QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 18, Issue 3, 930-942, March 2007

This Article Abstract Full Text (PDF) All Versions of this Article: E06-03-0217v1 18/3/930    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Majdalawieh, A. Articles by Ro, H.-S. Search for Related Content PubMed PubMed Citation Articles by Majdalawieh, A. Articles by Ro, H.-S.

Adipocyte Enhancer-binding Protein-1 Promotes Macrophage Inflammatory Responsiveness by Up-Regulating NF-B via IB Negative Regulation

Department of Biochemistry and Molecular Biology, Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada B3H 1X5

Submitted March 21, 2006; Revised December 18, 2006; Accepted December 22, 2006
Monitoring Editor: William Tansey

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Nuclear factor B (NF-B) subunits comprise a family of eukaryotic transcription factors that are critically involved in cell proliferation, inflammation, and apoptosis. Under basal conditions, NF-B subunits are kept under inhibitory regulation by physical interaction with NF-B inhibitors (IB subunits) in the cytosol. Upon stimulation, IB subunits become phosphorylated, ubiquitinated, and subsequently degraded, allowing NF-B subunits to translocate to the nucleus and bind as dimers to B responsive elements of target genes. Previously, we have shown that AEBP1 enhances macrophage inflammatory responsiveness by inducing the expression of various proinflammatory mediators. Herein, we provide evidence suggesting that AEBP1 manifests its proinflammatory function by up-regulating NF-B activity via hampering IB, but not IB, inhibitory function through protein–protein interaction mediated by the discoidin-like domain (DLD) of AEBP1. Such interaction renders IB susceptible to enhanced phosphorylation and degradation, subsequently leading to augmented NF-B activity. Collectively, we propose a novel molecular mechanism whereby NF-B activity is modulated by means of protein–protein interaction involving AEBP1 and IB. Moreover, our study provides a plausible mechanism explaining the differential regulatory functions exhibited by IB and IB in various cell types. We speculate that AEBP1 may serve as a potential therapeutic target for the treatment of various chronic inflammatory diseases and cancer.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Nuclear factor B (NF-B) comprises a family of ubiquitously expressed, eukaryotic transcription factors that participate in the regulation of multiple immediate genes that are expressed at the onset of many vital biological processes such as cell growth, immunoregulation, inflammation, and apoptosis (Tak and Firestein, 2001; Yamamoto and Gaynor, 2001). NF-B exists in cells as a heterodimer of members of the Rel family of proteins (e.g., p50, p65 (RelA), RelB, c-Rel, ..., etc.), which share a high degree of structural similarity (May and Ghosh, 1997). Under basal conditions, most NF-B subunits are sequestered in the cytosol, where they are constitutively bound by members of the NF-B inhibitor family of proteins; mainly IB and IB (Baeuerle and Baltimore, 1989; Ghosh and Baltimore, 1990). Structurally, IB and IB posses a signal-induced kinase domain, six conserved ankyrin (ANK) repeats, and a PEST domain (May and Ghosh, 1997). IB proteins inhibit NF-B nuclear translocation by their ability to interact through their ANK repeats with NF-B subunits. This protein–protein interaction masks the nuclear localization signal (NLS) located within the Rel homology domain (RHD) of NF-B subunits, which remain trapped in an inactive state in the cytosol (Baeuerle and Baltimore, 1989; Ghosh and Baltimore, 1990).

On stimulation, IB proteins become phosphorylated by a specific, cytosolic serine-protein-kinase multisubunit complex that includes two IB kinases (IKK and IKK) along with other nonenzymatic regulatory components (Regnier et al., 1997; DiDonato et al., 1997; Zandi et al., 1997). Subsequently, IB proteins become ubiquitinated, rendering them susceptible to 26S proteasome-dependent degradation (Alkalay et al., 1995; Chen et al., 1995). Degradation of IB proteins leads to unmasking of NLS on NF-B subunits, which translocate to the nucleus and bind selectively as dimers to specific B DNA-binding elements within the regulatory regions of their target genes leading to transcriptional transactivation (Kunsch et al., 1992). Indeed, NF-B activation also leads to rapid induction of IB mRNA synthesis due to the presence of B DNA-binding element within IB promoters (Sun et al., 1993; Chiao et al., 1994). This negative feedback regulatory loop promotes NF-B retention to the cytosol and ceases NF-B–mediated transactivation of target genes (Sun et al., 1993; Chiao et al., 1994), setting a molecular switch that ensures rapid, transient activation of target genes.

The adipocyte enhancer-binding protein 1 (AEBP1) is a ubiquitously expressed transcriptional repressor with carboxypeptidase and DNA-binding activities (He et al., 1995; Muise and Ro, 1999). AEBP1 is involved in several biological processes including adipogenesis, macrophage cholesterol homeostasis, and inflammation (Park et al., 1999; Kim et al., 2001; Zhang et al., 2005; Majdalawieh et al., 2006). In macrophages, AEBP1 was shown to promote the expression of IL-6, TNF-, MCP-1, and iNOS (Majdalawieh et al., 2006), whose expression is tightly regulated by NF-B activity (Tak and Firestein, 2001; Yamamoto and Gaynor, 2001). In this study, we focus on investigating a potential positive relationship between AEBP1 and NF-B in an attempt to shed light on the exact molecular mechanism by which AEBP1 enhances macrophage inflammatory responsiveness. Herein, we provide experimental evidence suggesting that AEBP1 up-regulates NF-B in macrophages by hampering IB inhibitory function via protein–protein interaction, which renders IB susceptible to increased phosphorylation and proteolytic degradation. Noteworthy, AEBP1-mediated up-regulation of NF-B activity in macrophages is independent of alterations in the kinetic activity of the IKK complex. Our study provides a novel molecular mechanism by which NF-B activity can be regulated by AEBP1 through altering the susceptibility of IB phosphorylation and proteolytic degradation via protein–protein interaction.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mice
Generation of AEBP1^TG (Zhang et al., 2005) and AEBP1^–/– (Ro et al., 2007) mice was previously described. Phenotype characterization of AEBP1^–/– mice was previously described by Ro et al. (2007). Age-matched mice were kept on a 12-h light cycle in the Carleton Animal Care Facility at Dalhousie University. Mice were fed chow diet and were killed by cervical dislocation at 8–12 wk of age to isolate thioglycollate-elicited peritoneal macrophages for protein and RNA analyses.

