Reductions in IB and Changes in NF-B Activity during B Lymphocyte Differentiation
http://www.100md.com
免疫学杂志 2005年第2期
Abstract
The levels and stability of IB have been examined in unstimulated and stimulated splenic B cells and compared with that of IB and IB. Primary murine splenic B cells but not T cells were found to contain high levels of IB protein, equivalent to levels of the abundant IB. Most agents that activate IB and IB degradation do not induce rapid degradation of IB. Interestingly, however, the levels of IB, but not of IB or IB, are dramatically reduced upon the stimulation of B cells both in vivo and in vitro. Since IB exhibits substrate specificity for NF-B Rel homodimers, this suggested the possibility that changes in NF-B-responsive genes might also occur during this transition. Consistent with this hypothesis, we found that a NF-B reporter construct sensitive to p65/RelA homodimers is activated at the time that IB levels decline following B cell stimulation. In IgG+ B cell lines, which contain low levels of IB, this same reporter construct was inactive, suggesting that the increases in Rel homodimer activity that accompany B cell stimulation are transient. However, there are differences in the level of expression of NF-B-responsive genes in these IgG+ B cell lines compared with their IgM+ counterparts. From these data, we conclude that there are transient changes in NF-B activity due to reductions in IB, which might contribute to long-term, persistent changes that accompany B cell differentiation. We propose an important role for IB in the differential regulation of nuclear NF-B activity in stimulated B cells.
Introduction
The NF-B/Rel family of proteins forms homo- and heterodimeric complexes that play a major role in controlling the expression of genes involved in immune, inflammatory and acute phase responses (1, 2, 3, 4, 5). NF-B activity is primarily regulated through nuclear translocation. In their inactive form, the complexes are sequestered in the cytoplasm, bound by members of the IB family of NF-B inhibitor proteins. Activation occurs in response to a diverse array of stimuli that lead to rapid degradation of IB followed by translocation of NF-B to the nucleus. The proteasomal degradation of IB proteins is triggered by the phosphorylation of two highly conserved serine residues within their N-terminal domains conducted by a high molecular mass IB kinase complex (IKK)3 (5, 6, 7, 8). The IB proteins identified to date that are the targets of this pathway include IB, IB, and IB, the members of the IB family that control the transcriptionally active NF-B complexes.
Although NF-B activation is constitutive in B lymphocytes, much of the NF-B remains sequestered in the cytoplasm, allowing for significant increases of nuclear NF-B upon B cell stimulation by agents such as LPS, CD40L, or anti-IgM Abs (9, 10, 11, 12, 13, 14, 15). The genes regulated by NF-B play important roles in B cell development, differentiation, and function (4, 11, 16, 17, 18, 19, 20, 21, 22, 23, 24). However, despite the importance of NF-B to B cells, little is understood about the mechanism(s) of constitutive and inducible NF-B activation in these cells or the role of individual IB proteins in regulating NF-B activity. Although the continual degradation of IB and/or IB is generally believed to be the mechanism for constitutive NF-B activation in IgM+ B cell lines (25, 26, 27, 28), studies from our laboratory have suggested that the activation of NF-B might be a dynamic process incorporating distinct mechanisms, depending on the B cell phenotype. We found that IgG+ B cell lines, which contain nuclear NF-B, do not display accelerated degradation of IB proteins, in striking contrast with their IgM+ counterparts, and have significantly reduced levels of IB (29). IB appears to differ from other IB proteins in two important ways. First, IB functions predominantly in the cytoplasm to sequester p65/RelA and/or cRel homodimers and is relatively inefficient at inhibiting transcription of genes regulated by p50/p65 heterodimers (30, 31). IB has been suggested to play an important role in the transient activation of a subset of genes regulated by Rel homodimers (31). Second, IB does not contain a C-terminal PEST sequence and therefore its expression may be regulated by mechanisms other than by its rapid degradation following activation.
Given the differences between IB and other IB proteins, we reasoned that IB might be an important regulator of NF-B/Rel activity in B lymphocytes capable of controlling, transiently or persistently, changes in gene expression in these cells. To address this issue, we examined IB levels in normal B cells at different stages of differentiation, determined whether transcriptional or posttranscriptional regulation was responsible for changes in abundance of IB, and examined the consequences of these changes on the expression of NF-B-responsive genes. Our results indicate that the regulation of IB expression and degradation is distinct from that of other IB proteins and suggest an important role for IB in the differential regulation of nuclear NF-B activity in B cells during differentiation.
Materials and Methods
B and T cell separation and culture
Single-cell suspensions were prepared from thymus or spleen of C57BL/6 (B6) mice (purchased from The Jackson Laboratory). B cell-enriched populations were prepared by depletion of T cells by a mixture of cytotoxic Abs, including anti-CD4 (GK1.5), anti-CD8 (53.6.72), and anti-Thy-1.2 (13-4), followed by rabbit complement as described previously (32). The resulting cells were >85% B cells with some monocyte contamination as assessed by staining with anti-B220 and anti-Mac1 Abs (BD Pharmingen). To further enrich B cells, cells from this population (2.5 x 106/ml) were cultured overnight in the presence of IL-4 (25 ng/ml; R&D Systems); IL-4 has been shown to have no effect on nuclear NF-B activity (20). The nonadherent cells were recovered and found to be >95% B220+ B cells and are referred to as "B – selection" (see Fig. 1). B cells were also enriched by positive selection by using anti-B220-coated paramagnetic beads (Miltenyi Biotec) in the presence of Fc block (BD Pharmingen) and then column purified according to the manufacturer’s protocol at 4°C ("B + selection"). This method yielded B cell preparations from normal spleen that were found to be 94–98% B220+, 79–88% IgM+, 9–18% IgG+, and 1% CD69+. T cells were enriched from spleen cells as described using nylon wool filtration (33), resulting in populations that were 90–95% T cells as assessed by staining with anti-CD3 and essentially devoid of B cells as assessed by B220 staining.
