| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-Specific Up-Regulation of Neutrophil Gelatinase-Associated Lipocalin Is Controlled by I
B-
1
* Department of Hematology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark; and
Department of Molecular and Cellular Biochemistry, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
, but not by TNF-
, despite an induction of NF-
B binding to the NGAL promoter by both cytokines. In this study, we present evidence that the IL-1 specificity is caused by a requirement of the NGAL promoter for the NF-B-binding cofactor IB-
for transcriptional activation. Up-regulation of NGAL expression in A549 cells following IL-1 stimulation was dependent on de novo protein synthesis and was greatly diminished by a small interfering against IB- mRNA. Cotransfection of A549 cells with a plasmid expressing IB- made TNF- capable of inducing NGAL transcription, indicating that IB- induction is the only factor discriminating between IL-1 and TNF- in their ability to induce NGAL expression. Coexpression of the cofactor Bcl-3, which is closely related to IB-, did not enable TNF- to induce NGAL transcription. A functional NF-B site of the NGAL promoter was required for IB- to exert its effect. The human defensin 2 gene also required IB- for its IL-1-specific induction in A549 cells. Our findings indicate that a common regulatory mechanism has evolved to control expression of a subset of antimicrobial proteins expressed in epithelial cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
B is a major regulator of this process. It exists as a dimeric complex composed of one or two of the five members of the NF-B family: RelA (p65), RelB, c-Rel, NF-B1(p50), and NF-B2(p52) (1). In unstimulated cells, NF-B is sequestered in the cytosol through association with a member of the IB protein family: usually IB- or IB- (2). After activation of the cell with an appropriate stimulus, IB- and IB- are degraded, followed by a translocation of the NF-B complex to the nucleus, where it can bind to the promoters of its target genes and induce transcription (1, 2, 3).
A number of different stimuli may lead to an inflammatory response and NF-B activation (e.g., infection, exposure to UV light, hypoxia (1, 3, 4)). NF-B is furthermore known to activate a large variety of genes that encode proteins with quite diverse biological functions such as antiapoptotic proteins, extracellular adhesion molecules, cytokines and chemokines, and antimicrobial proteins (1, 3). Because so many different stimuli converge in the activation of NF-B, it is believed that some additional mechanisms must have evolved for the cell to respond appropriately to the stimuli that evoke an inflammatory response. A number of reports indicate that this fine-tuning of gene expression is achieved in part by binding of NF-B-specific cofactors to the NF-B complex associated with the target gene promoter and in part by coinduction of other transcription factors (e.g., through the MAPK pathway) that cooperate with the NF-B factor on the target promoter (2, 5, 6, 7, 8).
Recently, an IL-1-inducible cofactor of NF-B, named IB- (9) (also termed MAIL (10) and INAP (11)), was identified. IB- is highly homologous to Bcl-3 (9, 11), another cofactor of NF-B belonging to the IB family. In contrast to transcriptional repressors such as IB-, Bcl-3 acts as a positive activator of transcription (1). The function of IB- may be similar to that of Bcl-3, although examples of transcriptional repression by IB- have also been reported (9, 12, 13, 14). IB- is rapidly induced in response to stimulation with IL-1 and LPS, but not with TNF-, with peak transcript levels observed 1–2 h after stimulation (9, 10, 12). When analyzing peritoneal macrophages from mice with a targeted disruption of IB-, it was found that a number of LPS-responsive genes could no longer be up-regulated (12). One of the affected genes encoded 24p3 (or lipocalin 2), which is the mouse homologue of neutrophil-gelatinase-associated lipocalin (NGAL)3 (12).
