Monocyte 15-Lipoxygenase Gene Expression Requires ERK1/2 MAPK Activity

Abstract

IL-13 induces profound expression of 15-lipoxygenase (15-LO) in primary human monocytes. Our studies have defined the functional IL-13R complex, association of Jaks with the receptor components, and the tyrosine phosphorylation of several Stat molecules in response to IL-13. Furthermore, we identified both p38MAPK and protein kinase Cδ as critical regulators of 15-LO expression. In this study, we report an ERK1/2-dependent signaling cascade that regulates IL-13–mediated 15-LO gene expression. We show the rapid phosphorylation/activation of ERK1/2 upon IL-13 exposure. Our results indicate that Tyk2 kinase is required for the activation of ERK1/2, which is independent of the Jak2, p38MAPK, and protein kinase Cδ pathways, suggesting bifurcating parallel regulatory pathways downstream of the receptor. To investigate the signaling mechanisms associated with the ERK1/2-dependent expression of 15-LO, we explored the involvement of transcription factors, with predicted binding sites in the 15-LO promoter, in this process including Elk1, early growth response-1 (Egr-1), and CREB. Our findings indicate that IL-13 induces Egr-1 nuclear accumulation and CREB serine phosphorylation and that both are markedly attenuated by inhibition of ERK1/2 activity. We further show that ERK1/2 activity is required for both Egr-1 and CREB DNA binding to their cognate sequences identified within the 15-LO promoter. Furthermore, by transfecting monocytes with the decoy oligodeoxyribonucleotides specific for Egr-1 and CREB, we discovered that Egr-1 and CREB are directly involved in regulating 15-LO gene expression. These studies characterize an important regulatory role for ERK1/2 in mediating IL-13–induced monocyte 15-LO expression via the transcription factors Egr-1 and CREB.

The Th2 lymphocyte-derived cytokine IL-13 is a potent activator of monocyte and macrophage function. IL-13 induces the expression of the lipid-oxidizing enzyme 15-lipoxygenase (15-LO) in peripheral blood monocytes (1, 2). Lipoxygenases are a family of lipid-peroxidating enzymes that catalyzes the oxygenation of free and esterified polyunsaturated fatty acids to form the corresponding hydroperoxy fatty acid derivatives (3, 4). 15-LO catalyzes the conversion of 15(S) H(p)ETE from arachidonate and 13(S) H(p)ODE from linoleate (5, 6). These products are potent mediators of inflammation (7) and are found in human atherosclerotic lesions (8-10). Several studies have suggested the involvement of 15-LO in the development of atherosclerosis, asthma, diabetes, cancer, renal injury, osteoporosis, and neurodegenerative disorders (3, 11–18). Recently, it was also shown that expression of the lipoxygenase gene Alox 15, which encodes 12/15-LO, induces cardiac inflammation and plays a crucial role in the development of heart failure (19). Because of its diverse clinical implications, we have pursued studies to understand the regulation of IL-13–mediated 15-LO expression in primary human monocytes.

Our previous studies have characterized the functional monocyte IL-13R complex and the downstream signaling events in primary monocytes (20). We have demonstrated the IL-13–induced heterodimerization of IL-13Rα1– and IL-4Rα–chains leading to the activation/phosphorylation of Jak2 and Tyk2 and tyrosine phosphorylation of the receptor components (2, 20). We have also reported the activation/tyrosine phosphorylation of specific Stat molecules Stat1, Stat3, Stat5, and Stat6 in response to IL-13 stimulation (20). These studies have defined a regulatory signal transduction pathway from the receptor to the nucleus in human monocytes.

Our recent data indicated that IL-13 induces Stat1 and Stat3 Ser⁷²⁷ phosphorylation in addition to tyrosine phosphorylation (21, 22). Our studies also demonstrated the activation of p38MAPK and subsequent phosphorylation of Ser⁷²⁷ residues on Stat1 and Stat3 molecules as critical regulators in IL-13–induced 15-LO expression in primary human monocytes (21). We further observed the involvement of protein kinase C (PKC) δ in forming a molecular complex with tyrosine-phosphorylated Stat3 and subsequent PKCδ-dependent Stat3 Ser⁷²⁷ phosphorylation for optimal expression of 15-LO in response to IL-13 stimulation (22).

One major mechanism involved in the regulation of inflammatory processes is the activation of ERK1/2 (23–25). They are members of the MAPK superfamily (p44MAPK/p42MAPK) and play a major role in cell proliferation and differentiation (26, 27). In response to a variety of stimuli, they are activated/phosphorylated (at specific Thr and Tyr residues) and can either act as an upstream kinase for several cytosolic proteins that participate in different signal transduction pathways (28) or translocate to the nucleus, where they can directly or indirectly activate several transcription factors (28, 29), to regulate the expression of specific genes (30, 31).

In the current study, we investigated the contribution of the MEK–ERK1/2-mediated signaling pathway in IL-13 induction of 15-LO expression through the activation of transcription factors early growth response-1 (Egr-1) and CREB in primary human monocytes. Our results indicate that MEK–ERK1/2-regulated stimulation of Egr-1 nuclear accumulation and CREB activation are directly linked to IL-13–induced 15-LO expression in human monocytes. We demonstrate that activation of ERK1/2 is dependent on Tyk2 and independent of the Jak2, p38MAPK, and PKCδ pathways. Thus, the MEK–ERK1/2-mediated signaling pathway represents a bifurcation of signaling downstream of the monocytes IL-13R that works in parallel with the previously identified Jak2, p38MAPK, and PKCδ signaling pathways to regulate IL-13–stimulated 15-LO expression. Hence, this pathway is required but not sufficient for IL-13–induced 15-LO expression. This work is also the first report, to our knowledge, on the involvement of transcription factors like CREB and Egr-1 in regulating the expression of 15-LO gene in IL-13–stimulated primary human monocytes.

Materials and Methods

Reagents

Recombinant human IL-13 was purchased from BioSource International (Camarillo, CA). Ab against rabbit reticulocyte 15-LO, cross-reacting with human 15-LO, was raised in sheep and was kindly provided by Dr. Joseph Cornicelli, Parke-Davis (New York, NY). Anti–phospho-p38MAPK, anti–phospho-ERK1/2, anti–phospho-(Ser¹³³)–CREB, anti-phosphothreonine 505 PKCδ, anti-CREB, and anti-PCNA (mouse monoclonal) Abs were purchased from Cell Signaling Technology (Beverly, MA) and diluted 1:1000 according to the manufacturer’s protocol. The other primary Abs used in this study were: rabbit anti-human Egr-1 (C-19) and rabbit anti-human PKCδ (C-20) from Santa Cruz Biotechnology (Santa Cruz, CA) and anti-ERK1/2 from Millipore (Billerica, MA). Pharmacological inhibitors such as PD98059 and SB202190 were purchased from Calbiochem (La Jolla, CA). U0126 and Rottlerin were purchased from Biomol (Butler Pike, PA). The inhibitors were dissolved in dimethyl sulfoxide and stored at −20°C as concentrated stock solutions.

Isolation of human monocytes

Human peripheral blood monocytes (PBMs) were isolated either by separation of mononuclear cells followed by adherence to bovine calf serum (BCS)-coated flasks as described earlier (32) or by Ficoll-Hypaque sedimentation followed by countercurrent centrifugal elutriation (33, 34). PBMs purified by these two methods were identical in response to IL-13 and consistently >95% CD14⁺. These studies complied with all relevant federal guidelines and institutional policies regarding the use of human subjects.

