 mRNA in cardiac fibroblasts. Confluent fibroblasts were stimulated with or without 50 ng/ml TNF-␣ for the indicated times. A, cytoplasmic IB isoforms and nuclear p65. Cytoplasmic extracts were prepared and immunoblotted for IB-␣ and IB-␤ as described in Fig. 4. Nuclear extracts were also prepared and subjected to immunoblotting for p65 (2.5 g of total protein loaded per lane; primary antibody from Santa Cruz Biotechnology, Inc., catalog no. sc-372, used at 1:200 dilution). B, AT 1 mRNA. Total RNA was isolated from the cells, and AT 1 mRNA was quantified by slot-blotting as described under "Experimental Procedures." All plots were derived from three independent experiments (mean Ϯ S.E.). diac fibroblasts with TNF-␣ increased IKK activity 10 -15ϫ during the first 5-15 min (Fig. 6, A and B). Kinase activity then rapidly diminished by 20 min (Fig. 6A). However, IKK activity did not return completely to basal levels, remaining 2-4ϫ that of unstimulated cells over a 12-h time course (Fig. 6B). Low but persistently elevated IKK activity could continue to phosphorylate IB, especially after it rebinds NF-B in the nucleus. Phosphorylated IB would be degraded, sustaining NF-B nuclear translocation and preventing IB isoforms from returning to control values. DISCUSSION Most of the effects of angiotensin II that are involved in extracellular matrix remodeling are mediated by the AT 1 receptor whose density is increased in the hearts of experimental animals post-MI (15,16). Previous work from our laboratory demonstrated that cytokines such as TNF-␣ and IL-1␤, but not other factors in the post-MI heart, increase AT 1 mRNA levels in cultured neonatal rat cardiac fibroblasts (20). The present studies were performed to elucidate the signal transduction pathways involved in this cytokine-induced AT 1 up-regulation. The major finding is that activation of the transcription factor NF-B is required for AT 1 mRNA up-regulation to occur. Furthermore, in response to TNF-␣ stimulation, persistent activation of the kinases that lead to degradation of inhibitory IB proteins appears responsible for the sustained nuclear translocation of NF-B that maintains AT 1 mRNA up-regulation.
Many signaling pathways are activated upon binding of TNF-␣ or IL-1␤ to its receptor (for reviews, see Refs. 36 and 47). We screened several pharmacologic compounds that are known to inhibit such signaling pathways to determine which might be responsible for AT 1 up-regulation. The profile of inhibition of the pharmacologic compounds was nearly identical for TNF-␣ and IL-1␤, suggesting that both cytokines up-regulate AT 1 by utilizing a common pathway. All compounds that demonstrated significant inhibition of cytokine-induced AT 1 mRNA up-regulation were known to affect the functioning of NF-B. In contrast, compounds that altered other signaling pathways failed to inhibit AT 1 up-regulation. Since both TNF-␣ and IL-1␤ ac-tivate NF-B downstream of receptor engagement (25,35) the observation that both cytokines produced similar pharmacologic profiles is not surprising.
To verify the necessity of NF-B in AT 1 up-regulation, a degradation-resistant mutant IB-␣ was utilized that more specifically blocked NF-B. For these experiments it was necessary to use IL-1␤ since cardiac fibroblasts that had been infected with Ad-IB␣M were killed when treated with TNF-␣ at concentrations and durations known to increase AT 1 mRNA levels. TNF-␣-induced apoptosis has been well documented in other cell types in which NF-B has been inhibited (31,48,49). Preventing transcription of NF-B-dependent antiapoptotic genes unmasks the proapoptotic effects of TNF-␣ in these situations. However, IL-1␤ treatment of IB␣M-expressing fibroblasts did not induce apoptosis. In these cells both NF-B nuclear translocation and AT 1 mRNA up-regulation were completely blocked.
Although the present studies indicate that the trans-acting factor NF-B is required for cytokine-induced up-regulation of AT 1 mRNA in rat cardiac fibroblasts, the cis-acting DNA elements in the AT 1A gene that are responsible for this effect have yet to be identified. Preliminary computer analysis of the AT 1A 5Ј-flanking region (GenBank TM accession number S66402) has revealed two putative NF-B binding sites at Ϫ365 and Ϫ2540 (50). Studies to ascertain the involvement of these putative sites are ongoing.
