Interleukin-1β and Tumor Necrosis Factor-α Induce MUC5AC Overexpression through a Mechanism Involving ERK/p38 Mitogen-activated Protein Kinases-MSK1-CREB Activation in Human Airway Epithelial Cells*

Mucin hypersecretion is commonly observed in many inflammatory diseases of the respiratory tract. MUC5AC is generally recognized to be a major airway mucin because MUC5AC is highly expressed in the goblet cells of human airway epithelium. Moreover, it is regulated by various inflammatory cytokines. However, the mechanisms by which the interleukin (IL)-1β and tumor necrosis factor (TNF)-α induce MUC5AC gene expression in normal nasal epithelial cells, and the signal molecules involved, especially in the downstream signaling of mitogen-activated protein (MAP) kinases, remain unclear. Here we show that pharmacologic or genetic inhibition of either ERK or p38 MAP kinase pathway abolished IL-1β- and TNF-α-induced MUC5AC gene expression in normal human nasal epithelial cells. Our results also indicate that the activation of mitogen- and stress-activated protein kinase 1 (MSK1) and cAMP-response element-binding protein and cAMP-response element signaling cascades via ERK and p38 MAP kinases are crucial aspects of the intracellular mechanisms that mediate MUC5AC gene expression. Taken together, these studies give additional insights into the molecular mechanism of IL-1β- and TNF-α-induced MUC5AC gene expression and enhance our understanding on mucin hypersecretion during inflammation.

It is well documented that mitogen-activated protein (MAP) kinase pathways are thought to be most important in transmitting inflammatory signals from the cell surface to the nucleus (30). After being triggered by growth factors, cytokines, UV rays, or other stress-inducing agents, a signal is delivered down the MAPKKK 3 MAPKK (in the cases of ERK, JNK, and p38, the signal is delivered through MEK1/2, MKK4/7, and MKK3/6, respectively) to the MAP kinase cascade. The MAP kinases play a role in cell proliferation, differentiation, apoptosis, cytoskeletal remodeling, and the cell cycle (31)(32)(33)(34)(35)(36). Mitogen-and stress-activated protein kinase 1 (MSK1) is a recently identified enzyme that is widely distributed in mammalian cells (37)(38)(39). MSK1 is activated in vitro and in vivo by two different classes of MAP kinases, ERK and p38 MAP kinases (37). Moreover, MSK1 is localized in the nuclei of stimulated or unstimulated cells (39), and two potential in vivo substrates are the cAMP-response element-binding protein (CREB) and the closely related activating transcription factor 1 (ATF1) (39).
Recently, reactive oxygen species were reported to increase the expression of the MUC5AC gene, by activating the ERK MAP kinase pathway (25), and non-typeable Haemophilus influenzae was reported to regulate MUC5AC transcription via p38 MAP kinase in human epithelial cells (40). In a study of MUC2, Pseudomonas aeruginosa was found to activate NF-B through Ras-MAPK-pp90rsk, which led to the increased expression of MUC2, but p38 was not involved in this pathway (41). However, the mechanism of MUC5AC gene expression during inflammation in normal airway epithelial cells, and the signal molecules involved, especially in the downstream signaling of MAP kinases have not yet been demonstrated.
Because MUC5AC hypersecretion during inflammation plays an important role in the pathogenesis of airway diseases, we hypothesized that major inflammatory cytokines, IL-1␤ and TNF-␣, up-regulate MUC5AC gene expression by activating specific signal transduction pathways in airway epithelial cells. Here we show that two different MAP kinases, ERK and p38 MAP kinases, are essential for IL-1␤-and TNF-␣-induced MUC5AC gene expression in normal human nasal epithelial (NHNE) cells. We also show that MSK1 mediates the IL-1␤and TNF-␣-induced phosphorylation of CREB and the transcription of MUC5AC. Furthermore, the cAMP-response element (CRE) in the MUC5AC promoter appears to be important for IL-1␤-and TNF-␣-induced MUC5AC gene expression in NCI-H292 cells. These pathways provide new insights into molecular mucous hypersecretion and may open up novel targets for therapeutic intervention.
