INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

Nitric oxide (NO)¹ acts as an extra- and intercellular messenger participating in vascular homeostasis, neurotransmission, and defense against infectious agents (1). NO is generated during the oxidation of L-arginine to L-citrulline catalyzed by at least three different isoforms of the enzyme nitric oxide synthase (NOS) (2-4). Neuronal and endothelial NOS are constitutively synthesized but dormant until activated by transient increases in Ca²⁺, which are necessary for the binding of calmodulin by the NOS. In marked contrast, the third isoform, inducible NOS (iNOS), is not present in resting cells. Instead, the cell must be activated to express the enzyme; its activity is independent of elevated Ca²⁺ concentrations or exogenous calmodulin (4, 5). iNOS is widely expressed in every type of tissue and cell after transcriptional induction following exposure to a vast array of immunologic and inflammatory stimuli. Once synthesized, this isoform is active for many hours or days and generates large amounts of NO that have both cytotoxic and cytoprotective effects (1, 6). Activity of iNOS has been associated with tissue damage in arthritis, nephritis, insulitis, and septic shock (7-11).

iNOS is subject to predominantly transcriptional regulation. The molecular basis for induction of the human iNOS (hiNOS) gene is only partially understood. Increased production of the iNOS gene splice variants in response to cytokines has been demonstrated (12, 13). Cloning of the promoter of the hiNOS gene has opened a route for molecular analysis of iNOS induction. Functional characterization of 8296-bp of the 5'-flanking DNA of the hiNOS gene revealed that a region upstream from 3.7 kb of the transcription initiation site confers inducibility by cytokines, whereas no effects of cytokines via the proximal part of the promoter have been found (14-16). This finding is in marked contrast to those with the murine iNOS promoter, which, despite 47% identity with the human sequence, requires only 1 kb of the proximal 5'-flanking region to augment transcriptional activity (17).

Computer-assisted analysis of the 8296-bp 5'-untranslated region of the hiNOS gene identified potential cytokine-responsive transcriptional elements. These include multiple copies of IFN- response elements, two copies of activator protein 1 (AP-1), and two copies of nuclear factor B (NF-B) response elements (14). AP-1 and NF-B are ubiquitous transcription factors and pleiotropic regulators of the inducible expression of many genes that encode proteins involved in the modulation of inflammatory and host defense processes in eucaryotic cells (18-20). Protein components of NF-B and AP-1 are encoded by a set of genes called "immediate early genes" whose transcription is rapidly induced, independently of de novo protein synthesis, following cell stimulation. The AP-1 transcription factor is a complex composed of proteins of the fos and jun proto-oncogene families, which need to dimerize to promote binding of the complex to the AP-1 recognition site (21). AP-1 has been shown to alter gene expression in response to growth factors, cytokines, tumor promoters, and carcinogens (20). Members of NF-B transcription factor family share a conserved amino-terminal region of approximately 300 amino acids known as the NF-B/rel/dorsal (NRD) homology region and include p50, p52, Rel A (p65), Rel B, c-Rel, v-Rel, dorsal and Dif proteins (22). NF-B is sequestered in the cytoplasm through its association with its inhibitors, p105 or IB-like proteins (23). Activation of NF-B by cytokines or oxidative stress requires either the degradation of its cytoplasmic inhibitor IB- or proteolytic cleavage of p105 (23). Free NF-B dimers translocate to the nucleus and activate target genes. This process is transient and terminated through delayed NF-B-mediated IB- induction (24).

Since the expression of several inducible genes is regulated through AP-1- (25) and NF-B-binding sites (23, 24), the present study was undertaken to explore the involvement of NF-B and AP-1 response elements in transcriptional regulation of the hiNOS gene in human lung epithelial cells. Constructs with mutated NF-B and AP-1 sites were used to demonstrate roles of these DNA-binding motifs in the activation of the hiNOS promoter by a combination of three cytokines (IL-1, IFN-, and TNF-). In addition, specific members of the NF-B and AP-1 transcription factor family that bound to each of the response elements in the hiNOS promoter were identified, and their regulation was determined.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Cell Culture and Cytokine Induction-- A549 cells (American Type Culture Collection (ATCC) CCL 185), a human alveolar type II epithelium-like lung adenocarcinoma cell line, were grown in Ham's F-12 K medium supplemented with 10% heat-inactivated fetal bovine serum, 2 mM glutamine, penicillin (100 units/ml), and 100 µg/ml streptomycin (all from Biofluids). To induce hiNOS promoter activity, cells were treated with a mixture of recombinant human cytokines (CM) containing interferon- (IFN-), 100 units/ml, interleukin-1 (IL1-B), 0.5 ng/ml, and tumor necrosis factor- (TNF-), 10 ng/ml. IL-1 was from Genzyme, and the others were from Boehringer Mannheim.