FIGURE 1. IB protein is expressed at high levels in primary B cells. A, Left panel, Western blots were prepared from whole cell extracts of murine thymocytes (Thymus), splenocytes (Spleen), T cell depleted splenocytes (T-depleted), or enriched splenic T cells (T). Right panel, Western blots were prepared from whole cell extracts of splenocytes (Spleen), B cells purified by T cell depletion and monocyte adhesion (B –selected), and B cells prepared by positive selection on paramagnetic beads coated with anti-B220 Ab (B + selected). The same extract from splenic T cells and an extract from an IgM+ B cell line (CH27) were used for comparison. Blots were sequentially probed with Abs raised against IB, IB, and calnexin as loading control. B, Left panel, Western blot containing a purified primary B cell extract and 1, 2, or 4 μl of COS extracts after transfection with recombinant FLAG-tagged murine IB (rmIB) or recombinant FLAG-tagged human IB (rhIB) was probed simultaneously with anti-IB and anti-IB (sc-7155) Abs. Right panel, Western blot prepared with 1 μl of the respective COS extracts containing IB or IB was probed with an anti-FLAG Ab.
All cells were cultured in complete DMEM containing 7% FBS (Invitrogen Life Technologies) and 50 μM 2-ME (Sigma-Aldrich) and supplemented as previously described (34). In some experiments, cells (2 x 106/ml) were treated with cycloheximide (CHX, 50 μg/ml; Sigma-Aldrich) or stimulated with LPS (50 μg/ml, Escherichia coli 055:B5; Difco), CD40L, or CD40L plus anti-IgM Ab (10 μg/ml, F(ab')2 goat anti-mouse; Jackson ImmunoResearch Laboratories) for various periods of time as described in Results. CD40L stimulation was conducted using a CD40L-CD8 fusion protein in combination with an anti-CD8 mAb for cross-linking (both reagents kindly provided by A. Marshak-Rothstein, Boston University School of Medicine, Boston, MA) as previously described (35). For studies to determine the half-life of IB proteins, LPS and anti-IgM treatments were initiated 2 h before the addition of CHX, while CD40L was added simultaneously with the CHX, because of the rapid induction of IB degradation following CD40L addition.
Infection of mice with LP-BM5 virus
Mice were inoculated i.p. with 0.1 ml of LP-BM5 murine leukemia virus pools at 6–8 wk of age, and the progression of disease was assessed as previously described (36). Mice were sacrificed at 4, 8, or 12 wk after infection, spleen weights were determined, and protein extracts were prepared as described below. Purified B cells were obtained by positive selection from spleens 4 or 8 wk after infection using paramagnetic beads as described above. The phenotypic characteristics of these B cells are described in Results. Mice were cared for and handled at all times in accordance with National Institutes of Health and institutional guidelines.
Preparation of protein extracts, phosphatase treatment, and immunoblotting
Whole cell extracts were prepared exactly as previously described (29), except in one experiment in which spleen cells from mice 12 wk after BM5 infection were analyzed. Spleen cells from these mice (or uninfected control mice) were prepared by washing splenocytes in 1x PBS (pH 7.4), followed by lysis with 4 pellet volumes of a buffer containing 10 mM HEPES (pH 7.9), 1.5 mM MgCl2, 10 mM KCl, 0.5% Nonidet P-40, 0.5 mM DTT, 30 μg/ml leupeptin, and 1 mM benzamidine. Nuclei were removed by centrifuging for 20 min at 1600 x g at 4°C and supernatants were collected as cytoplasmic extracts. All protein concentrations were determined by Bradford assay (Bio-Rad).
Lysates from equivalent amounts of protein (as indicated in the figure legends) or equivalent cell numbers (for CHX experiments) were fractionated on 10% SDS-PAGE, gels were transferred to nitrocellulose membranes (Micron Separations), and the membranes were stained with Ponceau S (Sigma-Aldrich) to ensure equivalent loading and transfer. Membranes were probed with the appropriate Abs and developed with the Renaissance System (New England Nuclear). Primary Abs included rabbit Abs against IB (sc-371), IB (sc-945), and IB (sc-7155 and sc-7156; identical results were obtained with both Abs), all obtained from Santa Cruz Biotechnology, anti-FLAG (M2; Eastman Kodak), and rabbit anti-calnexin (StressGen Biotechnologies) which served as an additional loading control. Autoradiographs were quantified using a Molecular Dynamics densitometer with ImageQUANT software (Molecular Dynamics).
Whole cell extracts were prepared in the presence of phosphatase inhibitors, 50 mM sodium fluoride and 1 mM sodium orthovanadate. However, in experiments designed to evaluate the phosphorylation status of IB protein, extracts were prepared in the absence of phosphatase inhibitors. Extracts were divided in two and treated with phosphatase (20 U/μl extract; New England Biolabs) for 30 min at 30°C, with or without the addition of sodium fluoride and sodium orthovanadate, according to the manufacturer’s instructions. Reactions were stopped by addition of 2x sample buffer (100 mM Tris (pH 6.8), 4% SDS, 20% glycerol, 0.1% bromphenol blue, and 700 mM 2-ME) and analyzed by SDS-PAGE, along with a sample from untreated cells.
The stability of IB proteins was determined following CHX treatment of cells as described elsewhere (29). Cells were cultured in the presence of CHX (50 μg/ml) for various periods of time, extracts were prepared, and IB expression was determined by Western blot analysis. This method yields results identical to those of conventional pulse-chase analysis (29).