NGAL is a 25-kDa glycoprotein, first identified as a matrix protein of specific granules of human neutrophils (15). Expression of NGAL has been observed in epithelial cells, where it is strongly induced during inflammation (16, 17, 18, 19, 20). NGAL belongs to the lipocalin superfamily, whose members share a barrel-shaped tertiary structure with a hydrophobic pocket that can bind lipophilic molecules (21). NGAL’s ligand is bacterial ferric siderophores, which are used by bacteria for uptake of the essential nutrient iron (22, 23). Targeted disruption of the gene encoding the murine homologue of NGAL (24p3) demonstrated that the bacteriostatic effect of NGAL against a clinical strain of Escherichia coli measured in vitro (22) is also important in vivo (23). Expression of NGAL in epithelial cells is dependent on NF-B, but the NGAL gene is only up-regulated by IL-1 (and LPS) and not TNF-, even though both induce activation and promoter binding of NF-B (12, 19, 20, 23). In this study, we demonstrate that this IL-1 specificity is caused by a requirement for the cofactor IB- (which itself is induced specifically by IL-1 and LPS) for activation of the NGAL promoter through the NF-B signaling pathway.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
A549 (ATCC CCL-185) cells were obtained from the American Type Culture Collection and grown in HAM F12 (Invitrogen Life Technologies) supplemented with 10% FCS (Invitrogen Life Technologies), 100 U/ml penicillin, and 100 µg/ml streptomycin (100/100 P/S) (Invitrogen Life Technologies) at 37°C in a humid atmosphere with 5% CO2. For transfection and/or induction with IL-1 and TNF- (both Sigma-Aldrich) or incubation with cycloheximide (Sigma-Aldrich), A549 cells were grown to 70–80% confluence and shifted to HAM F12 medium with 0.5% FCS and 100/100 P/S.
RNA isolation and Northern blot
Total RNA was prepared with TRIzol (Invitrogen Life Technologies), according to the manufacturer’s recommendations, and the concentration was determined by spectrophotometric measurement. For Northern blotting, 5 µg of RNA was run on a 1% agarose gel, transferred to a Hybond-N membrane (Amersham Biosciences), and hybridized, as described (19). The membranes were washed, as described (19), and developed by a Fuji BAS2500 phosphor imager. The size of the mRNAs was determined by reference to 18S and 28S ribosomal RNA. The membranes were stripped by boiling in 0.1% SDS before rehybridization. The NGAL, IL-8, and -actin cDNA probes have been described earlier (19). Probes for IB- and Bcl-3 were generated by PCR amplification of the entire coding region for the proteins with the following primer pairs: 1) 5'-GAAAGGACTTTGATTTGTGGC-3' and 5'-ATCTAGTCCAATGACTGAAGC-3', and 2) 5'-TAAAGCTTGCCACCATGGACGAGGGGCCCGTG-3' and 5'-GAGGATCCTCAGCTGCCTCCTGGAGC-3', respectively, using: 1) cDNA from IL-1-stimulated A549 cells, and 2) IMAGE clone 5806689 (Bcl-3) (MRC Geneservice) as templates. The PCR products were cloned in pCR2.1-TOPO (Invitrogen Life Technologies) and verified by sequencing. The probes were radiolabeled with [-32P]dCTP using the Random Primers DNA Labeling System (Invitrogen Life Technologies). For quantitative assessments, the intensities of the IB-, NGAL, and IL-8 signals were normalized to the hybridization intensity from a probe against -actin.
Quantitative TaqMan PCR
Quantitative PCR analysis was performed on the Applied Biosystems 7500 Real-Time PCR system (Applied Biosystems) with the commercial gene expression assay Assay-on-demand (Applied Biosystems). The assays used were: DEFB4 (gene encoding human defensin 2 (hBD2)), Hs00175474_m1; IL-8, Hs00174103_m1; and -actin, Hs99999903_m1.
Plasmids and small interfering RNA (siRNA)
A human IB--specific 21-nt siRNA Silencer-predesigned siRNA (identification number 33380: sense, 5'-GGAAAAAGGUAAAUACAGCTT-3' and antisense, 5'-GCUGUAUUUACCUUUUUCCTT-3' (Ambion)) was used to knock down IB- mRNA. Silencer Negative Control 1 siRNA (Ambion) was included as control to analyze for any nonspecific effects of the siRNA treatment. The promoter constructs pNGP1695CAT, pNGP1695(NF-B)CAT, pNGP183CAT, and pCAT3basic (Promega) are described elsewhere (19). Expression vectors for IB- and Bcl-3 were generated by excision of the coding region from pCR2.1-TOPO(IB-) and pCR2.1-TOPO(Bcl-3) described above by restriction with EcoRI and XhoI. The DNA fragments were cloned in pcDNA3.1+ (Invitrogen Life Technologies) digested with the same enzymes.