Northern Analysis of Cytokine-stimulated hiNOS mRNA in Cells-- Total RNA (10 µg/lane) from A549 cells, incubated with and without cytokines, was subjected to formaldehyde-agarose gel electrophoresis, transferred to Nytran membranes, hybridized with a 3.1-kb, [-³²P]dATP-labeled hiNOS cDNA probe (encompassing the sequence of +1 to +3101), and then evaluated by autoradiography (13). Before RNA was transferred to Nytran membranes, agarose gels were stained with ethidium bromide and photographed using an Eagle Eye II Still Video System (Stratagene). The intensity of bands on autoradiograms or photographs was determined by densitometry (Personal densitometer SI; Molecular Dynamic). The intensity of hiNOS mRNA bands was normalized to that of ribosomal 28 S RNA.

Transient Transfection and Luciferase Assay-- A549 cells were transfected using LipofectAMINE Reagent (Life Technologies, Inc.) with constructs containing the luciferase reporter gene. At 80% confluency, after washing in serum-free medium, cells were incubated with 1 µg of DNA and 8 µl of LipofectAMINE Reagent/well (six-well dishes, Costar Corp., Cambridge, MA) for 5 h at 37 °C; fetal bovine serum was then added to a final concentration of 10%. After 24 h, the culture medium was replaced with fresh medium containing cytokines. Cells were harvested 12 h later, and luciferase activity was determined using a Luciferase Assay System Kit (Promega). Briefly, cells were washed twice in phosphate-buffered saline (PBS) and lysed by adding 200 µl of a 1× lysis buffer (Promega). After 10 min at room temperature, the lysate was removed from the plate and centrifuged at 12,000 × g. To 40 µl of the supernatant, 100 µl of a luciferase assay reagent (Promega) were added, and luciferase activity was immediately measured using a Monolight 2010 luminometer (Analytical Luminescence Laboratory). Cell lysates were analyzed for protein content using the BCA (bicinchoninic acid) protein assay (Pierce), and luminescence units were normalized for total protein content. Luciferase activities are reported as means of values from four independent experiments, each performed in triplicate.

Generation of hiNOS Luciferase Reporter Constructs-- The plasmid pGL3-8.3, containing the full-length hiNOS promoter cloned into the pGL3-basic luciferase reporter gene vector (Promega, Madison, WI), was used for oligonucleotide-directed mutagenesis of the NF-B- and AP-1-binding sites using the Chameleon^TM Double-stranded Site-directed Mutagenesis KIT (Stratagene). By using the selection primer a SalI site in the vector was changed to a SacII 2010 site. After the annealing of selection and mutagenesis primers, the strands were completed with T7 DNA polymerase and circularized with T4 DNA ligase. The parent plasmid was linearized with SalI, and the mixture was transformed in repair-deficient bacteria (XLmutS). DNA was further linearized with SalI to enhance the mutation efficiency, and the resultant DNA digest was transformed into Epicurian coli XL1-Blue competent bacteria cell line. Colonies were expanded, and DNA was harvested and sequenced. Mutagenesis of the deletion construct pGL3-5.574 was performed by a similar procedure, leading to disruption of the AP-1 upstream site.

