Neutrophil Gelatinase-Associated Lipocalin Is Up-Regulated in Human Epithelial Cells by IL-1β, but Not by TNF-α

Jack B. Cowland, Ole E. Sørensen, Maxwell Sehested and Niels Borregaard
J Immunol December 15, 2003, 171 (12) 6630-6639; DOI: https://doi.org/10.4049/jimmunol.171.12.6630

Abstract

Synthesis of the antimicrobial protein neutrophil gelatinase-associated lipocalin (NGAL) increases dramatically in bronchial epithelial cells and alveolear type II pneumocytes during lung inflammation. IL-1β induces a >10-fold up-regulation of NGAL expression in the type II pneumocyte-derived cell line A549 cells, whereas TNF-α, IL-6, and LPS had no effect. Similar IL-1β selectivity was demonstrated in primary bronchial epithelial cells and epidermal keratinocytes and for an NGAL promoter fragment transfected into A549 cells. By deletion and substitution analysis of the NGAL promoter, a 40-bp region containing an NF-κB consensus site was found to control the IL-1β-specific up-regulation. Involvement of the NF-κB site was demonstrated by site-directed mutagenesis, by transfection with a dominant-negative inhibitor of the NF-κB pathway, and by EMSA. TNF-α activation of NF-κB, in contrast, did not increase NGAL synthesis, even though induced binding of NF-κB to the NGAL promoter was observed in vitro. IL-1β specificity was not contained within the NF-κB site of the NGAL promoter, as determined by exchanging the NGAL promoter′s NF-κB-binding sequence with that of the IL-8 promoter or with the NF-κB consensus sequence and by testing the NF-κB-binding sequence of the NGAL promoter against the heterologous SV40 promoter. Selectivity for the IL-1 pathway was substantiated by demonstrating that NGAL promoter activity could be induced by LPS stimulation of A549 cells transiently expressing Toll-like receptor 4, which use the same intracellular signaling pathway as the IL-1R. Together, this demonstrates a selective up-regulation of NGAL by the IL-1 pathway.

Neutrophil gelatinase-associated lipocalin (NGAL) 3 is a 25-kDa glycoprotein first found in the specific granules of the human neutrophil (1). NGAL belongs to the lipocalin superfamily whose members share a barrel-shaped tertiary structure with a hydrophobic pocket that can bind small lipophilic molecules (2). Recently, crystallographic analysis has indicated that the ligand of NGAL is bacterial ferric siderophores, which are used by bacteria for uptake of the essential nutrient iron (3). This is substantiated by the observation that NGAL acts as a potent bacteriostatic agent under iron-limiting conditions (3). Siderophores have a very high affinity toward ferric iron and are able to acquire iron bound to mammalian proteins, such as hemoglobin, transferrin, and lactoferrin (4). Because NGAL, in contrast to antimicrobial proteins such as lactoferrin and transferrin (4), does not bind iron directly, but rather indirectly via the bacterial iron-transporting molecule, this lipocalin represents a novel and important iron-depleting antimicrobial defense strategy by the organism.

During our previous characterization of the NGAL gene, we found that epithelial tissues, and especially lung and trachea, express high amounts of NGAL (5). Previous reports also suggested, but far from demonstrated, that NGAL is up-regulated in response to inflammation (6, 7). This prompted us to examine the expression of NGAL in epithelial cells under normal growth conditions and following stimulation with proinflammatory cytokines. In this study, we demonstrate that NGAL is found in bronchial goblet cells and alveolar type II pneumocytes and that the expression of NGAL is increased in bronchial and alveolar cells of inflamed lungs. Examination of the type II pneumocyte-derived cell line A549, as well as primary epithelial cell cultures from bronchus and skin, demonstrated a constitutive synthesis of NGAL, which is specifically up-regulated by IL-1β. By promoter studies, we show that IL-1β induces NGAL expression at the transcriptional level by an NF-κB-dependent pathway and that activation of NF-κB by TNF-α is insufficient to increase NGAL expression, indicating that an additional (transcription) factor, induced by IL-1β, but not TNF-α, must act in conjunction with NF-κB to up-regulate NGAL synthesis.

Materials and Methods

Cell culture

A549 (ATCC CCL-185) and HL-60 (ATCC CCL-240) cells were obtained from the American Type Culture Collection (Manassas, VA). A549 was grown in HAM F12 (Life Technologies, Rockville, MD) and HL-60 in RPMI 1640 (Life Technologies). The medium was supplemented with 10% FCS (Life Technologies), 100 U/ml penicillin, and 100 μg/ml streptomycin (Life Technologies). Normal human bronchial epithelia (NHBE; with retinoic acid) and normal human epidermal keratinocytes (NHEK) were purchased from Clonetics (Walkersville, MD) and grown in bronchial/tracheal epithelial growth medium and keratinocyte growth medium-2 with bronchial/tracheal epithelial growth medium and keratinocyte growth medium-2 (without insulin) growth supplements, respectively (Clonetics). Cells were grown at 37°C in a humid atmosphere with 5% CO2. For transfection and/or induction with LPS (Klebsiella pneumoniae and Escherichia coli J5; Sigma-Aldrich, St. Louis, MO) and the cytokines IL-1β (Sigma-Aldrich), TNF-α (Sigma-Aldrich), and IL-6 (Sandoz, Basel, Switzerland), A549 cells were seeded at 3 × 105 cells/ml and grown in HAM F12 with 0.5% FCS (IL-1β, TNF-α, and IL-6) or 10% FCS (LPS) and 100 U/ml penicillin, and 100 μg/ml streptomycin. NHBE and NHEK were grown to confluence and transfected and/or induced in NHBE and NHEK medium with supplements, as above.

Isolation of neutrophils and bone marrow (BM) cells

Mature neutrophils (PMNs) were isolated from peripheral blood by Lymphoprep (Nygaard, Oslo, Norway) density centrifugation, and neutrophil precursors (myelocytes and metamyelocytes) were isolated by Percoll density centrifugation of BM cells, as described earlier (8).

RNA isolation and Northern blot

Total RNA was prepared with TRIzol (Life Technologies), according to the manufacturer’s recommendations, and the concentration was determined by spectrophotometric measurement. Total RNA from trachea, lung, colon, and brain was purchased from Clontech (Palo Alto, CA). For Northern blotting, 10 μg of RNA was run on a 1% agarose gel, transferred to a Hybond-N membrane (Amersham Biosciences, Little Chalfont, U.K.), and hybridized, as described (8). The membranes were washed, as described (8), and developed by a Fuji BAS2500 phosphor imager. The sizes of the mRNAs were determined by reference to 18S and 28S RNA. The membranes were stripped by boiling in 0.1% SDS before rehybridization. The hybridization probes were a 663-bp NGAL cDNA fragment (8) and a 254-bp IL-8 cDNA fragment, which was PCR amplified with 5′-ACAAGCTTCTAGGACAAGAGC-3′ and 5′-GTTGGCGCAGTGTGGTCC-3′ using cDNA from IL-1β-stimulated A549 cells and cloned in pCRII (Invitrogen, San Diego, CA). The probes were radiolabeled with [α-32P]dCTP using the Random Primers DNA Labeling System (Life Technologies). For quantitative assessments, the intensities of the NGAL and IL-8 signals were normalized to the hybridization intensity from a probe against β-actin (8).