Cell Cultures-The culture system used for the normal human nasal epithelial (NHNE) cells was as described previously (42). The human lung mucoepidermoid carcinoma cell line (NCI-H292) was purchased from the American Type Culture Collection (CRL-1848; Manassas, VA) and cultured in RPMI 1640 (Invitrogen) supplemented with 10% fetal bovine serum in the presence of penicillin/streptomycin at 37°C in a humidified chamber with 5% CO 2 . For serum deprivation, confluent cells were washed twice with phosphate-buffered saline and recultured in RPMI 1640 with 0.2% fetal bovine serum.
RT-PCR-Total RNA was isolated using TRIzol (Invitrogen) from NCI-H292 cells treated with IL-1␤ (10 ng/ml) or TNF-␣ (10 ng/ml). cDNA was synthesized with random hexamers (PerkinElmer Life Sciences and Roche Applied Science) using Moloney murine leukemia virus-reverse transcriptase (PerkinElmer Life Sciences). Oligonucleotide primers for PCR were designed based on the GenBank TM sequence of MUC5AC (GenBank TM accession number AJ001402, 5Ј primer CGA-CAACTACTTCTGCGGTGC; 3Ј primer GCACTCATCCTTCCTGTC-GTT). The following PCR conditions used involved 35 cycles: denaturation at 94°C for 30 s, annealing at 60°C for 30 s, and polymerization at 72°C for 30 s. The oligonucleotide primers for ␤ 2 -microglobulin (used as a control gene for the RT-PCR) were designed based on the Gen-Bank TM human sequence (GenBank TM accession number XM007650, 5Ј primer CTCGCGCTACTCTCTCTTTCTGG; 3Ј primer GCTTACAT-GTCTCGATCCCACTTAA). PCR parameters used involved 23 cycles as follows: denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and polymerization at 72°C for 30 s. The PCR products were run in a 1.5% agarose gel and visualized with ethidium bromide under a transilluminator.
Real Time-PCR Primers and probes were designed with PerkinElmer Life Sciences Primer Express software and purchased from PE Biosystems. Commercial reagents (TaqMan PCR Universal PCR Master Mix, PerkinElmer Life Sciences) and conditions according to the manufacturer's protocol were applied. An amount of 1 g of cDNA (reverse transcription mixture) and oligonucleotides at a final concentration of 800 nM of primers and 200 nM of TaqMan hybridization probes were analyzed in a 25-l volume. The probe of real-time PCR was labeled with carboxyfluoroscein (FAM) at the 5Ј end and with the quencher carboxytetramethylrhodamine (TAMRA) at the 3Ј end. The following primers and TaqMan probes were used: MUC5AC, forward 5Ј-CAGCCACGTCCCCTTCAATA-3Ј and reverse 5Ј-ACCGCATTTGG-GCATCC-3Ј; TaqMan probe 6FAM-CCACCTCCGAGCCCGTCACTGA-G-TAMRA; ␤ 2 -microglobulin, forward 5Ј-CGCTCCGTGGCCTTAGC-3Ј and reverse 5Ј-GAGTACGCTGGATAGCCTCCA-3Ј; and TaqMan probe 6FAM-TGCTCGCGCTACTCTCTCTTTCTGGC-TAMRA. Real time RT-PCR was performed on a PerkinElmer Life Sciences ABI PRISM 7700 Sequence Detection System. The thermocycler (ABI PRISM 7700 Sequence Detection System) parameters were 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. All reactions were performed in triplicate. Relative quantity of MUC5AC mRNA was obtained using a comparative cycle threshold method and was normalized using ␤ 2 -microglobulin as an endogenous control.
Western Blot Analysis-NCI-H292 cells were grown to confluence in 6-well plates. After 15 or 45 min of treatment with IL-1␤ or TNF-␣, respectively, the cells were lysed with 2ϫ lysis buffer (250 mM Tris-Cl (pH 6.5), 2% SDS, 4% ␤-mercaptoethanol, 0.02% bromphenol blue, 10% glycerol). Equal amounts of whole cell lysates were resolved by 10% SDS-PAGE and transferred to a polyvinylidene difluoride membrane (Millipore, Bedford, MA). Membranes were blocked with 5% skim milk in Tris-buffered saline (50 mM Tris-Cl (pH 7.5), 150 mM NaCl) for 2 h at room temperature. This blot was then incubated overnight with primary antibody in TTBS (0.5% Tween 20 in Tris-buffered saline). After washing with TTBS, the blot was further incubated for 45 min at room temperature with anti-rabbit or anti-mouse antibody (Cell Signaling) in TTBS and then visualized by using the ECL system (Amersham Biosciences).