Preparation of Cell Extracts-- Cytosolic and nuclear extracts were obtained from cells grown to 85% confluency in 100-mm² dishes and treated with cytokine mixture (CM) for 1, 3, 6, or 24 h. All extraction procedures were performed on ice with ice-cold reagents. Cells were washed twice with PBS, harvested by scraping into 4 ml of PBS, and centrifuged (500 × g, 5 min). The pellet was dispersed in 1 packed cell volume of hypo-osmotic buffer (10 mM Hepes-KOH, pH 7.9, 10 mM KCl, 1.5 mM MgCl₂, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, aprotinin, pepstatin, and leupeptin, each 2 µg/ml). After 15 min on ice, Nonidet P-40 was added to a final concentration of 0.6% (v/v), and nuclei were pelleted by centrifugation (5000 × g, 5 min). Supernatants containing cytoplasmic proteins were stored at 70 °C. The pelleted nuclei were dispersed in a high salt buffer (20 mM Hepes-KOH, pH 7.9, 420 mM NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 25% glycerol, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, aprotinin, pepstatin, and leupeptin, each 2 µg/ml) to solubilize DNA-binding proteins. The suspended nuclei were gently shaken for 30 min at 4 °C and centrifuged in a microcentrifuge (12,000 × g, 20 min). The clear supernatants containing nuclear proteins were stored in small portions at 70 °C until used for EMSA. Protein concentrations were determined using a Bradford assay kit with bovine serum albumin as standard.

Oligonucleotides Used in EMSA-- Complementary oligonucleotides were synthesized on an Applied Biosystems (Foster City, CA) 380B DNA Synthesizer. The positive strands of the double-stranded oligonucleotides corresponding to the identical regions of the hiNOS promoter are as follows (mutated (mt) sequences are underlined and nomenclature used in the text is indicated in bold): wt NF-Bu (8287 to 8270), 5'-CCCTGGGGAACTCCTGCA-3'; mt NF-Bu, 5'-CCCTATAGAACTAATGCA-3'; wt NF-Bd (102 to 119), 5'-GCTGGGGACACTCCCTTT-3'; mt NF-Bd, 5'-GCTGATAACACTAACTTT-3'; wt AP-1u (5307 to 5290), 5'-CCAGCTTGAGTCACACTC-3'; mt AP-1u, 5'-CCAGCTTAATTAACACTC-3'; wt AP-1d (5121 to 5104), 5'-TTTGTGTGACTCACGCCC-3'; mt AP-1d, 5'-TTTGTGTAATTAACGCCC-3'.

Electrophoretic Mobility Gel Shift Assay-- NF-B- and AP-1-binding activity in nuclei of uninduced and induced cells was determined by electrophoretic mobility gel shift assay using the Promega Gel Shift Assay System. Samples (5-8 µg) of nuclear proteins were incubated with the respective radiolabeled oligonucleotides 20 min at room temperature (~25 °C). Specificities of the binding reactions were tested in competition assays in which a 100-fold excess of unlabeled wild-type or mutated oligonucleotide was added 15 min before the labeled probe. Protein-nucleotide complexes were separated by electrophoresis in a 6% DNA retardation gel (Novex) with Tris borate/EDTA (50 mM Tris-HCl, 50 mM boric acid, 1 mM EDTA, pH 8.3) at constant current (30 mA) at 4 °C. Photographic film was exposed to dried gels at 70 °C.

Western Blot Analysis-- Protein samples (20-30 µg) were mixed with an equal volume of 2× SDS sample buffer (Novex), boiled for 5 min, and separated by SDS-polyacrylamide gel electrophoresis in 10% gels (Novex). After electrophoresis and electrophoretic transfer of proteins to nitrocellulose membranes (Bio-Rad) using the Novex Xcell Mini-Gel System, the membranes were incubated overnight in 5% non-fat milk, rinsed, and incubated for 2 h at room temperature with primary antibodies at dilutions of 1:500 for rabbit polyclonal antibodies and 1:1000 for goat polyclonal antibodies (Santa Cruz, CA) in PBS containing 0.05% Tween 20 (TPBS) with 3% non-fat milk. Primary antibody was removed by washing five times in TPBS. Peroxidase-labeled secondary antibodies were added at a dilution of 1:3000 (anti-rabbit IgG-Amersham Pharmacia Biotech; anti-goat IgG, Santa Cruz, CA). After 1 h at room temperature and five washes in TPBS, blots were incubated in enhanced chemiluminescence reagent (ECL; Amersham Pharmacia Biotech) and exposed to x-ray film.

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

Effect of Cytokines on hiNOS mRNA in A549 Cells-- The effect of cytokines on hiNOS mRNA was evaluated in A549 cells incubated with a CM (TNF-, IL-1, and IFN-) for 2, 4, 6, 8, 12, 24, or 48 h (Fig. 1). By Northern analysis, hiNOS mRNA first appeared at 4 h (16%), was highest at 12 h (100%), and had declined to 20% of maximal levels by 48 h.