Plasmids for promoter studies

A 1746-bp fragment of the human NGAL promoter region (−1695 to +51) was PCR amplified with primers A, 5′-TGCAAGCTTCTCGAGGATCTCGGCTCAC-3′, and B, 5′-TGCAGATCTGCAGGCGCTGTGGTGGC-3′, using genomic DNA as template. The BglII-digested PCR fragment was cloned in SmaI/BglII-restricted pCAT3-basic (Promega, Madison, WI), resulting in pNGP1695CAT. Deletions from the 5′ end were generated by digesting pNGP1695 with SacI and XhoI, followed by exonuclease III treatment (Erase-a-base; Promega) or by PCR amplification using a specific upstream primer and primer B (see above). The PCR products were cloned in pCAT3-basic. Point mutations were generated by PCR amplification with mutated primers. The mutants were generated either by direct PCR amplification with a mutated primer or by use of the QuikChange protocol (Stratagene, La Jolla, CA). The DNA fragment containing the mutated region was inserted in pNGP1695CAT by substitution cloning in which the wild-type sequence was exchanged for the mutated sequence. The following point mutations were made: 1) pNGP-183MCAT (with a 3-bp substitution in the −180/−171 NF-κB site (GGG→AAT) of the −183 promoter deletion); 2) pNGP1695(NF-κB)CAT, which has the same 3-bp substitution as pNGP-183MCAT in the context of the 1695 full-length promoter; 3) pNGP1695(C/EBP-1)CAT (with a 2-bp substitution of the −148/−139 C/EBP (CCAAT/enhancer binding protein) site: TT→ CC); 4) pNGP1695(C/EBP-2)CAT (with a 2-bp substitution of the −130/−121 C/EBP site: AA→GG); 5) pNGP1695(NF-Y)CAT (with a 2-bp substitution of the −143/−139 NF-Y site: CCAAT→GCAGT); 6) pNGP1695(AP-1)CAT (with a 2-bp substitution of the −56/−50 AP-1 site: CA→TG); 7) pNGP1695(NF-κB(IL-8))CAT (with a 6-bp substitution of the NGAL NF-κB site (underlined) and its 3 flanking bases on each side to the homologous region of the IL-8 NF-κB site: TCCGGGAATGTCCCTC→TCGTGGAATTTCCTCT); 8) pNGP-183(IL-8) with the same substitution as in pNGP1695(NF-κB(IL-8))CAT, but in the context of the −183 substitution; and 9) pNGP-183(cons) with a change of the NF-κB sequence of −183 to the NF-κB consensus sequence (GGGAATGTCC→GGGAATTCCC).

Twelve plasmids with a 20-bp substitution of the NGAL promoter were made by PCR amplifying the entire plasmid using 5′-GCGAATTCTCCGCTCXXXXX-3′ and 5′-GCGAATTCTTTGATTXXXXX-3′, which each contained an internal EcoRI restriction site (underlined) and a sequence specific for the bases flanking the 20-bp region that was to be substituted (the flanking sequences, which varied for each substitution, are denoted by the XXXXX). Following PCR amplification, the DNA product was digested with EcoRI and self ligated, resulting in the formation of the 20-bp substitution sequence (5′-GAGCGGAGAATTCTTTGATT-3′) with an internal EcoRI site. The DNA fragment containing the mutated region was excised and inserted in pNGP1695CAT by substitution cloning, as above. The 12 substitutions covered the region from −304 to −120 with a 5-bp overlap between each successive mutant starting with pNGP1695CAT (304/285), in which the −304- to −285-bp region is substituted, and ending with pNGP1695CAT (139/120) (see Fig. 5⇓ for further details). Oligonucleotides containing the 10-bp NF-κB recognition sequence of the NGAL or IL-8 promoters with, or without, the flanking 5 bp on each side. For each oligonucleoide pair, the sense and antisense strands were synthesized with additional bases at their ends such that the annealed oligos could be cloned directly into NheI/BglII-restricted pCAT3 promoter (Promega). The −180/+44 region of the IL-8 promoter was PCR amplified from human genomic DNA with 5′-CGCGCTAGCAAAGAAAACTTTCGTCATACTCC-3′ and 5′-GCGCTCGAGAGCTTGTGTGCTCTGCTGTC-3′ and NheI/XhoI restricted for cloning in NheI/XhoI-digested pCAT3-basic, resulting in pIL-8CAT. The vector pIL-8(NF-κB)CAT (with a 6-bp substitution of the IL-8 promoter NF-κB site: TGGAATTTCC→GTTAACTTAA) was constructed by PCR-based mutation using pIL-8CAT as template. The mutated IL-8 promoter fragment was cloned in pCAT3-basic. The cDNA insert of pcDNA3.1-MD2 was amplified with 5′-CGCAAGCTTGCCACCATGTTACCATTTCTGTTTTTTTC-3′ and 5′-CGCCTCGAGCTAATTTGAATTAGGTTGGTGTA-3′ using cDNA generated by reverse transcription (SuperScript II; Life Technologies) of total RNA from PMNs. The PCR product was HindIII/XhoI restricted cloned in pcDNA3.1/zeo (Invitrogen) digested with the same enzymes. The 5′ end points of the deletion mutants and the correctness of all cloned PCR fragments were assured by sequencing. The vectors pCMV-IκBα and pCMV-I-κBαM were purchased from Clontech. pcDNA3-TLR4 (Toll-like receptor 4) was kindly donated by R. Medzhitov (Yale University, New Haven, CT). pcDNA3-dnMyD88 was kindly provided by M. Muzio, Discovery Research Oncology (Pharmacia, Nerviano, Italy) and R. Dziarski, Indiana University School of Medicine (Gary, IN).

Cell transfection and reporter enzyme assay

Transfection was performed by use of Effectene (Qiagen, Hilden, Germany) with 0.8 μg CAT promoter construct. To compensate for differences in transfection efficiency, the cells were cotransfected with 0.2 μg pcDNA3-β-Gal, a plasmid encoding β-galactosidase (β-Gal) (a gift from J. Bundgaard, Rigshospitalet, Denmark). All transfections were performed in quadruplicate with DNA from two independent purifications. Expression of the reporter enzymes was quantitated by CAT and β-Gal ELISA (Roche Diagnostics, Somerville, NJ), according to the manufacturer’s recommendations. For each sample, the CAT activity was normalized to β-Gal activity. A weak, but significant CAT and β-Gal activity could be measured 8 h posttransfection, demonstrating transcriptional initiation from the transfected plasmids 6–7 h after transfection. For this reason, induction of the transfected promoters was initiated 6 h posttransfection, and the cells were harvested 24 h later. However, for cells transfected with plasmids expressing TLR4 and/or MD2, induction was first initiated 18 h posttransfection to allow synthesis of these membrane proteins before administration of LPS. In this case, the level of CAT activity was determined at the time of induction and subtracted from the CAT activity measured 24 h later.

Nuclear extracts and EMSA

Nuclear proteins were prepared from A549 cells stimulated for 1 h with IL-1β or TNF-α, as previously described (9). Double-stranded oligonucleotides (5′-ACTCCGGGAATGTCCCTCAC-3′ + 5′-GTGAGGGACATTCCCGGAGT-3′ (probe/specific competitor) and 5′-ACTCCAATAATGTCCCTCAC-3′ + 5′-GTGAGGGACATTATTGGAGT-3′ (unspecific competitor)) were labeled with [γ-32P]ATP and incubated with nuclear extracts for 30 min at room temperature in 20 mM HEPES buffer (pH 7.9), containing 50 mM KCl, 1 mM EDTA, 2 mM DTT, 10% (w/v) glycerol, 2 μg poly(dI-dC), 0.1% Nonidet P-40, 1 mg/ml nuclease-free BSA, and 0.5 mM PMSF. If competitor oligos were included, the nuclear extract was added following a 20-min preincubation period. For supershift analysis, 1 μl polyclonal antisera against C/EBP-ε or the NF-κB subunits p50, p52, Rel-A, Rel-B, and c-Rel (all Santa Cruz Biotechnology, Santa Cruz, CA) were added. All incubations were subjected to electrophoresis on a 4% nondenaturing polyacrylamide gels and subsequently dried and autoradiographed.

Immunohistochemistry

Specimens of human bronchial and alveolar tissues were obtained from recipient lungs from lung transplant patients with either emphysema due to α1-anti-trypsin deficiency (uninflamed tissue, four patients) or severe bronchopulmonary infection due to cystic fibrosis (inflamed tissue, three patients). The tissue samples were embedded in paraffin, and 5-μm sections were dried overnight at 37°C. Paraffin wax was dissolved in xylene (2 × 5 min), and the sections were hydrated through decreasing concentrations of ethanol. Endogenous peroxidase was blocked with 3% H2O2 for 5 min. The samples were microwave treated in Tris-EDTA buffer (10 mM Tris, 1 mM EDTA, pH 9.0) and incubated with the endogenous avidin/biotin blocking kit (Zymed, San Francisco, CA), according to the manufacturer’s recommendations. The sections were analyzed with affinity-purified polyclonal rabbit anti-NGAL Abs (0.4 mg/ml) diluted 1/1000 in TBS (50 mM Tris, 150 mM NaCl, pH 7.6) containing 2.5% BSA. As a control of the immune reaction, samples were incubated in parallel with anti-NGAL Abs that had been preincubated with 0.8 μg/ml rNGAL (10) for 30 min at 37°C before use. Incubation with preimmune rabbit IgG (DAKO, Glostrup, Denmark; X-0936) was also included as a negative control. The sections were incubated at room temperature for 30 min. The primary Abs were detected by biotinylated goat Ab to mouse and rabbit Ig, followed by incubation with the streptABComplex/HRP Duet, mouse/rabbit kit (DAKO), according to the manufacturer’s instructions. All steps were followed by three washes in TBS. Sections were developed with 3,3′-diaminobenzidine tetrahydrochloride (Kem-En-Tec, Copenhagen, Denmark) for 20 min and counterstained with Mayer’s hematoxylin (Bie & Berntsen, Rødovre, Denmark).