In Vitro p38 Kinase Assay-p38 kinase activity was measured using a p38 MAP kinase assay kit (Cell Signaling) according to the manufacturer's instructions. Briefly, confluent cells were rendered quiescent for 24 h and then incubated with or without 20 M SB203580 for 1 h prior to being stimulated with IL-1␤ or TNF-␣ for 15 min. Cells lysates were scraped off the dish with 500 l of lysis buffer and 1 mM PMSF, sonicated 4 times for 5 s each on ice, and centrifuged for 10 min at 4°C. Supernatants were then transferred to a new tube, and 400 g of the cell lysates and 20 l of immobilized phospho-p38 MAP kinase monoclonal antibody were incubated with gentle rocking overnight at 4°C. The pellet was washed twice with lysis and kinase buffer and then resuspended in kinase buffer containing 200 M ATP and 2 g of activating transcription factor 2 (ATF2) fusion protein. It was then incubated for 30 min at 30°C and immunoblotted with phospho-ATF2 antibody.
Preparation of Inducible Dominant-negative Mutant Stable Cell Lines-Plasmid encoding the kinase-deficient MEK1 mutant (pcDNA5-MEK1DN) was cut with BamHI and ligated with pBluescript (Stratagene, La Jolla, CA). This clone was cut with HindIII, filled in with Klenow, cut with SacII (Promega), and then ligated with pTRE vector. Plasmid encoding kinase-inactive p38 mutant (pcDNA3-p38(AGF)) was cut with BamHI, filled in with Klenow, cut with XbaI (Promega, Madison, WI), and then ligated with pTRE vector (Clontech, Palo Alto, CA). NCI-H292 cells were cotransfected with pTet-off (Clontech) regulation vector and pTRE-p38DN or pTRE-MEK1DN (1:20 ratio of regulation vector to expression vector) using FuGENE 6 transfection reagent (Roche Applied Science), following the procedure recommended by the manufacturer. Stably transfected cell lines were selected with 200 g/ml G418 (Calbiochem), and the medium was replaced with G418 and doxycycline every 3 days.
Plamids, Transient Transfection, and Luciferase Assay-Cells were transiently transfected with plasmids containing wide-type MSK1, Nterminal kinase dead MSK1 mutant (D195A), C-terminal kinase dead MSK1 mutant (D565A), CREB DN (S133A), and reporter constructs, and the MUC5AC 5Ј-flanking region was fused to a luciferase reporter gene using FuGENE 6 transfection reagent (Roche Applied Science) according to the manufacturer's instructions. Deletion mutants covering promoter regions of MUC5AC were generated by PCR using pairs of primers bearing specific restriction sites at their 5Ј and 3Ј ends (Table  I) and then were constructed in promoterless pGL3 basic vector. Cells were incubated for 48 h, harvested, and assayed for luciferase activity, using a luciferase assay system (Promega), according to the manufacturer's instructions. ␤-Galactosidase activity was also assayed to standardize the transfection efficiency of each sample.

IL-1␤ and TNF-␣ Can Induce the Gene Expression of MUC5AC through ERK and p38 MAP Kinases
Signaling in NHNE Cells-To determine whether IL-1␤ and TNF-␣ can induce MUC5AC gene expression within NHNE cells, we carried out RT-PCR after treatment with IL-1␤ or TNF-␣ for 24 h. The results showed that MUC5AC mRNA was significantly increased after treatment with IL-1␤ or TNF-␣ in NHNE cells (Fig. 1A). No corresponding change was found in the expression of internal control, ␤ 2 -microglobulin. As a next step, to determine which MAP kinase signal pathway is activated within NHNE cells stimulated by IL-1␤ and TNF-␣, we performed a Western blot analysis using phospho-specific antibodies. ERK and p38 MAP kinases were maximally activated at 15 min, and this effect decreased after 45 min (Fig. 1B). However, no change was detected in the activation of JNK. A549 cells were used as a positive control for JNK activation. It thus appeared that stimulation by IL-1␤ and TNF-␣ induced the ERK and p38 MAP kinases pathways in NHNE cells. In order to investigate the possible involvement of ERK and p38 MAP kinases in IL-1␤-and TNF-␣-induced MUC5AC gene expression, 20 M PD98059, specific MEK1/2 inhibitor, or 20 M SB203580, p38 inhibitor, were applied before treatment with IL-1␤ and TNF-␣. The Western blot and in vitro kinase assays showed that PD98059 and SB203580 clearly inhibited ERK and p38 MAP kinases, respectively, in NHNE cells (Fig. 2, A and B). Under this experimental condition, we checked the expression level of MUC5AC by performing real time quantitative PCR analysis. These results showed that pretreatment with PD98059 or SB203580 for 1 h inhibited MUC5AC gene expression (Fig. 2, C and D). Interestingly, the inhibition of either ERK or p38 MAP kinase pathway inhibited MUC5AC mRNA in NHNE cells. Thus, the activation of ERK and p38 MAP kinases appeared to be closely related to the signaling pathways activated by IL-1␤ or TNF-␣.