Immunoprecipitation and immunoblotting

PBMs (5 × 10⁶ cells/well in 2 ml 10% BCS/DMEM) were either directly treated with IL-13 (1 or 2 nM) or pretreated with pharmacological inhibitors (30 min) followed by IL-13 treatment for different time intervals as indicated. Total, cytosolic, and nuclear extracts were prepared by previously published protocols (20, 35, 36). For the preparation of cytosolic and nuclear extracts, we washed primary human monocytes with 10 ml PBS (two times) to remove the traces of BCS/DMEM and pelleted by centrifugation. The cell pellet was resuspended in 200 μl cold buffer A (10 mM HEPES [pH 7.9], 10 mM KCl, 0.1 mM EDTA, 1 mM DTT, 1mM PMSF, 0.1% Nonidet P-40, and protease inhibitor mixture [1:50 dilution]), kept on ice for 5–10 min, and centrifuged for 1 min in a microfuge. The supernatant (cytosolic extract) was transferred to a fresh tube. The nuclear pellet was resuspended in 100 μl ice-cold buffer C (20 mM HEPES [pH 7.9], 0.4 M NaCl, 1 mM EDTA, 1 mM DTT, 1mM PMSF, and protease inhibitor mixture [1:50 dilution]), vortexed for 10–15 s, kept on ice for 15 min, and centrifuged at 10,000 × g for 10 min. The supernatant (nuclear extract) was saved at −80°C. Using our method of fractionation, we found that proliferating cell nuclear Ag (PCNA) or cyclin appeared in the nuclear fraction but not in the cytosolic fraction, whereas β-tubulin appeared in the cytosolic fraction but not in the nuclear fraction (22). After determining the protein concentration using the Bio-Rad protein assay reagent (Hercules, CA), lysate proteins (50 μg/lane) were resolved by 8% SDS-PAGE, transferred to a polyvinylidene difluoride membrane, blocked with 5% BSA in PBS with 0.1% Tween 20 and subjected to immunoblotting with p-Ser¹³³ CREB, Egr-1, p-p38MAPK, p-ERK1/2, and p-Thr⁵⁰⁵ PKCδ Ab overnight. The hybridization signal was detected using SuperSignal West Pico Chemiluminescent substrate (Pierce, Rockford, IL). 15-LO protein was detected on Western blots following a previously described protocol (2). For immunoprecipitation experiments, the lysates were incubated with anti-PKCδ Abs for 2 h at 4°C with constant rotation and precipitated with prewashed protein A-Sepharose beads (Sigma-Aldrich, St. Louis, MO) at 4°C overnight. Immunoprecipitates were thoroughly washed with a lysis buffer containing 1% Triton X-100, 150 mM NaCl, 50 mM NaF, 30 mM β-glycerophosphate, 0.5 mM phosphoserine, 0.5 mM phosphotyrosine, 1.0 mM phosphothreonine, 1.5 mM p-nitrophenylphosphate, 50 mM Tris (pH 7.4), 1 mM sodium orthovanadate, 500 μM PMSF, and protease inhibitor mixture (Sigma-Aldrich), and the immune complexes were released by boiling the beads in SDS sample buffer and were then subjected to Western blot analysis as described earlier (20). Immunoblots were stripped and reprobed to assess equal loading according to our previously published protocol (20).

Treatment of monocytes with sense/scrambled and antisense oligodeoxyribonucleotides for Jak2, Tyk2, and ERK1/2

The antisense oligodeoxyribonucleotide (ODN) sequences for human Jak2 and Tyk2 were selected based on our previously published literature (2). Control ODNs for Jak2 and Tyk2 consisted of complementary sense ODNs. The antisense ODN sequence for human ERK1/2 was also selected based on the previous studies by Lee et al. (37). A scrambled ODN (for ERK antisense) was used as a control. All ODNs were end-modified (phosphorothioated, three bases of 5′ and 3′) oligodeoxyribonucleotides to limit DNA degradation, and all were HPLC-purified preuse (Invitrogen, Carlsbad, CA).

The sequences of the ODNs are as follows: Jak2 antisense, 5′-TCT TAA CTC TGT TCT CGT TC-3′; Jak2 sense, 5′-GAA CGA GAA CAG AGT TAA GA-3′; Tyk2 antisense, 5′-CCA ACT TTA TGT GCA ATG TG-3′; Tyk2 sense, 5′-CAC ATT GCA CAT AAA GTT GG-3′; ERK1/2 antisense, 5′-AGC AGA TAT GGT CAT TGC-3′; and ERK1/2 scrambled, 5′-TCG TCT ATA CCA GTA ACG-3′.

Primary human monocytes (5 × 10⁶ cells/well) were plated in six-well culture plates overnight. Cells were then transfected with Jak2 and Tyk2 sense and antisense ODNs or with ERK1/2 scrambled and antisense ODNs at 2 μM concentration using Mirus TransIt-Oligo Transfection Reagent (Mirus Bio, Madison, WI) according to the manufacturer’s protocols, and the incubation was continued for 48 h. For the transfection control, monocytes were incubated with the transfection reagent alone for 48 h. After this treatment, monocytes were exposed to IL-13 for another 30 min or 24 h to study either the activation/phosphorylation of p44/p42 MAPK (ERK1/2) or 15-LO gene expression.

Decoy oligonucleotide transfection

Double-stranded decoy ODN containing the conserved promoter binding site of the CREB, early growth response element-1 (Egr-1), and a scrambled sequence were prepared from complementary single-stranded phosphorothioate-modified oligonucleotides (ordered from Invitrogen) by melting at 95°C for 5 min, followed by a cooldown phase overnight. Human monocytes were plated in six-well culture plates overnight. Cells were then transfected with decoy ODNs using Superfect Transfection Reagent (Qiagen, Valencia, CA) according to the manufacturer’s instructions for 24 h. Monocytes were then incubated in the absence or presence of IL-13 for another 24 h for 15-LO mRNA quantification or 48 h for 15-LO protein detection. The single-stranded sequences of the decoy ODNs were as follows: CREB (consensus): sense, 5′-AGA GAT TGC CTG ACG TCA GAG AGC TA-3′; Egr-1 (consensus): sense, 5′-CCC GGC GCG GGG GCG ATT TCG AGT C-3′; and scrambled: sense, 5′-AAC AGA AGC CAG GAA CCC TCC TCT-3′, adapted from Grote et al. (38).

RNA extraction and quantitative real-time PCR analysis

Monocytes (5 × 10⁶ in 2 ml 10% BCS/DMEM) were plated in six-well culture plates. Two hours after plating, cells were treated with IL-13 for 24 h. In some experiments, monocytes were pretreated with MEK inhibitors (PD98059 and U0126) for 30 min or transfected with decoy ODNs as described above followed by IL-13 treatment for 24 h. Total cellular RNA was extracted using the RNeasy mini kit from Qiagen. The cDNA was prepared by reverse transcription of 1 μg total RNA using random hexamers as primers (Roche, Branchburg, NJ). One hundred nanograms cDNA was used for quantification by RT-PCR using specific primers listed in Table I. Human GAPDH gene was used as an internal standard for sample normalization. RT-PCR reactions were performed in duplicate on an ABI PRISM 7000 sequence detection system (Applied Biosystems, Foster City, CA) using SYBR Green PCR core reagent according to established protocols (21). The relative levels of 15-LO mRNA were calculated using the comparative cycle threshold method (ΔΔCt) (39).

Table I.

Primer list

Gene	Forward Primer	Reverse Primer
15LOX	5′-GCTGGAAGGATCTAGATGACT-3′	5′-TGGCTACAGAGAATGACGTTG-3′
GAPDH	5′-CACCAACTGCTTAGCACCCC-3′	5′-TGGTCATGAGTCCTTCCACG-3′

TransMax EGR activation assay

A TransMax EGR transcription factor assay kit from Genlantis (San Diego, CA) was used to allow the rapid analysis of EGR-1 activation (DNA binding activity) from nuclear extracts using the manufacturer’s protocol. This kit offers a chemiluminescence-based assay that specifically recognizes EGR-1 DNA binding to its consensus DNA binding sequence (5′-CGCCCCCGC-3′) using a streptavidin-alkaline phosphatase conjugate and a chemiluminescent substrate. The amount of signal (light intensity) generated is proportional to the amount of EGR-1 in the nuclear extract and was detected using a microplate luminometer (Multilabel Counter Victor³, PerkinElmer, Wellesley, MA). Two reagent controls were included in every assay. The positive reagent control provides the maximum signal for each site, and the negative reagent control provides the background signal for each site. The results are expressed as relative light units after background was subtracted.

Transcription factor ELISA assay to quantitate the binding activity of EGR1

A TF ELISA EGR1 kit from Panomics (Panomics, Santa Clara, CA) was used to assess the binding activity of EGR1 to the cognate sequence identified within the 15-LO promoter according to the manufacturer’s instructions. In brief, activated EGR1 molecules from nuclear extracts bound to an EGR1 consensus binding site (EGR1 probe, 5′-GCGGGGGCG-3′) on a biotinylated oligonucleotide were immobilized on a streptavidin-coated 96-well assay plate. The EGR1, bound to the oligonucleotide, was detected by an Ab against EGR1, followed by an HRP-conjugated secondary Ab reaction with the tetramethylbenzidine substrate to provide a colorimetric readout, which was taken at 450 nm. Nonlabeled EGR1 Cold Probe (EGR1 CP) was used in this assay as a negative control to measure competition of the labeled probe.

For competitive binding experiments, oligodeoxyribonucleotides corresponding to the cognate EGR1-binding site (as predicted by Genomatix, http://www.genomatix.de) in the context of 15-LO promoter flanking sequences (EGR1 BS-15LO) were used in this assay. The single-stranded sequence of this competitor was as follows: EGR1 BS-15LO sense, 5′-GAG AAC AGC AGG GGC GGC GGG GGA-3′. As a control competitor, we used the consensus EGR1-binding site in the context of flanking sequences present in the 15-LO promoter (EGR1 BSC-15LO). The sequence of the sense strand of this control competitor (EGR1 BSC-15LO sense) was 5′-GAG AAC AGC GGG GGC GGC GGG GGA-3′. As a noncompetitor, a mutated (MT) sequence of the cognate EGR1-binding site in the context of 15-LO promoter flanking sequences (EGR1 BS MT-15LO) was used. The sequence of the sense strand of this non-competitor (EGR1 BS MT-15LO sense) was 5′-GAG AAC AAA TTG GGC GGC GGG GGA-3′. All of these ODNs were purchased from Integrated DNA Technologies (Coralville, IA). The biotinylated EGR1 probe concentration was 250 nM in the final reaction, whereas the final concentration for the competitors and noncompetitors was 10 μM.