Quantitation of NGAL and IL-8

The cells were lysed in dilution buffer (0.5 M NaCl, 3 mM KCl, 8 mM Na2HPO4/KH2PO4, pH 7.2, 1% Triton X-100, 1% BSA) before the ELISA measurements. NGAL was quantitated by ELISA, as described previously (10). IL-8 was quantitated with the IL-8 optEIA ELISA kit (BD PharMingen, San Diego, CA), according to the manufacturer′s recommendations.

SDS-PAGE and immunoblotting

SDS-PAGE and immunoblotting were performed according to the instructions given by the manufacturer (Bio-Rad, Hercules, CA). For immunodetection, the polyvinylidene fluoride membranes (Millipore, Bedford, MA) were blocked for 1 h with 5% skimmed milk in PBS after the transfer of proteins from 14% polyacrylamide gels and incubated overnight with protein A-purified rabbit anti-NGAL Abs. The following day, the membranes were incubated for 2 h with peroxidase-conjugated porcine Abs to rabbit Igs (DAKO; P-0217) and visualized by diaminobenzidine/metal concentrate and stable substrate buffer (Pierce, Rockford, IL).

Results

NGAL is expressed in human airway epithelial cells

During our previous characterization of the NGAL gene, we screened 50 human tissues for the presence of NGAL mRNA by dot-blot hybridization (5). Besides bone marrow, which hybridized strongly due to its large content of NGAL-producing neutrophilic precursors (8), strong expression of NGAL was found in most epithelial tissues, especially in lung and trachea (5). To validate these dot-blot data, we analyzed RNA from epithelial cells by Northern blot hybridization (Fig. 1⇓A). A single prominent band of ∼1000 bp was detected in RNA from trachea and a fraction of bone marrow cells (myelocytes and metamyelocytes) known to express the NGAL gene (8). A weaker, but still significant, hybridization signal was obtained from lung and colon tissue and from A549 cells (a type II pneumocyte-derived cell line) (11), NHBE cells (primary human bronchial epithelial cells), and NHEK cells (primary human keratinocyte epithelial cells). No expression of NGAL was observed in HL-60 cells (a promyelocytic cell line), peripheral blood PMNs, and brain, in accordance with previous observations (8, 12).

FIGURE 1.
FIGURE 1.

Expression of NGAL in various tissues and cell lines. A, Total RNA (10 μg) from trachea, lung, colon, brain, PMN, MC + MM (a myelocyte- and metamyelocyte-enriched fraction of bone marrow), HL-60 cells, NHEK cells, NHBE cells, and A549 cells hybridized to a 32P-labeled probe against NGAL. B, Medium from uninduced A549 cells was collected after 24 h and analyzed by immunoblotting with polyclonal anti-NGAL Abs. A single 25-kDa band was detected in the medium from A549 cells. No immunoreactivity was found in fresh medium. PMNs (1 × 106 cells/ml) were included as a positive control.

Due to the strong signal from the cells of the respiratory tract, we decided to examine these tissues further. To ensure that the NGAL signal originated from the epithelial cells of the pulmonary system, we analyzed uninflamed bronchial and alveolar tissue samples for NGAL by immunohistochemical staining (Fig. 2⇓). Because strong induction of NGAL previously was shown in inflamed colon epithelium (6), we included tissue samples from patients with severe pulmonary inflammation. Staining for NGAL was observed in bronchial goblet cells of uninflamed lungs (Fig. 2⇓a). A much stronger staining was, however, observed in the bronchial epithelial cells of heavily inflamed lungs (Fig. 2⇓d), and in this case not only the goblet cells, but also the ciliated epithelial cells expressed NGAL. A weak staining of some alveolar cells was seen in uninflamed lungs (data not shown), but, similar to bronchi, a considerably higher expression was found in inflamed alveolar tissue (Fig. 2⇓, g and j). The stained alveolar cells were characterized as type II pneumocytes (also known as surfactant cells) on the basis of their characteristic morphology (rounded nuclei with a prominent nucleolus and vacuolated cytoplasm) (13). Thus, NGAL was induced in human airway epithelium during inflammation in vivo.

FIGURE 2.
FIGURE 2.

NGAL synthesis is up-regulated in inflammatory diseases of the lung. a, Immunohistochemistry using a polyclonal Ab against NGAL on lung sections reveals staining for NGAL in bronchial goblet cells of uninflamed lungs. The specificity of the reaction is demonstrated by the lack of staining using NGAL-blocked Abs (b) or preimmune serum (c). Strong staining for NGAL is seen in bronchial epithelial cells (d) and alveolar type II pneumocytes (g) of inflamed lungs. Lack of specific tissue staining was seen when blocked Abs (e, h) or preimmune serum (f, i) were used. Higher magnification of the alveolar stains in g and i is shown in j and k, respectively. Magnifications: a–i, ×400; j and k, ×1000.

NGAL is expressed in A549 cells

Based on the above findings, we decided to examine NGAL expression in A549 cells, as this is a well-characterized airway-derived cell line with a gene expression profile very similar to that of normal airway epithelial cells (14). However, to ensure that our findings in A549 were not artifacts caused by the use of an immortalized cell line, primary epithelial cells from human bronchus (NHBE) and skin (NHEK) were also included where appropriate to confirm the major findings in A549 cells.

We measured the amount of NGAL in unstimulated A549 cells by ELISA to 37 ng per 109 cells, whereas the same amount of cells secreted 8 μg NGAL into the medium over a 24-h period, which is >200-fold more NGAL than stored within the cells. This indicates that NGAL is transported out of the cell by constitutive secretion. A single band at 25 kDa, identical in size with NGAL from neutrophils, was identified by Western blotting of A549 medium (Fig. 1⇑B). No NGAL was detected in A549 cells by immunoblotting (data not shown), supporting the lack of intracellular NGAL storage in these cells. In a similar manner, NGAL was synthesized and exported to the medium by NHBE and NHEK cells (Fig. 3⇓).

FIGURE 3.
FIGURE 3.

Time course of NGAL and IL-8 production by A549, NHBE, and NHEK cells stimulated with inflammatory mediators. The amount of NGAL in the medium of A549 (A), NHBE (C), and NHEK (E) cells was determined at the indicated time points after addition of fresh medium, or medium supplemented with IL-1β, TNF-α, IL-6, or LPS. To ensure that IL-1β and TNF-α induced a proper cellular response, IL-8 was measured in all the samples (B, D, and F). IL-6 does not induce IL-8 synthesis and was included to ensure that no unspecific induction of the cells took place. The concentrations of NGAL and IL-8 from A549 cells are shown as the mean ± SD of four independent experiments.

NGAL synthesis is up-regulated by IL-1β

Because intense staining for NGAL was observed in inflamed lung tissue, we examined whether NGAL synthesis could be up-regulated in A549 cells in response to inflammatory mediators. A549 cells were incubated with either IL-1β (100 pg/ml), TNF-α (20 ng/ml), IL-6 (50 ng/ml), or LPS (50 μg/ml), and the amount of NGAL in the medium was quantitated at different time points over a 48-h period (Fig. 3⇑A). A strong induction of NGAL synthesis was observed when the cells were incubated with IL-1β and resulted in a 10- to 12-fold increase of NGAL in the medium at 24–48 h postinduction relative to that measured from uninduced cells at the same time points. In contrast, an increase of NGAL over time, similar to that from uninduced A549 cells, was measured in the medium of A549 cells incubated with TNF-α, IL-6, and LPS, demonstrating that these inflammatory mediators do not increase NGAL synthesis.