Expression vectors and transfections
The IB expression vector containing a 5' FLAG tag followed by the complete coding region of human IB under the control of a CMV promoter was obtained from Dr. D. Ballard (Vanderbilt University, Nashville, TN) (37). The IB expression vector (pC3 FLAG-IB) was constructed by attaching a 5' FLAG tag sequence (Eastman Kodak) to the 1.1-kb coding region of murine IB via PCR using a plasmid containing the 2.2-kb IB cDNA as template (pC3 IB; obtained from D. Thanos, Columbia University, New York, NY) (30). COS cells were grown in 10-cm dishes in IMDM (Invitrogen Life Technologies) supplemented with 7% FBS, antibiotic/antimycotic solution (Invitrogen Life Technologies), and 2 mM glutamine. Cells were transfected with expression vectors for IB and IB using DEAE-dextran (Amersham/Pharmacia) as previously described (38).
A20 and M12 cells were transfected using DEAE-dextran (Amersham/Pharmacia) with the following luciferase reporter constructs: an ELAM luciferase reporter containing 3 B sites (obtained from D. Golenbock, University of Massachusetts Medical Center, Worcester, MA) (39); an IL-8 promoter construct (generously provided by Dr. K. LeClair, Antigenics, Inc., Woburn, MA), which contains a single B site that responds only to Rel homodimers and not to conventional NF-B heterodimers (40, 41, 42), and a mutant IL-8 construct (mut 2) that has a nonfunctional B site. A -galactosidase reporter construct was used as an internal control for transfection efficiency (Clontech). Whole cell lysates were prepared 24–49 h following transfection as analyzed according to the luciferase detection assay kit (Promega).
RNA analysis
RNA was isolated from cell populations using TRIzol (Invitrogen Life Technologies) according to the manufacturer’s specifications or using Qiagen/lithium chloride/urea/mRNA. Northern blots were performed as previously described (43) and probed with a 1.1-kb IB fragment, comprising the entire coding region (30) (obtained from D. Thanos). Northern blots were also probed with NF-B1 (p50) and c-rel (both described in Ref.38) and Ig (clone sc33; Ref.44) and CHO-B (45) as a loading control. All probes were labeled by random priming.
For semiquantitative RT-PCR, RNA isolated from B cells or B cell lines was isolated using the RNeasy kit according to the manufacturer’s specifications (Qiagen). RNA samples were reversed transcribed and then subjected to varying numbers of cycles (21, 22, 23, 24, 25, 26, 27) of PCR using the Superscript One-Step RT-PCR System (Invitrogen Life Technologies). The products were then resolved on agarose gels and quantitated by densitometry (Ultraviolet Products BioImaging). For this analysis, the following forward and reverse primers (Invitrogen Life Technologies), respectively, were used: for IB, 5'-AGAGTGACTCTGGTTCTGTT-3' and 5'-GGCAGCCGCTTTGGGATG-3' or 5'-GCTATTCTGTTGCTTGGC-3' and 5'-GTACATCAATGTCAGCTC-3'; for IB, 5'-GGCCTGGACTCCATGAAG-3' and 5'-GGTCTGCGTCAAGACTGC-3'; and for calnexin, 5'-CCAAGCCTCTCATTGTTC-3' and 5'-CAGTCATCTGGCTTGACAG3'. The relative abundance of each product was normalized using calnexin as a standardization control.
Results
Splenic B lymphocytes contain high levels of IB
We first evaluated the expression of IB in normal lymphocyte populations, including splenic B cells. As shown in Fig. 1A (left panel), high levels of IB were present in the spleen, in agreement with previous studies (46). Because significantly lower levels of IB were found in the thymus, we compared the expression of IB in enriched populations of splenic B and T cells. IB expression was high in B cell-enriched populations from which T cells were depleted (T depleted) and low when B cells were removed (T). Similar differences were obtained by analyzing the steady-state distribution of IB mRNA in the various lymphocyte subpopulations by Northern blot analysis (data not shown). We also confirmed that the multiple IB bands detected were phosphoisoforms (data not shown), as expected from other studies (31).
To prove that normal splenic B cells had high levels of IB, we subjected these cells to further purification by T cell depletion and the removal of adherent cells (B – selected) or isolated by positive selection using paramagnetic beads (B + selected). Both populations contained >95% B cells as assessed by FACS. As shown in Fig. 1A (right panel), B cells isolated by either method contained significantly higher levels of IB protein than the enriched T cell population and comparable levels of IB to those found in the IgM+ B cell lymphoma CH27. Note that while IB is expressed in higher concentrations in B cells than in T cells, this is not true for other IB proteins, such as IB, which is expressed at similar levels in T and B cells (Fig. 1A). We conclude that B cells are the major source of IB within normal unstimulated splenic lymphocyte populations.
The results in Fig. 1A suggest that IB levels might be very high in B lymphocytes, perhaps as high as IB levels. To obtain a comparative estimate of the relative expression of IB and IB in splenic B cells, we used recombinant FLAG-tagged IB and IB from transfected COS cells to determine whether the Abs raised against IB and IB detected similar levels of proteins. A Western blot was prepared containing 20 μg of whole cell extract from purified splenic B cells and different amounts of extracts from COS cells transiently transfected with the FLAG-tagged IB or IB expression vectors. The blot was probed simultaneously with IB and IB Abs. As shown in Fig. 1B (left panel), the IB and IB signals obtained from the B cell extract corresponded to those obtained from 2 to 4 μl of the respective COS extract. Probing with an anti-FLAG Ab demonstrated that the IB and IB levels in the two COS extracts were similar (Fig. 1B, right panel). These results demonstrate that the level of expression of IB and IB can be compared from the Western blots. From these data, we conclude that IB is highly expressed in normal B cells at levels comparable to those of IB. Therefore, IB is not a "minor" IB protein in splenic B cells, but a major inhibitory protein regulating a significant proportion of NF-B activity.