Cell transfection and reporter enzyme assay
For experiments with siRNA, transfection with LipofectAMINE 2000 (Invitrogen Life Technologies) was performed according to the manufacturer’s instructions. Transfections involving only plasmid DNA were performed by use of Effectene (Qiagen). For promoter studies, 0.8 µg of CAT plasmid promoter construct was cotransfected with 0.2 µg of pcDNA3--Gal (encoding -galactosidase (-Gal)) to compensate for differences in transfection efficiency. Expression of the reporter enzymes was quantitated by CAT and -Gal ELISA (Roche Diagnostic Systems), according to the manufacturer’s recommendations. For each sample, the CAT activity was normalized to -Gal activity.
Quantitation of NGAL, IL-1, and IL-8 in medium
NGAL was quantitated by ELISA, as described previously (24). IL-1 and IL-8 were quantitated with the IL-1 and IL-8 optEIA ELISA kits (BD Pharmingen), 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). For immunodetection, the polyvinylidene fluoride membranes (Millipore) were blocked for 1 h with 5% skimmed milk in PBS after the transfer of proteins from 14% polyacrylamide gels, and incubated overnight with rabbit anti-IB- Abs (12) (1:1000) and anti--tubulin (Sigma-Aldrich; T-9026) (1:4500). The following day, the membranes were incubated for 2 h with peroxidase-conjugated porcine anti-rabbit Abs (DakoCytomation; P-0217) and visualized ECL (Amersham Biosciences).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
B- is specifically up-regulated by IL-1 in A549 cells
We have reported earlier that NGAL is up-regulated by IL-1, but not by TNF-, in human epithelial cells in an NF-B-dependent manner, despite identical ability of both cytokines to induce binding of NF-B to the NGAL promoter (19). It was recently reported that a nuclear binding partner of NF-B (named IB-) is selectively induced by IL-1 in a number of tissues (9). Furthermore, it was reported that IB- could act both as an activator and a repressor of NF-B-dependent transcription (9, 12, 13, 14). To investigate whether IB- could be responsible for the IL-1-specific up-regulation of NGAL, we analyzed the transcript profile of IB- in the lung epithelial cell line A549 that we have previously used as model cell line (19). Strong induction of IB- was observed in IL-1-stimulated cells with peak transcript level at 1.5 h (Fig. 1). No IB- mRNA was observed in uninduced cells, and only weak induction was seen following administration of TNF-. In contrast, the mRNA level of the closely related cofactor Bcl-3 (9, 11), which has also been associated with transcriptional modulation of NF-B (1, 5), was unaltered following stimulation with each of these cytokines (Fig. 1).
|
induction of NGAL synthesis
These data indicated that IB- could function as a positive activator of NF-B on the NGAL promoter. To determine whether a newly synthesized cofactor was required for activation of the NGAL promoter, we stimulated A549 cells with IL-1 in the presence of cycloheximide, which blocks new protein synthesis. Inclusion of cycloheximide abrogated the strong induction of NGAL mRNA normally seen 4–6 h after administration of IL-1 (Figs. 1 and 2). In contrast, induction of the transcripts for IL-8 and IB- was unaffected by the presence of cycloheximide; in fact, an increase in mRNA levels was observed for the latter two targets, indicating a stabilization of these two transcripts under these circumstances. Taken together, this demonstrated a need for de novo protein synthesis for IL-1 induction of the NGAL promoter. It is further seen in Fig. 3C that IL-1 induces an accumulation of IB- in A549 cells.
|
|
B- expression results in decreased NGAL production
To further validate that IB- could be the cofactor required for NGAL induction by IL-1, we transfected A549 cells with a specific siRNA against IB-. Northern blot analysis of a sample collected 2 h after IL-1 stimulation demonstrated a 3-fold reduction of IB- mRNA in the presence of IB--siRNA, whereas addition of scrambled RNA (scRNA), which was used as control for the siRNA, did not down-regulate IB- mRNA levels (Fig. 3). The IL-8 mRNA also peaked 2–3 h following IL-1 stimulation (Figs. 1 and 2), but the latter transcript was unaffected by the presence of both scRNA and siRNA, demonstrating the specificity of the IB--siRNA (Fig. 3). Following IL-1 stimulation, NGAL mRNA levels accumulate for 24–48 h (Fig. 1). We therefore decided to examine the effect of IB--siRNA on NGAL mRNA levels after 24 h. The level of NGAL mRNA was greatly diminished in IB--siRNA-treated cells compared with untreated or scRNA-treated cells. Again, no difference was seen for IL-8 mRNA levels between scRNA- and siRNA-treated cells. This indicates that the IB- cofactor remains active for a long period when first synthesized (11).