Both ERK and p38 MAP Kinases Are Essential for IL-1␤-or TNF-␣-induced MUC5AC Gene Expression-When the same experiments (Figs. 1 and 2) were performed using NCI-H292 cells, human lung mucoepidermoid carcinoma cell line, the results were the same in the NCI-H292 cells as in the normal cells (data not shown). In order to confirm the significance of ERK and p38 kinases upon the cellular level of MUC5AC gene expression, cells stably expressing dominant-negative (DN) mutant under the control of the Tet-off system were generated. The p38 DN mutant was generated by replacing Thr 180 and Tyr 182 by Ala and Phe, respectively (43). This inactive form of p38 MAP kinase binds endogenous substrates, thereby inhibiting signaling by the endogenous p38 MAP kinase pathway. After removing doxycycline to induce MEK1DN, stimulation with IL-1␤ or TNF-␣ for 15 min decreased the phosphorylation of ERK (Fig. 3A). However, no change in ERK expression was observed. Real time quantitative PCR showed a decrease in MUC5AC mRNA after 24 h for IL-1␤ and TNF-␣. In a similar way, we investigated the role of p38 MAP kinase on IL-1␤-or TNF-␣-induced MUC5AC gene expression using p38DN. An in vitro kinase assay showed that the activation of p38 MAP kinase in this mutant stable cell line was decreased following IL-1␤ and TNF-␣ stimulation (Fig. 3B). Under this condition, MUC5AC mRNA was decreased 24 h after treatment with IL-1␤ and TNF-␣. These results showed that ERK and p38 MAP kinases are essential for IL-1␤-and TNF-␣-induced  MUC5AC gene expression in NCI-H292 cells.
IL-1␤-or TNF-␣-induced Activation of MSK1 Is Mediated by Both ERK and p38 MAP Kinases-To determine which molecules are involved in the downstream signaling of the ERK and p38 MAP kinases within the signal pathway of MUC5AC gene expression induced by IL-1␤ or TNF-␣, we investigated the phosphorylation of MSK1 by phospho-specific antibody. MSK1 is widely distributed in mammalian cells and can be activated by MAPK/ERK and SAPK2a/p38 (37)(38)(39). The result showed that the phosphorylation of MSK1 by IL-1␤ or TNF-␣ reached a maximum at 30 min and decreased at 60 min after IL-1␤ and TNF-␣ stimulation (Fig. 4, A and B). Pretreatment with 20 M PD98059 and/or 20 M SB203580 inhibited IL-1␤-or TNF-␣induced MSK1 phosphorylation (Fig. 4, C and D), indicating that MSK1 is regulated by ERK and/or p38 MAP kinase(s). Furthermore, to determine whether MSK1 influences MAP kinases, cells were transfected with DNA expression constructs encoding a mutant MSK1 (NT-KD) that the N-terminal kinase domain was inactivating by a point mutation and a further mutant (CT-KD) that the C-terminal kinase domain was inactivating (39). Two MSK1 mutants did not affect IL-1␤-and TNF-␣-induced MAP kinases activation (Fig. 4E), indicating that MSK1 appears to be controlled by MAP kinases. These results showed that MSK1 acts as a downstream signaling mediator of ERK and p38 MAP kinases.