An autoregulatory feedback loop exists in many cells to avoid prolonged activation of NF-B that can cause chronic inflammation. After strong activation, IKKs show a rapid loss of activity that is thought to be caused by autophosphorylation of Cterminal serine residues (51). In addition, transcription of the IB-␣ gene is NF-B-dependent (52)(53)(54). Upon activation of NF-B, transcription of IB-␣ increases, and IB-␣ protein levels quickly rise. IB-␣ can resequester NF-B and remove it from the nucleus via chromosome region maintenance 1-dependent nuclear export (55). Lowered IKK activity and resynthesis of IB-␣ prevent prolonged activation of NF-B. Our results indicate that stimulation of cardiac fibroblasts with TNF-␣ produces a rapid activation of IKK, subsequent degradation of IB-␣/␤, and nuclear translocation of NF-B. IKK activity rapidly declines thereafter, and IB-␣ is resynthesized. However, over a 12-h period both IKK activity and IB-␣/␤ proteins do not return fully to unstimulated levels. Low but persistent IKK activity and IB degradation could be responsible for persistence of NF-B activity in the continued presence of TNF-␣. This sustained NF-B activation appears to be responsible for the sustained increase in AT 1 mRNA that was observed in our previous study (20). Whether a 2-4ϫ elevation of IKK activity is sufficient to maintain degradation of IB⅐NF-B complexes still needs to be addressed. Signal-induced nuclear proteasome activity has been implicated in degrading IB-␣ that has resequestered NF-B in 293T and HeLa cells, thus prolonging NF-B nuclear translocation (55,56). Lowered but nuclear localized IKK activity may be sufficient to maintain nuclear localization of NF-B in cardiac fibroblasts.
There also is evidence that, after its signal-induced degradation, IB-␤ can be resynthesized as a hypophosphorylated form that can bind nuclear NF-B and protect it from resequestration by IB-␣ (43,57,58). IB-␤ is degraded substantially in cardiac fibroblasts by TNF-␣ treatment (see Fig. 4). Although we have not studied its contribution to prolonging NF-B activation in these cells, resynthesis of hypophosphorylated IB-␤ may play a role. However, this mechanism seems unlikely for two reasons. 1) The effect in cardiac fibroblasts is dependent on the continued presence of cytokine since nuclear p65 levels drop if TNF-␣ is removed from the culture medium (Fig. 5A). 2) IB-␣ seems to predominate over IB-␤ from 1-12 h postcytokine stimulation in the fibroblasts (Fig. 4A).
There is evidence that AT 1 receptor density is increased on cardiac fibroblasts post-MI in the peri-infarction zone and in distant regions of myocardium (15,16). Following a MI there is also an increase in the level of cytokines in the heart including TNF-␣ and IL-1␤ (21)(22)(23). These cytokines may be responsible for the observed up-regulation of AT 1 . Binding of angiotensin II to AT 1 stimulates cardiac fibroblasts, for example, to proliferate, to produce extracellular matrix proteins, and to secrete transforming growth factor-␤ 1 (4 -6, 59). TNF-␣-induced upregulation of fibroblast AT 1 receptors has been shown to augment angiotensin II-mediated responses such as production of inositol phosphates (20), collagen synthesis, and tissue inhibitor of metalloproteinases-1 secretion. 3 Therefore, by up-regulating AT 1 cytokines appear to enhance cultured cardiac fibroblast properties that would contribute to post-MI fibrosis. This possibility is supported by observations made in transgenic mice with cardiac-specific overexpression of TNF-␣. These animals develop a heart failure phenotype that includes increased interstitial fibrosis (60,61).
AT 1 mRNA is increased in cultured neonatal rat cardiac fibroblasts by treating with TNF-␣ or IL-1␤, an effect that requires NF-B. Sasamura et al. (62) have shown that IL-1␤ but not TNF-␣ up-regulates AT 1 mRNA 2.3-fold in rat vascular smooth muscle cells. This contrast exemplifies the cell type-specific regulation of AT 1 and may reflect a difference in function of the receptor between vascular smooth muscle cells and fibroblasts. NF-B is involved typically in the regulation of genes controlling apoptosis and inflammation. It is well known that NF-B can up-regulate cell surface adhesion receptors such as E-selectin, vascular cell adhesion molecule-1, and intercellular adhesion molecule-1 that are involved in inflammatory responses (63). That NF-B can also up-regulate the cell surface AT 1 receptor in cardiac fibroblasts suggests an important and novel role of NF-B in modulating post-MI cardiac remodeling.