Degradation of IB proteins in stimulated B cells
To compare the turnover of IB with that of other IB proteins in B cells, we stimulated primary B cells with agents that activate NF-B, including LPS, CD40L, anti-IgM Ab, or a combination of CD40L and anti-IgM (3, 25) in the presence of CHX. At various times after stimulation, whole cell extracts were prepared and analyzed by Western blot for IB, IB, and IB expression (Fig. 2A). The blots were reprobed with calnexin to monitor loading, the bands were analyzed by densitometry, the signal intensities normalized to the calnexin signal, plotted, and the half-lives were determined (Fig. 2B).
FIGURE 2. Half-lives of IB, IB, and IB in primary B cells after stimulation with B cell mitogens. A, Primary B cells were cultured overnight in IL-4 and subsequently stimulated with LPS, CD40L or a combination of CD40L and anti-IgM. CHX was then added to the cultures (see Materials and Methods) and whole cell extracts were prepared from equivalent cell numbers at 0, 30, 60, and 120 min after CHX addition. Extracts were analyzed by Western blot with the Abs indicated (calx, calnexin). Note that the gel in A, of extracts from cells stimulated with CD40L plus anti-IgM, was not run as far as the other three gels shown and therefore the different phosphoisoforms of IB are not resolved. B, Signals from autoradiographs shown in A were quantitated by densitometry. The intensities of the calnexin signals were used to correct for differences in loading and blotting efficiency. The normalized signal intensities of IB (?), IB (), and IB () (top four panels) were plotted and half-lives were calculated by linear regression; these values are shown in tabular form. The bottom left panel compares the degradation of IB induced by different stimuli.
IB, IB, and IB were relatively stable in untreated B cells with half-lives estimated to be between 90 and 130 min. The slow turnover is not surprising since nuclear NF-B levels are relatively low in resting B cells (11, 14). LPS stimulation led to a significant increase in IB and IB turnover, with half-lives estimated to be 30 and 45 min, respectively. In contrast, the turnover of IB was only slightly faster than in unstimulated B cells. When B cells were activated by CD40L, IB, and IB showed dramatic increases in turnover, with half-lives of only 10–15 min. Surprisingly, however, no increase in turnover of IB was detected in CD40L-stimulated B cells. Treatment with anti-IgM Abs to engage the BCR also had no effect on IB degradation (data not shown). When CD40L stimulation was combined with anti-IgM treatment, accelerated degradation of IB was observed, with a half-life of 45 min. Nevertheless, the turnover of IB following CD40L plus anti-IgM treatment was still 3- to 4-fold slower than the turnover of IB and IB in these same cells. Taken together, these results demonstrate a surprising and unprecedented specificity of NF-B-activating signals in primary B cells for IB and IB and a significantly reduced susceptibility of IB to degradation.
IB levels decrease in stimulated B cells
Although B cell stimuli do not induce rapid degradation of IB, previous studies from our laboratory have shown that levels of IB may be modulated in B cells since IgG+ B cell lines contain very low levels of IB compared with IgM+ B cell lines (29). We therefore sought to determine whether splenic B cells also contained low levels of IB following stimulation. Given the difficulties in obtaining populations of cells containing sufficient numbers of IgG+ cells from normal or immunized mice for Western blot analysis, even after immunization (see, for example, Ref.47), we used an in vivo model of B cell differentiation that makes use of the replication-defective LP-BM5 murine leukemia virus. This virus causes a progressive lymphoproliferation and immunodeficiency, also known as murine AIDS, that is characterized by B cell activation, proliferation, and differentiation, including Ig class switching (48, 49). The activation and differentiation of B cells in this model is dependent on CD4+ T cell stimulation and CD40-CD40L interactions (50, 51), and thus has many of the hallmarks of B cell activation by T-dependent Ags. Spleens of infected mice contain significantly increased numbers of IgG+ B cells (36, 52).
To determine whether levels of IB were lower in these B cells, we first compared the abundance of IB in cytoplasmic extracts from total splenocytes from control uninfected mice (Fig. 3, Ctrl.) and from three individual mice 12 wk after infection with the LP-BM5 virus. As shown in Fig. 3A, levels of IB were significantly lower in extracts from each of the three infected mice compared with uninfected controls. The decrease in IB expression was specific for IB, since IB levels (and IB levels, data not shown) remained largely unchanged. As a measure of B cell differentiation that accompanies infection, we compared levels of IgM and IgG in cell extracts. The decrease in IB was paralleled by an increase in IgG expression while IgM levels in the three mice postinfection were significantly reduced compared with their normal counterparts. Since B cells are the major source of IB in the spleen, these initial results suggested that levels of IB declined in the B cell population.
FIGURE 3. IB expression in primary splenocytes or purified splenic B cells after polyclonal B cell activation caused by LP-BM5 infection. A, Western blots of 20-μg cytoplasmic extracts from pooled splenocytes from normal, uninfected mice (Ctrl.) or from three mice 12 wk after infection with LP-BM5 virus (BM5 #1, 2, and 3) were sequentially probed with anti-IB, anti-IB, anti-IgG, or anti-IgM Abs. B, Western blots of 20-μg whole cell extracts from sorted splenic B cells from uninfected mice (Ctrl.) or from mice 4 or 8 wk after infection with LP-BM5 virus (BM5 4wk, 8wk) were sequentially probed with anti-IB, anti-IB, anti-IgG, or anti-calnexin (calx) Abs. For comparison, Western blots contained equivalent amounts of whole cell extracts from the IgM+ B cell line CH27 and the IgG+ B cell line M12. M12 expresses low levels of IgG (29 ) that is detectable on longer exposures of the blot (data not shown).