Effects of MSK1 and CREB on IL-1␤-and TNF-␣-induced MUC5AC Gene Expression-To examine the role of MSK1 on the induction of MUC5AC gene expression, cells were transiently transfected with DNA expression constructs encoding wild-type MSK1 (W/T), NT-, or CT-KD MSK1 mutant. IL-1␤ or TNF-␣-induced MUC5AC gene expression increased in cells transfected with W/T MSK1, whereas overexpression of NT-or CT-KD MSK1 markedly suppressed IL-1␤ or TNF-␣-induced MUC5AC gene expression (Fig. 5, A and B). These results showed that MSK1 appears to be closely related in the MUC5AC gene expression by IL-1␤ and TNF-␣.
Because MSK1 is currently the best candidate for the mediation of cytokine-induced CREB phosphorylation at Ser 133 (37-39), we investigated the possible implication of CREB in IL-1␤or TNF-␣-induced MUC5AC gene expression. To examine whether IL-1␤ or TNF-␣ can induce the phosphorylation of endogenous CREB via MAP kinases and MSK1 in airway epithelial cells, we performed Western blot using phospho-specific CREB (Ser 133 ) antibody. A transient phosphorylation of CREB was observed upon the stimulation with IL-1␤ and TNF-␣, reaching the maximum peak at 30 min (Fig. 6, A and B). In addition, pretreatment with PD98059 and/or SB203580 or NTor CT-KD MSK1 remarkably inhibited the phosphorylation of CREB (Fig. 6, C and D). Next, to determine whether CREB plays a role in MUC5AC gene expression, we used forskolin (an activator of adenylate cyclase) and 3-isobutyl-1-methylxanthine (IBMX; an inhibitor of AMP phosphodiesterase). A transient phosphorylation of CREB was observed upon the stimulation with forskolin and IBMX, reaching the maximum peak at 10 min (Fig. 7A). The cAMP pathway-induced CREB phosphorylation increased MUC5AC gene expression (Fig. 7B). Furthermore, IL-1␤-and TNF-␣-induced MUC5AC gene expression was significantly suppressed in cells transfected with plasmid encoding mutant CREB (pCREB S133A) (Fig. 7C). These findings suggested that the activation of MSK1 and CREB is essential for IL-1␤-and TNF-␣-induced MUC5AC gene expression via ERK and p38 MAP kinases.
Identification of the Binding Complex between CREB and CRE in Response to IL-1␤ and TNF-␣-To analyze the DNA binding activity of IL-1␤-and TNF-␣-activated CREB, we performed EMSA using nuclear extracts from NCI-H292 cells after IL-1␤ or TNF-␣ treatment. As shown in Fig. 8, A and B, the activity of consensus CRE oligonucleotide (CREc) and MUC5AC specific CRE (CREs) remarkably increased in response to IL-1␤ or TNF-␣ but not by mutant CRE of MUC5AC promoter (CREm) oligonucleotide. To distinguish any specific CRE-binding complexes, competition and supershift analysis were performed using 50-fold excesses of non-radiolabeled (cold) CREs oligonucleotide and anti-phospho-CREB antibody, respectively. The specific band was found to be selectively inhibited by the specific CRE competitor and was supershifted by anti-phospho-CREB antibody. These results indicated that activated CREB binds to a cis-acting element, CRE, in the MUC5AC promoter.
Identification of IL-1␤-and TNF-␣-responsive Regions within MUC5AC Promoter-Cells were then transiently transfected with the various deletion mutants and treated with IL-1␤ (40 ng/ml) or TNF-␣ (40 ng/ml) for 24 h, respectively. As shown in Fig. 9A, IL-1␤ and TNF-␣ selectively increased luciferase activity of Ϫ929/ϩ4 region of MUC5AC promoter. No effect was seen on fragments covering Ϫ1376/ϩ4, Ϫ776/ϩ4, and Ϫ486/ϩ4 regions, indicating that the Ϫ929/Ϫ776 region of MUC5AC promoter may be necessary to observe a response to IL-1␤ or TNF-␣. To further know whether CRE within the Ϫ929/Ϫ776 region of the MUC5AC promoter, identified using the TRANSFAC 4.0 data base, critically acts as cis-element, cotransfection with plasmid expression construct encoding mutant CREB was performed to study its effect on MUC5AC transcription activity. CREB DN suppressed luciferase activity of Ϫ929/ϩ4 region of MUC5AC promoter (Fig. 9B). Moreover, we examined whether activation of CRE is required for IL-1␤and TNF-␣-induced MUC5AC transcription by performing selective mutagenesis of the CREB-binding site. As a shown in Fig. 9C, mutant constructs M1, M2, and M3 abolished responsiveness of wild-type MUC5AC promoter construct (Fig. 9C). These results showed that CRE in the regulatory region of MUC5AC promoter was critical for the up-regulation of the transcriptional activity of MUC5AC induced by IL-1␤ or TNF-␣. DISCUSSION Mucin hypersecretion causes many clinical problems, such as rhinorrhea, nasal stuffiness, and sputum in the respiratory tract. It has been reported that MUC5AC is the major mucin in human airways (22)(23)(24)44). The mechanism of the regulation of MUC5AC secretion by inflammatory cytokines in airway is very important, and the understanding of this mechanism may offer new therapeutic strategies for the inhibition of airway mucus hypersecretion.
