Rapid Up-regulation of IκBβ and Abrogation of NF-κB Activity in Peritoneal Macrophages Stimulated with Lipopolysaccharide*

Lipopolysaccharide (LPS) administration to mice elicited the activation of nuclear factor κB (NF-κB) in several tissues including liver and macrophages. Maximal activation was observed 1 h after treatment but declined at 3 and 6 h. The levels of IκBα and IκBβ were analyzed during this period in an attempt to correlate NF-κB activity with IκB resynthesis. Degradation of IκBα was very rapid and was followed by recovery 1 h after LPS administration. IκBβ degradation, which has been associated with persistent NF-κB activation, was complete at 1 h. However, a rapid recovery of IκBβ in these tissues was observed at 3 h in parallel with the abrogation of NF-κB activity. Immunolocalization of newly synthesized IκBβ by confocal microscopy revealed its preferential accumulation in the cytosol. Analysis of IκBβ by Western blot using high resolution polyacrylamide gel electrophoresis showed the presence of two bands in cytosolic extracts of LPS-treated macrophages at 3 h, but only one band with the same mobility as the control was detected at 6 h. Moreover, treatment of extracts of resynthesized IκBβ with alkaline phosphatase resulted in the accumulation of the protein of slightly higher electrophoretic mobility, indicating the prevalence of a rapid phosphorylation of the newly synthesized IκBβ. At the mRNA level, up-regulation of IκBβ was observed in macrophages stimulated for 1 h with LPS. When the effect of pro-inflammatory cytokines was investigated, tumor necrosis factor α, but not interleukin-1 or interferon-γ, promoted an important degradation of IκBβ followed by an increase in the mRNA at 1 h. These results suggest the existence of LPS- and tumor necrosis factor α- specific pathways involved in a rapid IκBβ degradation and resynthesis and might explain the transient period of activation of NF-κB in these tissues upon stimulation with these factors. This rapid control of NF-κB function may contribute to the attenuation of the inflammatory response of these cells.

Lipopolysaccharide (LPS) administration to mice elicited the activation of nuclear factor B (NF-B) in several tissues including liver and macrophages. Maximal activation was observed 1 h after treatment but declined at 3 and 6 h. The levels of IB␣ and IB␤ were analyzed during this period in an attempt to correlate NF-B activity with IB resynthesis. Degradation of IB␣ was very rapid and was followed by recovery 1 h after LPS administration. IB␤ degradation, which has been associated with persistent NF-B activation, was complete at 1 h. However, a rapid recovery of IB␤ in these tissues was observed at 3 h in parallel with the abrogation of NF-B activity. Immunolocalization of newly synthesized IB␤ by confocal microscopy revealed its preferential accumulation in the cytosol. Analysis of IB␤ by Western blot using high resolution polyacrylamide gel electrophoresis showed the presence of two bands in cytosolic extracts of LPS-treated macrophages at 3 h, but only one band with the same mobility as the control was detected at 6 h. Moreover, treatment of extracts of resynthesized IB␤ with alkaline phosphatase resulted in the accumulation of the protein of slightly higher electrophoretic mobility, indicating the prevalence of a rapid phosphorylation of the newly synthesized IB␤. At the mRNA level, up-regulation of IB␤ was observed in macrophages stimulated for 1 h with LPS. When the effect of pro-inflammatory cytokines was investigated, tumor necrosis factor ␣, but not interleukin-1 or interferon-␥, promoted an important degradation of IB␤ followed by an increase in the mRNA at 1 h. These results suggest the existence of LPS-and tumor necrosis factor ␣specific pathways involved in a rapid IB␤ degradation and resynthesis and might explain the transient period of activation of NF-B in these tissues upon stimulation with these factors. This rapid control of NF-B function may contribute to the attenuation of the inflammatory response of these cells.
