Introduction

Top Abstract Introduction Methods Results Discussion References

The NF-B/Rel family of proteins includes NFB1 (p50), NFB2 (p52), RelA (p65), RelB and c-Rel, among others. These comprise a family of transcription factors that exist as homodimeric or heterodimeric complexes in a variety of tissues. In most cell types, they remain inactive in the cytoplasm, through association with either an unprocessed dimeric subunit, such as p105, or a third protein, IB (Miyamoto and Verma, 1995). A number of stimuli are known to activate NF-B, including cytokines (Griffin et al., 1989), LPS (Sen and Baltimore, 1986), phorbol ester (Sen and Baltimore, 1986), reactive oxygen intermediates (Schreck et al., 1991) and physical stress (Stein et al., 1989). Despite the fact that these agents likely use a number of different signal transduction pathways, all induce the phosphorylation of IB and its proteolytic degradation (Whiteside et al., 1995).

Although inducible phosphorylation is required for degradation of IB and has been mapped to two amino-terminal serines, the responsible kinase is presently unknown. Several kinases are known to phosphorylate IB, including PKC (Ghosh and Baltimore, 1990) and casein kinase II (Janosch et al., 1996). However, these phosphorylations are located in the carboxyl-terminal end of the molecule and not at the critical S32/S36 (Brown et al., 1995). Recently, investigators identified novel kinases that are potential candidates for the IB kinase (Cao et al., 1996; Chen et al., 1996). Thus, these enzymes may play a critical role in catalyzing the phosphorylation at these critical serines, flagging the protein for subsequent ubiquitination and degradation via the 26S proteasome (Li et al., 1995; Palombella et al., 1994; Traenckner et al., 1994). Once IB has been degraded, the active NF-B dimer translocates to the nucleus, where it binds to its cognate enhancer element and induces the expression of a number of immediate-early genes involved in inflammatory and immune responses (Muller et al., 1993).

It is clear that NF-B plays a key role in the regulated expression of a large number of proinflammatory mediators, including cytokines such as IL-6 and IL-8 (Liberman and Baltimore, 1990; Mukaida et al., 1990), cell adhesion molecules such as intercellular adhesion molecule and vascular cell adhesion molecule (Kawai et al., 1995; Ledebur and Parks, 1995; Müller et al., 1995; Shu et al., 1993) and inducible nitric oxide synthase (Adcock et al., 1994a; Xie et al., 1994). Such mediators are known to play a role in the recruitment of leukocytes at sites of inflammation and, in the case of inducible nitric oxide synthase, may lead to organ destruction in some inflammatory and autoimmune diseases (Kleemann et al., 1993; McCartney-Francis et al., 1993).

The importance of NF-B in inflammatory disorders is further strengthened by models of airway inflammation (Adcock et al., 1994b; Blackwell et al., 1994; Haddad et al., 1996) and in rheumatoid arthritis synovium (Fujisawa et al., 1996; Handel et al., 1995) in which NF-B has been shown to be activated. This activation may underlie the increased cytokine production and leukocyte infiltration characteristic of these disorders. In addition, inhibition of NF-B may be the mechanism mediating the efficacy of steroids in the treatment of these disorders. Glucocorticoids have recently been shown to be potent inhibitors of NF-B (Auphan et al., 1995; Caldenhoven et al., 1995; Mukaida et al., 1994; Scheinman et al., 1995) and, as such, may suppress inflammatory mediator production through this mechanism. Thus, a novel inhibitor of NF-B activation would be of great therapeutic use in inflammatory disorders.

In this study, we examined the effects of the marine natural product hymenialdisine on NF-B activation in U937 cells, a cell of monocyte lineage. This compound was originally isolated from the marine sponges Axinella verrucosa and Acanthella aurantiaca (Cimino et al., 1982). The closely related analog debromohymenialdisine, also a marine natural product (Sharma et al., 1980), was shown to be effective in a model of adjuvant-induced arthritis (DiMartino et al., 1995). Its antiinflammatory properties were reported to be due to inhibition of PKC (DiMartino et al., 1995). Using various in vitro models and diverse stimuli, we show that hymenialdisine inhibits the activation of NF-B in a PKC-independent manner in U937 cells and that this effect is associated with inhibition of IL-8 production.