A similar cytokine specificity was observed for the primary epithelial cells. IL-1β (20 ng/ml) induced an almost 3-fold increase of exported NGAL relative to uninduced and TNF-α (20 ng/ml)-induced NHBE cells (Fig. 3⇑C). This modest increase of NGAL is probably due to a higher basal activity of NGAL and/or a weaker response to IL-1β by NHBE cells (15). A slightly larger effect of IL-1β stimulation was observed in NHEK cells in which a 5- to 6-fold increase in NGAL was measured in the medium of cells induced for 48 h with IL-1β (20 ng/ml) relative to that from unstimulated cells and cells induced by TNF-α (20 ng/ml) at the same time points (Fig. 3⇑E). Stimulation of primary cells with LPS and IL-6 was omitted due to the limited amount of cells available. The biological activities of IL-1β and TNF-α were validated by their ability to induce IL-8 synthesis (Fig. 3⇑, B, D, and F). The seemingly stronger induction of IL-8 synthesis than of NGAL synthesis by IL-1β in NHBE cells (if measured as fold induction relative to that secreted from uninduced cells) is probably due to the combined effect of the transcriptional activating signals generated by IL-1β (and TNF-α) and the concomitant relieve of the transcriptional silencing of the IL-8 promoter by NF-κB-repressing factor that occurs in uninduced cells (16).

IL-1β induces a steady increase of NGAL mRNA levels

To investigate the kinetics and specificity of NGAL up-regulation at the transcriptional level, we purified RNA from A549, NHBE, and NHEK cells harvested at different time points (0–48 h) after IL-1β and TNF-α induction. As controls, we purified RNA from unstimulated cells harvested at the same time points after addition of fresh medium. The RNA was analyzed by Northern blotting and hybridization to probes for NGAL, IL-8, and the housekeeping protein β-actin for normalization (Fig. 4⇓). A large rise in NGAL mRNA levels was seen only in response to IL-1β in contrast to IL-8 mRNA levels, which increased in response to both IL-1β and TNF-α. The IL-1β-induced expression of NGAL increased over time for all cell types and was most pronounced after 48 h of stimulation.

FIGURE 4.
FIGURE 4.

Kinetics of IL-1β-induced NGAL and IL-8 transcription in A549, NHBE, and NHEK cells. Left, Cells were harvested at the indicated time points after addition of fresh medium (unstimulated) or medium supplemented with IL-1β or TNF-α. RNA was isolated and hybridized to 32P-labeled NGAL, IL-8, and β-actin cDNA probes. Right, A schematic representation of the relative NGAL and IL-8 transcript levels from unstimulated cells and following stimulation with IL-1β and TNF-α. The band intensities were quantified on a phosphor imager, and the time 0 levels of NGAL and IL-8 mRNA, respectively, were given the value 1. The levels of the other bands are shown relative to these. For both NGAL and IL-8, the transcript levels were normalized to the intrinsic transcript level of β-actin to compensate for loading differences. The data are representative of two independent experiments.

IL-1β specifically induces expression of the CAT reporter gene under control of the NGAL promoter

To test whether the in vivo induction pattern of NGAL could be reproduced in vitro, we transfected A549 cells with a CAT reporter plasmid under control of a 1746-bp fragment of the NGAL 5′-upstream region (−1695 to +51). Following stimulation with IL-1β, TNF-α, IL-6, or LPS for 24 h, the cells were harvested and the CAT activity (Fig. 5⇓A), as well as the amount of NGAL and IL-8 in the medium (Fig. 5⇓B), was determined. As shown in Fig. 5⇓A, only IL-1β was able to induce an increased CAT synthesis. The level and specificity of IL-1β induction demonstrated that the NGAL promoter construct was able to reproduce the induction pattern found for endogenous NGAL. The data furthermore support that stimulation of NGAL synthesis by IL-1β occurs at the transcriptional level.

FIGURE 5.
FIGURE 5.

Activity of the NGAL promoter in stimulated A549 cells. Top, DNA sequence and putative regulatory consensus elements of the −200 to 72 region of the NGAL gene (5 ). Underlined sequences denote putative transcription factor recognition sites, and the numbers above the sequence show the end points of the −183 to −63 deletion mutants. The transcriptional and translational start sites are indicated with dots and Met, respectively. The extent of the replaced sequences of the five most downstream substitutions is shown with bars under the sequence. A, A CAT-reporter plasmid carrying the −1695 to +51 NGAL promoter region was transfected into A549 cells and stimulated with IL-1β, TNF-α, IL-6, or LPS for 24 h (as in Fig. 3⇑). Transfected (unstim.) and untransfected cells (untrans.) grown in medium without inflammatory mediators were included as controls. B, NGAL and IL-8 were measured in the medium of all the cells to ensure that proper stimulation had occurred. C, Relative CAT activities of 5′ deletion mutants of the NGAL promoter ranging from −1695 to −63 and of the −183 deletion with a mutated NF-κB binding site (183M). The activities are shown relative to that of the −1695 deletion, which was given the value 1. Background activity was determined by including the reporter plasmid pCAT-basic with no promoter insert (basic). D, Fold induction in CAT activity of the deletion mutants following stimulation with IL-1β. E and F, Fold induction in CAT activity of the deletion (E) and substitution mutants (F) following stimulation with IL-1β and TNF-α. All results are the mean ± SD of four independent transfections. In all cases, the CAT activity was normalized to the β-Gal activity from the cotransfected vector pcDNA3-β-Gal.

Deleting the −183 to −153 region of the NGAL promoter abolishes transcriptional induction by IL-1β

To determine the regions of the NGAL promoter that are required for transcriptional induction by IL-1β, we tested 5′ deletions of the NGAL promoter (Fig. 5⇑C). The basal expression was reduced ∼50% when truncating from −1057 to −640 and then remained constant until the region between −153 and −120 was deleted and resulted in a further 5-fold reduction of CAT activity. The −90 and −63 deletions were at background level. This indicates that cis elements important for basal activity are found between −1057 and −640 and between −153 and −90. Our focus was, however, on the elements responsible for the transcriptional induction by IL-1β. Therefore, we also measured the CAT activity of transfected cells following induction with IL-1β. A 17- to 28-fold induction of CAT activity was found for the −1695 to −391 deletions, and an 8-fold induction was measured for the −183 deletion, whereas no induction was observed for deletions from −153 and downward (Fig. 5⇑B). Inspection of the sequence between −183 and −153 revealed a sequence 5′-GGGAATGTCC-3′ (−180 to −171) with homology (7 of 8 nt match) to the NF-κB consensus sequence 5′-GGG(A/G)NNT(C/T)CC-3′ (17). To test the importance of this sequence motif, we mutated 3 bases of the putative NF-κB site in the −183 deletion, which are highly conserved at NF-κB binding sites (GGGAATGTCC→AATAATGTCC) (17). This mutation (183M) did not alter the basal CAT activity, but completely abolished induction by IL-1β (Fig. 5⇑, C and D).

The region between −196 and −155 is required for full IL-1β responsiveness

As the induction of the −183 deletion (8-fold) was significantly lower than the −391 deletion (18-fold), we made additional deletions in the −400 to −200 region to narrow down the region required for complete IL-1β responsiveness. Induction of the −200 and −183 deletions was 8- to 10-fold, whereas all deletions from −250 to −427 were up-regulated 17- to 22-fold in response to IL-1β (Fig. 5⇑E). This indicated that part of the promoter between −250 and −200 was required for full IL-1β induction. No induction of promoter activity was observed following TNF-α stimulation, demonstrating that the IL-1β specificity was retained in these deletions. To further pinpoint the extent of the NGAL promoter imposing IL-1β responsiveness, we made a number of constructs in which a 20-bp sequence of the −1695 promoter was substituted with a 20-bp nonsense sequence. The substitutions ranged from −304 to −120 with a 5-bp overlap between each successive substitution (Fig. 5⇑, top). A 20- to 30-fold induction was seen for the substitutions, except for those covering the −214 to −150 region (Fig. 5⇑F). As the −214/−195 and the −154/−135 substitutions retained the ability for strong induction, the part of the NGAL promoter required for full IL-1β response must be located between −196 and −155. Again, none of the constructs was induced by TNF-α.