To determine the effect of the siRNA on the proteins encoded by the above-mentioned mRNAs, we made another experiment in which endogenous IB- protein levels were measured 2 h after IL-1 stimulation by Western blot of whole cell lysates. As was the case for the cognate transcript, the level of the IB- cofactor was greatly diminished in siRNA-treated cells (Fig. 3C). Although the level of IL-8 mRNA peaked after 2–3 h, a continuous accumulation of both IL-8 and NGAL protein in the cell culture medium occurred during the entire 24-h IL-1 stimulation period (19). For this reason, the content of NGAL and IL-8 in culture medium was measured at 24 h. The amount of NGAL in the medium of IB--siRNA-treated cells followed the same pattern as observed for the NGAL mRNA (Fig. 3, B and D). For IL-8, the amount of protein synthesized was marginally affected by the presence of either scRNA or siRNA (Fig. 3D).
Coexpression of IB- enables NGAL promoter activation by TNF- stimulation
We have demonstrated previously that both TNF- and IL-1 stimulation can induce binding of NF-B to the NGAL promoter, but up-regulation of NGAL promoter activity was only observed after stimulation with IL-1 (19). To validate that the coinduction of IB- by IL-1 can explain the IL-1 specificity of the NGAL promoter, we cotransfected A549 cells with an NGAL promoter construct and an expression plasmid encoding IB- and subsequently stimulated with TNF- to induce activation of NF-B. A slight induction was observed when A549 cells transfected with control vector were stimulated with TNF-, as observed earlier (19). In contrast, a 6- to 8-fold increase in NGAL promoter activity was seen in TNF--stimulated cells cotransfected with the IB--expressing vector (Fig. 4A). Increasing the amount of cotransfected IB--expressing vector caused a concomitant increase in the NGAL promoter activity, which was not the case when the amount of control vector was increased in a similar manner (Fig. 4B). Cotransfection with a Bcl-3-expressing vector did not have any effect on NGAL promoter activity following TNF- stimulation (Fig. 4A). Taken together, these data demonstrate that expression of IB- is sufficient to transform TNF- into a potent inducer of NGAL promoter activity. This indicates that it is the ability of IL-1 to induce both IB- expression and NF-B activation that allow it to up-regulate NGAL promoter activity in contrast to TNF-, which does not induce IB- expression.
|
B binding site is required for IB--mediated induction of the NGAL promoter
The NF-B binding site at position –180 to –171 has been shown to be required for IL-1 induction of the NGAL promoter (19). To determine whether the effect of IB- could be ascribed to an interaction between IB- and the p65:p50 NF-B heterodimer binding to this B element of the NGAL promoter, we examined the consequence of IB--siRNA on cells transfected with a 1695-bp NGAL promoter construct with either the wild-type B-binding sequence (pNGP1695) or a mutant of this promoter (pNGP1695M) in which the (–180/–171) B element is unable to bind the p65:p50 NF-B factor (19). As expected, siRNA against IB- caused 8- to 9-fold reduction of wild-type NGAL promoter activity following IL-1 induction compared with the control with scRNA (Fig. 5A). A small reduction of promoter activity by IB--siRNA compared with scRNA was also observed in uninduced cells for both the 1695-bp promoter and the 5' deletion mutant pNGP183, which terminate 3 bases upstream of the –180/–171 NF-B site, indicating slight activation of the NGAL promoter by NF-B:IB- under these circumstances. In contrast, the activities of the pNGP1695M promoter in IB--siRNA- and scRNA-treated cells were comparable: both in uninduced and IL-1-stimulated cells. Although the activity of pNGP1695M only increased 2- to 3-fold when stimulated by IL-1 (Fig. 5B), this activation did not appear to be dependent on IB-. Taken together, these data indicate that IB- only interacts with the NF-B complex bound to the –180/–171 site, as neither basal nor the residual IL-1-induced expression of a promoter without a functional NF-B element was affected. As discussed later, such interaction cannot be visualized by EMSA because the oligonucleotides used for this do not encompass both the NF-B binding site and the sequence information required for IB- binding.