Nuclear factor B (NF-B) 1 participates in the regulation of the expression of multiple immediate early genes involved in the immune, acute phase, and inflammatory responses (1). NF-B is a heterodimer of proteins of the Rel family of transcription factors. In mammalian cells, they include p65 (Rel A), Rel B, the proto-oncogene c-Rel, p50/p105 (NF-B1), and p52/ p100 (NF-B2) (1,2). NF-B proteins are constitutively present in the cell, but they are retained in the cytoplasm associated with inhibitory proteins known as IB (3,4). Activated NF⅐B complexes, typically composed of p50 and p65, are translocated to the nucleus in response to mitogens, cytokines (IL-1␤, IL-2, and TNF-␣), and bacterial lipopolysaccharide and lipopeptides (1,(5)(6)(7)(8). Activation of NF-B appears to require phosphorylation and degradation of the IB proteins, thereby allowing the rapid translocation of NF-B from the cytoplasm to the nucleus (7, 9 -11).
Several IB proteins have been characterized including IB␣, IB␤, IB␥, and the candidate oncogene Bcl-3 (3,4,12). All these proteins share a characteristic ankyrin repeat motif, which is required for the interaction with Rel proteins, and a C-terminal PEST sequence presumably involved in protein targeting and degradation (13). Different kinases have been involved in IB␣ phosphorylation, but the observation that antioxidants and alkylating agents inhibit the phosphorylation and subsequent degradation points to a common unidentified IB␣ kinase (14,15). Phosphorylation of specific residues seems to be the signal for ubiquitin conjugation followed by degradation via the 26 S proteasome (16,17). Degradation of IB␣ is rapidly followed by induction of IB␣ mRNA through a mechanism dependent on the binding of NF-B to the B sequences present in the promoter of the IB␣ gene (18,19). This newly synthesized IB␣ resets the NF-B switch in the cytoplasm and possibly in the nucleus (20), although in some cases IB␣ resynthesis is not sufficient to suppress nuclear NF-B activity (12,21). IB␤ has been cloned recently (12) and, together with IB␣, is the main regulator of NF-B activity through the interaction with the same Rel proteins (2). It has been proposed that IB␤ degradation causes a sustained activation of NF-B due to the large lag period of IB␤ resynthesis (12,21,22). To determine whether these observations are specific of some cell types or represent a general mechanism of NF-B activation, we investigated IB␣ and IB␤ turnover in an experimental model of murine septic shock and in cultured peritoneal macrophages triggered with different stimuli. Our results show a rapid IB␤ degradation followed by a fast recovery, both in liver and in macrophages. This recovery of IB␤ contrasts with the results obtained in LPS-stimulated lymphoid cells, where absence of IB␤ was observed for large periods of time (12). In our experimental model, an increase of IB␤ mRNA was detected as early as 1 h after stimulation and paralleled the resynthesis of IB␤ levels and the fall in NF-B activity.

MATERIALS AND METHODS
Chemicals-Cytokines were from Boehringer Mannheim. Polymerase chain reaction reagents were from Perkin-Elmer. Lipopolysaccharide (LPS) from Salmonella typhimurium and other reagents were from Sigma. Cell culture reagents were from BioWhittaker, Inc.(Walkersville, MD). Antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA).
Animal Treatment-Septic shock was induced in mice after intraperitoneal injection of 0.5 ml of a solution containing LPS (1 mg/kg of body weight) in saline. Animals were anesthetized with ether and immediately sacrificed. Tissues were processed immediately after extraction.
Preparation of Macrophages-Elicited peritoneal macrophages were prepared from male mice 4 days after intraperitoneal inoculation of 1 ml of 10% thioglycollate broth. Cells were seeded at 1.5 ϫ 10 6 in 6-cm plates and cultured with RPMI 1640 medium supplemented with 10% heat-inactivated fetal calf serum and antibiotics at 37°C in an atmosphere of humidified 5% CO 2 . After incubation for 1 h, nonadherent cells were removed, and remanent cells were cultured and stimulated for different periods of time in phenol red-free RPMI 1640 medium lacking serum.