Disruption of the NF-κB site in the −1695 promoter fragment abolishes transcriptional induction by IL-1β

To further validate the requirement for an intact NF-κB site for transcriptional induction by IL-1β, we decided to test the mutated NF-κB site in the context of the −1695 promoter. As both IL-1β and TNF-α are known to activate NF-κB, it is likely that the specific induction of NGAL by IL-1β is due to a cooperation between NF-κB and another transcription factor activated by IL-1β, but not by TNF-α. In addition to NF-κB, IL-1β also activates mitogen-activated protein (MAP) kinases, which again activate transcription factors such as members of the jun/fos (AP-1) (18), activating transcription factor/CREB (19), and C/EBP families (20). Although the sequence between −196 and −155 appeared to be sufficient for IL-1β induction, we decided to challenge this by testing the importance of a number of putative binding sites for transcription factors activated by MAP kinases in the region downstream of −200. By computer search, we found two putative C/EBP sites at −148 to −139 (C/EBP-1, 5′-CTTGCCCAAT3′) and −130 to −121; C/EBP-2, 5′-GGTGCAGAAA-3′) with 8 and 6 bases of 10 matching the C/EBP consensus sequence (5′-(A/G)TTGCG(T/C)AA(T/C)-3′) (21), respectively, and a possible AP-1 binding site at −56 to −50 (5′-TGAATCA-3′) with 6 of 7 bases matching the AP-1 consensus sequence (5′-TGA(G/C)T(C/A)A-3′) (22) (Fig. 6⇓). Specific mutations of the −1695 promoter construct were made in these sequences. Because a perfect consensus site for NF-Y (5′-CCAAT-3′) (23) was contained within the C/EBP-1 sequence, we decided to test two mutations of the C/EBP-1 site: one that should only affect binding of C/EBP factors (the C/EBP-1 mutation), and one that affects binding of both C/EBPs and NF-Y (the NF-Y mutation). For uninduced cells, the CAT activities of the NF-κB and C/EBP-2 mutations were similar to those of the wild-type promoter, whereas a 40–50% reduction in activity was seen for the NF-Y and C/EBP-1 mutations and a 75% decrease for the AP-1 mutation, indicating that the C/EBP-1 (or NF-Y) and AP-1 sites are important for basal expression (Fig. 6⇓A). However, with regard to the level of induction following IL-1β stimulation, only the NF-κB mutation showed no increase of CAT activity, whereas the activity of the wild-type promoter and the other point mutations increased by 30-fold or more (Fig. 6⇓B).

FIGURE 6.
FIGURE 6.

Point mutations of the NGAL promoter. A, Relative CAT activities of the −1695 NGAL promoter fragment with point mutations at different putative transcription factor recognition sites. Activities are shown relative to that of the −1695 wild-type promoter fragment (WT), which was given the value 1. B, Fold induction in CAT activity of the point mutants following stimulation with IL-1β. Transfections were made in quadruplicate and normalized to β-Gal, as in Fig. 5⇑.

A dominant-negative inhibitor of NF-κB represses transcriptional induction of the NGAL promoter by IL-1β

In order for NF-κB to be activated, the inhibitor IκB has to be phosphorylated at serines 32 and/or 36, which target it for proteolytic degradation (24). If these serines are mutated, IκB can no longer be degraded and will act as a superrepressor. To further substantiate the importance of NF-κB activation for IL-1β-induced NGALtranscription, we analyzed the effect of the IκB superrepressor (IκBM) on the transcriptional activation of the −1695 NGAL promoter fragment by IL-1β in A549 and NHEK cells (Fig. 7⇓). Induction of CAT activity was seen when the NGAL promoter was cotransfected with a nonexpressing vector or a vector expressing IκBα, whereas cotransfection with an IκBM-expressing vector almost abolished up-regulation by IL-1β. In accordance with previous data, no induction of the NGAL promoter was observed following administration of TNF-α or when the NF-κB binding site was corrupted. The IL-8 promoter was included as control and demonstrated strong transcriptional induction by both IL-1β and TNF-α when cotransfected with the empty vector or pCMV-IκB, and a more moderate induction when cotransfected with pCMV-IκBM. CAT activity above background could not be measured from the IL-8 promoter with a corrupted NF-κB binding site under any of the conditions used in this study.

FIGURE 7.
FIGURE 7.

Inhibition of NF-κB activation by a dominant-negative IκBα mutant. Top, A549 and NHEK cells were cotransfected with a, the −1695 wild-type or NF-κB-mutated NGAL promoter (0.8 μg); b, either an empty expression vector (pcDNA3.1), a vector expressing wild-type IκBα, or a vector expressing a dominant-negative IκBα mutant (0.05 μg); and c, a β-Gal-expressing vector for normalization of CAT values (0.2 μg). CAT activities were determined for uninduced cells and cells stimulated with IL-1β or TNF-α, as in Fig. 3⇑. Activities are shown relative to that of the uninduced −1695 wild-type promoter fragment, which was given the value 1. Bottom, As control, a parallel experiment was made with a wild-type or an NF-κB-mutated IL-8 promoter CAT reporter plasmid.

IL-1β specificity of the NGAL promoter does not reside in the NF-κB binding site

Having established that the NF-κB site is obligatory for NGAL induction by IL-1β, we wished to determine whether the IL-1β specificity also resided in this sequence. First, we examined nuclear extracts from A549 cells stimulated with IL-1β or TNF-α by EMSA using the NF-κB site of the NGAL promoter as probe. Similar EMSA complexes were formed following IL-1β and TNF-α stimulation, and in both cases the NF-κB complex was supershifted with Abs against p50 and Rel-A (Fig. 8⇓B). Although these EMSA data indicated no difference in NF-κB-binding specificity following IL-1β and TNF-α stimulation, we decided to examine this also by promoter studies. If the IL-1β specificity resided in the NF-κB-binding sequence, then this sequence by itself should be able to confer the same specificity to an unrelated basal promoter such as the SV40 promoter. This is not the case; a construct carrying four tandem repeats of the NGAL NF-κB site (including the 5 flanking bases on each side of the NF-κB site) did not induce any up-regulation of the SV40 promoter (NGAL (NF-κB × 4); Fig. 8⇓B), whereas a strong induction was seen for a construct containing four tandem repeats of the homologous 20-bp region of the IL-8 promoter (IL-8 (NF-κB × 4)) in response to both IL-1β and TNF-α. The same induction pattern was seen when constructs carrying four tandem repeats of the 10-bp NF-κB-binding sequence of the two promoters were tested (data not shown). Furthermore, if specificity resided in the NF-κB site, then an exchange of the NF-κB site in the NGAL promoter with another NF-κB site known to respond also to TNF-α stimulation should confer inducibility of the NGAL promoter to both IL-1β and TNF. Exchange of the 10-bp NF-κB site within the −183 deletion with the NF-κB consensus sequence (183(con)) resulted in an induction pattern similar to that observed for the wild-type −183 deletion (183(wt); Fig. 8⇓C). Exchange of the NGAL NF-κB site and the 3 flanking bases on each side with the homologous region of the IL-8 promoter (183(IL-8)), in contrast, almost abolished induction of the NGAL promoter construct to both IL-1β and TNF-α stimulation. This was also true when this mutation was tested in the context of the −1695 promoter (1695(IL-8)).

FIGURE 8.
FIGURE 8.

IL-1β specificity is not contained in the NF-κB site. A, EMSA with nuclear extracts from A549 cells stimulated with IL-1β or TNF-α using the NF-κB sequence of the NGAL promoter as probe. Extracts from unstimulated cells (unstim) were included to detect the background binding from uninduced cells (bands denoted by ∗). A specific competitor (identical with the probe (spec comp)) and a competitor with a mutated NF-κB site (unsp comp) were tested. Supershift was performed with Abs against the NF-κB subunits p50, p52, Rel-A, Rel-B, and c-Rel. Abs against the myeloid-specific transcription factor C/EBP-ε were included as a negative control. B, Fold induction in CAT activity of pCAT3-promoter (SV40) and the same construct with four tandem repeats of the NF-κB sequence from the NGAL (NGAL(NF-κB × 4)) or IL-8 (IL-8(NF-κB × 4)) promoters inserted upstream of the SV40 basal promoter. C, Fold induction in CAT activity of the 183 deletion (183(wt)) and mutants of the 183 deletion, in which the NF-κB site was changed to the consensus (183(con)) or IL-8 (183(IL-8)) sequences as well as the −1695 promoter and a mutant of the 1695 promoter containing the IL-8 NF-κB site (1695(IL-8)). Transfections were performed, as in Fig. 5⇑.