TransAM pCREB activity assay

ATransAM pCREB kit from Active Motif (Carlsbad, CA) was used to evaluate the cAMP-responsive element (CRE) promoter-binding activity of activated (phosphorylated) CREB according to the manufacturer’s instructions. Nuclear extracts were incubated in a 96-well assay plate precoated with immobilized oligonucleotides containing the CREB consensus sequence (5′-TGACGTCA-3′). The wild-type (WT) consensus oligonucleotide was provided with the kit, which prevents CREB binding to the probe immobilized on the plate. Conversely, the MT oligonucleotide was provided as a control and expected to have a limited effect on CREB binding.

For competitive binding experiments, different oligodeoxyribonucleotides corresponding to the cognate CRE-binding sites (as predicted by Genomatix, http://www.genomatix.de) in the context of 15-LO promoter flanking sequences (CRE BS1-15LO, CRE BS2-15LO, and CRE BS3-15LO) were used in this assay. The single-stranded sequences of these competitor ODNs were as follows: CRE BS1-15LO sense, 5′-TCC ATG CCA TGA AGT TTA TGT TAG TAT TC-3′; CRE BS2-15LO sense, 5′-ACA CGT GCA TAA CTC CTA CCC CCA-3′; and CRE BS3-15LO sense, 5′-TCT CCC TCC CGT CAA GAT AG-3′. As a control competitor, we used the consensus CRE-binding site in the context of the same flanking sequences present in CRE BS3-15LO (CRE BS3C-15LO). The sequence of the sense strand of this control competitor (CRE BS3C-15LO sense) was 5′-TCT CCC TGA CGT CAA GAT AG-3′. As a non-competitor, a mutated sequence of the cognate CRE-binding site in the context of the same flanking sequences present in CRE BS3-15LO (CRE BS3 MT-15LO) was used in this assay. The sequence of the sense strand of this noncompetitor (CRE BS3 MT-15LO) was 5′-TCT CCC GAC CAT AGA GAT AG-3–. The detection of activated CREB was carried out by monitoring the colorimetric readout at 450 nm with a reference wavelength of 655 nm using a Synergy HT Multi-Mode Microplate Reader from BioTek (Winooski, VT).

Data analysis

The number of experiments analyzed is indicated in each figure. Band intensities were quantified by densitometric analyses using a laser densitometer (Microtek ScanMaker 8700, Microtek Lab, Cerritos, CA) and National Institutes of Health ImageQuant software (Bethesda, MD). Differences among experiment groups were analyzed using one-way ANOVA, followed by Tukey’s multiple comparison tests to identify each group differences. Comparisons between two groups were done with Student t test analysis. All statistical analyses were performed with a GraphPad Prism4 program (GraphPad, San Diego, CA), and p < 0.05 was considered statistically significant.

Results

IL-13 stimulates the phosphorylation of ERK1/2 in primary human monocytes

Our previous studies demonstrated the involvement of different Ser/Thr kinases, including p38MAPK and PKCδ in regulating 15-LO expression in response to IL-13 stimulation (21, 22, 40). We were also interested in exploring the role of ERK1/2 (p44/p42 MAPK) signaling on IL-13 induction of 15-LO expression in primary human monocytes. To investigate the phosphorylation/activation status of ERK1/2 poststimulation of primary monocytes with recombinant human IL-13 for 5, 15, 30, 60, and 120 min, we used a phospho-ERK1/2 Ab that specifically recognized the phosphorylated (Thr²⁰² and Tyr²⁰⁴) form of ERK1/2. Our Western analysis data of phospho-ERK1/2 and total-ERK1/2 are shown in Fig. 1A. We have also provided the densitometric analysis data for the ratio of phospho-ERK1/2 to total-ERK1/2 in a time-dependent manner (Fig. 1B). Our time-course experiment for IL-13 induction of ERK1/2 phosphorylation (Fig. 1A) demonstrated that the level of active ERK1/2 increased rapidly after 5 min of IL-13 stimulation, declined after 5 min, and a second peak of activation appeared after 60 min of IL-13 stimulation. We observed no significant difference in total ERK1/2 expression levels upon IL-13 stimulation across the time points. These data suggest that in addition to p38MAPK and PKCδ (21, 40), a rapid biphasic activation of ERK1/2 (another important member of the MAPK family) is also mediated by IL-13 stimulation of primary human monocytes.

FIGURE 1.

IL-13 induces phosphorylation of ERK1/2 in primary human PBMs. To determine the role of IL-13 incubation time on ERK1/2 phosphorylation, human PBMs were incubated with 2 nM IL-13 for 5, 15, 30, 60, and 120 min. The monocyte cell lysates were immunoblotted with Abs that recognize the phosphorylated Thr²⁰² and Tyr²⁰⁴ of ERK1/2 followed by stripping and reprobing the same blot with an Ab recognizing total ERK1/2. A, Western blot image of phosphorylated ERK1/2 and total ERK1/2 (representative experiment). B, The densitometric quantification of phospho-ERK1/2 to total ERK1/2 in IL-13–treated monocytes relative to untreated monocytes. The data represent the mean ± data range from two independent experiments.

ERK1/2 regulates IL-13–stimulated 15-LO gene expression in monocytes

Because IL-13 stimulated ERK1/2 activation in primary human monocytes, we next examined whether ERK1/2 activity was required for IL-13–induced expression of 15-LO mRNA and protein. To investigate the role of ERK1/2 activity in regulating IL-13 induction of 15-LO expression at the mRNA level, we used two inhibitors of the upstream kinase that regulates ERK1/2 phosphorylation and activation. Monocytes were pretreated with either MEK inhibitor PD98059 (Fig. 2A) or U0126 (Fig. 2B) at several indicated doses for 30 min followed by incubation with IL-13 for an additional 24 h. Total RNA was extracted and subjected to quantitative real-time RT-PCR using the primers listed in Table I to quantify the expression of 15-LO mRNA. As expected, the quantitative real-time RT-PCR experiments showed that incubation of monocytes with IL-13 for 24 h substantially induced 15-LO mRNA expression (>4000-fold [2¹²]). Treatment with PD98059 and U0126 both had profound, dose-dependent inhibitory effects on 15-LO mRNA levels. Cell viability was not affected by these inhibitors at the specified doses and times of treatment. The highest indicated doses of those inhibitors down-regulated 15-LO mRNA expression almost completely after 24 h of IL-13 treatment (Fig. 2A, 2B). The mRNA levels of GAPDH (used as control) were nearly identical in all the samples, indicating specificity of the response.

FIGURE 2.

ERK1/2 activity/expression is required for IL-13–induced 15-LO gene expression in primary human monocytes. Primary human monocytes (5 × 10⁶/group) were pretreated with MEK inhibitors PD98059 (A, C, D) or U0126 (B, D) at various indicated doses for 30 min, followed by stimulation with 1 nM IL-13 for 24 h. In A and B, total cellular RNA was extracted, and RNA (1 μg) from each sample was subjected to quantitative real-time RT-PCR analysis. Amplification plots of real-time RT-PCR analysis showed the regulatory effect of ERK1/2 activity on IL-13 induction of 15-LO mRNA expression. Postnormalization with GAPDH amplification, the fold induction of 15-LO mRNA expression for different groups was plotted. Data were collected from two independent experiments and shown as the mean ± data range. In C and D, monocytes were harvested and lysed; 50 μg lysates were resolved by 8% SDS-PAGE, and 15-LO protein expression was detected on Western blots with a 15-LO–specific Ab. The blots were then stripped and reprobed with β-tubulin (C, D, lower panels) to assess equal loading. The arrows indicate the positions of 15-LO and β-tubulin based on the migration of 15-LO positive control and m.w. markers in adjacent lanes. Data in C and D are from representative experiments of three identical experiments that were performed. To show the direct involvement of ERK1/2 on IL-13–induced 15-LO mRNA and protein expression, human blood monocytes were transfected with or without 2 μM antisense or scrambled ODNs to ERK1/2 (E, F) according to protocols described in Materials and Methods prior to the addition of IL-13 (2 nM) for 24 h. In E, Total of 50 μg monocyte lysates were separated by SDS-PAGE and immunoblotted with a 15-LO–specific Ab (upper panel). The blot was reprobed with a total ERK1/2 Ab to examine the effect of antisense ODN on total ERK1/2 expression (middle panel). The same blot was then stripped and reprobed with β-tubulin Ab (lower panel) as a loading control. Data are from a representative experiment of three repeat experiments that showed similar results. In F, 15-LO mRNA expression of monocytes was detected. The fold induction of 15-LO mRNA expression for different groups was shown after correcting for the GAPDH mRNA levels. 15-LO mRNA and GAPDH mRNA (internal control) were quantified by real-time quantitative PCR analysis. Results are from a representative experiment of three performed in which data are the mean ± SD (n = 3). Significant differences were determined by comparing the antisense or scrambled ODN (to ERK1/2)-treated groups to the transfection control (IL-13 treated). *p < 0.025.