RNA Extraction and Northern Blot Analysis-Total RNA was extracted from 1.5 ϫ 10 6 cells or from 50 mg of tissues following the guanidinium thiocyanate method (23). After electrophoresis in a 0.9% agarose gel containing 2% formaldehyde, the RNA was transferred to a Nytran membrane (NY 13-N; Schleicher & Schuell, Inc., FRG) with 10 ϫ SSC (1.5 M NaCl, 0.3 M sodium citrate, pH 7.4). The membranes were prehybridized, and the levels of different mRNAs were determined with specific labeled probes. A 422-base pair IB␤ fragment was obtained by reverse transcription-polymerase chain reaction using 1 g of mouse testis RNA as template and oligonucleotides, based on the published sequence (12). The following primers were used for amplification: 5ЈGGACACAGCCCTGCACTTGG3Ј (forward, nucleotides 247-266) and 5ЈGTAGCCTCCAGTCTTCATCA3Ј (reverse, nucleotides 668 -648). The polymerase chain reaction fragment was cloned in a pGEM-T vector (Promega, Madison, WI) and sequenced (Sequenase, Amersham Life Science, Inc.), exhibiting the expected published sequence (12). For Northern blot analysis of IB␣, a BamHI/HindIII fragment (1.4 kilobases) of the IB␣ cDNA was used. The plasmid containing the cDNA for IB␣ was a gift of Dr. Moscat (24). Probes were labeled with [␣-32 P]dCTP using a commercial kit (Boehringer Mannheim). The membranes were washed with 0.1 ϫ SSC and 0.1% SDS at 42°C for 30 min followed by exposure to an x-ray film (Kodak X-OMAT). Quantitation of the films was performed by laser densitometry (Molecular Dynamics, Sunnyvale, CA) using the hybridization with a ribosomal 18 S probe as internal standard. Various exposition times of the micrograph films were used to ensure that bands were not saturated. Results are expressed in arbitrary units as the ratio of IB/ribosomal 18 S RNA level.
Preparation of Cytosolic and Nuclear Extracts-A modified procedure based on the method of Schreiber (25) et al. was used. Cells (1.5 ϫ 10 6 ) were washed with PBS and collected by centrifugation. Cell pellets were homogenized with 100 l of buffer A (10 mM Hepes, pH 7.9, 1 mM EDTA, 1 mM EGTA, 10 mM KCl, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 2 g/ml aprotinin, 10 g/ml leupeptin, 2 g/ml TPCK, 5 mM NaF, 1 mM NaV0 4 , 10 mM Na 2 MO 4 ). After 10 min at 4°C, Nonidet P-40 was added to reach a 0.5% concentration. The tubes were gently vortexed for 15 s, and nuclei were collected by centrifugation at 8.000 ϫ g for 15 min. Tissues (100 mg) were homogenized in 8 volumes of buffer A containing 0.5 M sucrose. The supernatants were stored at Ϫ80°C (cytosolic extracts), and the pellets were resuspended in 50 l of Buffer A supplemented with 20% glycerol, 0.4 M KCl and gently shaken for 30 min at 4°C. Nuclear protein extracts were obtained by centrifugation at 13,000 ϫ g for 15 min, and aliquots of the supernatant were stored at Ϫ80°C. Protein content was assayed using the Bio-Rad protein reagent. All steps of cell fractionation were carried out at 4°C. To dephosphorylate proteins, extracts were treated for 1 h at 30°C with 1 unit of agarose-immobilized alkaline phosphatase/g of protein. Appropriate controls of heat-inactivated alkaline phosphatase were used to ensure the specificity of the reaction.
Electrophoretic Mobility Shift Assays (EMSAs)-Oligonucleotides were synthesized in a Pharmacia oligonucleotide synthesizer. The oligonucleotide sequence corresponding to the consensus NF-B binding site (nucleotides Ϫ978 to Ϫ952) of the murine iNOS promoter was 5ЈTGCTAGGGGGATTTTCCCTCTCTCTGT3Ј (26). Oligonucleotides were annealed with their complementary sequence by incubation for 5 min at 85°C in 10 mM Tris-HCl, pH 8.0, 50 mM NaCl, 10 mM MgCl 2 , 1 mM dithiothreitol. Aliquots of 50 ng of these annealed oligonucleotides were end-labeled with Klenow enzyme fragment in the presence of 50 Ci of [␣-32 P]dCTP and the other unlabeled dNTPs in a final volume of 50 l. 5 ϫ 10 4 dpm of the DNA probe were used for each binding assay of nuclear extracts as follows: 3 g of protein were incubated for 15 min at 4°C with the DNA and 2 g of poly(dI⅐dC), 5% glycerol, 1 mM EDTA, 100 mM KCl, 5 mM MgCl 2 , 1 mM dithiothreitol, 10 mM Tris-HCl, pH 7.8, in a final volume of 20 l. The DNA-protein complexes were separated on native 6% polyacrylamide gels in 0.5% Tris borate-EDTA buffer (27). Supershift assays were carried out after incubation of the nuclear extract with the antibody (0.5 g) for 1 h at 4°C followed by EMSA. Anti-p50 (human) and anti-c-Rel (human) were a generous gift of Dr. N. R. Rice (27); anti-p65 (murine), anti-IB␣ (murine), and anti-IB␤ (murine) polyclonal Abs were from Santa Cruz.