View larger version (60K): [in this window] [in a new window]   Fig. 1.   Time course of effect of CM on hiNOS mRNA in A549 cells. A, hybridization with 32P-labeled hiNOS probe of hiNOS mRNA in samples of total RNA (10 µg/lane) from resting cells and cells after incubation with CM containing TNF- (10 ng/ml), IL-1 (0.5 ng/ml), and IFN- (100 units/ml) for 2-48 h. B, ribosomal 28 S RNA (stained with ethidium bromide) in the indicated samples.

Effect of Mutation in NF-B and AP-1 Binding Sequences of the iNOS Gene on Cytokine-stimulated Induction of Promoter Activity-- In each of the decameric NF-B sequence motifs, five nucleotides were changed (Fig. 2). Similarly, constructs bearing a three-base mutation in the AP-1-binding motif were generated (Fig. 2). A549 cells transfected with wild-type and mutated iNOS promoters were incubated for 12 h with a combination of cytokines IFN-, IL-1, and TNF- (CM). After transient transfection, CM induction of the luciferase reporter driven by 8296 nucleotides upstream of the transcriptional start site of the hiNOS gene (wt-8.3) was 48-fold (Fig. 3B). As expected, the promoterless vector containing the luciferase gene (Fig. 3, Bas) was unaffected by CM treatment, confirming that these effects were specific to the hiNOS promoter. A single mutation in the AP-1 heptad at position 5301 (AP-1u) decreased induction by CM to 4.6 from 48.2 for wt-8.3. In contrast, disruption of the AP-1-binding site at 5115 (AP-1d) reduced responsiveness to cytokines by only 45% relative to wt-8.3, indicating functional nonidentity of the two AP-1 elements. Induction by CM in cells transfected with a construct bearing mutations in both AP-1 sites (AP-1u/d) was comparable to that with AP-1u, with a 94% reduction in response. These results are consistent with the hypothesis that a regulatory region in the hiNOS promoter is located between 5.7 and 3.7 kb (14, 15). To confirm further the importance of the upstream AP-1 site, it was mutated in a deletion construct of 5574 bp, which has been shown to be the shortest construct conferring cytokine inducibility on the hiNOS promoter (14, 15). Disruption of this site resulted in a marked reduction in CM response with activity being 10% of the 5574-wild-type construct (Fig. 4).

View larger version (9K): [in this window] [in a new window]   Fig. 2.   NF-B- and AP-1-binding sites in the proximal 8.3 kb of the human iNOS promoter. Putative binding elements for upstream and downstream NF-B and AP-1 are illustrated as follows: NF-Bu at 8283, NF-Bd at 115, AP-1u at 5301, and AP-1d at 5115. wt indicates wild-type; mt, mutated sequences.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effect of mutations in the hiNOS gene NF-B and AP-1 sites on responses to CM. A549 cells transfected with the indicated construct were incubated for 12 h with CM before assay of luciferase activity. Data are means of values from four experiments (± S.E.). A, open bars represent basal expression and solid bars that after CM treatment. B, values are expressed as the fold increase in luciferase activity seen following cytokine treatment relative to the luciferase activity seen with no added cytokines.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Effect of mutation in the hiNOS gene deletion construct 5574 AP-1 upstream site on responses to CM. A549 cells transiently transfected with the indicated construct were incubated for 12 h with CM before assay of luciferase activity. Data are means of values from three experiments (± S.E.). A, open bars represent basal expression and solid bars that after CM treatment. B, values are expressed as the fold increase in luciferase activity seen following cytokine treatment relative to the luciferase activity seen with no added cytokines.

Cytokines Induce Binding of NF-B and AP-1 in A549 Cells-- The binding of proteins to 18-bp oligonucleotides, with sequences based on the upstream and downstream AP-1 and NF-B sites, was evaluated by electrophoretic mobility gel shift assay. A549 cells were incubated without or with CM for 1, 3, 6, or 24 h. This time course was performed to determine whether the effects on binding of the transcription factors paralleled the induction of the hiNOS gene. Specific slow-moving DNA-protein complexes that bind to AP-1 downstream (Fig. 5I, A) and AP-1 upstream sequences (Fig. 5I, B) were observed in control cells. CM treatment slightly increased amounts of DNA-protein complexes at 3 and 24 h with oligonucleotides containing either the downstream or upstream AP-1 element. In contrast to AP-1, NF-B showed different patterns of binding activity after incubation of cells with CM; with a probe for either of the NF-B-binding sites (Fig. 5II, A and B), complexes were observed only with the nuclear fraction from the cells stimulated by CM. Binding activity was highest after 1 h of CM treatment and then gradually decreased with longer incubation. All DNA-protein complexes were specifically reduced by a wild-type oligonucleotide but not by a 100× molar excess of oligonucleotides with mutated NF-B or AP-1 sites.