We next evaluated the effect of MEK/ERK1/2 inhibitors PD98059 and U0126 on IL-13–induced 15-LO protein expression. Primary monocytes were incubated for 24 h, with or without IL-13, in the presence or absence of these pharmacological inhibitors. Postincubation, the monocytes were harvested and lysed, and 15-LO protein expression was detected on Western blots. The results, shown in Fig. 2C and 2D, indicate that both MEK inhibitors suppressed expression of 15-LO protein in a dose-dependent manner. PD98059 treatment at 25–50 μM concentration and U0126 treatment at 20 μM concentration caused almost total inhibition of IL-13–stimulated 15-LO protein expression. These findings present evidence that ERK1/2 activation is required for 15-LO gene expression in IL-13–treated human PBMs.

To further test the direct involvement of ERK1/2 in regulating IL-13–induced 15-LO gene expression, we blocked ERK1/2 expression by treating monocytes with ERK1/2-specific antisense ODNs before IL-13 addition. As a control, scrambled ODNs to ERK1/2 were used. Total cellular RNA or proteins were extracted for real time RT-PCR or Western blot analysis, respectively. Our results, presented in Fig. 2E, indicate that the antisense ODN (2 μM) against ERK1/2 profoundly inhibited the IL-13–stimulated 15-LO protein expression (~60%), whereas the scrambled ODNs and the vehicle control had no effect on IL-13–induced 15-LO protein expression (Fig. 2E, upper panel). Downregulation of ERK1/2 protein expression level by the antisense ODNs was also verified by reprobing the same blot with an Ab against total ERK1/2 (Fig. 2E, middle panel). Our results further show that IL-13 induction of 15-LO mRNA expression was significantly inhibited (*p < 0.025) by the treatment of monocytes with ERK1/2-specific antisense ODNs (Fig. 2F). The transfection control as well as the scrambled ODNs to ERK1/2 showed no inhibition of IL-13–induced 15-LO mRNA expression (Fig. 2F). These data confirm the results presented in Fig. 2A–D, in which pharmacological inhibitors were employed and indicate that ERK1/2 expression/activity is required for IL-13–stimulated 15-LO gene expression.

Tyk2 is required for IL-13–mediated ERK1/2 activation

Results from our previously published studies clearly established the requirement of both Jak2 and Tyk2 kinases for the IL-13–mediated signaling, leading to the expression of 15-LO in primary human monocytes (2). In the heterodimeric IL-13R, Tyk2 is associated with IL-13Rα1 and Jak2 is associated with IL-4Rα. Because we demonstrated that IL-13 stimulated ERK1/2 activation in human monocytes and activation of ERK1/2 was required for IL-13–stimulated 15-LO gene expression, we next investigated the requirement for Jak2 and Tyk2 kinases for the IL-13–induced signaling pathways leading to the activation of ERK1/2 in primary monocytes. For these studies, monocytes were treated with antisense or sense ODN against Jak2 or Tyk2 kinases for 48 h in presence of Mirus TransIt-oligo transfection reagent (Mirus Bio). Posttreatment, the cells were exposed to IL-13 for another 30 min. Total cell lysates were extracted, and ERK1/2 phosphorylation/activation was evaluated. The results in Fig. 3A show that the antisense ODN (2 μM) against Tyk2 kinase inhibited the IL-13–mediated phosphorylation/activation of ERK1/2, whereas the sense ODN and the vehicle controls had no effect on the activation of ERK1/2. Antisense ODN inhibition of Tyk2 protein expression level was also verified by reprobing the same blot with an Ab against Tyk2 kinase (Fig. 3A, middle panel). These data thus indicate that Tyk2 is the upstream receptor-associated tyrosine kinase of ERK1/2 phosphorylation/activation in human monocytes in response to IL-13 stimulation.

FIGURE 3.

Tyk2 but not Jak2 antisense treatment inhibits phosphorylation/activation of ERK1/2 poststimulation with IL-13 in human monocytes. Monocytes were pretreated with antisense or sense ODNs to either Tyk2 (A) or Jak2 (B) according to protocols described in Materials and Methods prior to the addition of IL-13 (2 nM) for 30 min. The cells were lysed, and 50 μg cell extracts (from each sample group) were separated by SDS-PAGE and immunoblotted with anti–phospho-ERK1/2 Ab (A, B, upper panels). Blots were then reprobed with either Tyk2 or Jak2 Abs to examine the effect of antisense ODN on Tyk2 and Jak2 expression (A, B, middle panels, respectively). The same blots were stripped and reprobed with an Ab recognizing total ERK1/2 to assess equal loading (A, B, lower panels). The results are representative of three independent experiments.

Jak2 is not involved in regulating the IL-13–stimulated activation of ERK1/2

Next, we examined whether inhibition of Jak2 by the antisense ODN against Jak2 kinase also altered the IL-13–induced activation of ERK1/2. Although our results indicated substantial inhibition of Jak2 expression in monocytes treated with antisense to Jak2 (2 μM) (Fig. 3B, middle panel), the reduced expression level of Jak2 had no inhibitory effect on the IL-13–induced activation of ERK1/2 (Fig. 3B, upper panel). The bottom panel in Fig. 3B indicates nearly equal level of total ERK1/2 in all the lanes. These results thus indicate that Jak2 is not required for IL-13–mediated activation of ERK1/2 and ERK1/2-mediated regulation of 15-LO expression in IL-13–treated primary human monocytes.

ERK1/2 does not regulate p38MAPK or PKCδ activation/phosphorylation in human monocytes

Previously, we demonstrated that activation of PKCδ and p38MAPK pathways are independent, but they work in parallel to regulate IL-13–induced 15-LO expression in human monocytes (40). To investigate whether ERK1/2 activity has any role in controlling the activation of p38MAPK and/or PKCδ and vice versa in IL-13–stimulated monocytes, we performed a series of experiments using the selective pharmacological inhibitors of ERK1/2, p38MAPK, and PKCδ followed by IL-13 stimulation in primary human monocytes. Cells were preincubated with various doses of MEK/ERK1/2 inhibitors PD98059 and U0126 prior to IL-13 stimulation. Total cell lysates were extracted, and p38MAPK phosphorylation/activation was evaluated. As presented in Fig. 4A, pretreatment with U0126 or PD98059 at a dose (10–20 μM and 50 μM, respectively) that sub-stantially inhibited ERK1/2 activation (41, 42) and 15-LO expression (Fig. 2) had no detectable effect on IL-13–induced p38 MAPK phosphorylation (Fig. 4A). These data suggest that p38MAPK is not a downstream target of ERK1/2 in the IL-13–induced signaling pathway in human monocytes.

FIGURE 4.

Activation of ERK1/2 by IL-13 is independent of both the p38MAPK and PKCδ pathways. Human monocytes 5 × 10⁶/group (A, B, D) and 10 × 10⁶/group (C) were pretreated with MEK inhibitors U0126 (A, C) and PD98059 (A, lower panel), p38MAPK inhibitor SB202190 (B), and the PKCδ selective inhibitor rottlerin (D) at different indicated doses for 30 min, pre-exposure to IL-13 (1 nM) for 15 min (A, C) or 30 min (B, D). A total of 50 μg cell lysate was subjected to Western blot analysis (A, B, D) using anti–phospho-p38MAPK Ab (A) and anti–phospho-ERK1/2 Ab (B, D). The membrane was stripped and reprobed with anti-ERK1/2 Ab to assess equal loading (B, D, lower panels). Whole-cell lysates were immunoprecipitated with an Ab against PKCδ (C). The immunoprecipitates were then resolved by 8% SDS-PAGE for immunoblotting with an anti–phospho-Thr⁵⁰⁵ PKCδ Ab (C, upper panel). The blot was stripped and reprobed with PKCδ Ab (C, lower panel) to confirm equal immunoprecipitation. Data are from a representative experiment of four repeat experiments that showed similar results.

To determine whether p38MAPK was upstream of ERK1/2, we treated cells with the p38MAPK inhibitor SB202190 prior to IL-13 activation and evaluated ERK1/2 activation/phosphorylation. Pretreatment with SB202190 at a dose that significantly inhibited p38MAPK activity and IL-13–induced 15-LO expression (21, 43) caused no inhibition of IL-13–induced activation/phosphorylation of ERK1/2 (Fig. 4B).

In a similar experiment, we preincubated cells with various indicated doses of U0126 followed by IL-13 activation. Pretreatment of monocytes with U0126 (even at 20 μM concentration) could not block the IL-13–induced Thr⁵⁰⁵ phosphorylation of PKCδ (Fig. 4C). In a reciprocal experiment, we pretreated cells with the PKCδ selective inhibitor rottlerin (different doses) for 30 min before IL-13 stimulation for another 30 min. Total cell lysates were extracted, and ERK1/2 phosphorylation/activation was evaluated. As presented in Fig. 4D, pretreatment with rottlerin at a dose (5 μM) that profoundly inhibited PKCδ activation (44) and 15-LO expression (40) had no significant effect on IL-13–induced ERK1/2 activation/phosphorylation. These results together with our prior published work (40) indicate that IL-13–mediated activation of ERK1/2 is independent of p38MAPK and PKCδ activation, yet both are required to act in parallel to regulate 15-LO expression.