Reagents and Plasmids
Thioglycollate broth medium was purchased from Sigma (St. Louis, MO). Enhanced chemiluminescence (ECL) reagent was purchased from Amersham Biosciences (Buckinghamshire, England). Protein A/G plus agarose beads were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). DNA oligonucleotides for RT-PCR and gel shift assay were purchased from Sigma. [-³²P]ATP and [-³²P]ATP were purchased from Amersham Biosciences. Poly dI:dC was purchased from Promega (Madison, WI). Lipopolysaccharide (LPS) isolated from Escherichia coli 055:B5 and cycloheximide (CHX) were purchased from Sigma. The pcDNA-AEBP1 expression derivative constructs are derived from the pcDNA vector. Detailed plasmid construction is available upon request. The NF-B-luciferase construct is a kind gift from Dr. Kirill Rosen (Dalhousie University, Halifax, Nova Scotia, Canada). The GST-IB (1-54) construct is a kind gift from Dr. Michael Karin (University of California, San Diego, California).

Antisera
Anti-AEBP1 antibody, generated in rabbits against recombinant mouse AEBP1, was affinity-purified from whole serum using recombinant mouse AEBP1 protein immobilized on nitrocellulose, as previously described (Olmsted, 1981). Primary antibodies directed against p65, IB, phospho-IB, IB, IKK, NIK, MEKK1, and c-Myc were purchased from Santa Cruz Biotechnology. Anti-p53 antibody from Santa Cruz Biotechnology was kind gift from Dr. Patrick Lee (Dalhousie University). Anti-actin antibody was purchased from Sigma. Phycoerythrin (PE)-conjugated anti-mouse TLR4 and PE-conjugated mouse IgG1 isotype control antibodies were purchased from eBioscience (San Diego, CA). Purified rat anti-mouse CD14, rat IgG1 isotype control, mouse IgG2a isotype control, FITC-conjugated anti-rat IgG1 antibodies (BD Biosciences PharMingen, San Diego, CA), purified anti-mouse FcR II/III (CD32/CD16) antibody (Cedarlane Labs, Hornby, Ontario, Canada), and FITC-conjugated anti-mouse IgG isotype control antibody were kindly provided by Dr. Jean Marshall (Dalhousie University). Anti-AKT antibody and horseradish peroxidase (HRP)-conjugated secondary antibodies were purchased from Cell Signaling Technology (Beverly, MA).

Peritoneal Macrophage Isolation and Culture
Peritoneal macrophages were isolated as previously described (Miles et al., 2000). Briefly, mice were injected intraperitoneally with 3 ml sterile 4% Brewer's thioglycollate broth solution (Sigma). Five days later, mice were killed by cervical dislocation, and peritoneal exudate cells were isolated by peritoneal lavage using sterile, cold RPMI-1640 medium. Peritoneal exudate cells were obtained by centrifugation for 10 min at 4°C at 1100 rpm, resuspended and cultured in RPMI-1640 medium supplemented with 10% heat-inactivated fetal calf serum, 1% penicillin-streptomycin, and 10 mM HEPES under standard culture conditions.

Cell Culture and Transfection
Rat C6 glioma cells (kindly provided by Dr. Kenneth Renton) were cultured in DMEM supplemented with 5% fetal calf serum and 1% penicillin-streptomycin. C6 glioma cells were transiently transfected at 90% confluency using Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. 3T3-L1 (Neo-1) and anti-sense (AS/Neo-11) stably transfected preadipocytes were cultured in DMEM supplemented with 10% calf serum, 1% penicillin-streptomycin, and 200 µg/ml G418. J774 monocytes/macrophages (kindly provided by Dr. Timothy Lee, Dalhousie University) were cultured in RPMI-1640 medium supplemented with 10% heat-inactivated fetal calf serum, 1% penicillin-streptomycin, and 10 mM HEPES. All cell lines were cultured under standard culture conditions. Where applicable, cells were treated with 1 µg/ml LPS (Sigma), or phosphate-buffered saline (PBS) as a control, for 30 min to induce NF-B activation. Where applicable, J774 cells were treated with 100 nM phorbol 12-myristate 13-acetate (PMA; Sigma) for 48 h to induce their differentiation into macrophages. For protein stability studies, cells were treated with 50 µg/ml CHX (Sigma) or DMSO as a control, for the indicated time points.