View larger version (63K): [in this window] [in a new window]   Fig. 5.   Time course of AP-1 (I) and NF-B (II) binding following incubation of A549 cells with CM. A549 cells were incubated for the indicated time with medium alone (Rest) or with CM before preparation of nuclear extracts for gel retardation analysis with 32P-labeled double-stranded oligonucleotides for either downstream- (A) or upstream (B)-binding sites. Specificity of the binding reactions was tested by adding a 100-fold excess of unlabeled wild-type (wt AP-1 and wt NF-B) or mutated (mt AP-1 and mt NF-B) oligonucleotide 15 min before the labeled probe. Specific shifted bands are indicated with arrows. The autoradiographs are representative of three experiments.

Identification of NF-B- and AP-1-binding Proteins in Nuclei of A549 Cells Treated with CM-- Mobility shift assays were performed to identify proteins binding to the AP-1 and NF-B sites. For AP-1, antibodies raised against proteins of the Jun (c-Jun, Jun B, and Jun D) and Fos (Fos B, c-Fos, Fra-1, and Fra-2) families were used. As shown in Fig. 6, with oligonucleotides bearing sequences of downstream (Fig. 6A) and upstream (Fig. 6B) AP-1 elements, Jun D- and Fra-2-specific antibodies produced marked supershifts, suggesting that these proteins are major components of the DNA-binding complex. Control serum and other antibodies did not supershift the complexes formed with A549 nuclear extracts. Interestingly, each B site bound different complements of NF-B/Rel family members. As shown in Fig. 7, DNA-protein complexes containing the downstream NF-B site were shifted by Rel A (p65) and p50. Slight binding of p50 to the upstream site was observed. However, the extent of p50 binding to the downstream NF-B site relative to p65 was significantly greater than that observed with the upstream site, compatible with Rel A/Rel A homodimer interaction at the upstream site (Fig. 7B) and Rel A/p50 heterodimer binding to the downstream site (Fig. 7A). Therefore, differences in the nucleotide sequence of the B motif result in distinct compositions of nucleoprotein complexes, consistent with a functional difference between these two sites.

View larger version (139K): [in this window] [in a new window]   Fig. 6.   Identification of AP-1-binding proteins in nuclei of A549 cells treated with CM. Nuclear extracts were prepared from cells treated with CM for 3 h; no significant differences in binding activity of nuclear extracts were observed following 3, 6, and 24 h of incubation with cytokines; all three were significantly greater than controls. Nuclear extracts were incubated overnight without (no IgG) or with preimmune serum (r, rabbit; g, goat IgG) or indicated antibodies against proteins of the Jun and Fos families after they had bound oligonucleotide of a downstream (A) or upstream (B) AP-1 site. Solid arrow indicates specific AP-1 complexes. Supershifted bands are indicated by open arrows. Results shown are representative of three experiments.

View larger version (105K): [in this window] [in a new window]   Fig. 7.   Identification of NF-B-binding proteins in nuclei of A549 cells treated with CM. Nuclear extracts prepared from cells treated with CM for 3 h were incubated overnight without (no IgG) or with preimmune serum (r, rabbit; g, goat IgG) or with indicated antibodies against the amino terminus of Rel A (Rel A (N)), the carboxyl terminus of Rel A (Rel A (C)), p50, p52, c-Rel, or Rel B overnight after the binding reaction with oligonucleotide containing downstream NF-B (A) or upstream NF-B (B). Solid arrow indicates specific NF-B complexes. Supershifted bands are indicated by open arrows. Results are representative of three experiments.