|
stimulation is essential for induction of NGAL synthesis, but can be partially compensated by TNF- stimulation at later stages
NGAL accumulated in the medium of IL-1-stimulated A549 cells during the entire 48-h induction period, whereas no increase was observed for TNF--stimulated cells compared with uninduced cells, not even at the end of the experiment. TNF- is known to induce a rapid de novo production of a number of cytokines, as exemplified by IL-8 (Figs. 1 and 6). If TNF- likewise induced IL-1 production in A549 cells, a delayed, but measurable induction of NGAL synthesis by this newly synthesized IL-1 would be expected. To determine whether TNF- induced IL-1 synthesis, we measured IL-1 levels at different time points following stimulation. As demonstrated in Fig. 6A, no increase in IL-1 levels was observed in medium from TNF-- or uninduced cells. Furthermore, a high level of IL-1 was still present in the medium of IL-1-stimulated cells at the end of the experiment, probably explaining how NGAL and IL-8 synthesis could be sustained for 48 h.
|
, it is possible that this stimulus is required only during formation of an active transcription complex composed of NF-B and IB-. Because expression of IB- peaks after 1.5–2 h (Figs. 1 and 2), it is feasible that further stimulation with IL-1 is required only to maintain NF-B in an active state. If this was the case, stimulation with IL-1 for 3 h followed by stimulation with TNF- should result in an NGAL synthesis similar to that obtained with IL-1 alone. As demonstrated in Fig. 6C, increased NGAL production was observed in cells prestimulated with IL-1 for 3 h before TNF- stimulation, compared with cell stimulated with TNF- alone. The amount of NGAL synthesized was 32 and 23% of that produced by cells stimulated with IL-1 for 48 h and 3 + 45 h, respectively. However, a higher NGAL expression was also observed in cells prestimulated 3 h with IL-1 and then changed to medium with or without IL-1 compared with cells growing under the same conditions without prestimulation. Likewise, a significant increase in IL-8 synthesis was observed under all three growth conditions for cells prestimulated with IL-1. Taken together, these data indicate that with regard to NGAL synthesis, TNF- can compensate to some degree for the signal generated by IL-1 from 3 h and beyond.
IL-1-specific up-regulation of hBD2 is also dependent upon IB- expression
In an earlier publication, we argued that the IL-1-specific up-regulation of the antimicrobial protein NGAL might in fact reflect an adaptation to the TLR pathway, which uses the same intracellular signaling pathway as the IL-1R (19). This was verified by demonstrating that induction of NGAL promoter activity could be achieved through TLR-4 by the bacterial ligand LPS (19). If the IL-1 specificity does reflect such an adaptation, one could expect that genes for other antimicrobial proteins would be regulated in the same manner as the NGAL gene, i.e., requires the induction of both IB- and NF-B for transcriptional activation. Recently, we demonstrated that both NGAL and hBD2 were up-regulated by IL-1, but not by TNF- or IL-6 in human keratinocytes (19, 20). To determine whether the same specificity applied for A549 cells, we first tested whether hBD2 was also specifically up-regulated by IL-1 in these cells and found that this was indeed the case (Fig. 7A). hDB2 transcripts accumulate during the entire period of stimulation (20), and hBD2 mRNA levels were therefore measured 24 h poststimulation in A549 cells, similar to NGAL transcripts. Next, we analyzed the effect of treating the cells with IB--siRNA. At 24 h of IL-1 stimulation, the level of hBD2 transcript in IB--siRNA-treated cells was <20% of that measured in cells not receiving external RNA- and scRNA-treated cells. As control, we measured the amount of IL-8 mRNA 2 and 24 h after stimulation with IL-1 and found, as before, no effect of either siRNA or scRNA on IL-8 transcript levels (Fig. 7B). This indicates that a common regulatory mechanism exists for the two antimicrobial proteins NGAL and hBD2.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
B pathway is more complex than just the mere binding of an NF-B factor to the B site(s) of NF-B-responsive genes. Involvement of NF-B-binding cofactors and NF-B-interacting transcription factors such as C/EBP, c-jun, or AP-1 induced by the MAPK pathway (6, 7) has added an additional layer of complexity to the regulatory mechanism used by cells to tailor their response to the biological condition causing the inflammatory response. A subset of NF-B target genes encodes antimicrobial proteins such as NGAL and hBD2, and these should only be induced during a microbial challenge. In this study, we present evidence that the specific induction of these two innate immune defense proteins is critically dependent on induction of the NF-B cofactor IB-.