Flow Cytometry
Flow cytometry was performed as previously described (Jawdat et al., 2004). Briefly, 0.5 x 10⁶ macrophages were washed twice with immunofluorescence buffer (2% BSA, 0.02% NaN₃ in PBS). Macrophages were resuspended in immunofluorescence buffer and incubated for 20 min on ice with 5 µg/ml anti-mouse FcR II/III (CD32/CD16) antibody (Cedarlane) to block free antibody-binding sites. Subsequently, macrophages were incubated for 20 min on ice with 1 µg/ml PE-conjugated anti-mouse TLR4 antibody, purified anti-mouse CD14 antibody, or isotype control antibodies (mouse IgG1 and rat IgG1, respectively). Macrophages were then washed three times with immunofluorescence buffer. For CD14 detection, macrophages were incubated with 1 µg/ml FITC-conjugated anti-rat IgG1 antibody for 20 min on ice. Subsequently, stained macrophages were washed three times with immunofluorescence buffer and finally were fixed in 1% paraformaldehyde in PBS for 30 min on ice. Fixed macrophages were washed twice with immunofluorescence buffer and resuspended in 300 µl immunofluorescence buffer. For flow cytometric analysis, macrophages were acquired with an FACSCalibur (BD Biosciences) and analyzed with CellQuest (BD Biosciences). Positive staining using specific antibodies was evaluated relative to negative staining using isotype control antibodies.

RNA Isolation and RT-PCR
Total RNA was isolated and purified from peritoneal macrophages using RNA STAT-60 isolation reagent (TEL-TEST, Friendswood, TX) following the manufacturer's recommendations. RNA, 1 µg, was subjected to reverse transcription using the Omniscript reverse transcriptase kit (Qiagen, Chatsworth, CA) along with oligo (dT)_12–18 primers (Invitrogen) according to the manufacturer's directions. Specific murine PCR primers were used to amplify PCR products corresponding to the genes of interest using Hotstar Taq DNA polymerase kit (Qiagen). The following primer sets were used to amplify the genes of interest using the MJ Research PTC-100 Thermal Cycler (Scientific Support, Waltham, MA): for p65, 5'-CTCTGGGGCGGCACGTTTTAC-3' and 5'-CATCCCACCTGTTCCCCTTGG-3'; IB, 5'-CAGGACTGGGCCATGGAGGG-3' and 5'-TGGCCATTGTAGTTGGTGGC-3'; IB, 5'-GGACACAGCCCTGCACTTGG-3' and 5'-GTAGCCTCCAGTCTTCATCA-3'; AEBP1/ACLP, 5'-GTAGTAGCCCCAAGGAGGAC-3' (PCR4) and 5'-CGATCCACTGGGTCTGCGAC-3' (PCR7); and -actin, 5'-CTGGCACCACACCTTCTACA-3' and 5'-TCGTACTCCTGCTTGCTGATC-3'. The PCR conditions and number of thermal cycles were optimized for each gene examined.

Whole-Cell and Nuclear Protein Extraction and Western Blot Analysis
For whole-cell protein extraction, cells were washed with cold PBS and subsequently lysed in cold RIPA lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM dithiothreitol [DTT], 0.25% sodium deoxycholate, 0.1% NP-40) containing 1 mM phenylmethysulfonyl fluoride (PMSF), 50 mM sodium pyrophosphate, 1 mM Na₃VO₄, 1 mM NaF, 5 mM EDTA, 5 mM EGTA, and protease inhibitors cocktail (Roche, Indianapolis, IN). Cell lysis was performed on ice for 30 min. Clear protein extracts were obtained by centrifugation for 30 min at 4°C. For nuclear protein extraction, the Nuclear Protein Extraction Kit (Active Motif) was used according to the manufacturer's recommendations. Protein extraction from white adipose tissue (WAT) was performed as previously described (Zhang et al., 2005). Protein concentration was determined by a Bradford protein assay using BSA for standardization. Proteins were resolved on 8.5% polyacrylamide gels and subsequently transferred onto nitrocellulose membranes (Amersham). For immunoblotting, nitrocellulose membranes were incubated with specific antibodies recognizing target proteins overnight at 4°C. The membranes were then incubated with HRP-conjugated secondary antibody (1:3000) for 1 h at room temperature and subsequently analyzed by ECL detection system (Amersham).

Coimmunoprecipitation Analyses
For immunoprecipitation analysis, whole-cell protein extracts were obtained by shearing cells in cold RIPA lysis buffer as described above. Protein extracts (0.5 mg) were first incubated with protein A/G plus agarose beads (Santa Cruz Biotechnology) for 2 h at 4°C. The precleared protein extracts were subsequently incubated with 2 µg/ml normal IgG or the indicated specific antibodies for 16 h at 4°C and then with protein A/G plus agarose beads for 4 h at 4°C to allow precipitation of target proteins. Then, the protein A/G plus agarose beads were collected and washed twice with cold RIPA lysis buffer, once with cold RIPA lysis buffer containing 500 mM NaCl, and finally once with cold RIPA lysis buffer. Subsequently, 20 µl sample buffer was added to the immunoprecipitate samples, which were then boiled for 5 min and spun at 14,000 rpm for 10 min. Finally, the immunoprecipitate samples were resolved on 8.5% polyacrylamide mini-gels, transferred onto nitrocellulose membranes, and subjected to immunoblotting as described above.

Luciferase Reporter and -Galactosidase Assays
Luciferase reporter activity was assessed using the luciferase assay system (Promega) according to the manufacturer's directions. In brief, C6 glioma cells were transiently cotransfected with NF-B-luciferase construct (kindly provided by Dr. Kirill Rosen), pCMV--galactosidase expression vector (pHermes-lacZ), and pcDNA-AEBP1 expression plasmids in 12-well plates. Forty-eight hours later, transfectants were washed twice in cold PBS and subsequently lysed in 120 µl passive lysis buffer (Promega). Protein extract, 30 µl, was used to measure reporter activity using the BMG FLUOstar Galaxy microplate reader (BMG Lab Technologies, Offenburg, Germany). Protein extract, 70 ml, was used to measure -galactosidase activity as previously described (He et al., 1995). Luciferase activity was adjusted based on -galactosidase activity to normalize for transfection efficiency.