Western Blot Analysis-After determining the protein content of cytosolic extracts, samples were boiled in 250 mM Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, and 2% ␤-mercaptoethanol. Proteins (15 g) were size-separated in minigels (7 cm) of 10% SDS-polyacrylamide gel electrophoresis. When a higher resolution of the proteins was required, the proteins were separated in a 15-cm gel, allowing the 36.5-kDa band of the prestained molecular weight markers to reach the border of the gel. Gels were blotted onto a polyvinylidene fluorescein membrane (Amersham) and processed as recommended by the supplier of the antibodies for IB␣ and IB␤ (Santa Cruz). Proteins recognized by the antibodies were revealed following the ECL technique (Amersham). Autoradiographs were quantified by laser densitometry (Molecular Dynamics), and several time expositions were analyzed to ensure the linearity of the band intensities. At the end of the experiment, the membranes were treated with Ponceau S reagent to confirm the protein charge after blotting.
Immunolocalization of IB by Confocal Microscopy-Macrophages were stimulated for the indicated period of time, and the cell layers were washed with ice-cold PBS and fixed for 2 min with methanol (Ϫ20°C). Fixed cells were blocked for 1 h at reverse transcription with blocking solution (3% bovine serum albumin in PBS) and then treated for 1 h with a 1:50 dilution of anti-IB␣ or anti-IB␤ Abs. After two washes with PBS, the cells were incubated for 30 min at reverse transcription with fluorescein-labeled goat-anti-rabbit IgG Ab (Cy3, Amersham) diluted 1:150 in PBS. The dishes were washed three times with PBS and analyzed in a MRC-100 confocal microscope (Bio-Rad). The fluorescence of the cell was digitalized using the Cosmos software (Bio-Rad). Results were expressed as the total fluorescence intensity associated to the cytosol or nuclear compartments, respectively.
Statistical Analysis-Results are expressed as the mean Ϯ S.E. of the indicated number of experiments. Statistical significance was estimated using Student's t test for unpaired observations. A P value of Ͻ0.05 was considered significant.

IB␤ Is Up-regulated after LPS Treatment in Vivo and ex
Vivo-It has been suggested that the persistent NF-B activation observed in B lymphocytes in response to LPS is due to a selective degradation and delayed resynthesis of IB␤ (12). Indeed, administration of LPS to mice induces NF-B activation in several tissues, including resident peritoneal macrophages and liver (5,26,27). Using this animal model of septic shock, we analyzed the process of NF-B activation together with IB degradation and synthesis. As Fig. 1A shows, a peak of NF-B binding was observed 1 h after LPS administration both in peritoneal macrophages and in liver. This response decreased at 3 and 6 h and was completely absent in samples obtained at 18 h. Because NF-B activation is largely dependent on IB degradation, we analyzed the levels of IB␣ and IB␤ at several sampling times. In liver, IB␣ was lost at 20 min and rapidly recovered at 3 h (Fig. 1B). Interestingly, IB␤ degradation and resynthesis was delayed with respect to IB␣, recovering basal levels at 6 h. However, the recovery of IB␣ and IB␤ in peritoneal macrophages from these animals was more rapid (Fig. 1B), suggesting a differential kinetic control of this process among cells.