Immunoblot Analysis for the Proteins of the AP-1 and NF-B Complexes Expressed in A549 Cells Upon Cytokine Treatment-- In view of the different patterns in the protein binding activities of NF-B and AP-1 oligonucleotides in EMSA experiments, cytokine-induced expression of protein constituents of NF-B and AP-1 nucleoprotein complexes were examined, in nuclear and cytoplasmic compartments of A549 cells. Jun D protein was constitutively expressed in untreated cells. Induction was not observed after CM stimulation in either nuclear extracts or cytoplasm (Fig. 8I, A and B). c-Jun was significantly increased in the nuclear extract 1 h after CM; only slight increases were observed in Jun-B (Fig. 9). Our finding is consistent with previous reports where it was shown that expression of Jun D, in contrast to that of other members of the Jun proto-oncogene family (e.g. c-jun, jun B), is relatively insensitive to most agents that control cellular proliferation and differentiation (28, 29).

View larger version (27K): [in this window] [in a new window]   Fig. 8.   Immunoblot analysis for Jun D (I), Fra-2 (II), p50 (III), and Rel A (IV) in nuclear extract (A) or cytoplasm (B) and IB- (V) in cytoplasm. A549 cells were incubated for the indicated time with medium alone (Rest) or CM. Cell extracts were subjected to Western analysis with respective antibodies. Results are representative of three experiments. Arrows indicate bands of the respective proteins.

View larger version (50K): [in this window] [in a new window]   Fig. 9.   Immunoblot analysis for c-Jun (I), Jun B (II), Fra 1 (III), c-Fos (IV), and Fos B (V) in nuclear extract. A549 cells were incubated for the indicated time with medium alone (Rest) or CM. Cell extracts were subjected to Western analysis with respective antibodies. Arrows indicate bands of the respective proteins. Results are representative of two experiments.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

NO produced by iNOS has been shown to be beneficial through its antitumor and antimicrobial activities but is thought also to have negative effects through its contributions to inflammatory conditions, carcinogenesis, autoimmune diseases, and neurological disorders (34). Thus, iNOS gene expression must be tightly controlled. Considering that many different agents up- or down-regulate iNOS expression (reviewed in Ref. 35), it is not surprising that multiple positive and negative regulatory elements, responsive to numerous transcription factors, have been identified in the hiNOS promoter. The studies reported here demonstrate potential roles of AP-1 and NF-B cis-elements in the hiNOS promoter in iNOS induction by a mixture of three proinflammatory cytokines IFN-, TNF-, and IL-1.

Closely positioned AP-1 sites differed greatly in their ability to confer cytokine-mediated inducibility. Based on mutational analysis, with both the 8.3 and 5.574-kb promoter, the cis-element encompassing the upstream AP-1 sequence is critical for hiNOS activation by CM, whereas the downstream AP-1 site is less important. Both AP-1 sites were mutated to the same sequence, yet mutation of the promoter at the downstream site resulted in partial reduction in promoter activity compared with the upstream mutation which almost abolished promoter activity; these data indicate that location is crucial and suggest that other factors binding to adjacent domains may be important in regulating AP-1 activity.

The presence of a -IRE juxtaposed with the AP-1 site at 5301 was of interest since inducible expression of several AP-1-regulated genes, including collagenase (36) and glutathione transferase P (37), involves juxtaposed motifs. Mutation of the -IRE outside of the overlapping AP-1 site did not reduce cytokine responsiveness of the promoter, indicating that the AP-1 site is presumably responsible for the functional importance of this region.

Using oligonucleotides with sequences corresponding to AP-1 sites from the hiNOS promoter, two proteins, Jun D and Fra-2, were identified as components of AP-1-DNA nucleoprotein complexes in A549 cells. Conceivably, regulation of the hiNOS promoter in A549 cells may be mediated, in part, through a homodimer of Jun D or a Jun D/Fra 2 heterodimer. The human involucrin gene (38) and the orphan receptor gene nur 77 (39) are transcriptionally activated via an AP-1-response element that binds Jun D.