	Methods

Top Abstract Introduction Methods Results Discussion References

Materials. Hymenialdisine (SK&F 108752) was obtained from Suntory Ltd. (Japan). The selective PKC inhibitor RO 32-0432 was synthesized by the Department of Medicinal Chemistry, SmithKline Beecham Pharmaceuticals (King of Prussia, PA), according to the reported synthesis (Bit et al., 1993). TNF- was prepared at SmithKline Beecham Pharmaceuticals, as previously described (Chen et al., 1987). PMA and LPS were purchased from Sigma Chemical (St. Louis, MO).

Cell culture. The U937 human histiocytic lymphoma cell line was purchased from the American Type Culture Collection (CRL 1593; Rockville, MD) and was maintained in RPMI 1640 with 10% fetal bovine serum (Hyclone Laboratories, Logan, Utah). U937 clones permanently transfected with reporter plasmids (see below) were cultured in the above medium with the addition of 250 µg/ml geneticin (G418 sulfate; Life Technologies, Grand Island, NY).

Plasmid construction. The pHIVlucneo luciferase reporter plasmid was prepared by insertion of a 619-bp region comprising the HIV LTR into the Nhe I/HindIII sites of pGL2-basic (Promega, Madison, WI) directly upstream of the luciferase gene. To confer resistance for stable transfection, the neomycin resistance cassette was excised from pMAMneo (Clontech, Palo Alto, CA) by BamHI digestion and ligated into a like site in pHIVluc. A luciferase reporter plasmid containing the IL-8 core promoter, termed pIL8lucneo, was prepared by first isolating genomic DNA from U937 cells with the RapidPrep Micro Genomic DNA Isolation Kit (Pharmacia Biotech, Piscataway, NJ). The IL-8 promoter region spanning 317 to +7 was amplified by PCR and, after purification, ligated into the Kpn I/HindIII sites of pGL2-basic. The neomycin resistance cassette was inserted as described above.

Stable transfection. Stable transfections were performed using the electroporation method. Briefly, 20 µg of plasmid DNA, linearized with Sal I, was combined with 2 × 10⁷ U937 cells in 0.25 ml of PBS without Ca⁺⁺ and Mg⁺⁺. The mixture was electroporated (200 V, 960 µF) and then resuspended in culture medium and incubated for 1 hr at 37°C in 5% CO₂. A count of viable cells was performed by trypan blue exclusion, and the cells were diluted in culture medium to 1.5 × 10⁴ viable cells/ml. The cells were divided into 0.2-ml aliquots in 96-well tissue culture plates and allowed to incubate for 48 hr at 37°C in 5% CO₂. After the incubation, 150 µl of culture medium was removed from each well and replaced with an equal amount of fresh medium containing Geneticin so that the final concentration was 500 µg/ml. The plates were incubated at 37°C in 5% CO₂ until positive growth was seen. Wells containing positive growth were amplified and assayed for luciferase production (see below).

Luciferase reporter assay. Transfected U937 clones were twice centrifuged at 300 × g for 5 min and resuspended in RPMI 1640 with 10% FBS to a density of 1 × 10⁶ cells/ml. One-milliliter aliquots were added to the wells of 24-well plates. Inhibitor or DMSO carrier (1 µl) was added to the appropriate wells, and the plates were incubated at 37°C in 5% CO₂ for 30 min. The stimulus was added (5 ng/ml TNF-, 100 ng/ml LPS or 0.1 µM PMA), and the samples were incubated for 5 hr at 37°C in 5% CO₂, transferred to 1.9-ml polypropylene tubes and centrifuged at 200 × g for 5 min. The supernatants were collected and stored frozen at 20°C until assayed for IL-8 (see below). The cell pellets were washed twice in 1 ml of PBS without Ca⁺⁺ and Mg⁺⁺ and centrifuged as indicated above. The resulting cell pellets were lysed in 50 µl of 1× lysis buffer (Promega), vortexed and incubated for 15 min at room temperature. A 20-µl aliquot of each lysate was transferred to an opaque white 96-well plate (Wallac, Gaithersburg, MD) and assayed for luciferase production in a MicroLumat LB 96 P luminometer (EG&G Berthold, Bad Wilbad, Germany). The luminometer dispensed 100 µl of luciferase assay reagent (Promega) into each well and recorded the integrated light output for 20 sec. Light output was measured in RLUs.