Electrophoretic Mobility Gel Shift Assay
The probe containing the B DNA-binding consensus oligonucleotide (5'-AGTTGAGGGGACTTTCCCAGGC-3') was radiolabeled with [-³²P]ATP (Amersham) by Klenow fill-in reaction. Nuclear protein extracts (2 µg) were incubated with the B-radiolabeled probe and 1 µg poly dI:dC (Promega) for 20 min at room temperature in binding reaction buffer (10 mM Tris-HCl, pH 7.5, 10 mM KCl, 5 mM MgCl₂, 1 mM DTT, 2.5% glycerol). Specific (unlabeled oligonucleotide) or nonspecific (unrelated oligonucleotide) competitors were used at 50x excess. The DNA–protein complexes were resolved on 5% 0.25x TBE polyacrylamide mini-gels, which were then dried and subjected to autoradiography.

In Vitro Kinase Assay
Endogenously expressed IKK was immunoprecipitated from 0.5 mg precleared protein extracts by incubation with 2 µg/ml anti-IKK antibody (Santa Cruz Biotechnology) for 16 h at 4°C and then with protein A/G plus agarose beads for 4 h at 4°C. The immunoprecipitate samples were washed twice with cold RIPA lysis buffer, once with cold RIPA lysis buffer containing 500 mM NaCl, and finally once with cold kinase reaction buffer (50 mM Tris-HCl, pH 8.0, 10 mM MgCl₂, 1 mM DTT, 0.1% NP-40, 1 mM Na₃VO₄, 1 mM PMSF, and 0.3 mM cold ATP). Kinase assays were performed by adding 1 µg of bacterially expressed GST-IB (1-54), as a substrate, and 5 µCi of [-³²P]ATP (Amersham) to the IKK immunoprecipitate samples in a total of 20 µl kinase reaction buffer. The kinase reaction was allowed to take place for 30 min at 30°C. The reaction was stopped by the addition of 2x SDS sample loading buffer followed by boiling for 5 min and centrifugation for 10 min at 14,000 rpm. The samples were subjected to SDS-PAGE, transferred to nitrocellulose membranes, analyzed by autoradiography, and finally subjected to immunoblotting using specific antibodies against the indicated proteins.

Statistical Analysis
Data are expressed as the mean ± SEM of the indicated number of experiments. Statistical significance was determined using Student's t test for unpaired observations. p < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
AEBP1 Up-Regulates NF-B Activity in Macrophages
We have previously demonstrated that AEBP1 expression in macrophages positively correlates with IL-6, TNF-, MCP-1, and iNOS (Majdalawieh et al., 2006). Hence, we suspected that AEBP1 overexpression and deficiency in macrophages would lead to enhanced and inhibited NF-B activity, respectively. To examine this likely possibility, we assessed NF-B activity upon AEBP1 overexpression in unstimulated and LPS-stimulated C6 glioma cells, which express LPS-responsive toll-like receptor 4 (TLR4; Jack et al., 2005) and its coreceptor CD14 (Deininger et al., 2003), by luciferase reporter assay, due to repeatedly unsuccessful attempts to transfect THP-1, U937, and J774 monocyte/macrophage cell lines using various transfection reagents. C6 glioma cells were cotransfected with NF-B-luciferase construct along with increasing amounts of the pcDNA-AEBP1 plasmid expressing AEBP1. As shown in Figure 1A, AEBP1 enhances NF-B activity in a dose-responsive manner.

View larger version (39K): [in this window] [in a new window]   Figure 1. AEBP1 up-regulates NF-B activity in vitro and in vivo. (A) Exogenously expressed AEBP1 enhances NF-B activity in a dose-responsive manner in C6 glioma cells cotransfected with NF-B-luciferase reporter and increasing amounts of a pcDNA-AEBP1 plasmid. Relative luciferase activity was set to 100% for the 0-ng control sample. Statistical significance was determined relative to 0-ng transfection under control and LPS-stimulatory conditions (n = 6). (B–D) AEBP1–/– macrophages exhibit significantly lower NF-B activity. (B) Nuclear protein extracts isolated from unstimulated and LPS-stimulated macrophages were subjected to SDS-PAGE and immunoblotting for the indicated proteins. (C) A histogram illustrates densitometric analysis of protein levels shown in B. Nuclear protein levels were normalized based on actin expression (n = 6). (D) Nuclear protein extracts isolated from unstimulated and LPS-stimulated macrophages were subjected to EMSA in presence of specific (unlabeled NF-B probe) or nonspecific (unlabeled, unrelated probe) competitors. This data are representative of three independent experiments. (E) Equal amounts of cytoplasmic and nuclear protein extracts isolated from unstimulated and LPS-stimulated macrophages were subjected to SDS-PAGE and immunoblotting for the indicated proteins. This data are representative of three independent experiments. (F) Flow cytometry shows that AEBP1+/+ and AEBP1–/– macrophages express equivalent levels of TLR4 and CD14 on their cell surface.

TLR4 has been shown to function as the functional receptor responsive to signaling initiated by bacterial LPS (Poltorak et al., 1998) after specific binding to CD14 and LPS-binding protein (LBP; Wright et al., 1990) that are expressed on various cell types including macrophages. Thus, it was important to compare the expression of TLR4 and CD14 on the cell surface of AEBP1^+/+ and AEBP1^–/– macrophages to eliminate the possibility that the differential NF-B activity displayed by those macrophages is not due to altered expression of TLR4 or CD14. Flow cytometric analysis revealed that AEBP1^+/+ and AEBP1^–/– macrophages express equivalent cell surface TLR4 and CD14 levels (Figure 1F), suggesting indiscriminative LPS-induced signaling cascade leading to NF-B activation. Taken together, these results suggest that AEBP1 is capable of enhancing NF-B activity in vitro and in vivo under basal and LPS-stimulatory conditions.