To assess whether this extremely rapid IB␤ up-regulation was a peculiarity of the in vivo experimental model of septic shock, cultured peritoneal macrophages were stimulated with LPS, and NF-B activity and IB levels were determined. As Fig. 2A shows, NF-B activation exhibited a maximum at 1 h followed by a rapid decrease of binding. Characterization of the proteins retained in EMSA by supershift analysis revealed the presence of p50/p50 (Fig. 2B, lower band) and p50/p65 dimers (Fig. 2B, upper band), and a negligible content in p52 or c-Rel (Fig. 2B). The amount of IB␤ present in the cytosol was determined by Western blot, and the protein levels were barely detectable at 1 h and increased at 3 h (28% of the control level; Fig. 2A). Newly synthesized IB␤ seems to require phosphorylation to interact with and to inhibit Rel proteins (22). Since the kinetics of this phosphorylation could depend on the cell type analyzed, we investigated the phosphorylation state of the IB␤ after resynthesis. Using high resolution SDS-polyacrylamide gels, two bands of IB␤-immunodetected protein were observed at 3 h after LPS treatment of the cells, the upper band corresponding to a phosphorylated IB␤ species in view of the increased mobility of the protein after treatment with immobilized alkaline phosphatase (Fig. 2C). Interestingly, the main IB␤ species detected in control cells corresponded to the phosphorylated state of the protein as deduced by the shift after phosphatase treatment. Quantitation of the relative intensity of the two bands observed in samples from cells treated for 3 h with LPS showed a 78 and 22% distribution for the upper and lower bands, respectively. However, in samples analyzed after 6 h of LPS treatment, the upper band systematically represented Ͼ90% of the distribution, indicating the prevalence of the phosphorylated form in the cytosol. The presence of nonphosphorylated IB␤ species has been related to the formation of IB␤-protected NF-B active complexes both in the cytosol and in the nucleus (22). Therefore, to investigate whether these protected ternary complexes could be present in the nuclei of LPS-activated cells, a supershift EMSA was performed with anti-IB␤ Ab after incubation of the nuclear extracts from cells treated for 1 or 3 h with LPS. However, the electrophoretic profile of NF-B binding was not affected. Moreover, the minimal amount of IB␤ detected in the nuclear extracts corresponded to the phosphorylated form. These results indicate that the level of IB␤ present in the nucleus was very low, always corresponding to the active form of the protein (not shown, see next section).
To determine the contribution of the proteasome to IB␤ degradation, experiments were done in the presence of several proteasome inhibitors, and the amount of IB␤ was determined  by Western blot. As Table I shows, these inhibitors prevented IB␤ degradation at the time that prevented NF-B activation (not shown).

IB␤ Accumulates in the Cytosol of LPS-treated Macrophages-
The degradation and resynthesis of IB␣ and IB␤ in cells treated with LPS was also investigated in situ using fluorescence confocal microscopy. As Fig. 3 shows, IB␣ and IB␤ were undetectable at 30 min. Newly synthesized IB␣ was detected at 1 h after LPS treatment, and at 4 h the protein was present both in the cytosol and in the nucleus. A quantitative analysis of the subcellular distribution of the fluorescence is shown in Fig. 4. When the immunofluorescence asso-ciated with IB␤ was analyzed (Fig. 3), a complete absence of staining was observed in cells treated for 1 h with LPS, followed by a resynthesis of the protein that accumulates in the cytosol. These time courses of resynthesis were in agreement with the immunodetection of the protein by Western blot analysis. Whereas IB␣ was observed both in the cytosol and nucleus, IB␤ was detected in the cytosol in agreement with the rapid phosphorylation of this protein, which blocks the nuclear localization signal domain of the NF-B⅐Rel complexes. The quantitative analysis of these data is reported in Fig. 4.
IB␤ mRNA Levels Are Increased by LPS-The steady-state levels of IB␣ and IB␤ mRNA were determined in order to better assess the recovery of the corresponding proteins. As  Fig. 5, A and B, show and in agreement with previous reports, IB␣ mRNA was rapidly up-regulated after treatment of cultured macrophages with LPS (12,27). Interestingly, IB␤ mRNA also increased in response to LPS, although the changes were lower and delayed with respect to the IB␣ levels (peak values were obtained at 1 and 4 h for IB␣ and IB␤, respectively). To confirm the specificity of the mRNA detected by the probes, total RNA from selected murine tissues was examined by Northern blot, and cross-hybridization with both probes was accomplished. As Fig. 5C shows, IB␣ and IB␤ mRNAs were expressed at different levels in several tissues in agreement with a previous report (12). IB␣ was very abundant in spleen, whereas IB␤ exhibited a high expression in testis. However, both IB mRNA increased in the liver of animals after LPS treatment for 4 h (Fig. 5C), supporting the results shown in Fig.  1 at the protein level.