The molecular mechanism by which MUC5AC is up-regulated by IL-1␤ and TNF-␣ remains poorly understood. In the present study, we undertook to determine the involvement of IL-1␤ and TNF-␣ in the up-regulation of MUC5AC gene expression in normal human airway epithelial cells.
The fact that more than one MAP kinase may be necessary for the IL-1␤-and TNF-␣-induced MUC5AC gene expression in NHNE and NCI-H292 cells is an interesting finding of the present study (Figs. 2 and 3). Although MUC5AC is regulated by various inflammatory cytokines such as neutrophil elastase (27), IL-9 (28), and IL-4 (29), it was not shown which mechanisms are essential for cytokine-induced MUC5AC gene expression. Recently, Takeyama et al. (25) reported that epidermal growth factor increased MUC5AC gene expression via ERK MAP kinase but not p38 MAP kinase in NCI-H292 cells. Moreover, Wang et al. (40) showed that non-typeable H. influenzae regulated MUC5AC transcription via p38 MAP kinase but did not mediate ERK MAP kinase. In the present study, we showed that both ERK and p38 MAP kinase, but not JNK signaling, are essential for IL-1␤-and TNF-␣-induced MUC5AC gene expression. These suggest that the signaling pathways leading to MUC5AC gene expression are distinct, depending on the type of stimuli and cell lines used. We do not yet know how both pathways intervene in the activation of cytokine-induced MUC5AC gene expression. Recently, TNF-␣induced matrix metalloproteinase (MMP)-1 and MMP-3 gene expression is known to be regulated through AP-1-dependent transcriptional activation via the ERK pathway and AP-1independent enhancement via p38 MAPK by mRNA stabilization in human skin fibroblasts (44). Interestingly, TNF-␣ has an effect on the stability of MUC5AC mRNA in NCI-H292 cells (45,46). Thus, taken together, it is conceivable that the intracellular signaling coordination controlled by ERK in combination with p38 MAP kinase may be essential for IL-1␤-and TNF-␣-induced MUC5AC gene expression.
To date, signal molecules involved in the downstream signaling of MAP kinases for MUC5AC gene expression have not been yet demonstrated. The role of MSK1 and CREB in the downstream signaling of MAP kinases in the IL-1␤-and TNF-␣-induced MUC5AC gene expression in airway epithelial cells is a major finding of the present study. MSK1 is known to be regulated by MAPK/ERK and SAPK2a/p38 and is currently the best candidate for the mediation of cytokine-induced CREB phosphorylation at Ser 133 (37)(38)(39)47). Although CREB activation by MSK1 has been established by previous studies in other cells (37)(38)(39), it has remained unclear in airway epithelial cells. In addition, little is known about the involvement of CREB in MUC5AC gene expression. In this study, our results show that CREB activation is involved in the downstream signaling of MAP kinases and MSK1 for IL-1␤-and TNF-␣-induced MUC5AC gene expression.