Luciferase enzyme assay. Luciferase, isolated from Photinus pyralis (Boehringer-Mannheim Biochemicals, Indianapolis, IN), was resuspended in 0.5 M tris-acetate buffer, pH 7.5, at a stock concentration of 1 mg/ml. An aliquot of the stock was further diluted, and 50 µl (6.25 µg) was added per well to a 96-well opaque white plate. Inhibitor or DMSO carrier (0.5 µl) was added to the enzyme solution and allowed to incubate for 5 or 30 min. The samples were then assayed on a luminometer as described above.

Preparation of cellular and nuclear extracts. U937 cells were cultured to a density of 1 × 10⁶/ml. The cells were harvested by centrifugation, washed in PBS without Ca⁺⁺ and Mg⁺⁺ and resuspended in PBS with Ca⁺⁺ and Mg⁺⁺ at 1 × 10⁷ cells/ml. To examine the effect of hymenialdisine on the activation of NF-B, the cell suspensions were treated with various concentrations of drug or vehicle (DMSO, 0.1%) for 30 min at 37°C before stimulation with TNF- (5.0 ng/ml) for an additional 15 min. Cellular and nuclear extracts were prepared as previously described (Dignam et al., 1983; Osborn et al., 1989). Briefly, at the end of the incubation period, the cells (1 × 10⁷ cells) were washed twice in PBS without Ca⁺⁺ and Mg⁺⁺. The resulting cell pellets were resuspended in 20 µl of buffer A (10 mM HEPES, pH 7.9, 10 mM KCl, 1.5 mM MgCl₂, 0.5 mM DTT and 0.1% NP-40) and incubated on ice for 10 min. The nuclei were pelleted by microcentrifugation at 3500 rpm for 10 min at 4°C. The resulting supernatant was collected as the cellular extract, and the nuclear pellet was resuspended in 15 µl buffer C (20 mM HEPES, pH 7.9, 0.42 M NaCl, 1.5 mM MgCl₂, 25% glycerol, 0.2 mM EDTA, 0.5 mM DTT and 0.5 mM PMSF). The suspensions were mixed gently for 20 min at 4°C and then microcentrifuged at 14,000 rpm for 10 min at 4°C. The supernatant was collected and diluted to 60 µl with buffer D (20 mM HEPES, pH 7.9, 50 mM KCl, 20% glycerol, 0.2 mM EDTA, 0.5 mM DTT and 0.5 mM PMSF). All samples were stored at 80°C until analyzed. The protein concentration of the extracts was determined according to the method of Bradford (1976) with BioRad (Hercules, CA) reagents.

EMSA. The effect of hymenialdisine on transcription factor activation was assessed in an EMSA using nuclear extracts from treated cells as described above. The double-stranded NF-B consensus oligonucleotide (Santa Cruz Biotechnology, Santa Cruz, CA) (5-AGTTGAGGGGACTTTCCCAGGC-3), C/EBP concensus oligonucleotide (Santa Cruz Biotechnology) (5-TGCAGATTGCGCAATCTGCA-3) and an oligonucleotide representing a C/EBP site reported to be found in the HIV LTR (Mondal et al., 1995) (5-GATCGCTTGCTACAAGGCTTGCTACAAGG-3) were labeled with T₄ polynucleotide kinase and [-³²P]ATP. The binding mixture (25 µl) contained 10 mM HEPES-NaOH, pH 7.9, 4 mM Tris·HCl, pH 7.9, 60 mM KCl, 1 mM EDTA, 1 mM DTT, 10% glycerol, 0.3 mg/ml bovine serum albumin and 1 µg of poly(dI-dC)·poly(dI-dC). The binding mixtures (10 µg of nuclear extract protein) were incubated for 20 min at room temperature with 0.5 ng of ³²P-labeled oligonucleotide (50,000-100,000 cpm) in the presence or absence of unlabeled competitor, after which the mixture was loaded onto a 4% polyacrylamide gel prepared in 1× Tris borate/EDTA and electrophoresed at 200 V for 2 hr. After electrophoresis, the gels were dried and exposed to film for detection of the binding reaction.