TNF-␣ but Not IL-1␤ or IFN-␥ Induces IB␤ Degradation and IB␤ mRNA Up-regulation in Macrophages-NF-B
activation requires phosphorylation, targeting, and degradation of the IB components of the heteromeric complexes. Therefore, the measurement of the IB␣ and IB␤ mRNA and protein levels provides useful criteria for the assessment of their rate of resynthesis and the turn-off of the NF-B activation process. To investigate the effect of pro-inflammatory cytokines on the levels of IB␤, macrophages were stimulated with TNF-␣, IFN-␥, IL-1␤, or a combination of them, and the amount of IB␤ was quantified by Western blot. As Fig. 6 shows, only cells treated with TNF-␣ exhibited a decrease in IB␤ levels (46% of the control value) 1 h after stimulation, whereas recovery was observed at 4 h. Analysis of IB␤ on a high resolution gel showed the presence of 16% nonphosphorylated protein at 4 h (not shown). Simultaneous triggering with TNF-␣, IL-1␤, and The specificity of the IB␣ and IB␤ mRNA detected was determined by cross-hybridization of a membrane containing 20 g of total RNA of the indicated tissues with each probe. Results were not affected by the order of the sequential hybridization. When the effect of LPS on liver IB mRNA levels was measured, this was intraperitoneal-injected at 1 mg/kg, and liver samples were collected after 4 h (panel C). Results show a representative experiment out of three. * denotes p Ͻ 0.005 with respect to the corresponding values at zero time.
IFN-␥, a condition that appears to potentiate the expression of some genes dependent on NF-B activation, did not modify the response to TNF-␣ alone. The mRNA levels of IB␣ and IB␤ were measured at 1 and 4 h (Fig. 7). In agreement with the effects of TNF-␣ at the protein level, an increase of IB␤ mRNA at 1 h (3-fold) and at 4 h (5.4-fold) was observed. Interestingly, challenge of macrophages with TNF-␣, IFN-␥, and IL-1␤ resulted in a synergistic effect on IB␣ mRNA up-regulation but not on IB␤, suggesting a different transcriptional control of both genes. DISCUSSION Activation of NF-B constitutes an important step in the course of several immune and inflammatory responses, including septic shock (1,2,8,11). Two main regulatory mechanisms of NF-B activity have been recognized. One is the precise nucleotide sequence of the B motif to which NF-B binds, the important differences existing in the transcriptional activity depending on variations in the consensus sites and in the flanking regions (28,29). The other involves the association of NF-B with inhibitory subunits such as the various forms of IB proteins and the formation of an inactive complex in the cytosol (3,11). Specific interactions between IB and NF-B proteins have been described. For example, IB␣ and IB␤ strongly bind to p65 and c-Rel but not to the p50 component of the complex (3,4). In addition to this, a cell-specific pattern of expression of members of the IB family has been observed (12), and therefore, NF-B activity depends ultimately on the balance between the rates of degradation and resynthesis of each IB protein. We have used the B sequence corresponding to the murine iNOS promoter, a gene for which transcription requires NF-B activation and that exhibits a transient expression in the tissues examined (26,27). Our data show a complete degradation of IB␤ in liver and in peritoneal macrophages of animals challenged with LPS as well as in cultured macrophages. This degradation was very rapid since a complete loss of immunodetected protein in the Western blot was observed less than 1 h after stimulation. The best known pathway of IB degradation is that of IB␣. Phosphorylation of IB␣ is a necessary requisite for its proteolytic degradation that is essential for in vivo NF-B activation (10). The identification of the enzymes involved in IB␣ phosphorylation points to several protein kinases, including casein kinase II and mitogen-activated protein kinase (10, 29 -31). Interestingly, these data suggest that IB␣ might be targeted by several protein kinases depending on the extracellular stimuli. However, the mechanisms that control IB␤ targeting and degradation still remain unidentified, although the proteasome is required for this process (Ref. 12 and this data).