To prove that the B cells contained reduced levels of IB, B cells were isolated from the spleens of mice 4 or 8 wk after infection and IB levels in these B cells were compared with levels in cells from control, uninfected mice. FACS analysis confirmed that 94–98% of the isolated cells were B cells, as assessed by B220 staining. As shown in Fig. 3B, IB levels decreased significantly in B cells during the course of infection, such that there was almost as little IB in B cells 8 wk after infection as in the IgG+ B cell line, M12. In contrast, levels of IB were similar in all B cells. Interestingly, the decrease in IB levels preceded the increase in total IgG levels as determined by Western blot analysis (Fig. 3B) and surface IgG staining. FACS analysis revealed that IgG+ B cells increased from 3 to 8% in B cells from uninfected animals to 28–36% of the B cells from individual mice 8 wk after infection and continued to increase up to 12 wk when 30–54% of splenic B cells were IgG+. These data indicate that there appears to be a significant difference in the regulation of IB expression in differentiating B cells compared with primary splenic IgM+ B cells. However, it is not known from these data whether the changes in IB occur soon after activation of responding IgM+ B cells or at later times during their differentiation.
IgG+ B cells do not contain increased levels of functional Rel homodimers
Unlike other IB proteins, IB demonstrates substrate specificity for NF-B in vivo, in that it selectively retains Rel (p65/RelA and c-Rel) homodimers but not conventional p50/p65 heterodimers (Refs.30, 31 and our unpublished results). Given that IgG+ B cell lines have low levels of IB, it was possible that higher levels of transcriptionally active Rel homodimers might be present in these cells. To test this, we transfected the IgG+ B cell lines M12 and A20 with two different NF-B luciferase reporter constructs, one driven by the ELAM promoter containing three conventional B sites (39) and the second under the control of an IL-8 promoter containing a single B site. The IL-8 reporter provides a specific monitor for the presence of functional nuclear RelA homodimers (40, 41, 42). Cells were transfected with one of the NF-B reporter constructs along with a -galactosidase reporter construct as an internal control for transfection efficiency and normalization and luciferase activity was measured 24 h later. As shown in Fig. 4, the IL-8 promoter was not functional in either cell line. This was not due to the fact that the IL-8 promoter contains only a single B site and might therefore be insensitive to activation (see below). In contrast with the IL-8 promoter, the ELAM promoter was fully active in both IgG+ B cell lines, confirming the presence of functional NF-B. These results indicate that while conventional p50/p65 heterodimers are present in the nucleus of these cells, Rel homodimers are not, despite the low levels of IB. Consequently, if reductions in IB lead to increased Rel homodimer activity, these increases are likely to be transient rather than long-term, stable increases in this subset of NF-B complexes.
FIGURE 4. NF-B reporter activity in two IgG+ (M12 and A20) B cell lines. Cells (107) were transiently transfected with either the ELAM-luciferase or the IL-8-luciferase reporter constructs along with a -galactosidase reporter construct as a control for transfection efficiency. Whole cell lysates were prepared, promoter activity was measured (Promega) with a luminometer, and corrected for transfection efficiency.
Reductions in IB and increased Rel homodimer activity after B cell stimulation
The data in Fig. 3 suggested that IB levels decline before differentiation of stimulated B cells into IgG+ B cells. To determine when IB might decline during B cell differentiation, we monitored IB protein levels in purified splenic B cells at various times after stimulation with LPS or CD40L and anti-IgM. As shown in Fig. 5, IB levels declined precipitously within the first 24 h after stimulation with LPS. Levels of IB were calculated by densitometry and found to be only 2% of initial levels at this time. In contrast, IB levels were still 60% of control levels after 24 h of stimulation and never declined to the same extent as IB. Similar results were obtained in cells stimulated with CD40L and anti-IgM. In these cells, IB protein levels were reduced by 70 and 85% at 24 and 48 h of stimulation, respectively, while levels of IB were not reduced in the first 24 h and exhibited only a modest 30% reduction at 48 h of stimulation. Therefore, despite the fact that IB is not degraded nearly as rapidly as IB following stimulation (Fig. 2), levels of this protein decline more dramatically following stimulation, perhaps as the result of transcriptional changes. The sharp declines in IB levels suggest that selective and transient increases in nuclear Rel homodimer activity might be generated during this period of time.
FIGURE 5. Steady-state levels of IB and IB after stimulation of primary B cells. Purified splenic B cells were stimulated with either LPS (A) or CD40L and anti-IgM (B) for 72 h. Whole cell lysates were prepared at 0, 24, 48, and 72 h. Western blots of 15 μg of whole cell lysates were sequentially probed with anti-IB, anti-IB, or anti-calnexin (calx) Abs.
To investigate this possibility, we used the inducible IgM+ B cell line CH12-LBK, which serves as a useful model for the differentiation of IgM+ B cells (34, 53, 54). LPS stimulation resulted in significant reductions in steady-state levels of IB over a 48-h period, with reductions in both IB and a predominant phosphoisoform showing similar reductions of 50% in the first 24 h and almost 90% by 48 h after stimulation (Fig. 6, A and B). In contrast, IB levels remained essentially unchanged over the same period of time. To determine whether these changes led to increased in functional Rel homodimer activity, CH12 cells were transiently transfected with one of several NF-B reporter constructs, and reporter activity in unstimulated and LPS-stimulated cells was compared 24 h later. A representative example of eight independent experiments is shown in Fig. 6C. IL-8 reporter activity was significantly increased in LPS-stimulated compared with unstimulated cells, even though the overall NF-B activity was essentially unchanged, as monitored by the ELAM promoter. The mut 2 construct, in which the single IL-8 B site has been mutated (41), showed only limited activity, indicating that most of the increase on the IL-8 promoter was due to activation by NF-B, specifically Rel homodimers. As expected, cotransfection of an IB expression vector with the IL-8 reporter construct into LPS-stimulated B cells resulted in the total inhibition of luciferase activity (data not shown). This confirms that IB is capable of retaining Rel homodimers and inhibiting the activation of the IL-8 promoter in these cells. Although overexpression studies must be interpreted with caution, these results are consistent with the suggestion that the decline in IB levels following B cell stimulation is responsible for increases in nuclear Rel homodimer activity.