Noteworthy, AEBP1 gene extends over 10 kb and gives rise to two transcripts by an alternative splicing mechanism (Ro et al., 2001). One transcript is translated into an 82-kDa protein, AEBP1, whereas the other transcript is translated into a 175-kDa protein, ACLP (Aortic Carboxypeptidase-Like Protein), the latter having an additional 380-amino acid coding sequence at its N-terminus (Ro et al., 2001). ACLP was initially identified in human aortic smooth muscle cells, and its expression was shown to be up-regulated during smooth muscle cell differentiation (Layne et al., 1998). Although its transcript can be detected in several tissues including the kidney, colon, aorta, brain, stomach, spleen, white adipose tissue, and thymus, ACLP protein can only be detected in the aorta (Layne et al., 1998) and adipose tissue (Layne et al., 2001). Unlike AEBP1, ACLP is an extracellular matrix (ECM)-associated secretory protein that is excluded from the nuclear and cytoplasmic compartments of mouse aortic smooth muscle cells (Layne et al., 2001). Thus, it is inconceivable that ACLP can have any effect on NF-B activity in macrophages or C6 glioma cells, despite the fact that ACLP possesses the exact same amino acid sequence of AEBP1 at its C-terminus (Ro et al., 2001). Most importantly, experimental evidence clearly indicates that neither ACLP mRNA nor ACLP protein is expressed in primary macrophages, J774 macrophages, or C6 glioma cells (Figure 2, A and B, respectively), despite their detection in white adipose tissue (Figure 2). Equally important, experimental evidence indicates that targeted overexpression of AEBP1 in macrophages isolated from AEBP1^TG mice leads to significant induction of NF-B activity (data not shown) and macrophage inflammatory responsiveness (Majdalawieh et al., 2006), indicating that NF-B activity can only be regulated by AEBP1, not ACLP, because overexpression of AEBP1 in AEBP1^TG macrophage is driven by a transgene carrying the cDNA of AEBP1 (Zhang et al., 2005). Therefore, unlike AEBP1, ACLP is incapable of modulating NF-B activity in macrophages.

View larger version (46K): [in this window] [in a new window]   Figure 2. ACLP, unlike AEBP1, is not expressed in macrophages and C6 glioma cells. (A) Total RNA obtained from AEBP1+/+ and AEBP1–/– macrophages, PMA-treated J774 macrophages, C6 glioma cells, as well as AEBP1+/+ and AEBP1–/– white adipose tissue (WAT) was subjected to RT-PCR using PCR4 and PCR7 primer set (refer to Ro et al., 2001) to assess AEBP1 and ACLP mRNA levels (top panel). Ear genomic DNA from AEBP1+/+ mouse was used as a control. Using PCR4 and PCR7 primer set, the expected amplicon size corresponding to genomic DNA, AEBP1 transcript, and ACLP transcript is 955, 787, and 300 base pairs, respectively. -actin expression served as a control (bottom panel). (B) Whole-cell protein extracts obtained from AEBP1+/+ and AEBP1–/– macrophages, PMA-treated J774 macrophages, and C6 glioma cells, as well as AEBP1+/+ and AEBP1–/– white adipose tissue (WAT) were subjected to SDS-PAGE and immunoblotting using anti-AEBP1 antibody (top panel). Actin protein levels are shown as an input control (bottom panel). This data are representative of three independent experiments.

AEBP1 Modulates NF-B Activity Independent of Its Transcriptional Repression Function

View larger version (44K): [in this window] [in a new window]   Figure 3. AEBP1 promotes IB phosphorylation and proteolytic degradation in macrophages. (A and B) AEBP1+/+ and AEBP1–/– macrophages express equivalent p65, but higher IB and IB, mRNA levels. (A) Total RNA obtained from macrophages treated with 1 µg/ml LPS or PBS (control) for 1 h was subjected to RT-PCR to assess p65, IB, and IB mRNA levels. (B) A histogram demonstrates densitometric analysis of RT-PCR results shown in A based on -actin expression. (C and D) AEBP1–/– macrophages display reduced IB phosphorylation and elevated IB, but not IB, protein levels. (C) Whole-cell protein extracts isolated from macrophages were subjected to SDS-PAGE and immunoblotting for the indicated proteins. (D) A histogram illustrates densitometric analysis of the immunoblotting analysis shown in C. Statistical significance was determined relative to protein levels in unstimulated or LPS-stimulated AEBP1+/+ macrophages. (E and F) AEBP1–/– macrophages display prolonged IB, but not IB, protein stability. (E) Whole-cell protein extracts isolated from macrophages treated with vehicle (ethanol) or 50 µg/ml CHX were subjected to SDS-PAGE and immunoblotting for the indicated proteins. (F) A line-curve illustrates densitometric analysis of protein levels shown in E. Protein levels at 0 min was taken to be 100% for every group, and protein levels at other time points are expressed relative to that at 0 min within the same group.