IL-8 determination. U937 cells were treated as described above. At the end of the incubation period, the supernatants were collected. IL-8 in the U937 cell supernatants was quantified using an IL-8 ELISA kit from BioSource International (Camarillo, CA).

RT-PCR. U937 cells were cultured as described above. The cells were harvested by centrifugation and resuspended in culture medium at 1 × 10⁷ cells/ml. Cell samples (1 × 10⁷) were treated with various concentrations of hymenialdisine or vehicle (DMSO, 0.1%) for 30 min at 37°C in 5% CO₂, followed by stimulation with TNF- (5.0 ng/ml) for 3 hr. Total RNA was extracted from the samples using TRIzol reagent (Life Technologies). All of the RNA samples were treated with DNase (deoxyribonuclease I, amplification grade, Life Technologies) before use. The RT portion was carried out using the Reverse Transcription System (Promega) and the DNA amplification with Taq DNA polymerase (Fisher Scientific, Pittsburgh, PA), according to the manufacturers' instructions. A human IL-8 amplimer set yielding a 289-bp product and a human G3PDH amplimer set yielding a 983-bp product (Clontech) were used according to the manufacturers' instructions. In addition, amplimers were made to PAI-1, as previously described (Cobb et al., 1996). The primer sequences for PAI-1 (forward and backward) are: 5-CTACTTCAACGGCCAGTGGAAGCATC-3 and 5-GAGGCCAAGGTCTTGGAGACAGATCT-3. They yield an 801-bp product. Duplicate samples were run through the entire assay with no RT. As positive controls for amplification, full-length cDNAs for IL-8 and G3PDH (Clontech) (1 ng) were included in the PCR portion of the assay. Aliquots (15 µl) of the PCR samples were electrophoresed in 1.0% agarose gels in Tris acetate/EDTA buffer. The bands were visualized by ethidium bromide staining.

IB immunoblot analysis. Cellular extracts were subjected to SDS-PAGE on 10% gels (BioRad), and the proteins were transferred to nitrocellulose sheets (Hybond ECL, Amersham Corp., Arlington Heights, IL). Immunoblot assays were performed using a polyclonal rabbit antibody directed against a carboxyl-terminal portion of IB or a polyclonal rabbit antibody directed against a carboxyl-terminal portion of IB (Santa Cruz Biotechnology) at a 1:500 dilution for 1.5 hr, followed by a peroxidase-conjugated donkey anti-rabbit secondary antibody (Amersham). Immunoreactive bands were detected using the ECL assay system (Amersham).

	Results

Top Abstract Introduction Methods Results Discussion References

Hymenialdisine inhibits the activation of NF-B. Hymenialdisine was evaluated for its effects on the activation of NF-B using the luciferase reporters under the control of either the HIV-LTR or the IL-8 promoter. Both the HIV-LTR and IL-8 promoter are routinely used to study the regulation of NF-B-driven gene transcription. The transcriptional control of LTR has been widely studied. These studies have demonstrated that the major determinant of LTR activity is the binding of transcription factors to its enhancer region. The enhancer element contains a TATA box, three Sp1 sites and a strong enhancer composed of two tandemly arranged binding sites for NF-B (Jones et al., 1986; Nabel and Baltimore, 1987). Mutations of the B sites in the HIV enhancer have been reported to eliminate inducibility of the HIV enhancer (Osborn et al., 1989), suggesting that the binding of NF-B to its consensus motif is critical for LTR-mediated transcription. Similar results are seen with respect to the IL-8 promoter. Kunsch et al. (1994) demonstrated that NF-B plays a critical role in the regulation of gene transcription mediated by the IL-8 promoter. Mutation of the NF-B sites in the promoter results in a loss of inducible expression. As such, the U937/pHIVlucneo and U937/pIL8lucneo reporter constructs were chosen to investigate the effect of hymenialdisine on the activation of NF-B.

View larger version (K): [in this window] [in a new window]   Fig. 1.   Effect of different stimuli on luciferase production from the U937 luciferase reporters. U937/pHIVlucneo (A) and U937/pIL8lucneo (B) (1 × 106 cells/sample) were left untreated or stimulated with either 5 ng/ml TNF-, 100 ng/ml LPS or 0.1 µM PMA for 5 hr at 37°C in 5% CO2. Cellular extracts were prepared and measured for luciferase activity. Each bar represents the mean ± S.D. of 3 determinations.