Ectopic expression of TLR4 in A549 cells enables LPS induction of NGAL promoter activity

Because IL-1 and TLRs use the same intracellular signal transduction pathways (25), it was surprising that LPS was unable to up-regulate NGAL synthesis in A549 cells despite the use of high levels of LPS (Fig. 3⇑). Northern blot analysis, however, demonstrated that A549 cells do not express TLR4 (Fig. 9⇓), which is required for the cells to respond to LPS (26). This allowed us to validate whether the specific induction of NGAL was due to an activation of the intracellular IL-1 (and TLR) signaling pathway. To test this, we cotransfected A549 cells with the −1695 NGAL promoter construct and plasmids expressing TLR4 and MD2 (a secreted cofactor required for optimal TLR4 activity (27)) and analyzed whether the NGAL promoter activity was increased following ligation of TLR4. As shown in Fig. 9⇓A, this conferred LPS responsiveness to the transfected cells and the ability of LPS to induce NGAL promoter activity. No induction was seen when the TLR4-expressing plasmid was omitted or when an NGAL promoter with a corrupted NF-κB site was used. To further demonstrate the use of a shared signaling pathway, A549 cells were cotransfected with a vector encoding a dominant-negative inhibitor of MyD88 (dnMyD88) along with plasmids expressing TLR4 and MD2 and the −1695 NGAL promoter construct. MyD88 is an adaptor protein in the Toll/IL-1R family signaling pathway that is involved in activation of NF-κB (28). This activation is inhibited by dnMyD88 (29). A strong inhibitory effect of dnMyD88 is demonstrated both for IL-1β and LPS induction of the NGAL promoter (Fig. 9⇓B), which substantiates a use of a shared Toll/IL-1R signaling pathway for up-regulation of NGAL.

FIGURE 9.
FIGURE 9.

Up-regulation of NGAL promoter activity by LPS ligation of TLR4. A, A549 cells were cotransfected with a, the −1695 wild-type (1695) or NF-κB-mutated NGAL promoter (1695M), and b, either the empty pcDNA3.1 expression vector (vector) or a vector expressing TLR4 (TLR4). An expression vector encoding the TLR4 cofactor MD2 was included in all transfections. CAT activities were determined for uninduced cells and cells stimulated with LPS from K. pneumoniae and E. coli J5 (100 ng/ml). Activities are shown relative to that of the uninduced −1695 wild-type promoter (value = 1). B, A549 cells were transfected with the −1695 NGAL promoter and vectors expressing TLR4, MD2, and either the empty pcDNA3.1 expression vector (vector) or a vector expressing dnMyD88. Activities are shown as fold induction relative to uninduced cells following stimulation with IL-1β (100 pg/ml) or LPS (100 ng/ml). C, Total RNA (10 μg) from PMNs and A549 hybridized to a 32P-labeled probe against TLR4.

Discussion

Recently, NGAL has been shown to be preferentially expressed in cells and tissues exposed to microorganisms (5, 7). In this study, we found large amounts of NGAL in bronchial goblet cells and low amounts in alveolar type II pneumocytes by immunohistochemical staining of uninflamed lungs. This cellular distribution matches the expression pattern of the NGAL gene in which the amount of NGAL transcript in trachea (which is morphologically similar to bronchus) was considerably higher than in lung (alveolar) tissue (Fig. 1⇑) (5). Besides the lower level of NGAL gene expression, type II pneumocytes may also stain weakly for NGAL due to a low storage capacity for secretory proteins. Type II pneumocytes are specialized in synthesizing antimicrobial proteins such as surfactant protein (SP)-A and SP-D. SP-A is constitutively secreted from type II pneumocytes before its accumulation in intracellular storage granules (31), and the data presented in this work demonstrate that the majority of newly synthesized NGAL in A549 cells likewise is constitutively secreted to the medium. Bronchial goblet cells, in contrast, have the ability to store large quantities of protein and therefore stain very strongly for NGAL. Positive immunostaining for NGAL in mucin-producing bronchial epithelium cells of normal lungs has also recently been described by Friedl et al. (7).

We also examined NGAL in inflamed lungs by immunohistochemistry. A much stronger staining was seen both in bronchi and alveoli, indicating that synthesis of NGAL is up-regulated in response to inflammatory stimuli. This was also the case for the type II pneumocyte-derived cell line A549 and for primary epithelial cells from bronchus and skin. Stimulation of A549 and NHEK cells with IL-1β caused a dramatic increase in NGAL and IL-8 synthesis (both at transcript and protein level), whereas a more modest increase was observed in NHBE cells. However, as the level of IL-8 induction in NHBE (and A549) cells was as previously reported under these growth conditions (15), this indicates that the NHBE cell line is less responsive to IL-1β than NHEK and A549 cells. The reason for the lower NGAL expression in NHBE than in bronchus is probably that this cell culture does not form goblet cells.

An increase of NGAL has previously been demonstrated in inflammatory bowel diseases (6) and in adenocarcinomas of tissues exposed to high numbers of microorganisms (7). Recently, it has been described that NGAL has a strong specificity for bacterial ferric siderophores and that NGAL acts as a bacteriostatic agent by sequestering iron, which is vital for bacterial growth (3). A similar iron-depleting antimicrobial effect is ascribed to lactoferrin (4), which besides limiting bacterial growth also can inhibit the formation of bacterial biofilms (32). The difference between lactoferrin and NGAL is that lactoferrin binds iron directly, whereas NGAL mops up the iron-loaded bacterial transporter, the siderophore (which has a higher affinity for iron than lactoferrin, and thus may compromise the effect of lactoferrin). Expression of the murine NGAL homologue, 24p3, is also up-regulated in response to inflammatory stimuli (33) and in the uterus at the time of parturition, where it is believed to be part of a local inflammatory response associated with birth (34). This, together with the data presented in this work, suggests an important role for NGAL as part of the innate immune response to bacterial challenges. Although neutrophils, which store NGAL in their specific granules, provide the organism with a mobile source of NGAL, the epithelial cells are important for the local defense against infection, as illustrated by the strong induction of NGAL by these cells in response to inflammation.

IL-1β and TNF-α are usually regarded as prototypic proinflammatory mediators, which both induce NF-κB, a transcription factor that is central for gene induction by these cytokines (35, 36). It was therefore surprising that only IL-1β was able to induce NGAL synthesis in the three epithelial cell lines examined in this study. In this study, we demonstrate that IL-1β-induced up-regulation of NGAL is strongly dependent upon NF-κB activation, as both site-directed mutation of the NF-κB binding site and cotransfection with an IκB superrepressor severely reduced or completely abrogated the otherwise 20- to 40-fold increase in promoter activity observed in A549 and NHEK cells. Administration of TNF-α, in contrast, did not increase NGAL synthesis or promoter activity, even though NF-κB was clearly activated by this cytokine (Figs. 7⇑ and 8⇑).

As our promoter studies demonstrated that TNF-α did not induce a repressor of the NGAL promoter, we believe that IL-1β induces or modulates another transcription factor besides NF-κB, which is required for transcriptional induction of the NGAL promoter. By deletion and substitution mutations, we identified a 40-bp region of the NGAL promoter between −196 and −155, which was required for full IL-1β response. Although IL-1β can activate the MAP kinase cascade, no effect on the IL-1β specificity was observed when mutating putative binding sites for transcription factors activated by MAP kinases. Our data also indicate that the IL-1β specificity does not reside in the NF-κB-binding sequence, as similar EMSA patterns were seen after stimulation with IL-1β and TNF-α, and as exchange of the NF-κB element of the NGAL promoter with the consensus sequence or that from the IL-8 promoter did not confer TNF-α inducibility to the NGAL promoter. Introduction of the IL-8 NF-κB element with its 3 flanking bases (in which 4 of the 6 bases surrounding the NF-κB element were changed) did in fact almost abolish inducibility of the NGAL promoter to IL-1β. This observation, combined with the finding that four tandem repeats of the NGAL NF-κB site were unable to cause up-regulation of SV40 promoter activity following IL-1β stimulation (Fig. 8⇑), suggests that the sequence surrounding the NGAL NF-κB site is required for both NF-κB function and IL-1β specificity. Such a direct effect of another transcription factor on the activity of NF-κB has been reported, for e.g., the IL-8 and IFN-β promoters in which NK-κB-repressing factor is involved in transcriptional silencing in the absence of cytokine stimulation and activation following stimulation (16, 37). An analogous factor that activates NF-κB on the NGAL promoter only following IL-1β stimulation may thus be envisioned.