AEBP1 Negatively Influences IB Protein Stability in Macrophages

Because IB phosphorylation renders it susceptible to proteolytic degradation (Alkalay et al., 1995; Chen et al., 1995), we wanted to evaluate IB and IB protein stability in AEBP1^+/+ and AEBP1^–/– macrophages. To this end, macrophages were treated with 50 µg/ml CHX for various time points, and whole-cell protein extracts were subjected to SDS-PAGE. As shown in Figure 3, E and F, IB protein half-life is significantly higher in AEBP1^–/– macrophages compared with AEBP1^+/+ macrophages (t_1/2 = 250 min vs. t_1/2 = 50 min, respectively; Figure 3F). Noteworthy, IB protein levels in AEBP1^–/– macrophages upon CHX treatment were comparable to those in AEBP1^+/+ macrophages (Figure 3, E and F). As a control, tumor suppressor p53 protein level was assessed in AEBP1^+/+ and AEBP1^–/– macrophages upon CHX treatment. As shown in Figure 3E, p53 protein stability trend is very similar in both types of macrophages (i.e., t_1/2 = 50 min). This IB and IB protein stability study was also performed in control 3T3-L1 preadipocytes (Neo-1) and stably transfected AEBP1 anti-sense (AS/Neo-11) preadipocytes, and very similar results were obtained (data not shown). Taken together, these findings strongly suggest that lack of AEBP1 promotes IB protein stability, significantly inhibiting its turnover.

Because modulation of IKK activity is associated with modulation of IB phosphorylation status, and subsequently IB proteolytic degradation (Li et al., 1999), it is possible that AEBP1 may serve as a direct or indirect up-regulator of IKK activity, leading to increased phosphorylation of IB in macrophages. To examine this possibility, we assessed IKK kinetic activity in AEBP1^+/+ and AEBP1^–/– macrophages by in vitro kinase assay. To this end, endogenously expressed IKK complex was immunoprecipitated from whole-cell protein extracts obtained from unstimulated and LPS-stimulated AEBP1^+/+ and AEBP1^–/– macrophages using anti-IKK antibody, and IKK kinetic activity was assessed using a bacterially expressed GST-IB (1-54) fusion protein as a substrate in an in vitro kinase assay. Clearly, IKK kinetic activity is indistinguishable in AEBP1^+/+ and AEBP1^–/– macrophages (Figure 4A), indicating that AEBP1 does not alter the enzymatic activity of IKK in macrophages. Equivalent GST-IB (1-54) and IKK and IKK immunoprecipitate levels was confirmed by blotting the membrane with anti-GST, anti-IKK, and anti-IKK antibodies, respectively (Figure 4A).

View larger version (54K): [in this window] [in a new window]   Figure 4. AEBP1 does not influence the kinetic activity of the IKK complex and is not part of it. (A and B) AEBP1+/+ and AEBP1–/– macrophages exhibit similar IKK kinetic activity in vitro. (A) In vitro kinase assay using immunoprecipitated, endogenously expressed IKK complex and GST-IB (1-54) fusion protein as a substrate was performed followed by autoradiography. The IKK complex was immunoprecipitated using anti-IKK antibody, and normal IgG was used as a negative control. Anti-GST antibody was used to detect GST-IB (1-54) as a loading control, and antisera against the indicated proteins were used to confirm successful immunoprecipitation of the indicated IKK complex components. (B) AEBP1 and actin protein levels are shown as an input control. (C and D) AEBP1 is not a component of the IKK complex. (C) Whole-cell protein extracts isolated from unstimulated and LPS-stimulated macrophages were subjected to coimmunoprecipitation analysis using anti-AEBP1 antibody or normal IgG as a control. SDS-PAGE was performed and followed by immunoblotting with specific antisera against the indicated proteins. (D) Whole-cell protein extracts were immunoblotted for the indicated proteins as an input control. This data are representative of two independent experiments.

AEBP1 Is Not a Component of the IKK Complex
It is possible that AEBP1 may be part of the IKK complex by means of physical interaction with one or more of its known or unknown components, and thus, it may be directly involved in regulating IB phosphorylation and proteolytic degradation. To examine this possibility, AEBP1 was immunoprecipitated from unstimulated and LPS-stimulated AEBP1^+/+ and AEBP1^–/– macrophages, and Western blot analysis was performed on the immunoprecipitate samples to check for IKK and IKK coimmunoprecipitation. Protein extracts obtained from LPS-stimulated AEBP1^+/+ macrophages were incubated with normal IgG to serve as a negative control for AEBP1 coimmunoprecipitation analysis. Figure 4C suggests that neither IKK nor IKK coimmunoprecipitates along with AEBP1, despite successful AEBP1 immunoprecipitation. Figure 4D represents immunoblotting of whole-cell protein extracts for the indicated proteins as an input control. Consistently, AEBP1 does not coimmunoprecipitate along with IKK, IKK, NIK, and MEKK1, major components of the IKK complex (Figure 4A). These findings strongly suggest that AEBP1 is not a component of the IKK complex in macrophages.

AEBP1 Physically Interacts with Endogenous IB, but not IB, in Macrophages
In an attempt to shed light on the molecular mechanism by which AEBP1 contributes to IB protein instability, we hypothesized the AEBP1 may be capable of physically interacting with IB in a way that renders IB more prone to phosphorylation, and subsequently, proteolytic degradation. To examine this possibility, we performed coimmunoprecipitation studies in which endogenously expressed IB was immunoprecipitated from whole-cell protein extracts isolated from unstimulated and LPS-stimulated AEBP1^+/+ and AEBP1^–/– macrophages using anti-IB antibody. Protein extracts obtained from LPS-stimulated AEBP1^+/+ macrophages were incubated with normal IgG to serve as a negative control for IB coimmunoprecipitation analysis. Figure 5A, top left panel, clearly demonstrates that endogenous AEBP1 physically interacts with endogenous IB in macrophages. Importantly, no AEBP1 band was detectable in the normal IgG control sample or IB immunoprecipitate samples from AEBP1^–/– macrophages. Notably, coimmunoprecipitation analysis suggests that AEBP1 physically interacts with IB to the same extent under basal and LPS-stimulatory conditions Figure 5A, top left panel. Reciprocal coimmunoprecipitation analysis revealed detection of 37-kDa bands corresponding to IB when endogenously expressed AEBP1 was immunoprecipitated from LPS-stimulated AEBP1^+/+ macrophages (Figure 4C), confirming AEBP1-IB physical interaction.