FIGURE 6. Increase in Rel homodimer activity accompanies decreases in IB following LPS stimulation. A, Western blots of whole cell lysates from the IgM+ B cell line CH12 LBK after stimulation with LPS. Blots were sequentially probed with anti-IB, anti-IB, or anti-calnexin (calx) Abs. B, Signals from autoradiographs shown in A were quantitated by densitometry. The intensities of the calnexin signals were used to correct for differences in loading and blotting efficiency. These values were then plotted on a log scale (, IB; , IB; ?, IB-p). C, Normalized luciferase reporter activity in CH12LBK IgM+ B cells. Cells were stimulated with LPS for 24 h () or left untreated (). The cells were then transiently transfected with either the ELAM-luciferase (with three B sites) or IL-8-luciferase (with a single B site) reporter constructs. mut 2 is a mutant IL-8 construct in which the B site has been mutated. Cells were also cotransfected with a -galactosidase reporter construct as a control for transfection efficiency. Whole cell lysates were prepared 24 h following transfection and luciferase activity was measured.
Selective reductions in IB mRNA and changes in gene expression following B cell stimulation
We next investigated the basis for the selective reductions in IB expression. Although IB does not degrade as rapidly as IB in response to B cell stimulation (Fig. 2), IB levels are reduced much more dramatically after stimulation than IB (Figs. 3 and 5). Although IB expression, like that of IB, has been reported to be regulated by NF-B (31), we considered the possibility that IB might be regulated at the RNA level. RNA from the inducible B cell line CH12-LBK or from primary B cells were subjected to semiquantitative RT-PCR. As shown in Fig. 7A, levels of IB mRNA declined by 50% following stimulation of CH12-LBK cells. These data were corroborated by results of DNA microarray data (Affymetrix GeneChip), which consistently showed a 50% reduction in IB levels following LPS stimulation (data not shown). In contrast, IB levels remained steady over the course of the experiment. Similar results were obtained using RNA from primary B cells stimulated with LPS (Fig. 7B). IB mRNA levels declined by at least 50% within 48 h after stimulation and fell by 80% or more by 72 h after stimulation. In contrast, IB levels increased during the first 24–48 h and then returned to baseline levels.
FIGURE 7. Selective decrease in IB mRNA levels in stimulated B cells. A, Semiquantitative RT-PCR of total RNA from the IgM+ B cell line CH12-LBK after LPS stimulation. RNA was isolated at various time points following LPS stimulation and subjected to increasing rounds of RT-PCR (1, 2, and 3 represent 21, 24, and 27 cycles, respectively). B, Semiquantitative RT-PCR of total RNA from primary B cells. RNA was isolated at various time points following LPS stimulation and subjected to increasing rounds of RT-PCR as in A. C, Northern blot analysis of NF-B-dependent genes from two IgM+ (WEHI231 and CH27) and two IgG+ (A20 and M12) B cell lines. Twenty micrograms of total RNA was separated on 1.2% denaturing agarose gels, transferred to nitrocellulose membrane, and sequentially probed with 32P-labeled cDNA probes for IB, NF-B1 (p50), c-rel, Ig, and CHO B as a loading control.
To determine whether reductions in IB expression in stimulated B cells might result in selective changes in the expression of a subset of NF-B-regulated genes, mRNA expression levels of four genes known to be regulated by NF-B were compared in IgM+ (WEHI231 and CH27) and IgG+ (A20 and M12) B cell lines. These included c-Rel (55), NF-B1 (56), and Ig L chain (10) in addition to IB. As shown in Fig. 7C, IB mRNA was abundant in IgM+ B cells, but present at much lower levels in the two IgG+ B cells. However, not all NF-B target genes showed similar reductions. The levels of c-Rel and Ig were not reduced in IgG+ B cells relative to their IgM+ counterparts, while NF-B1 levels were reduced, albeit to a lesser degree than IB. These data indicate that the decreased expression of IB in IgG+ B cells is not simply a reflection of overall reduced NF-B transactivation potential, but instead may reflect a selective reduction in a specific subset of NF-B-regulated genes.
Discussion
Our data are consistent with a role for IB in the differential modulation of NF-B-responsive genes during B cell differentiation. IB is expressed at very high levels in IgM+ B cells, levels comparable to IB. Levels of IB decline significantly following B cell activation to levels much lower than IB. However, in contrast to IB, whose levels remain high at all stages of B cell differentiation, IB levels may remain low, since little IB (mRNA or protein) is expressed in IgG+ B cell lines, and IB levels are low in chronically stimulated splenic B cells. This reduction may have the potential for profoundly altering the nature of the NF-B target genes activated following B cell stimulation.
Hoffmann et al. (57) demonstrated that IB (and IB) function to normalize NF-B activity in responding cells that would otherwise be extinguished by the re-expression of IB. These investigators also predicted that this would result in changes in specific gene expression in cells in which NF-B was transiently activated. Whiteside et al. (31) also suggested that IB might regulate a distinct subset of NF-B-responsive genes, based in part on its substrate specificity. IB, like IB and IB, retains NF-B complexes containing c-Rel and RelA, the transcriptionally active NF-B/Rel proteins in the cytoplasm (30, 31, 46). However, unlike IB and IB, IB shows substrate specificity when expressed at normal levels in vivo and has strong binding affinity for c-Rel and RelA homodimers (Refs.30, 31 ; K. M. Daley and R. B. Corley, unpublished results), IB does not inhibit the more common heterodimeric complexes formed with NF-B p50 except at high concentrations (30). Rel homodimers are known to have specificity for certain promoters (40, 41, 42). During the differentiation of B cells, at a time that IB levels first decline (Figs. 5 and 6), promoters sensitive to Rel homodimers would expected to be stimulated, as demonstrated by activity on the IL-8 promoter (Fig. 6). Interestingly, in IgG+ B cells, there is no increase in Rel homodimer activity (Fig. 4), suggesting that this change is transient and results in a new steady-state level of NF-B. It is possible that this initial transition is required for the changes in the levels of NF-B-responsive genes that emerge in IgM+ and IgG+ B cells (Fig. 7), but additional studies linking these events will be required. Nevertheless, we conclude from these data that, in addition to the transient changes in NF-B activity observed within the first 1–2 days after B cell stimulation, long-term, persistent changes also occur, as evidenced by the decreased expression of some NF-B-responsive genes, including IB itself.