Although it is acknowledged that the cellular responses to IL-1β and TNF-α are not identical (35, 36), only a few genes are known to be specifically activated by IL-1β and not by TNF-α. Interestingly, the few reports describing such an induction pattern involve expression of effectors or modulators of innate immunity. This is the case for human β-defensin 2 (hBD-2) in epithelial colon cells (38) and keratinocytes (39), inducible NO synthase in rat insulinoma cells (40), IL-6 in human enterocytes (41), and IκB-ζ in mouse macrophages (42). It has furthermore been demonstrated that NF-κB activation is required for IL-1 induction of hBD-2 (38). Although NGAL, hBD-2, and inducible NO synthase have very different antimicrobial activities (38, 43), the expression of these important effector molecules in innate immunity seems to be regulated similarly. Most genes activated by NF-κB are modulators of the immune response, such as chemokines, cytokines, and adhesion molecules, and as such have to be induced by many proinflammatory cytokines (44). In contrast, the primary function of NGAL and other antimicrobial proteins is to combat microorganisms, and therefore the activation of these genes should be tuned into a response toward a microbial challenge. This raises the possibility that their IL-1β-specific induction may actually reflect the existence of a parallel induction mechanism, which is stimulated by bacteria or other microorganisms, and share some intracellular signal transduction molecules with the IL-1 pathway. One such possibility is the family of TLRs, which are very homologous to the IL-1R (45). At present, 10 different human TLRs have been identified (46). The ligands for 5 of the TLRs are known, and 4 of these are of bacterial origin: lipoproteins (TLR2), LPS (TLR4), flagellin (TLR5), and bacterial DNA (TLR9) (25, 47, 48, 49). For TLR2 and TLR4, use of the IL-1 pathway has been demonstrated (25, 47), and for TLR5 and TLR9, ligation of the receptor has been shown to cause an activation of NF-κB (48, 49). As A549 cells do not express TLR4, no induction of NGAL was seen in response to LPS. This, however, does not exclude the possibility that other bacterial products may activate NGAL synthesis through another TLR and/or that the TLR4 receptor may be used for NGAL induction in other cells. The observation that NGAL was up-regulated in a NF-κB-dependent manner following LPS administration to A549 cells cotransfected with a TLR4-expressing vector demonstrates that NGAL synthesis can be induced by ligation of TLRs and may offer an alternative explanation for the evolution of an apparent IL-1-specific induction of NGAL.

Acknowledgments

The expert technical assistance of Inge Kobbernagel and Hanne Kidmose is greatly appreciated. We thank G. Sandra Tjabringa and Pieter S. Hiemstra (Leiden University Medical Center, Leiden, The Netherlands) for help with the primary bronchial cell experiments, and, along with Kim Theilgaard-Mönch, for critical review of the manuscript.

Footnotes

  • ↵1 This work was supported by grants from Danish Medical Research Council, Carlsberg Foundation, Alfred Benzon Foundation, Novo Nordic Foundation, Copenhagen University Hospital (H:S), Danish Medical Association Research Fund, and Danish Foundation for Cancer Research.

  • ↵2 Address correspondence and reprint requests to Dr. Jack B. Cowland, The Granulocyte Research Laboratory, Department of Hematology Rigshospitalet, 93.2.2, 9 Blegdamsvej, DK-2100 Copenhagen, Denmark. E-mail address: jcowland{at}rh.dk

  • ↵3 Abbreviations used in this paper: NGAL, neutrophil gelatinase-associated lipocalin; β-Gal, β-galactosidase; BM, bone marrow; dnMyD88, dominant-negative mutant of MyD88; hBD-2, human β-defensin 2; MAP, mitogen-activated protein; NHBE, normal human bronchial eipthelia; NHEK, normal human epidermal keratinocyte; PMN, neutrophil; SP, surfactant protein; TLR, Toll-like receptor.

  • Received August 2, 2002.
  • Accepted October 1, 2003.