IL-13 induces Egr-1 expression and nuclear accumulation in human monocytes

Because ERK1/2 activation by IL-13 stimulation was required for expression of 15-LO in primary monocytes, we investigated the effect of ERK1/2 in regulating transcription factors that might control 15-LO expression. In silico analysis of the 15-LO promoter (as predicted by Genomatix; http://www.genomatix.de) demonstrated that Elk1, Egr-1, and CREB transcription factor binding sites are present in the 15-LO promoter. We therefore investigated the role of IL-13 and ERK1/2 in stimulating these transcription factors in primary human PBMs.

We observed that Elk1 is poorly expressed in primary monocytes. It was difficult to detect by Western analysis (data not shown). Next, we analyzed whether Egr-1 is expressed in human blood monocytes and whether IL-13 induced Egr-1 expression in monocytes. Our results showed a significant level of Egr-1 expression in total cell lysates from monocytes (data not shown). To investigate the IL-13 induction of Egr-1 expression, we evaluated both cytosolic and nuclear fractions. Although IL-13 induced Egr-1 protein expression in the cytosol (Fig. 5A, upper panel), the level of expression was modest. In contrast, Egr-1 nuclear accumulation was substantially induced by IL-13 (Fig. 5A, upper panel). Time-course experiments demonstrated that IL-13 significantly stimulated Egr-1 nuclear accumulation in a time-dependent manner. The maximal level of accumulation of Egr-1 in the nucleus was reached at 1 h and then declined (Fig. 5B, upper panel). We verified nuclear localization of Egr-1 by evaluating the purity of our nuclear preparation as described previously (22). The same blot was reprobed with PCNA Ab (used as a nuclear marker) to assess equal loading (Fig. 5B, lower panel).

FIGURE 5.

IL-13 stimulates Egr-1 expression/nuclear accumulation, CREB serine phosphorylation, and nuclear translocation of ERK1/2. Freshly isolated human blood monocytes (10 × 10⁶/group) were left untreated or directly treated with IL-13 (1 nM) for either 1 h (A) or for different time intervals as indicated (B–D). Cytosolic (40 μg/lane) and nuclear extracts (10 μg/lane) were resolved by 8% SDS-PAGE and subjected to immunoblotting using Abs against Egr-1 (A, upper panel, B, upper panel), CREB (A, lower panel), phospho-CREB (Ser¹³³) (C, upper panel), and ERK1/2 (D). In B and C, the same blots were stripped and reprobed using Ab against PCNA (lower panels) to assess equal loading. Results are from a representative experiment of three performed.

CREB activation rather than expression is stimulated by IL-13 treatment in primary human monocytes

Like Egr-1, we also observed substantial levels of CREB expression in total cell lysates from monocytes (data not shown). To examine whether the expression/activation of CREB is modulated by IL-13, we first checked the effect of IL-13 treatment on CREB expression in monocytes. CREB was hardly expressed in the cytosol, whereas the expression level of CREB was significant in the nuclear extract. IL-13 treatment failed to induce CREB expression in either the cytosolic or nuclear fraction (Fig. 5A, lower panel). We next evaluated the ability of IL-13 to induce the activation of CREB (phosphorylation at Ser¹³³), which has been reported as one of the major downstream targets of ERK1/2 (25, 45-49). Our results clearly indicate that IL-13 stimulated Ser¹³³ phosphorylation of CREB in a time-dependent manner in the nuclear fraction (Fig. 5C, upper panel). We observed a transient phosphorylation of CREB that peaked at 1 h after IL-13 treatment, with the signal diminishing thereafter. Nuclear localization of CREB activation (Ser¹³³ phosphorylation) was further confirmed by evaluating our method of fractionation as described earlier (22). Equal loading was also verified by reprobing the blot with anti-PCNA Ab (Fig. 5C, lower panel).

IL-13 induces nuclear translocation of ERK1/2

Our results revealed that IL-13 treatment had no effect on the ERK1/2 protein expression levels in the cytosol (Fig. 5D, upper panel), but it strongly facilitated the nuclear import of ERK1/2 (Fig. 5D, lower panel). This result suggests that upon IL-13 stimulation, ERK1/2 molecules are dual-phosphorylated by upstream kinase (MEK1/2) and translocate from the cytoplasm to the nucleus. Our results demonstrated a significant increase of ERK1/2 protein in the nuclear extract within 1 h and remained stationary thereafter until the 4-h time point (Fig. 5D, lower panel). The maximal level of IL-13–driven nuclear import of ERK1/2 (between 1 and 4 h) coincides with the peak expression level of Egr-1 (1 h) and peak phosphorylation level of CREB (1 h) in the nuclear extract.

IL-13 induction of Egr-1 nuclear translocation requires ERK1/2 activity in primary human monocytes

To determine whether the MEK–ERK1/2 MAPK pathway was linked to IL-13–mediated upregulation of Egr-1 nuclear translocation, monocytes were pretreated with MEK/ERK1/2 inhibitors PD98059 (12.5–25 μM) and U0126 (20 μM) prior to incubation with IL-13. Our results, presented in Fig. 6A (lower panel) and 6B (upper panel), indicate that IL-13 substantially stimulated the level of accumulation of Egr-1 in the nucleus (~35–40-fold enhanced level of nuclear accumulation of Egr-1 compared with the untreated control). Monocytes pre-exposed to PD98059 displayed significant reduction of nuclear Egr-1 levels compared with cells treated with IL-13 alone (~72%; Fig. 6A, lower panel, 6B, upper panel). IL-13–induced nuclear accumulation of Egr-1 was almost completely abolished (~97%) in the presence of 20 μM concentration of U0126 (Fig. 6A, lower panel, 6B, upper panel). These results suggest that the MEK–ERK1/2 MAPK pathway is an important regulator of IL-13–mediated induction of nuclear Egr-1 levels. Interestingly, pretreatment of monocytes with the same pharmacological inhibitors (with the same indicated doses) did not inhibit the IL-13–stimulated Egr-1 protein expression level (~4-fold increased level of expression compared with the untreated control) in the cytosolic extracts (Fig. 6A, upper panel).

FIGURE 6.

IL-13–induced Egr-1 nuclear accumulation and Egr-1 activation (DNA binding activity) requires Erk1/2 activity in human monocytes. Human monocytes 10 × 10⁶/group (A, B) and 5 × 10⁶/group (D) were pretreated with MEK inhibitors PD98059 and U0126 (at different indicated doses) (A, B, D) and p38MAPK inhibitor SB202190 (10 μM) (B, D) for 30 min, followed by IL-13 stimulation (1 nM) for 1 h or directly stimulated with IL-13 for the same period of time (A, B, D). Nuclear (20 μg/lane) (A, lower panel, B, upper panel) and cytosolic (40 μg/lane) (A, upper panel) extracts were separated by SDS-PAGE and immunoblotted with an Ab against Egr-1 (A, upper panel of B). The same blot was stripped and reprobed with anti-PCNA Ab (B, lower panel) as a loading control. In A, human rEgr-1 was used as a positive control. In C, monocytes (10 × 10⁶/group) were pretreated with antisense or sense ODNs to Tyk2 and Jak2 according to protocols described in Materials and Methods prior to the addition of IL-13 (2 nM) for 1 h. Nuclear extracts (20 μg/lane) were separated by 8% SDS-PAGE and immunoblotted with an Ab against Egr-1 (C, upper panel). The same blot was stripped and reprobed with anti-PCNA Ab (C, lower panel) as a loading control. The results are representative of three independent experiments. To detect Egr-1 activity (D), 1 μg nuclear extracts/well were run in triplicate to measure the binding of Egr-1 to its consensus DNA binding sequence. The positive and negative reagent control wells were also included in each assay and run in triplicate. Data are the mean ± SD (n = 3). Significant differences were determined by comparing each group to the IL-13–treated group as the control according to the Data analysis section under Materials and Methods. *p < 0.001; ^§no significant difference.

To explore the specific role of ERK1/2 in regulating IL-13–stimulated Egr-1 nuclear accumulation, the effect of the p38MAPK specific inhibitor SB202190 (10 μM) was again investigated. SB202190 did not inhibit IL-13–induced Egr-1 nuclear accumulation in primary monocytes (Fig. 6B, upper panel). ERK1/2-dependent accumulation of Egr-1 in the nuclear compartment was verified as described previously (22). Equal loading of the nuclear fraction was assessed by reprobing the same blot with the nuclear marker PCNA (Fig. 6B, lower panel).