Resynthesis of IB␣ is directed by NF-B activation (18). However, the regulation of the transcriptional activity of the IB␤ promoter is poorly characterized, and only indirect data are available. In B cell lines, LPS and IL-1 promoted a persistent IB␤ degradation compatible with a sustained NF-B activation for at least 48 h, whereas phorbol esters and cytokines such as TNF-␣ did not affect IB␤ levels and produced only a transient activation of NF-B (12). Also, activation of human vascular endothelial cells with phorbol esters and TNF-␣ produced a persistent activation of NF-B (more than 20 h) that paralleled IB␤ degradation, whereas IL-1 produced only a transient activation associated with a rapid recovery of IB␤ (21); moreover, the co-stimulatory signal elicited by CD28 engagement in T cells produced a rapid and persistent degradation of IB␤ that contributed to the activation of several NF-B/Rel heterodimers (32). Opposite of these cases, our data clearly show that in peritoneal macrophages treated with LPS or TNF-␣ or in liver from animals suffering septic shock, a rapid resynthesis of IB␤ occurred concomitantly with the decrease of NF-B activity. However, a residual NF-B activity still persists in macrophages after 6 h of treatment, suggesting that the levels of IB are not sufficient to dissociate NF-B from the DNA either because the turnover of IB in the cytosol is rapid or because of an improved stability of the NF-B⅐DNA complexes present in the nucleus of these cells. At the mRNA level, up-regulation of IB␤ was observed at 1 h after stimulation, immediately following IB␣ induction. Taken together, these data suggest the existence of pathways (dependent on LPS and TNF-␣ in our models) that can rapidly regulate IB␤ transcription and in this way contribute to attenuate NF-Bdependent responses. Moreover, a certain cell specificity exists in the control of IB␤ degradation in view of the opposite results observed in preB cells (70Z/3 cells) and in endothelial cells upon challenge with the same array of pro-inflammatory cytokines and phorbol esters (12,21).
The rapid resynthesis of IB␤ observed in our experimental model parallels the fall in NF-B activity. However, recent results indicate that the phosphorylation state of the newly synthesized IB␤ protein might influence its interaction with NF-B (22,33). Resynthesis of IB␤ in 70Z/3 cells mainly corresponds to an unphosphorylated protein that exhibits a specific interaction with NF-B. This NF-B/IB␤ ternary complex retains NF-B activity since, in its unphosphorylated state, IB␤ is unable to mask the nuclear localization signal and DNA binding domains of the complex (22). In addition to this, unphosphorylated IB␤ prevents the interaction of NF-B with IB␣, and therefore it can contribute to a persistent NF-B activation (22,23,34). Opposite to this situation, the translated IB␤ in activated macrophages is rapidly phosphorylated, although this constitutive phosphorylation of IB␤ is unsufficient to induce degradation (34). Probably for this reason, phosphorylated IB␤ together with IB␣ efficiently participates in the blockage and retention of NF-B in the cytosol. In agreement with these data, only minimal amounts of IB␤ were detected in the nucleus, as confirmed by different techniques including the immunolocalization by confocal microscopy, the supershift EMSA or the immunodetection by Western blot using nuclear extracts (not shown). Taken together, these results are compatible with the presence of a functional, predominantly phosphorylated IB␤ in the cytosol. The diversity in the regulation of IB␤ is not unique. Indeed, the immunosuppression mediated by glucocorticoids was explained on the basis of an important up-regulation of IB␣ in lymphoid cells (35); however, glucocorticoids inhibit NF-B activation in epithelial cells through a mechanism independent of IB␣ upregulation and possibly involving an inhibition of the translocation process from the cytosol to the nucleus (36).
The up-regulation of IB␤ in the course of septicemia is in agreement with the observation of a transient activation of NF-B; this experimental model might provide additional clues in understanding the process of LPS-dependent IB␤ resynthesis. The role of IB␤ degradation in supporting a sustained NF-B activation has been confirmed in T cells infected with leukemia virus type 1, in which the permanent NF-B activation observed in infected cells is due to the inactivation of IB␤ by the Tax protein (37,38). Finally, the availability of animal models deficient in IB␤, as shown for the IB␣ counterpart (39), may provide additional clues to unravel IB␤ function in physiological and pathological situations.