References

  1. ↵
    Kjeldsen, L., D. F. Bainton, H. Sengeløv, N. Borregaard. 1994. Identification of neutrophil gelatinase-associated lipocalin as a novel matrix protein of specific granules in human neutrophils. Blood 83:799.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    Flower, D. R.. 1996. The lipocalin protein family: structure and function. Biochem. J. 318:1.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Goetz, D. H., M. A. Holmes, N. Borregaard, M. E. Bluhm, K. N. Raymond, R. K. Strong. 2002. The neutrophil lipocalin NGAL is a bacteriostatic agent that interferes with siderophore-mediated iron acquisition. Mol. Cell 10:1033.
    OpenUrlCrossRefPubMed
  4. ↵
    Finkelstein, R. A., C. V. Sciortino, M. A. McIntosh. 1983. Role of iron in microbe-host interactions. Rev. Infect. Dis. 5:(Suppl. 4):S759.
    OpenUrlCrossRefPubMed
  5. ↵
    Cowland, J. B., N. Borregaard. 1997. Molecular characterization and pattern of tissue expression of the gene for neutrophil gelatinase-associated lipocalin from humans. Genomics 45:17.
    OpenUrlCrossRefPubMed
  6. ↵
    Nielsen, B. S., N. Borregaard, J. R. Bundgaard, S. Timshel, M. Sehested, L. Kjeldsen. 1996. Induction of NGAL synthesis in epithelial cells of human colorectal neoplasia and inflammatory bowel diseases. Gut 38:414.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Friedl, A., S. P. Stoesz, P. Buckley, M. N. Gould. 1999. Neutrophil gelatinase-associated lipocalin in normal and neoplastic human tissues: cell type-specific pattern of expression. Histochem. J. 31:433.
    OpenUrlCrossRefPubMed
  8. ↵
    Cowland, J. B., N. Borregaard. 1999. The individual regulation of granule protein mRNA levels during neutrophil maturation explains the heterogeneity of neutrophil granules. J. Leukocyte Biol. 66:989.
    OpenUrlAbstract
  9. ↵
    Dignam, J. D., R. M. Lebovitz, R. G. Roeder. 1983. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11:1475.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Kjeldsen, L., C. Koch, K. Arnljots, N. Borregaard. 1996. Characterization of two ELISAs for NGAL, a newly described lipocalin in human neutrophils. J. Immunol. Methods 198:155.
    OpenUrlCrossRefPubMed
  11. ↵
    Lieber, M., B. Smith, A. Szakal, W. Nelson Rees, G. Todaro. 1976. A continuous tumor-cell line from a human lung carcinoma with properties of type II alveolar epithelial cells. Int. J. Cancer 17:62.
    OpenUrlCrossRefPubMed
  12. ↵
    Le Cabec, V., J. B. Cowland, J. Calafat, N. Borregaard. 1996. Targeting of proteins to granule subsets is determined by timing and not by sorting: the specific granule protein NGAL is localized to azurophil granules when expressed in HL-60 cells. Proc. Natl. Acad. Sci. USA 93:6454.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Burkitt, H. G., B. Young, J. W. Heath. 1993. Wheater′s Functional Histology Churchill Livingstone, Hong Kong.
  14. ↵
    Hellmann, G. M., W. R. Fields, D. J. Doolittle. 2001. Gene expression profiling of cultured human bronchial epithelial and lung carcinoma cells. Toxicol. Sci. 61:154.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Coulter, K. R., M. D. Wewers, M. P. Lowe, D. L. Knoell. 1999. Extracellular regulation of interleukin (IL)-1β through lung epithelial cells and defective IL-1 type II receptor expression. Am. J. Respir. Cell Mol. Biol. 20:964.
    OpenUrlCrossRefPubMed
  16. ↵
    Nourbakhsh, M., S. Kalble, A. Dorrie, H. Hauser, K. Resch, M. Kracht. 2001. The NF-κB repressing factor is involved in basal repression and interleukin (IL)-1-induced activation of IL-8 transcription by binding to a conserved NF-κB-flanking sequence element. J. Biol. Chem. 276:4501.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Muller, J. M., H. W. Ziegler Heitbrock, P. A. Baeuerle. 1993. Nuclear factor κB, a mediator of lipopolysaccharide effects. Immunobiology 187:233.
    OpenUrlCrossRefPubMed
  18. ↵
    Zhou, G., S. C. Lee, Z. Yao, T. H. Tan. 1999. Hematopoietic progenitor kinase 1 is a component of transforming growth factor β-induced c-Jun N-terminal kinase signaling cascade. J. Biol. Chem. 274:13133.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Sano, Y., J. Harada, S. Tashiro, M. R. Gotoh, T. Maekawa, S. Ishii. 1999. ATF-2 is a common nuclear target of Smad and TAK1 pathways in transforming growth factor-β signaling. J. Biol. Chem. 274:8949.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Lacorte, J. M., E. Ktistaki, A. Beigneux, V. I. Zannis, J. Chambaz, I. Talianidis. 1997. Activation of CAAT enhancer-binding protein δ (C/EBPδ) by interleukin-1 negatively influences apolipoprotein C-III expression. J. Biol. Chem. 272:23578.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Osada, S., H. Yamamoto, T. Nishihara, M. Imagawa. 1996. DNA binding specificity of the CCAAT/enhancer-binding protein transcription factor family. J. Biol. Chem. 271:3891.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Faisst, S., S. Meyer. 1992. Compilation of vertebrate-encoded transcription factors. Nucleic Acids Res. 20:3.
    OpenUrlFREE Full Text
  23. ↵
    Mantovani, R.. 1998. A survey of 178 NF-Y binding CCAAT boxes. Nucleic Acids Res. 26:1135.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Brown, K., S. Gerstberger, L. Carlson, G. Franzoso, U. Siebenlist. 1995. Control of IκB-α proteolysis by site-specific, signal-induced phosphorylation. Science 267:1485.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Kirschning, C. J., H. Wesche, A. T. Merrill, M. Rothe. 1998. Human Toll-like receptor 2 confers responsiveness to bacterial lipopolysaccharide. J. Exp. Med. 188:2091.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    Chow, J. C., D. W. Young, D. T. Golenbock, W. J. Christ, F. Gusovsky. 1999. Toll-like receptor-4 mediates lipopolysaccharide-induced signal transduction. J. Biol. Chem. 274:16.
    OpenUrl
  27. ↵
    Shimazu, R., S. Akashi, H. Ogata, Y. Nagai, K. Fukudome, K. Miyake, M. Kimoto. 1999. MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4. J. Exp. Med. 189:1777.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Medzhitov, R., P. Preston-Hurlburt, E. Kopp, A. Stadlen, C. Chen, S. Ghosh, C. A. Janeway, Jr. 1998. MyD88 is an adaptor protein in the hToll/IL-1 receptor family signaling pathways. Mol. Cell 2:253.
    OpenUrlCrossRefPubMed
  29. ↵
    Muzio, M., J. Ni, P. Feng, V. M. Dixit. 1997. IRAK (Pelle) family member IRAK-2 and MyD88 as proximal mediators of IL-1 signaling. Science 278:1612.
    OpenUrlAbstract/FREE Full Text
  30. Holmskov, U.. 1999. Lung surfactant proteins (SP-A and SP-D) in non-adaptive host responses to infection. J. Leukocyte Biol. 66:747.
    OpenUrlAbstract
  31. ↵
    Osanai, K., R. J. Mason, D. R. Voelker. 1998. Trafficking of newly synthesized surfactant protein A in isolated rat alveolar type II cells. Am. J. Respir. Cell Mol. Biol. 19:929.
    OpenUrlCrossRefPubMed
  32. ↵
    Singh, P. K., M. R. Parsek, E. P. Greenberg, M. J. Welsh. 2002. A component of innate immunity prevents bacterial biofilm development. Nature 417:552.
    OpenUrlCrossRefPubMed
  33. ↵
    Liu, Q., M. Nilsen Hamilton. 1995. Identification of a new acute phase protein. J. Biol. Chem. 270:22565.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Liu, Q., J. Ryon, M. Nilsen-Hamilton. 1997. Uterocalin: a mouse acute phase protein expressed in the uterus around birth. Mol. Reprod. Dev. 46:507.
    OpenUrlCrossRefPubMed
  35. ↵
    Martin, M. U., W. Falk. 1997. The interleukin-1 receptor complex and interleukin-1 signal transduction. Eur. Cytokine Network 8:5.
    OpenUrlPubMed
  36. ↵
    Darnay, B. G., B. B. Aggarwal. 1997. Early events in TNF signaling: a story of associations and dissociations. J. Leukocyte Biol. 61:559.
    OpenUrlAbstract
  37. ↵
    Nourbakhsh, M., H. Hauser. 1999. Constitutive silencing of IFN-β promoter is mediated by NRF (NF-κB-repressing factor), a nuclear inhibitor of NF-κB. EMBO J. 18:6415.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    O’Neil, D. A., E. M. Porter, D. Elewaut, G. M. Anderson, L. Eckmann, T. Ganz, M. F. Kagnoff. 1999. Expression and regulation of the human β-defensins hBD-1 and hBD-2 in intestinal epithelium. J. Immunol. 163:6718.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    Liu, A. Y., D. Destoumieux, A. V. Wong, C. H. Park, E. V. Valore, L. Liu, T. Ganz. 2002. Human β-defensin-2 production in keratinocytes is regulated by interleukin-1, bacteria, and the state of differentiation. J. Invest. Dermatol. 118:275.
    OpenUrlCrossRefPubMed
  40. ↵
    Hill, J. R., J. A. Corbett, A. C. Baldwin, M. L. McDaniel. 1996. Nitric oxide production by the rat insulinoma cell line, RINm5F, is specific for IL-1: a spectrophotometric IL-1 bioassay. Anal. Biochem. 236:14.
    OpenUrlCrossRefPubMed
  41. ↵
    Parikh, A. A., A. L. Salzman, J. E. Fischer, C. Szabo, P. O. Hasselgren. 1997. Interleukin-1β and interferon-γ regulate interleukin-6 production in cultured human intestinal epithelial cells. Shock 8:249.
    OpenUrlCrossRefPubMed
  42. ↵
    Yamazaki, S., T. Muta, K. Takeshige. 2001. A novel IκB protein, IκB-ζ, induced by proinflammatory stimuli, negatively regulates nuclear factor-κB in the nuclei. J. Biol. Chem. 276:27657.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    Vazquez-Torres, A., F. C. Fang. 2001. Oxygen-dependent anti-Salmonella activity of macrophages. Trends Microbiol. 9:29.
    OpenUrlCrossRefPubMed
  44. ↵
    Ghosh, S., M. Karin. 2002. Missing pieces in the NF-κB puzzle. Cell 109:S81.
    OpenUrlCrossRefPubMed
  45. ↵
    Muzio, M., N. Polentarutti, D. Bosisio, M. K. Prahladan, A. Mantovani. 2000. Toll-like receptors: a growing family of immune receptors that are differentially expressed and regulated by different leukocytes. J. Leukocyte Biol. 67:450.
    OpenUrlAbstract
  46. ↵
    Brown, P.. 2001. Cinderella goes to the ball. Nature 410:1018.
    OpenUrlCrossRefPubMed
  47. ↵
    Zhang, F. X., C. J. Kirschning, R. Mancinelli, X. P. Xu, Y. Jin, E. Faure, A. Mantovani, M. Rothe, M. Muzio, M. Arditi. 1999. Bacterial lipopolysaccharide activates nuclear factor-κB through interleukin-1 signaling mediators in cultured human dermal endothelial cells and mononuclear phagocytes. J. Biol. Chem. 274:7611.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, K. Hoshino, H. Wagner, K. Takeda, S. Akira. 2000. A Toll-like receptor recognizes bacterial DNA. Nature 408:740.
    OpenUrlCrossRefPubMed
  49. ↵
    Hayashi, F., K. D. Smith, A. Ozinsky, T. R. Hawn, E. C. Yi, D. R. Goodlett, J. K. Eng, S. Akira, D. M. Underhill, A. Aderem. 2001. The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 410:1099.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section