Although our current results do not establish a cause and effect relationship between the decline in IB and persistent changes in NF-B activity, the results of Memet et al. (58) are consistent with this hypothesis. They generated IB-deficient mice using a lacZ reporter that was included within a targeting construct 3' of the IB 5' regulatory sequences. By analyzing -galactosidase activity in cells from mutant mice, these authors found that very few B cells were positive, while T cells were positive. They concluded that IB was not highly expressed in B cells. However, direct analysis of IB expression shows that it is highly expressed in primary IgM+ B cells, not only based on the results of the current studies, but also from previous studies of IgM+ B cell lines (29, 46) and from gene expression profiling analysis on peripheral B cells (59). Thus, we suggest another interpretation of the results of Memet et al. (58): in the absence of IB, a different steady-state level of NF-B-responsive genes is expressed in B cells, one that does not induce high levels of IB transcription, but maintains high levels of IB, among other NF-B-responsive genes.
In B cells, IB is not subject to rapid degradation by stimuli that otherwise result in the degradation of IB and IB (Fig. 2). The differential regulation of IB protein degradation is not unprecedented. Like IB, IB has also been reported to be susceptible to a subset of activating signals in certain cells (60). In previous studies, IB was found to be of intermediate susceptibility to proteasomal degradation in a monocytic cell line in response to NF-B-activating signals such as LPS, being degraded faster than IB but less rapidly than IB (31). Surprisingly, IB is not degraded nearly as rapidly as IB in B cells in response to LPS (Fig. 2). This suggests that the stimulation of degradation of the different IB proteins not only depends on the stimulus, but exhibits cell-type specificity as well. Thus, the degree to which IB proteins are susceptible to degradation is not solely an intrinsic property of these proteins, but can be regulated depending on factors modified by the signal(s) given or on factors unique to particular cells. This difference could take place at the level of the IKK complexes (5, 61), for example, through the presence or absence of adaptor molecules or at the level of the IB, for example by a specific phosphorylation event that would render it either unsuitable as IKK substrate or that would prevent proteasomal degradation. In any event, our results suggest that IB is an ideal inhibitory protein for differentially regulating nuclear NF-B activity in B cells. In B cells, unlike IB and IB, IB is not subject to rapid degradation by stimuli that otherwise result in the degradation of IB and IB (Fig. 2). However, IB protein levels do decline after stimulation (Figs. 3, 5, and 6), but this might result from a combination of slow degradation and reductions in IB transcript levels. The slow degradation rates of IB following LPS stimulation are still sufficient to account for the significant reductions in IB levels 24 h later, assuming the abundance of IB mRNA is reduced as well.
Changes in IB levels results in the release of functional Rel homodimers to the nucleus, as evidenced by the activation of the IL-8 promoter within 48 h of stimulation of IgM+ B cells. This change in Rel homodimer activity appears to be a transient event because this same promoter is not active in IgG+ B cells, in which IB levels are low. Whether the genes activated following declines in the levels of IB contribute to isotype switching is unknown, but they are clearly insufficient based on the results of Strober and colleagues (62). They found that NF-B activated via CD40-mediated induction of IB degradation by the IKK complex was absolutely essential for isotype switching. Nevertheless, the transient promoter activation initiated by reductions in IB suggests that whatever subset of genes might be activated in response to Rel homodimers are likely to be induced only briefly. These results therefore support the prediction of Israel and colleagues (31) who suggested an important role for IB in the transient regulation of a subset of late genes. The transient activation of this set of genes is likely to result in the stable expression of a new subset of genes in activated B cells, including those regulated by NF-B.
Acknowledgments
We thank Lia Luus for technical assistance; Dr. Herbert Morse (National Institute of Allergy and Infectious Diseases, Bethesda, MD) for LP-BM5-infected mice; Dr. Ann Marshak-Rothstein (Boston University School of Medicine, Boston, MA) for the reagents for T cell depletion and the CD40L-CD8 fusion protein and anti-CD8 Ab; Dr. Dimitris Thanos (Columbia University, New York, NY) for the IB plasmid; Dr. Dean Ballard (Vanderbilt University, Nashville, TN) for the FLAG-tagged IB expression construct; D. Golenbock for the ELAM reporter construct; and Dr. Kenneth LeClair (Antigenics, Inc., Woburn, MA) for the IL-8-luciferase reporter constructs.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants AI31209 and CA36642 (to R.B.C.).
2 Address correspondence and reprint requests to Dr. Ronald B. Corley, Department of Microbiology, 715 Albany Street, L504, Boston University School of Medicine, Boston, MA 02118. E-mail address: rbcorley@bu.edu
3 Abbreviations used in this paper: IKK, IB kinase complex; CHX, cycloheximide; CHO, Chinese hamster ovary.
Received for publication November 14, 2002. Accepted for publication November 5, 2004.
References
Baeuerle, P. A., T. Henkel. 1994. Function and activation of NF-B in the immune system. Annu. Rev. Immunol. 12:141
Baldwin, A. S., Jr. 1996. The NF-B and IB proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649.
Ghosh, S., M. J. May, E. B. Kopp. 1998. NF-B and rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16:225.
Sha, W. C.. 1998. Regulation of immune responses by NF-B/Rel transcription factors. J. Exp. Med. 187:143
Karin, M., Y. Ben-Neriah. 2000. Phosphorylation meets ubiquitination: the control of NF-B activity. Annu. Rev. Immunol. 18:621.