Tyk2 is involved in IL-13–induced nuclear accumulation of Egr-1

As Tyk2 is the upstream regulator of ERK1/2 phosphorylation/activation in IL-13– stimulated human monocytes, we further investigated whether Tyk2 is required for Egr-1 nuclear accumulation in response to IL-13 stimulation. Our results, presented in Fig. 6C (upper panel), indicate that antisense ODNs (2 μM) against Tyk2 almost completely inhibited the IL-13–induced Egr-1 nuclear accumulation, whereas the Tyk2 sense ODNs and the transfection control had no inhibitory effect on the level of accumulation of Egr-1 in the nucleus. By using the antisense ODNs against Jak2 in the same experiment, we further confirmed that Jak2 is not involved in IL-13–stimulated nuclear accumulation of Egr-1 (Fig. 6C, upper panel). Equal loading of the nuclear fraction was further assessed by reprobing the same blot with PCNA (Fig. 6C, lower panel). The specificity and effectiveness of these antisenses have been documented previously by our laboratory and are also shown in Fig. 3.

IL-13–induced DNA binding activity of Egr-1 transcription factor requires ERK1/2 kinase activity in human monocytes

We next investigated the requirement of ERK1/2 kinase activity in mediating the IL-13–stimulated binding of Egr-1 to its consensus DNA binding sequence. We used a chemiluminescence-based TransMax Egr transcription factor assay method to evaluate the binding activity of Egr-1 in response to IL-13 stimulation. Our results demonstrated that Egr-1 has a basal level of binding activity to its consensus DNA binding sequence that was enhanced (>2-fold) by IL-13 treatment (Fig. 6D). IL-13 induction of Egr-1 DNA binding activity was inhibited by the pretreatment of monocytes with MEK/ERK1/2 inhibitors PD98059 and U0126 in a dose-dependent manner. PD98059 at 25 μM and U0126 at 20 μM concentration brought back the IL-13–stimulated Egr-1 DNA binding activity almost to the basal level (Fig. 6D). In contrast, the p38MAPK inhibitor SB202190 at 10 μM concentration showed no inhibition on IL-13–stimulated binding of activated Egr-1 to its nascent DNA binding sequence in primary monocytes (Fig. 6D). These results implicate a specific role of ERK1/2 kinase activity in regulating Egr-1 DNA binding activity in response to IL-13 stimulation.

ERK1/2 activity is required for CREB activation in IL-13–stimulated monocytes

To determine whether ERK1/2 was involved in the IL-13–stimulated activation (Ser¹³³ phosphorylation) of CREB, we also investigated the effect of MEK/ERK1/2 inhibitors PD98059 and U0126 on IL-13 induction of CREB serine phosphorylation. The results of a representative experiment are shown in Fig. 7A (upper panel). Monocytes were pretreated with these pharmacological inhibitors for 30 min prior to the addition of IL-13 and were either treated or left untreated with IL-13 for 1 h and harvested. The serine phosphorylation status of CREB in the nuclear extract was determined by Western blots using a phosphoserine-CREB–specific Ab. The results, presented in Fig. 7A (upper panel), indicate that PD98059 inhibited IL-13–stimulated CREB Ser¹³³ phosphorylation in a dose-dependent manner. The degree of inhibition was significant (p < 0.0008) when we used 25 μM concentration of PD98059. U0126 at 10 mM concentration also significantly (p < 0.0006) downregulated Ser¹³³ phosphorylation of CREB in IL-13–induced monocytes. In contrast, the p38MAPK specific inhibitor SB202190 (10 μM) demonstrated no inhibitory effect on IL-13 induction of CREB Ser¹³³ phosphorylation in monocytes. These results support a specific regulatory role of ERK1/2 in mediating the activation (Ser¹³³ phosphorylation) of CREB in the nuclear extracts. Nuclear localization of IL-13–mediated CREB Ser¹³³ phosphorylation was verified by evaluating the purity of the nuclear extract as described previously (22). Equal loading of the nuclear fraction was further verified by reprobing the same blot with the nuclear marker PCNA (Fig. 7A, lower panel).

FIGURE 7.

ERK1/2 regulates IL-13–stimulated CREB serine phosphorylation and CREB activation (CRE promoter site binding activity). Human monocytes (5 × 10⁶/group) (A, C) were pretreated with MEK inhibitors PD98059 and U0126 (at different indicated doses) (A, C) and p38MAPK inhibitor SB202190 (10 μM) (A, C) for 30 min, followed by stimulation with 1 nM IL-13 for 1 h or directly exposed to IL-13 for the same period of time (A, C). In case of A, nuclear extracts (10 μg/lane) were resolved by 8% SDS-PAGE and subjected to immunoblotting using Abs against phospho-CREB (Ser¹³³) (upper panel). The same blot was stripped and reprobed with anti-PCNA Ab (lower panel) to assess equal loading. In B, monocytes (10 × 10⁶/group) were pretreated with antisense or sense ODNs to Tyk2 and Jak2 as in Fig. 6C. Nuclear extracts (20 μg/lane) were separated by SDS-PAGE and immunoblotted with an Ab against phospho-CREB (Ser¹³³) (B, upper panel). The same blot was stripped and reprobed with anti-PCNA Ab (B, lower panel) as a loading control. The results are representative of three repeat experiments. In case of pCREB activation (C), 5 μg nuclear extracts/well were run in duplicate to perform an immunodetection of activated CREB. Nuclear extracts from forskolin-stimulated WI-38 cells were used as a positive control for activated CREB. The WT and MT consensus oligonucleotides were used to monitor the specificity of the assay. The results are representative of three independent experiments. Values are mean ± SEM of three separate experiments. Significant differences were determined by comparing each group to the IL-13–treated monocytes as the control as described in the Data analysis section under Materials and Methods. *p < 0.05; **p < 0.001; ^§no significant difference.

Tyk2 is involved in IL-13–induced phosphorylation/activation of CREB

As ERK1/2 activity regulates CREB phosphorylation/activation in IL-13–induced human monocytes, we further investigated whether Tyk2 is required for CREB serine phosphorylation/activation in the nucleus after IL-13 stimulation. Our results, shown in Fig. 7B (upper panel), demonstrate that downregulation of Tyk2 by using antisense ODNs (2 μM) against Tyk2 significantly reduced the IL-13–induced CREB activation (Ser¹³³ phosphorylation), whereas the Tyk2 sense ODNs and the transfection control had no inhibitory effect on IL-13–mediated CREB serine phosphorylation/activation in the nucleus. In contrast, downregulation of Jak2 by using the antisense ODNs directed to Jak2 showed no inhibitory effect on IL-13 induction of CREB Ser¹³³ phosphorylation in monocytes (Fig. 7B, upper panel). Equal loading of the nuclear fraction was verified as before by reprobing the same blot with PCNA (Fig. 7B, lower panel).

ERK1/2 activity is involved in IL-13–induced DNA binding activity of CREB in primary human monocytes

After already demonstrating in Fig. 7A that IL-13 induction of CREB activation is regulated by the ERK1/2 kinase activity, we further examined the role of ERK1/2 in mediating the DNA binding activity of activated (phosphorylated) CREB in primary monocytes. Using an ELISA-based TransAM method and employing specific phospho-CREB primary Ab, we showed that IL-13 treatment facilitated DNA binding activity of IL-13–stimulated CREB transcription factor (Fig. 7C). Furthermore, pretreatment of monocytes with MEK/ERK1/2 inhibitors PD98059 and U0126 for 30 min before IL-13 stimulation significantly attenuated the extent of DNA binding activity of phosphorylated CREB in a dose-dependent manner. The WT consensus oligonucleotides (used as a competitor for CREB binding) also significantly reduced the CREB activation (Fig. 7C). Conversely, the p38MAPK inhibitor SB202190 at 10 μM concentration and the MT consensus oligonucleotides showed negligible effects on IL-13–induced DNA binding activity of CREB in primary monocytes. These results indicate the involvement of ERK1/2 kinase activity in regulating CREB activation and related DNA binding activity in response to IL-13 stimulation.

In monocytes treated with IL-13, Egr-1 and CREB specifically bind to their cognate sequences derived from the 15-LO promoter

To demonstrate that Egr-1 and CREB transcription factors can bind to their cognate DNA binding sites present in the 15-LO promoter after IL-13 stimulation, we performed competitive binding experiments using ELISA-based activation assays for Egr-1 and CREB. Our results indicated that the basal level of binding activity of Egr-1 to its consensus DNA binding sequence was enhanced (~3-fold) by IL-13 treatment (Fig. 8A). EGR1 CP completely abolished the elevated level of EGR1 activation after IL-13 stimulation. EGR1 BS-15LO, oligodeoxyribonucleotides corresponding to the cognate EGR1-binding site in the context of 15-LO promoter flanking sequences, was used as a competitor in this assay and significantly inhibited (*p < 0.002) the IL-13–induced EGR1 activation. EGR1 BSC-15LO, oligodeoxyribonucleotides containing the consensus EGR1-binding site in the context of flanking sequences present in the 15-LO promoter, used as a control competitor for EGR1 binding in this assay, also significantly downregulated the EGR1 activation (Fig. 8A). In contrast, EGR1 BS MT-15LO, a mutated sequence of the cognate EGR1-binding site in the context of 15-LO promoter flanking sequences, was used as a noncompetitor and showed no inhibition of IL-13–induced DNA binding activity of EGR1 in primary human monocytes (Fig. 8A).