We show in this study that the IL-1 selectivity in inducing NGAL and hBD2 expression in epithelia can be explained by the synthesis of IB- elicited specifically by IL-1 stimulation. Activation of NGAL requires de novo protein synthesis (Fig. 2) and is strongly diminished when the IB- transcript is destabilized by siRNA (Fig. 3). An effect of the IB--siRNA on the NGAL promoter was, however, only observed provided a functional B element at position –180/–171 was present (Fig. 5). If NF-B was unable to bind this B site, the IB--siRNA had no further effect on the NGAL promoter, indicating that IB- did not bind the NGAL promoter by itself nor interacted with an NF-B factor binding to another B element of NGAL promoter. The observation that forced expression of IB- could rescue the inability of TNF- to induce NGAL promoter activity strongly indicates that the feature that discriminates between nonactivation of the NGAL promoter by TNF- and activation by IL-1 is the ability of the latter cytokine to induce synthesis of IB- in addition to activating NF-B (Fig. 4).
The explanation for the IL-1 specificity of IB- expression is due to stabilization of the IB- mRNA when stimulating with IL-1 (and other ligands using the IL-1R signaling pathway such as LPS activation through TLR-4 (12)). Induction of IB- gene expression depends on activation of NF-B, but efficient synthesis of IB- furthermore requires stabilization of the transcript (14). TNF- stimulation does not result in stabilization of the IB- transcript. Stimulation by TNF-, therefore, does not result in increase of the IB- protein level despite its ability to induce IB- gene expression through NF-B activation (13). The signaling pathways leading to activation of NF-B and stabilization of the IB- mRNA thus appear to be separate. This notion is strengthened by the observation that stimulation with IL-17 causes a stabilization of the IB- transcript without activating NF-B (14). In accordance with the results shown in this study (Fig. 4), costimulation with TNF- and IL-17 is able to cause a 7- to 8-fold induction of the murine homologue of NGAL (lcn-2/24p3) in the preosteoblast cell line MC3T3-E1 compared with the combined stimulatory effect of TNF- and IL-17 by themselves (25). Analysis of the 24p3 promoter in MC3T3-E1 cells showed a 1.7- and 2.8-fold induction, respectively, by IL-17 and TNF- alone, and a 22.6-fold induction of promoter activity by IL-17 plus TNF- (25). In contrast, the combined effect of TNF- stimulation and constitutive coexpression of IB- demonstrated in this study resulted only in a 7- to 8-fold increase of promoter activity in A549 cells (Fig. 4). This, however, can be explained by the lack of an IB--mRNA-stabilizing signal under these experimental conditions compared with stimulation with IL-1 or IL-17. The cis-element causing destabilization of the IB- mRNA is contained within the open reading frame of the transcript (14), and thus is present in the expression construct used by us.
In addition to the in vitro data, a requirement of IB- for in vivo up-regulation of 24p3 (the murine homologue of NGAL) through the IL-1R/TLR-4 signaling pathway has recently been demonstrated by the lack of 24p3 expression in peritoneal macrophages from IB-–/– mice after LPS stimulation (12). This contrasts with the strong up-regulation of the 24p3 transcript seen in wild-type mice (12). Taken together, these data strongly indicate that the IL-1 specificity of the NGAL promoter is governed by IB-.