Fra-2 may have either suppressive or stimulatory effects on the transactivation by Jun family proteins; Fra-2/c-Jun suppresses, whereas Fra-2/Jun D transactivates. Additionally, Jun D/Fra-2 heterodimer is a more efficient activator of AP-1-dependent gene expression than is Jun D alone (40), since Jun proteins bind with much higher affinity when associated with members of the Fos family as Jun/Fos heterodimers (41). In fact, Fra-2 displays dimerization and DNA-binding properties similar to those of the Fos proteins, as it forms stable heterodimeric complexes with the Jun family proteins and binds specifically to AP-1 sites (40). Constitutive Jun D protein expression was unaffected by CM, in agreement with other studies in which Jun D was found to be constitutively expressed at high levels in many tissues, including lungs (28, 42). In the absence of other factors, unmodified Jun D forms unstable complexes with DNA (43, 44). A large potential pool of AP-1 activity in resting cells may be accessible through the post-translational modification and/or the action of other accessory components. In contrast to Jun D, Fra-2 was markedly increased in response to cytokines. Hence, this protein might determine the activity of the AP-1-binding sites. The AP-1-binding pattern with transient increases at 3 and 24 h did not mimic the time course of Fra-2 protein induction, presumably because AP-1 binding activity does not depend solely on expression of fos and jun genes, but also on post-translational modification.

The importance of NF-B in mediating cytokine-induced transcription of hiNOS was first shown in murine macrophages (45). Numerous investigators demonstrated that induction of iNOS was inhibited by the NF-B inhibitor pyrroldinedithiocarbamate (46-48). The downstream NF-B site has been already shown by others to be important in a 3.2-kb hiNOS promoter fragment that does not confer cytokine inducibility (49). Our results from mutational analysis of the 8296-bp promoter fragment indicate that although both B sites contribute to full hiNOS gene activity, the upstream site is less important. Interestingly, NF-B was shown to enhance gene transcription synergistically with AP-1 (50). There are, in addition, at least four other transcription control proteins with which members of the NF-B/Rel family are known to interact, namely SP-1 (51), NF-IL6 (52), HMGI(Y) (53), and TATA-binding protein (54). Since the hiNOS promoter comprises NF-B sites together with two AP-1, one SP-1, and numerous NF-IL6 response elements, the hiNOS gene response may be the result of a very complex mechanism in which AP-1 and NF-B are only a part of the transcription factor network.

The B-binding proteins in A549 cells, Rel A and p50, form distinct nucleoprotein complexes that bound, respectively, to downstream (Rel A/p50) and upstream (Rel A/Rel A) NF-B sites. It is now widely recognized that different hetero- and homodimeric combinations of NF-B/Rel family members require slightly different B sequences for optimal binding (55), and they selectively activate specific sequences (56, 57). Activation of NF-B by several agents, including lipopolysaccharides and cytokines, has been linked to the subsequent transcription of iNOS mRNA in diverse types of cells (45, 58-60).

The data reported here demonstrate that a surge of nuclear NF-B activity is a relatively early and transient response to CM. In agreement with the current model of NF-B regulation (23), NF-B binding activity as well as nuclear expression of Rel A peaked after 1 h of exposure to CM, a time when IB- was markedly reduced. Rapid accumulation of IB- observed after 3 h of cytokine stimulation reestablishes cytoplasmic pools of NF-B·IB complexes. Newly synthesized IB-, after translocation to the nucleus, sequesters free Rel A and thereby promotes dissociation of DNA-bound NF-B. IB-·Rel A inactive complexes are transferred back to the cytoplasm (22). Consistent with this model, in IB- / cells, high levels of NF-B persist in the nucleus for a long time following exposure to TNF- (61). Of interest, binding to the NF-B site increases dramatically by 1 h of treatment with CM, whereas hiNOS mRNA levels do not increase significantly until 4 h after treatment and reach maximal levels at 12 h. In view of the length of the hiNOS promoter, other sites may participate in induction. It is possible that another factor fitting the pattern of induction is critical, although none fitting this role was identified by Western analysis. This difference in time of induction has been observed in other systems (62, 63).

p50, which is synthesized in cells in form of a precursor protein p105 (23), was not induced by cytokines in A549 cells. It is still unclear whether or not processing of p105 into its mature form is regulated. Although some reports suggest up-regulation in response to extracellular signals (64, 65), others indicate that the conversion is a relatively slow, constitutive process (24, 66).

The findings reported here show that critical elements in the transcriptional induction of hiNOS by CM in A549 cells are located at quite a large distance from the transcription start site and include AP-1 and NF-B sites. The structure of the hiNOS promoter and our data indicate that the transcriptional response to CM is the final outcome of a very complex interplay of nuclear targets. The interactions among different transcription factors are likely to serve as an important source of regulatory diversity of the hiNOS gene.