FIGURE 8.

In IL-13–stimulated monocytes, Egr-1 and CREB bind to their cognate DNA binding sequences located in the 15-LO promoter. Human monocytes (5 × 10⁶/group) (A, B) were either untreated or directly stimulated with IL-13 (2 nM) for 1 h. To measure the binding of Egr-1 to its cognate DNA binding sequence present in the 15-LO promoter, 5 μg nuclear extracts/well were run in triplicate using Panomics TF ELISA kits (Panomics) for EGR1 (A). EGR1 BS-15LO and EGR1 BSC-15LO were used as competitors in this assay, whereas EGR1 BS MT-15LO was used as a noncompetitor. EGR1 CP was provided by the kit and used at a concentration of 10 μM in the final reaction. In B, 5 μg nuclear extracts/well were run in triplicate to detect activated CREB using a TransAM pCREB kit. CRE BS1-15LO, CRE BS2-15LO, CRE BS3-15LO, and CRE BS3C-15LO were used as competitors, whereas CRE BS3 MT-15LO was used as a noncompetitor. All of the competitors and noncompetitors were used at a concentration of 20 pmol/well. In both A and B, the results are representative of three independent experiments. Data are the mean ± SD (n = 3). Significant differences in both the experiments were determined by comparing each group to the IL-13–treated monocytes as the control as described in the Data analysis section under Materials and Methods. *p < 0.002 in A; *p < 0.001 in B.

For CREB activation, our results demonstrated that IL-13 stimulated the binding activity of CREB to its consensus DNA binding sequence (>3-fold compared with the untreated control) (Fig. 8B). For competitive binding, CRE BS1-15LO, CRE BS2-15LO, and CRE BS3-15LO (three different oligonucleotides corresponding to the cognate cAMP-responsive element [CRE]-binding sites in the context of flanking sequences present in the 15-LO promoter) were added prior to addition of the nuclear extract. Although CRE BS1-15LO and CRE BS2-15LO (used as competitors for CREB binding) showed some inhibitory effects on IL-13–induced DNA binding activity of CREB in primary monocytes, CRE BS3-15LO significantly attenuated the IL-13–induced CREB DNA binding (*p < 0.001) (Fig. 8B). CRE BS3C-15LO (oligodeoxyribonucleotides containing the consensus CRE-binding site in the context of the same flanking sequences present in CRE BS3-15LO) was also used as a competitor for comparison. Addition of CRE BS3C-15LO in this assay profoundly downregulated IL-13–stimulated CREB DNA binding (*p < 0.001). The WT consensus oligodeoxyribonucleotides also significantly reduced CREB DNA binding (Fig. 8B). Conversely, CRE BS3 MT-15LO (an MT sequence of the cognate CRE-binding site in the context of the same flanking sequences present in CRE BS3-15LO) and the MT consensus oligodeoxyribonucleotides were used as noncompetitors in this assay and caused essentially no inhibition of IL-13–induced DNA binding activity of CREB (Fig. 8B). These data show the direct binding of Egr-1 and CREB transcription factors to their cognate DNA binding sites in the context of flanking sequences present in the 15-LO promoter after IL-13 stimulation in primary monocytes.

Egr-1 and CREB regulate IL-13–induced 15-LO expression in primary human monocytes

We have demonstrated that IL-13 induced Egr-1 protein expression, CREB activation, and Egr-1 and CREB DNA binding activity in primary human monocytes (Figs. 5-7). Our results further reveal that inhibition of IL-13–induced ERK1/2 activation by MEK/ERK1/2 specific pharmacological inhibitors downregulated these effects (Figs. 6, 7). To investigate whether Egr-1 and CREB are the downstream target genes of ERK1/2 signaling that play a significant role on IL-13–induced 15-LO expression in human monocytes, we transfected monocytes with decoy ODNs specific for Egr-1 and CREB consensus sequences and evaluated their impact on 15-LO expression. After 24 h of transfection with the decoy ODNs, we stimulated the cells with IL-13 for an additional 24 or 48 h and quantified the mRNA and protein expression level of 15-LO relative to monocytes transfected with scrambled decoy using quantitative real-time PCR and Western blot analysis, respectively. 15-LO mRNA and protein expression levels were determined from these experiments and the results are shown in Fig. 9A and 9B, respectively.

FIGURE 9.

The effect of Egr-1 and CREB decoy ODNs on IL-13–induced 15-LO mRNA and protein expression. To determine the role of Egr-1 and CREB on IL-13–induced 15-LO mRNA and protein expression, human blood monocytes (5 × 10⁶/group) were transfected with or without 2 μM scrambled, Egr-1, or CREB decoy ODNs or a combination of both Egr-1 and CREB decoy ODNs at 1 or 2 μM for 24 h. Twenty-four hours posttransfection, cells were stimulated with 2 nM IL-13 for 24 h (mRNA) or 48 h (protein). A, 15-LO mRNA expression of monocytes was detected posttransfection with decoy ODNs as indicated. Data represent the percent expression of 15-LO relative to monocytes transfected with the scrambled decoy ODNs after correcting for the GAPDH mRNA levels. 15-LO mRNA and GAPDH mRNA (internal control) were quantified by real-time quantitative PCR analysis. B, 15-LO and β-tubulin (control) protein expression of monocytes are shown posttransfection with decoy ODNs as indicated. Data shown are representative of three independent experiments and are presented as the mean ± SEM. Significant differences were detected by comparing the groups either to the scrambled decoy treated group or to the Egr-1 decoy-treated (2 μM) group as described in the Data analysis section under Materials and Methods. *p < 0.001; **p < 0.05.

Fig. 9A illustrates that transfection of monocytes with Egr-1 and CREB decoy ODNs (2 μM each) significantly inhibited IL-13–induced 15-LO mRNA expression levels as compared with the levels in monocytes transfected with scrambled decoy ODN (50% for Egr-1 and 78% for CREB). Cotransfection with a combination of both Egr-1 and CREB decoy ODNs (2 μM each) inhibited 15-LO mRNA levels even further (90% inhibition with the combination) as compared with the scrambled decoy ODN transfected monocytes. Transfection of monocytes in the absence of the decoy ODNs (transfection control) as well as in the presence of the scrambled decoy ODNs followed by IL-13 stimulation showed no inhibition of 15-LO protein expression as compared with the nontransfected monocytes induced by IL-13 (Fig. 9B); however, similar to the 15-LO mRNA results shown in Fig. 9A, both Egr-1 and CREB decoy ODNs markedly attenuated IL-13–induced 15-LO protein expression (Fig. 9B). Cotransfection of Egr-1 and CREB decoy ODNs (1 μM each) significantly inhibited the IL-13–stimulated 15-LO protein expression as compared with scrambled decoy ODN transfected monocytes (Fig. 9B). These data provide the first direct evidence, to our knowledge, that the transcription factors Egr-1 and CREB are important regulators of IL-13–induced 15-LO gene expression in primary human monocytes (Fig. 10).

FIGURE 10.

Model of the bifurcating parallel signal transduction pathways downstream of the IL-13R that regulate 15-LO gene expression in primary human monocytes. The IL-13Rα1–Tyk2-mediated signaling pathway that is required for IL-13–inducible Egr-1 and CREB activation via MEK–ERK1/2 in primary human monocytes is shown in the dotted-line box.

Discussion

In this study, we demonstrated a crucial role for ERK1/2 in IL-13–mediated 15-LO gene expression in primary human monocytes. ERK1/2 are members of the MAPK family that has been implicated in IL-13R regulation (50), IL-13–induced eotaxin release (51-53), and type I collagen gene regulation (54). ERK1/2 MAPK activation has also been reported to play a critical role in mediating IL-13–induced inflammation and alveolar remodeling (23). Expression of 15-LO has been implicated in the biosynthesis of anti-inflammatory lipid mediators (55, 56) and in the pathogenesis of inflammation and atherosclerosis (3, 57). We, therefore, focused on understanding the mechanistic impact of the ERK1/2-dependent signaling cascade in IL-13–stimulated primary human monocytes.

Earlier, our group showed that IL-13 induced the tyrosine phosphorylation of Jak2 and Tyk2 (IL-13R—associated tyrosine kinases) in human monocytes and that activation of both of these Jak kinases was essential for downstream signaling by IL-13, leading to the induction of 15-LO expression (2). We further demonstrated that Jak2 and Tyk2 are associated with IL-4Rα and IL-13Rα1, respectively, in the heterodimeric IL-13R complex (20). The results of our Jak2/Tyk2 antisense experiment thus clearly indicate that an IL-13Rα1–Tyk2–ERK1/2-dependent pathway exists in human monocytes, which acts in parallel with the IL-4Rα–Jak2-dependent signaling cascade to regulate the 15-LO gene expression in monocytes in response to IL-13 stimulation.