Which feature of the NGAL promoter determines the requirement for the cofactor IB-? Comparisons of the B sites of genes that need IB- for induction through the IL-1R/TLR pathway show no obvious common pattern (12). Analysis for subunit specificity of the individual proteins of the dimeric NF-B factor likewise failed to identify any common characteristics in the examined B sites that could explain the binding of different homo- and heterodimers of NF-B (26). In fact, the sequence of the B site of NGAL (GGGAATGTCC) is also found in the promoter of the IB- gene, but in this case mutation of the B site had no effect on the promoter activity of the IB- gene (27, 28). A recent publication demonstrated that a single base difference in the B site of the MCP-1 gene was sufficient to alter the cofactor specificity of the NF-B molecule bound to this site from IFN regulatory factor-3 (IRF-3) to Bcl-3 (5). The authors speculated that cofactor specificity was determined by a specific structural conformation of the NF-B dimer, imposed upon the protein complex by the DNA sequence it bound to. A similar mechanism could determine the specificity of the p65:p50 dimer binding to the NGAL promoter for the cofactor IB-, as altering the sequence of the B element of the NGAL promoter to that of the IL-8 promoter is sufficient to abolished IL-1 induction of NGAL (19). In contrast, the sequence of the NGAL B element (and the five flanking bases on each side) was not sufficient to impose IL-1 specificity to a heterologous SV40 promoter (19). Increasing the number of flanking bases on each side of the NGAL B element to 20 likewise had no effect (data not shown), indicating, as suggested previously, that a second transcription factor binding 100–150 bases downstream of the NGAL (–180/–171) B site is also required for IL-1 responsiveness of the NGAL gene (19). This may explain why identical band patterns were observed in an EMSA with the NGAL B element following IL-1 and TNF- stimulation (19) (data not shown) because the sequence information for binding of IB- was probably not contained within the oligos used for the experiment.
A requirement for interaction with two nonadjacent transcription factors for cofactor binding to NF-B has been described for the MCP-1 and IP-10 promoters. In this case, two B sites were needed for NF-B to bind the cofactors IRF-3 and Bcl-3 (5). A second potential B site exists at –91/–82 of the NGAL promoter, but as mutation of this sequence influences neither promoter activity nor specificity (J.B. Cowland and N. Borregaard, manuscript in preparation), we do not believe that this DNA element regulates IB- binding.
The IB- mRNA level peaks 1.5–2 h after IL-1 stimulation and then rapidly declines (Fig. 1). This indicates that IB- synthesis is required only during assembly of a transcription complex on the NGAL promoter, and that continued IL-1 stimuli is needed to keep the NF-B complex associated with the NGAL promoter. This notion is supported by the observation that TNF- stimulation to some degree compensates for the signal generated by IL-1, as cells stimulated for 3 h with IL-1 and then with TNF- for 45 h produced 25–30% of the NGAL measured for cells stimulated with IL-1 for all 48 h (Fig. 6C). However, the level of NGAL synthesized by cells pretreated with IL-1 (3 h) and then changed to medium with TNF- (45 h) was only three times that produced by cells that were changed to medium without cytokines. This indicates that shortly after formation of the transcription complex, an inactivation phase follows, which is impeded partly by TNF- stimulation and efficiently by IL-1 stimulation.
Following activation of NF-B, a number of genes are induced, including that encoding IB- (3, 29). This causes a de novo production of IB- that can interact with NF-B complexes bound to DNA and mediate their transport back to the cytosol (3). This negative feedback loop will in many cases terminate NF-B-induced transcription. NF-B may, however, also interact with the related factor IB-. Stimulation with, for example, IL-1 also results in degradation of IB-, although with a slower kinetic than for IB- (29, 30). De novo synthesis of IB- is also induced, but in this case an unphosphorylated form is generated that can interact with DNA-bound NF-B and act as a chaperone that hinders binding of, and inactivation by, IB- (30, 31). This may cause persistent activation of the promoter by NF-B rather than rapid inactivation by IB-. Further stabilization of the DNA-bound NF-B complex can be obtained by acetylation of the p65 subunit that also blocks for binding of IB- (32). In cases in which IB- is involved (as for the NGAL promoter), a further level of complexity is introduced because IB- can act both as an activator and repressor of NF-B-mediated transcription (13). It has been demonstrated that IB- binds to the p50 subunit in a manner analogous to that of Bcl-3, but its mechanism of action is unknown (9).