View larger version (K): [in this window] [in a new window]   Fig. 2.   Effect of hymenialdisine on the U937 luciferase reporters. U937/pHIVlucneo and U937/pIL8lucneo (1 × 106 cells/sample) were pretreated for 30 min at 37°C in 5% CO2 with various concentrations of hymenialdisine and then stimulated for 5 hr with either 5 ng/ml TNF- (A), 100 ng/ml LPS (B) or 0.1 µM PMA (C). Cellular extracts were prepared and measured for luciferase activity. Each point represents the mean ± S.D. of 3 determinations.

Luciferase enzyme assay. To confirm that hymenialdisine was inhibiting the induction of luciferase production, and not the luciferase enzyme itself, various concentrations of hymenialdisine were combined with luciferase enzyme isolated from Photinus pyralis and incubated for 5 or 30 min before assessment of light output. Hymenialdisine showed no inhibitory effect on luciferase enzyme in the range of 0.1 to 10 µM (data not shown).

EMSA. The effect of hymenialdisine on the activation of NF-B was further monitored using an EMSA that allows the measurement of nuclear, hence active, NF-B in response to stimulation. Consistent with previous studies (Kaufman et al., 1991), U937 cells were found to contain significant levels of constitutive NF-B activity evidenced by the presence of a shifted complex in the resting cells (fig. 3). Stimulation with TNF- (5 ng/ml) for 15 min resulted in a 3-fold increase in NF-B proteins in the nuclear extract (4.1 density units in unstimulated cells vs. 13.0 density units in TNF--stimulated cells). Treatment of the U937 cells with hymenialdisine inhibited the activation of NF-B demonstrated by a decrease in NF-B protein. The effect was most evident at 1.0 µM hymenialdisine (fig. 3), at which ~50% inhibition of the stimulated binding reaction was seen (13.0 density units in control, stimulated cells vs. 8.3 density units in hymenialdisine-treated cells, basal unstimulated activity = 4.1 density units). Hymenialdisine had no effect in the absence of TNF- stimulation (data not shown). In addition, the inhibitory effect of hymenialdisine on the B motif-specific DNA binding proteins in the nuclei of U937 cells was confirmed by Western blot analysis of the nuclear extracts. In those studies, treatment of the cells with hymenialdisine before TNF stimulation was associated with a marked decrease in nuclear immunoreactivity for p50, p65 and cRel (data not shown).

View larger version (K): [in this window] [in a new window]   Fig. 3.   Effect of hymenialdisine on NF-B binding activity. Nuclear extracts were prepared from U937 cells treated with various concentrations of hymenialdisine for 30 min before stimulation with TNF- for 15 min. The extracts (10 µg) were tested for binding activity to a -32P-labeled NF-B concensus oligonucleotide.

View larger version (K): [in this window] [in a new window]   Fig. 4.   Effect of hymenialdisine on C/EBP binding activity. Nuclear extracts were prepared from U937 cells treated with various concentrations of hymenialdisine for 30 min before stimulation with TNF- for 15 min. The extracts (10 µg) were tested for binding activity to a -32P-labeled C/EBP concensus oligonucleotide (A) and a C/EBP oligonucleotide representing a binding site found in the HIV LTR (B).

Hymenialdisine inhibits U937 cell IL-8 production. IL-8 production was monitored in U937 cells as a cellular marker of NF-B activation. Hymenialdisine caused a concentration-dependent decrease in IL-8 production by U937 cells in response to all stimuli examined (fig. 5). The IC₅₀ values for inhibition of IL-8 production were in the range of 0.34 to 0.48 µM (table 1). Consistent with results obtained from the luciferase reporters, RO 32-0432 had no effect on TNF--stimulated IL-8 production but inhibited PMA-stimulated production, with an IC₅₀ value of 0.11 µM (table 2).