Although our work focused on the involvement of IL-13Rα1-associated protein tyrosine kinase Tyk2 in mediating IL-13–dependent activation of ERK1/2 in primary monocytes, the signaling machinery triggered downstream of Tyk2 is still unknown. In the context of Tyk2-mediated signal transduction, it has been shown that Ras and Raf-1 (a critical downstream effector of Ras) activity is essential for Tyk2-dependent mitogenesis and supports the general importance of the protein–tyrosine kinase/Ras effector pathways in mitogenic signal transduction (58). Taken together, a highly probable scenario is that the Tyk2-dependent ERK1/2 activation by IL-13 utilizes the Ras–Raf–MEK1/2–ERK1/2 pathway. This is the topic of future studies.

We showed that IL-13 activation of the MEK–ERK1/2 pathway increased Egr-1 levels in the nucleus as well as DNA binding activity and that both are required for maximal expression of 15-LO in primary monocytes. Egr-1 is a zinc finger transcription factor that plays a key master regulatory role in multiple cardiovascular pathological processes including atherosclerosis, cardiac hypertrophy, ischemia, and angiogenesis (59) and recently was described as a major link between infection and atherosclerosis (60, 61). Recent studies also demonstrated Egr-1 to be a key regulator in the pathogenesis of IL-13–induced inflammatory and remodeling responses (62). Egr-1 activation via the MEK–ERK1/2 pathway has been reported to be involved in tissue factor expression and TNF-α gene expression in human monocytes and THP-1 monocytic cells (63). Our results not only provide the first evidence, to our knowledge, that IL-13 induces Egr-1 expression/activity in primary human monocytes, but also demonstrate the detailed mechanism by which IL-13 stimulates Egr-1 nuclear accumulation and DNA binding activity. Our data clearly establish a signaling pathway involving IL-13R, receptor-associated tyrosine kinase Tyk2, and the activated MEK–ERK1/2 cascade in regulating IL-13 induction of Egr-1 accumulation in the nuclear compartment and binding of Egr-1 to its cognate DNA binding sequence located within the 15-LO promoter.

In this study, we also provide evidence that IL-13 induced 15-LO transcription by the activation of CREB transcription factor through Ser¹³³ phosphorylation and its regulation by the ERK1/2 MAPKs. CREB belongs to the basic leucine zipper family of transcription factors that binds to CREs within target gene promoters with activation initiated by phosphorylation at Ser¹³³. This activation of CREB facilitates the DNA binding activity of the protein in many cell types (25, 45, 49, 64, 65). Upon phosphorylation of CREB at Ser¹³³, the transcriptional coactivator CREB-binding protein, a histone acetyl transferase, is recruited to CREB. This, in turn, promotes the assembly of the basal transcriptional machinery (66-70). It is already reported that CREB plays a pivotal role in the progression of inflammation by regulating the expression of inflammatory genes, such as mucin and cyclooxygenase 2 (25, 71, 72). Our results in the current study clearly show that IL-13 activation of CREB requires ERK1/2 activity; the pharmacological inactivation of ERK1/2 MAPKs abolished the IL-13–induced Ser¹³³ phosphorylation and CRE promoter-binding activity of CREB (Fig. 7A, 7C). In addition, our results also demonstrated that IL-13 induces the translocation of ERK1/2 from the cytosol to the nucleus, implying that the translocated ERK1/2 MAPKs are involved in regulating CREB serine phosphorylation/activation and binding of CREB to its cognate sequences identified within the 15-LO promoter.

Several transcription factors have been reported to regulate 15-LO expression. Although many of these studies showed evidence for Stat6-dependent transcriptional regulation of 15-LO expression and activity (73-76), our previous studies demonstrated the involvement of Stat1 and Stat3 as well as Stat6 as transcriptional regulators of 15-LO in IL-13–stimulated monocytes (21) (A. Bhattacharjee and M.K. Cathcart, unpublished observations). Earlier GATA6-mediated transcriptional regulation of 15-LO expression was reported in human colorectal cancer cells (77, 78). Requirement of the transcription factor KU 70/80 lupus autoantigen (DNA helicase) was also demonstrated for IL-13 induction of 15-LO gene expression in human epithelial A549 cells (79). Recent studies further identified the transcription factor SPI1 (a member of the Ets family of transcription factors), which was responsible for increased transcriptional activation of the ALOX15 (reticulocyte-type 15-LO) gene in macrophages (80). The results of our decoy experiments in this study strongly implicate both Egr-1 and CREB as potent transcriptional regulators of the 15-LO gene in IL-13–stimulated primary human monocytes and provide insights into the regulation of 15-LO transcription in these important inflammatory cells.

We have previously reported that IL-13–mediated induction of 15-LO requires Jak2- and Tyk2-mediated tyrosine phosphorylation of several Stat family members in primary human monocytes (2, 20). Our previous studies also demonstrated the involvement of the serine-threonine kinases p38MAPK and PKCδ in regulating IL-13–driven 15-LO expression (21, 22, 40). Our recent work identified the formation of an IL-13–stimulated signaling complex containing PKCδ, p38MAPK, and tyrosine-phosphorylated Stat3 (22) (A. Bhattacharjee and M.K. Cathcart, unpublished observations) that is required for the serine phosphorylation of Stat3, probably by p38MAPK-mediated interaction between PKCδ and tyrosine-phosphorylated Stat3.

In this study, we investigated whether IL-13–mediated activation of ERK1/2 MAPKs were related to p38MAPK and PKCδ activation in the IL-13–induced signaling pathways in monocytes. By using selective pharmacological inhibitors of ERK1/2, p38MAPK, and PKCδ, we demonstrated that IL-13 activation of ERK1/2 MAPKs is independent of p38MAPK- and PKCδ-mediated signaling cascades in primary human monocytes. The existence of these parallel signal transduction pathways, that although are individually not sufficient for IL-13 induction of 15-LO expression, are all required for 15-LO expression, suggests the importance of tight regulation of 15-LO gene expression. As 15-LO is a very important enzyme for monocyte function and required in many cellular processes, cells probably operate their own machinery in such a way that several independent but parallel pathways must orchestrate simultaneously to stimulate the upregulation of 15-LO gene expression in an IL-13–dependent manner.

In summary, we report in this study the transcriptional regulation of 15-LO by IL-13–mediated activation of two different transcription factors Egr-1 and CREB and the involvement of a MEK–ERK1/2-dependent signaling pathway in regulating this process. Our results schematically presented in Fig. 10 indicate the IL-13 signaling in monocytes through the membrane-bound heterodimeric IL-13R complex (comprised of IL-13Rα1 and IL-4Rα) and the requirement of receptor-associated tyrosine kinases Jak2 and Tyk2, which are activated in response to IL-13 stimulation and influence the 15-LO expression in monocytes (2, 20). We suggest the existence of two distinct bifurcating parallel regulatory pathways downstream of the IL-13R. One is the IL-4Rα–Jak2-dependent signaling cascade that is required for IL-13–stimulated Stat3 tyrosine phosphorylation (A. Bhattacharjee and M.K. Cathcart, unpublished observations) followed by the formation of an IL-13–stimulated cytosolic signaling complex containing PKCδ, p38MAPK, and tyrosine-phosphorylated Stat3 (22) (A. Bhattacharjee and M.K. Cathcart, unpublished observations). This complex formation is required for the serine phosphorylation of Stat3, which is a critical regulatory step in Stat3-mediated 15-LO gene transcription. The other one is the IL-13Rα1–Tyk2-mediated signaling pathway that is required for ERK1/2 MAPK activation followed by increased accumulation of ERK1/2 MAPKs in the nuclear compartment (represented in a dotted-line box in Fig. 10). In the nucleus, activated ERK1/2, either directly or by modulating the activation of some other downstream target, regulates the expression, nuclear accumulation, and DNA binding activity of Egr-1 transcription factor to a cognate DNA binding site present in the 15-LO promoter. ERK1/2 activities also play a pivotal role in inducing 15-LO transcription by the activation of the CREB transcription factor through Ser¹³³ phosphorylation and by regulating IL-13–induced specific interaction between CREB and a cognate CRE site located within the 15-LO promoter (dotted-line box, Fig. 10). Altogether, our results unequivocally demonstrate the regulatory and indispensable role of Egr-1 and CREB transcription factors in IL-13–induced 15-LO gene expression by an ERK1/2–MAPK-mediated signaling pathway. The results of our current study, which establish Egr-1 and CREB as critical determining factors in IL-13–stimulated transcriptional regulation of 15-LO gene, not only increase our understanding about IL-13 signal transduction in primary human monocytes, but also have major clinical implications for the management of inflammation, atherogenesis, and several other pathological conditions.

Abbreviations used in this paper

BCS

bovine calf serum

CRE

cAMP response element

Egr-1

early growth response-1

EGR1 CP

nonlabeled EGR1 Cold Probe

15-LO

15-lipoxygenase

MT

mutated

ODN

oligodeoxyribonucleotide

PBM

peripheral blood monocyte

PCNA

proliferating cell nuclear Ag

PKC

protein kinase C

WT

wild-type