View larger version (69K): [in this window] [in a new window]   Figure 5. AEBP1 physically interacts with endogenous IB in macrophages via its DLD. (A) Whole-cell protein extracts isolated from unstimulated and LPS-stimulated macrophages were subjected to coimmunoprecipitation analysis using anti-IB antibody (top left panel), anti-IB antibody (top right panel), or normal IgG followed by immunoblotting with anti-AEBP1 antibody. IB and IB immunoprecipitation was confirmed by immunoblotting using anti-IB and anti-IB antibodies (middle panels). AEBP1 and actin protein levels are shown as an input control (bottom panel). (B) A schematic representation illustrates AEBP1 derivatives with expected protein size. The ability of AEBP1 derivatives to physically interact with endogenous IB is indicated (– or +). (C) Whole-cell protein extracts isolated from unstimulated and LPS-stimulated C6 glioma transfectants were subjected to SDS-PAGE and immunoblotting using anti-AEBP1 antibody to detect exogenously expressed AEBP1 derivatives. (D) IB immunoprecipitate samples obtained from unstimulated and LPS-stimulated C6 glioma transfectants were subjected to SDS-PAGE and immunoblotting using anti-AEBP1 (top panel) anti-IB (bottom panel) antibodies. An asterisk represents AEBP1 derivatives that coimmunoprecipitated with endogenous IB. Data are representative of three independent experiments.

To determine the region or domain of AEBP1 that is required for protein–protein interaction with IB, we performed coimmunoprecipitation analysis using C6 glioma cells that are transfected with various mutant derivatives of AEBP1 (Figure 5B). On transfection, whole-cell protein extracts were obtained from C6 transfectants and endogenously expressed IB was immunoprecipitated. Subsequently, whole-cell protein extracts and immunoprecipitate samples were subjected to SDS-PAGE followed by immunoblotting using anti-AEBP1 antibody. As shown in Figure 5C, almost similar levels of AEBP1 derivatives were exogenously expressed in transfectants. Interestingly, the WT form of AEBP1 specifically coimmunoprecipitated with IB, in which AEBP1 band was absent in the normal IgG control sample (Figure 5D, top panel). It is evident that all AEBP1 mutant forms, except N and CP, are capable of binding IB (Figure 5D, top panel), despite equivalent IB immunoprecipitation (Figure 5D, bottom panel). Because N and CP are the only mutant derivatives of AEBP1 that lack DLD, these findings suggest that AEBP1 physically interacts with IB via its DLD. Consistently, coimmunoprecipitation analysis using whole-cell protein extracts obtained from unstimulated and LPS-stimulated C6 glioma transfectants confirmed that LPS has no detectable effect on AEBP1-IB protein–protein interaction in vitro (Figure 5D, top panel). In agreement with a lack of protein–protein interaction between endogenously expressed AEBP1 and IB in macrophages, coimmunoprecipitation studies using C6 glioma transfectants failed to reveal any physical interaction between AEBP1 and IB (data not shown). Taken together, these findings suggest that AEBP1 physically interacts with IB, but not IB, in macrophages via its DLD in an LPS-independent manner.

DLD Is Essential for AEBP1-mediated IB Proteolytic Degradation and NF-B Up-Regulation.
Because AEBP1 mutant derivatives that lack DLD were unable to physically interact with IB, we next examined the effect of deleting DLD of AEBP1 on IB protein stability, as well as NF-B transcriptional activity. Whole-cell protein extracts obtained from unstimulated and LPS-stimulated C6 glioma cells that are transiently transfected with plasmids expressing various AEBP1 derivatives were subjected to SDS-PAGE and immunoblotting to assess IB and IB protein levels. As shown in Figure 6, A and B, overexpression of WT, NCP, C, and STPA, but not N and CP, forms of AEBP1 is accompanied by decreased protein levels of IB, but not IB, under basal and LPS-stimulatory conditions, despite equal loading. Moreover, C6 glioma transfectants that express AEBP1 derivatives retaining DLD (i.e., WT, NCP, C, and STPA) display about threefold higher levels of phosphorylated IB, under basal and LPS-stimulatory conditions (Figure 6, A and B).

To evaluate the physiological importance of DLD in AEBP1-promoted IB protein degradation, C6 glioma cells were transiently transfected with plasmids expressing the WT, N, and C derivatives of AEBP1 and subsequently treated with CHX to assess IB protein stability. Immunoblotting and densitometric analyses revealed that DLD deletion eliminates AEBP1's ability to promote IB proteolytic degradation, indicating that DLD is indispensable for AEBP1-promoted IB protein instability (Figure 6, C and D). Importantly, the C mutant form of AEBP1, which was used as a positive control, behaved similar to the WT form (Figure 6, C and D). In those cells expressing WT, N, or C derivatives, the half-life of IB protein turnover is estimated to be 60, >240, and 60 min, respectively. Notably, transfectants expressing various derivatives of AEBP1 displayed no significant differential patterns with respect to IB or p53 protein turnover (Figure 6C), substantiating AEBP1's function as an IB proteolytic degradation mediator.