Interestingly, MUC5AC mRNA was inhibited in pCREB S133A transfected cells treated by IL-1␤ and TNF-␣, and treatment with forskolin and IBMX activated the phosphorylation of CREB and increased MUC5AC gene expression (Fig. 7, A and  B). These suggest that CREB may be a transcription factor for IL-1␤-and TNF-␣-induced MUC5AC gene expression. However, the increase of MUC5AC expression by forskolin and IBMX was less than that induced by IL-1␤ and TNF-␣, suggesting that activation by CREB alone is insufficient to induce the full expression of MUC5AC. This result suggests that a transcription factor, other than CREB, may be required for full expression of IL-1␤and TNF-␣-induced MUC5AC gene expression. Previously, P. aeruginosa-induced MUC2 expression is found to be regulated by NF-B in NCI-H292 cells (41). In fact, we could find the putative NF-B-binding site at Ϫ273 and Ϫ956 in MUC5AC promoter (48). Therefore, NF-B may also regulate cytokine-induced MUC5AC gene expression with the cooperation of CREB. This suggestion was further supported by our recent finding that IL-1␤ or TNF-␣ can initiate IB degradation in NHNE and NCI-H292 cells and that pretreatment of caffeic acid phenethyl ester, which is known to specifically block the translocation of p65 without affecting IB␣ degradation (41), inhibits MUC5AC gene expression induced by IL-1␤ (data not shown). In fact, Gerritsen et al. (49) reported that p300 and CREB-binding protein (CBP) act as coactivators of p65 transactivation and may play an important role in the cytokine-induced expression of various immune and inflammation genes. Furthermore, Perrais et al. (50) reported that transcription factor Sp1 is essential for epidermal growth factorand TGF-␣-mediated MUC5AC up-regulation. Taken together, these findings suggest that CREB may interact directly or indirectly with other transcription factor(s) and that non-DNA binding transcriptional coactivators, such as p300 and CBP, FIG. 8. IL-1␤-or TNF-␣-induced nuclear binding of CRE. Confluent, quiescent cells were stimulated for 1 h with IL-1␤ (A) or TNF-␣ (B). Nuclear protein extracts from IL-1␤-or TNF-␣-treated NCI-H292 cells were subjected to EMSA. Nuclear proteins were incubated with CREc, CREs, and CREm, 50-fold excess of cold probe or anti-phospho-CREB antibody before EMSA. The labeled nuclear proteins were separated by electrophoresis on 5% polyacrylamide gels, and the gels were dried and exposed to autoradiography at Ϫ70°C overnight. C, control; Ab, antibody; CREc, consensus CRE; CREs, MUC5AC-specific CRE; CREm, MUC5AC-mutant CRE; NS, nonspecific. which were thought to function as bridging proteins between DNA-binding transcription factors and the basal transcription factors, play a role as integrators of diverse signaling pathways in the MUC5AC gene expression.
Whereas CREB has recently emerged as a potent regulator of mucins (MUC2, MUC5AC, MUC5B, and MUC6), gene expression in the p15 arm of chromosome 11 (11p15) (51), and cholera toxin A subunit, an activator of cAMP-dependent protein kinase, activates transcription of MUC5B promoter (52), little is known about the involvement of CRE in MUC5AC transcription. Our results showed that Ϫ929/ϩ4 region of MUC5AC promoter was sufficient to get a response to IL-1␤ or TNF-␣ and that CRE in Ϫ878 region of MUC5AC promoter was critical for the up-regulation of the transcriptional activity of MUC5AC induced by IL-1␤ or TNF-␣. However, Perrais et al. (50) reported that TNF-␣ did not have any significant effect of activity of Ϫ1366/ϩ4 region of the MUC5AC promoter, which was in accordance with our results in Ϫ1376/ϩ4 region of MUC5AC promoter. These results suggest that TNF-␣-responsive repressor(s) or negative regulatory element that represses inherent basal and cAMP-inducible promoter activity (53) may be located in Ϫ1366/Ϫ929 region of MUC5AC promoter. Thus, it seems necessary to explore further the involvement of IL-1␤or TNF-␣-responsive repressor(s) or negative regulatory element in IL-1␤-or TNF-␣-induced MUC5AC transcription.
In summary, our results showed that ERK and p38 MAP kinases, but not JNK signaling, are essential for IL-1␤-and TNF-␣-induced MUC5AC gene expression. Furthermore, the activation of MSK1 and CREB is a crucial aspect of the intracellular mechanisms that mediate MUC5AC gene expression. This study also demonstrated that CRE in the MUC5AC promoter might play a role in these processes by binding CREB. Further analysis of the signal pathways activated by various cytokines may yield deeper insights into the signal mechanism of MUC5AC gene expression.