View larger version (K): [in this window] [in a new window]   Fig. 5.   Effect of hymenialdisine on U937 IL-8 production. U937 cells (1 × 106 cells/sample) were pretreated for 30 min 37°C in 5% CO2 with various concentrations of hymenialdisine and then stimulated for 5 hr in similar conditions with either 5 ng/ml TNF-, 100 ng/ml LPS or 0.1 µM PMA, as shown above. Cellular supernatants were analyzed for IL-8, as described in the text. Each point represents the mean ± S.D. of 3 determinations.

Hymenialdisine inhibits U937 cell IL-8 mRNA production. To confirm that hymenialdisine was acting through inhibition of a transcription-related mechanism, RT-PCR was used to monitor U937 IL-8 message. TNF- stimulated the production of IL-8 mRNA, which was inhibited by hymenialdisine in a concentration-dependent manner (fig. 6). To serve as a negative control, the same RNA samples were amplified using primers for PAI-1, a gene reported to be activated independently of NF-B (Descheemaeker et al., 1992). Although TNF- was a strong activator of PAI-1 message, hymenialdisine had no significant inhibitory effect (data not shown). Likewise, the housekeeping gene G3PDH was unaffected by hymenialdisine (data not shown).

View larger version (K): [in this window] [in a new window]   Fig. 6.   Effect of hymenialdisine on U937 IL-8 mRNA. Total RNA was prepared from U937 cells treated with hymenialdisine before TNF- stimulation. RNA samples (1 µg) were subjected to RT-PCR as described in the text. Duplicate samples were run through the assay minus RT, as shown above. Samples were electrophoresed in a 1.0% agarose gel and visualized by ethidium bromide staining. Lanes 1 and 2, unstimulated U937 cells; lanes 3 and 4, 5 ng/ml TNF-; lanes 5 and 6, 0.1 µM hymenialdisine + TNF-; lanes 7 and 8, 1.0 µM hymenialdisine + TNF-; lanes 9 and 10, 10 µM hymenialdisine + TNF-. Left, DNA bp ladder.

IB phosphorylation and degradation. The activation of NF-B is dependent on the phosphorylation and subsequent degradation of IB. Thus, the effect of hymenialdisine on IB protein levels was investigated as a possible mechanism of action. Cellular extracts were prepared from U937 cells treated with various concentrations of hymenialdisine before TNF- stimulation. Western blot analyses of both IB and IB were then conducted. TNF- stimulation caused a marked decrease in IB protein compared with unstimulated controls (fig. 7). Hymenialdisine had no effect on the disappearance of IB (fig. 7), suggesting that its site of action was downstream of IB. In addition, Western blot analysis of IB showed that the levels of IB remained unchanged in all samples (fig. 7).

View larger version (K): [in this window] [in a new window]   Fig. 7.   Effect of hymenialdisine on U937 IB phosphorylation and degradation. Cellular extracts were prepared from U937 cells treated with various concentrations of hymenialdisine for 30 min before stimulation with TNF- for 15 min. Samples (50 µg) were separated on a 10% SDS-PAGE gel, followed by electrophoretic transfer to a nitrocellulose membrane. The proteins were immunoblotted with anti-IB polyclonal rabbit serum or anti-IB polyclonal rabbit serum as shown, and the bands were visualized as described in the text.

	Discussion

Top Abstract Introduction Methods Results Discussion References

The complexity of the NF-B activation process, in which the endogenous inhibitor of NF-B, IB, is phosphorylated, ubiquitinated and subsequently degraded, allowing migration of NF-B to the nucleus, presents many potential sites for pharmacological intervention. In light of the role of NF-B as a coordinating regulator in the expression of a variety of rapid-response genes involved in inflammatory and immune reactions, therapeutic regulation of NF-B would likely be of benefit in the treatment of immune and inflammatory disorders.