View larger version (49K): [in this window] [in a new window]   Figure 6. DLD is required for AEBP1-mediated enhanced IB phosphorylation and proteolytic degradation. (A) Whole-cell protein extracts obtained from unstimulated or LPS-stimulated C6 glioma transfectants were subjected to SDS-PAGE and immunoblotting for the indicated proteins. (B) A histogram illustrates densitometric analysis of protein levels shown in A. Statistical significance was determined relative to individual protein levels in unstimulated or LPS-stimulated, empty vector transfectants. (C) Whole-cell protein extracts isolated from control and CHX-treated C6 glioma transfectants expressing WT, N, and C forms of AEBP1 were subjected to SDS-PAGE and immunoblotting for the indicated proteins. (D) A histogram illustrates densitometric analysis of protein levels shown in C. Relative protein levels were determined based on actin expression, values at 0 min were taken to be 100% for every transfectant group, and statistical significance was determined relative to IB protein level in WT-transfected cells at each time point (n = 3).

View larger version (50K): [in this window] [in a new window]   Figure 7. AEBP1-mediated up-regulation of NF-B activity is DLD-dependent. (A) Nuclear protein extracts obtained from unstimulated or LPS-stimulated C6 glioma transfectants were subjected to SDS-PAGE and immunoblotting for p65. (B) A histogram illustrates densitometric analysis of nuclear protein levels shown in A, after normalization based on actin expression. Statistical significance was determined relative to protein level in unstimulated or LPS-stimulated, empty vector-transfectants. (C) Whole-cell protein extracts isolated from unstimulated or LPS-stimulated C6 glioma transfectants that are cotransfected with NF-B-luciferase reporter were subjected to luciferase reporter assay. Statistical significance was determined relative to luciferase activity in unstimulated or LPS-stimulated, empty vector-transfectants (n = 3–8). (D) Nuclear protein extracts isolated from unstimulated and LPS-stimulated C6 glioma transfectants were subjected to EMSA in presence of specific (unlabeled NF-B probe) or nonspecific (unlabeled, unrelated probe) competitors. This data are representative of three independent experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Herein, we provide experimental evidence establishing a positive relationship between AEBP1 expression and NF-B activity in macrophages (Figure 1). Indeed, nuclear p65 protein level is significantly reduced in AEBP1^–/– macrophages, compared with AEBP1^+/+ counterparts (Figure 1, B and C). Consistently, EMSA clearly illustrates that ablation of AEBP1 expression in macrophages correlates with markedly inhibited NF-B DNA-binding activity (Figure 1D). Interestingly, AEBP1 expression in macrophages negatively correlates with IB, but not IB, protein levels (Figure 3, C and D). Our data suggest that the negative regulation imposed by AEBP1 on IB function is independent of alterations in the IKK complex kinetic activity (Figure 4A) and is mediated by protein–protein interaction that requires DLD of AEBP1 (Figure 5D). It is clear that the half-life of IB is significantly prolonged in macrophages that lack AEBP1, compared with macrophages that express normal levels of AEBP1 (250 vs. 50 min, respectively; Figure 3, E and F), and that DLD of AEBP1 is crucial for such a biological regulation (Figure 6, C and D). Although modulation of IB expression has been previously shown to be a mechanism explaining altered NF-B activity (Azuma et al., 1999; Delerive et al., 2000; Saile et al., 2001), our study is the first of its kind to propose a molecular mechanism behind modulated NF-B activity by which IB protein stability is altered via protein–protein interaction and is independent of alterations in IKK complex kinetic activity.

Our findings suggest that AEBP1 physically interacts with IB via its DLD, which is known be a protein–protein interaction mediator (Johnson et al., 1993; Prag et al., 2004). Indeed, DLD of AEBP1 was found to be required for protein–protein interaction between AEBP1 and MAPK (Kim et al., 2001). Similarly, this study suggests that AEBP1 is capable of physically interacting with IB by means of its DLD, whose deletion eliminated AEBP1-IB interaction (Figure 5D). It is worth mentioning that despite the structural similarities between IB and IB, our data suggests that AEBP1 is capable of interacting with IB, but not IB, in macrophages (Figure 5A). Analysis of IB-IB amino acid sequence alignment reveals that there are three main structural differences between IB and IB. First, the first 12 amino acid residues in IB are absent in IB. Second, there is a 41-amino acid stretch located between the third and forth ANK repeat of IB that is not present in IB. Third, there is an 18-amino acid stretch at the C-terminus of IB that is absent in IB. Based on sequence analysis, it is conceivable that either the presence of the first 12 amino acid residues in IB is required for interaction with AEBP1 or that the presence of the extra amino acid stretches in IB allows the formation of a tertiary structure that does not permit protein–protein interaction with AEBP1. It is also conceivable that the extra amino acid stretches in IB somehow mask the region, or domain, that is necessary for protein–protein interaction with AEBP1. Identification of the region of IB that mediates protein–protein interaction with AEBP1, which will shed more light on the differential ability of AEBP1 to regulate IB and IB functions in vivo, is currently underway. Further investigation of the AEBP1-interacting region of IB will shed more light on how AEBP1 is capable of differentially regulating IB and IB functions in vivo.

Consistent with our proposal that AEBP1-IB protein–protein interaction provokes destabilization of IB shortening its half-life, N and CP mutant forms of AEBP1 had no effect on IB protein stability (Figure 6). Importantly, in contrast to WT form of AEBP1, N, and CP mutant forms possess minimal up-regulatory function toward NF-B activity (Figure 7), confirming that AEBP1-IB interaction is a key biological event that is crucial for AEBP1-mediated IB-induced degradation and subsequent NF-B up-regulation.

Despite their structural similarities, IB and IB are known to be differentially regulated in various cell types and under several stimulatory conditions (Thompson et al., 1995; Weil et al., 1997). For instance, PMA or TNF- treatment was shown to induce IB degradation, while I