Based on this information, the following model can explain the findings of this study: continued expression of NGAL during IL-1 stimulation is obtained by binding of IB- to the NF-B:IB- complex or by an exchange of IB- with IB-. The stability of this complex is challenged by the newly synthesized IB- that will try to displace IB- and/or IB- from NF-B. Continued stimulation with IL-1 retains a high level of IB-, and disassembly of the active transcription complex occurs slowly. Stimulation with IL-1 for 3 h allows the assembly of the IB-:NF-B:DNA complex. When the cells subsequently are exposed to fresh medium with TNF-, this cytokine will cause degradation of the newly synthesized IB- and thus delay disassembly of the NF-B:DNA complex. Conversely, a faster down-regulation of NGAL synthesis occurs when the cells receive fresh medium without cytokines, as IB- remains stable under these circumstances.
As alluded to before, there is a need for specificity and selectivity of the responses elicited by NF-B activation. This may be the reason that the genes encoding NGAL and hBD2 require IB- for activation, as this cofactor is induced by a number of bacterial TLR ligands (12). Other antimicrobial proteins such as human cationic antimicrobial protein of 18 kDa, hBD1, hBD3, and secretory leukocyte protease inhibitor are regulated in a different manner that does not in all cases involve the NF-B pathway (20), and therefore might reflect an adaptation of the cell to antimicrobial challenges that are recognized by other mechanisms than through the TLR system. Many of the NF-B target genes, in contrast, do not encode antimicrobial proteins, but rather antiapoptotic proteins, extracellular adhesion molecules, cytokines, or chemokines (1, 3). In the case of an aseptic inflammatory stimulus (e.g., UV radiation and hypoxia) (1, 4), it is plausible that IB- is not induced and that NF-B instead interacts with cofactors such as Bcl-3, IRF-3, IB-, or IB-
(2, 5) and/or other transcription factors (6, 7, 8). This would probably lead to the activation of a different subset of NF-B-responsive genes than those observed during infection. Further knowledge about the stimuli that regulate the association between NF-B and its different cofactors may give us the answer as to how the NF-B pathway adjusts to the many different biological functions it governs.
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by grants from the Danish Medical Research Council, the Lundbeck Foundation, the Novo Nordic Foundation, Copenhagen University Hospital (Hovedstadens Sygehusfaellesskab), the Danish Medical Association Research Fund, and the Danish Foundation for Cancer Research. 
2 Address correspondence and reprint requests to Dr. Jack B. Cowland, Department of Hematology L-9322, Granulocyte Research Laboratory, Rigshospitalet, 9 Blegdamsvej, DK-2100 Copenhagen, Denmark. E-mail address: jcowland{at}rh.dk
3 Abbreviations used in this paper: NGAL, neutrophil-gelatinase-associated lipocalin; -Gal, -galactosidase; hBD, human defensin; IRF-3, IFN regulatory factor-3; scRNA, scrambled RNA; siRNA, small interfering RNA.
Received for publication August 30, 2005. Accepted for publication February 14, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
B. Genes Dev. 18: 2195-2224. B-NF-B signaling module: temporal control and selective gene activation. Science 298: 1241-1245. B puzzle. Cell 109: S81-S96. [Medline]-B activation in blood-brain barrier endothelium under hypoxic/reoxygenation stress. J. Neurochem. 92: 203-214. B site can determine cofactor specificity for NF-B dimers. Cell 118: 453-464. [Medline]B activity and cell viability. Mol. Biol. Cell 13: 2933-2945. B. Trends Endocrinol. Metab. 16: 46-52. [Medline]B protein, IB-, induced by proinflammatory stimuli, negatively regulates nuclear factor-B in the nuclei. J. Biol. Chem. 276: 27657-27662. B protein that potentiates LPS-induced IL-6 production. FEBS Lett. 485: 53-56. [Medline]B-. Nature 430: 218-222. [Medline]B-mediated transcription by IB-, an inducible nuclear protein. J. Biol. Chem. 280: 7444-7451. B regulator, IB-, via Toll/interleukin-1 receptor is mediated by mRNA stabilization. J. Biol. Chem. 280: 1678-1687. , but not by TNF-. J. Immunol. 171: 6630-6639. -induced genes in bone cells. J. Leukocyte Biol. 77: 388-399. B/Rel transcription factors defines functional specificities. EMBO J