The present study focused on the marine natural product hymenialdisine. Debromohymenialdisine, an analog of hymenialdisine, has been shown to possess anti-inflammatory activity in a model of adjuvant-induced arthritis in the rat (DiMartino et al., 1995). Although these compounds possess relatively potent PKC inhibitory activity, the studies described herein suggest that the anti-inflammatory activity of these compounds, and of hymenialdisine in particular, may be related to their ability to inhibit the activity of the transcription factor NF-B. Evidence for the ability of hymenialdisine to inhibit the activity of NF-B was obtained in studies examining the effect of this compound in two separate NF-B-driven luciferase reporter assays. In these studies, the HIV LTR-driven and IL-8 core promoter-driven luciferase reporter vectors were stably transfected into U937 cells. Although both of these constructs contain essential NF-B sites, their binding sites differ in their affinity for the various NF-B dimers. Although the B sites present in the HIV LTR promoter prefer a heterodimer composed of p50/p65, that in the IL-8 promoter has been reported to bind to a dimer composed of c-Rel/p65 (Parry and Mackman, 1994). Thus, the finding that hymenialdisine was equipotent against both reporters suggest that hymenialdisine exerts its inhibitory effect at a site common to the activation of both sets of heterodimers. Furthermore, hymenialdisine did not exhibit any selectivity with respect to stimulus used to activate the cells. Luciferase activities induced by TNF-, LPS and PMA were equally inhibited by treatment with hymenialdisine.

Evidence of the ability of hymenialdisine to inhibit NF-B activity was also seen in its inhibitory activity in an EMSA. Nuclear extracts from TNF--stimulated U937 cells pretreated with hymenialdisine showed a slight decrease in binding to an NF-B consensus oligonucleotide compared with untreated cells. Furthermore, subsequent studies demonstrated that hymenialdisine had no effect on the ability of other transcription factors, namely, C/EBP, Sp1 and AP-1, to bind to their consensus motifs. These findings were particularly important in light of the fact that the HIV LTR contains consensus binding motifs for the transcription factors AP-1 and Sp1 (Franza et al., 1988; Jones et al., 1986) and for members of the C/EBP family (Mondal et al., 1995; Ruocco et al., 1996), whereas that of the IL-8 promoter contains sites for C/EBP (NF-IL-6) and AP-1 (Mukaida et al., 1989, 1990). The lack of an effect by hymenialdisine in an EMSA using oligonucleotides containing these various transcription factor binding motifs suggests that the inhibitory activity seen in the luciferase reporter assay is indeed mediated through an inhibition of NF-B. However, these studies do not clearly address a possible action of hymenialdisine on the synergistic activation by NF-B and C/EBP of these promoters (Kunsch et al., 1994; Mukaida et al., 1990; Ruocco et al., 1996; Stein and Baldwin, 1993).

Associated with the inhibition of NF-B seen in response to treatment with hymenialdisine was an inhibition of IL-8 production by U937 cells. Such inhibition indicated that inhibition of NF-B in these cells was reflected in a least one functional parameter (i.e., IL-8 production). The effect of hymenialdisine on IL-8 production was mediated at the level of gene transcription in that RT-PCR demonstrated a significant reduction in IL-8 mRNA levels. In contrast, hymenialdisine had no significant effect on G3PDH (a housekeeping gene) and PAI-1 (a TNF--stimulated, NF-B-independent gene) (Descheemaeker et al., 1992), indicating that the inhibition of IL-8 production and luciferase reporter activity were not mediated through nonspecific effects on gene transcription.

The effect of hymenialdisine observed in the EMSA suggested that this compound might be inhibiting the activation of NF-B; therefore, hymenialdisine was studied for its effects at the level of IB. These studies revealed that hymenialdisine had no effect on the breakdown of IB, suggesting that the inhibition of NF-B was not due to an inhibition of the putative IB kinase(s) or to an inhibition of IB degradation. As such, hymenialdisine appears to inhibit NF-B activity at a site downstream of IB. Although the exact mechanism by which hymenialdisine inhibits NF-B activation is unclear, PKC does not appear to play a direct role in the activation process in response to TNF-. The selective PKC inhibitor RO 32-0432 had no effect on TNF--stimulated luciferase reporter activity or on TNF--stimulated IL-8 production. The ability of this compound to inhibit PKC in U937 cells is demonstrated by its very potent inhibition of PMA-stimulated responses. These results are consistent with those of Meichle et al. (1990), who showed that TNF- activated NF-B in a PKC-independent manner.

In summary, the results of the present study demonstrate that the marine natural product hymenialdisine is an inhibitor of the transcription factor NF-B. The ability of this compound to inhibit NF-B-driven inflammatory gene products lends support to the proposal that an inhibitor of NF-B will provide an anti-inflammatory therapeutic agent.