Stancovski, I., D. Baltimore. 1997. NF-B activation: the IB kinase revealed?. Cell 91:299.
May, M. J., S. Ghosh. 1999. IB kinases: kinsmen with different crafts. Science 284:271.
Israel, A.. 2000. The IKK complex: an integrator of all signals that activate NF-B?. Trends Cell Biol. 10:129.
Sen, R., D. Baltimore. 1986. Inducibility of immunoglobulin enhancer binding protein NF-B by a posttranslational mechanism. Cell 47:921.
Sen, R., D. Baltimore. 1986. Multiple nuclear factors interact with the immunoglobulin enhancer sequences. Cell 46:705.
Liou, H.-C., D. Baltimore. 1993. Regulation of the NF-B/rel transcription factor and IB inhibitor system. Curr. Opin. Cell Biol. 5:477.
Miyamoto, S., M. J. Schmitt, I. M. Verma. 1994. Qualitative changes in the subunit composition of B-binding complexes during murine B-cell differentiation. Proc. Natl. Acad. Sci. USA 91:5056.
Grumont, R. J., S. Gerondakis. 1994. The subunit composition of NF-B complexes changes during B-cell development. Cell Growth Differ. 5:1321.
Francis, D. A., J. G. Karras, X.-y. Ke, R. Sen, T. L. Rothstein. 1995. Induction of the transcription factors NF-B, AP-1 and NF-AT during B cell stimulation through the CD40 receptor. Int. Immunol. 7:151.
Neumann, M., G. Wohlleben, S. Chuvpilo, B. Kistler, T. Wirth, E. Serfling, A. Schimpl. 1996. CD40, but not lipopolysaccharide and anti-IgM stimulation of primary B lymphocytes, leads to persistent nuclear accumulation of RelB. J. Immunol. 157:4862.
Grilli, M., J. J. S. Chiu, M. J. Lenardo. 1993. NF-B and Rel: participants in a multiform transcriptional regulatory system. Int. Rev. Cytol. 143:1.
Kopp, E. B., S. Ghosh. 1995. NF-B and rel proteins in innate immunity. Adv. Immunol. 58:1.
Schauer, S. L., Z. Wang, G. E. Sonenshein, T. L. Rothstein. 1996. Maintenance of nuclear factor-B/Rel and c-myc expression during CD40 ligand rescue of WEHI 231 early B cells from receptor-mediated apoptosis through modulation of IB proteins. J. Immunol. 157:81.
Schauer, S. L., R. E. Bellas, G. E. Sonenshein. 1998. Dominant signals leading to inhibitor B protein degradation mediate CD40 ligand rescue of WEHI231 immature B cells from receptor-mediated apoptosis. J. Immunol. 160:4398.
Phillips, R. J., S. Ghosh. 1997. Regulation of IB in WEHI 231 mature B cells. Mol. Cell. Biol. 17:4390.
Miyamoto, S., B. J. Seufzer, S. D. Shumway. 1998. Novel IB proteolytic pathway in WEHI231 immature B cells. Mol. Cell. Biol. 18:19.
Fields, E. R., B. J. Seufzer, E. M. Oltz, S. Miyamoto. 2000. A switch in distinct IB degradation mechanisms mediates constitutive NF-B activation in mature B cells. J. Immunol. 164:4762.
Doerre, S., R. B. Corley. 1999. Constitutive nuclear translocation of NF-B in B cells in the absence of IB degradation. J. Immunol. 163:267.
Simeonidis, S., S. Liang, G. Chen, D. Thanos. 1997. Cloning and functional characterization of mouse IB. Proc. Natl. Acad. Sci. USA 94:14372.
Whiteside, S. T., J.-C. Epinat, N. R. Rice, A. Israel. 1997. IB, a novel member of the IB family, controls RelA and cRel NF-B activity. EMBO J. 16:1413.
Rifkin, I. R., E. A. Leadbetter, B. C. Beaudette, C. Kiani, M. Monestier, M. J. Shlomchik, A. Marshak-Rothstein. 2000. Immune complexes present in the sera of autoimmune mice activate rheumatoid factor B cells. J. Immunol. 165:1626
Corley, R. B.. 1982. The regulation of T cell growth: requirements for the activation and replication of antigen-specific interleukin 2 producing T cells. Immunobiology 163:63.
Ovnic, M., R. B. Corley. 1987. Quantitation of cell surface molecules on a differentiating Ly-1+ B cell lymphoma. J. Immunol. 138:3075.
Rothstein, T. L., J. K. Wang, D. J. Panka, L. C. Foote, Z. Wang, B. Stanger, H. Cui, S. T. Ju, A. Marshak-Rothstein. 1995. Protection against Fas-dependent Th1-mediated apoptosis by antigen receptor engagement in B cells. Nature 374:163.
Hartley, J. W., T. N. Fredrickson, R. A. Yetter, M. Makino, H. C. Morse, 3d. 1989. Retrovirus-induced murine acquired immunodeficiency syndrome: natural history of infection and differing susceptibility of inbred mouse strains. J. Virol. 63:1223.
Brockman, J. A., D. C. Scherer, T. A. McKinsey, S. M. hall, X. Qi, W. Y. Lee, D. W. Ballard. 1995. Coupling of a signal response domain in IB to multiple pathways for NF-B activation. Mol. Cell. Biol. 15:2809.
Doerre, S., P. Sista, S.-C. Sun, D. W. Ballard, W. C. Greene. 1993. The c-rel protooncogene product represses NF-B p65-mediated transcriptional activation of the long terminal repeat of type 1 human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 90:1023.
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. 274:10689.
Kunsch, C., C. A. Rosen. 1993. NF-B subunit-specific regulation of the interleukin-8 promoter. Mol. Cell. Biol. 13:6137.
Wechsler, A. S., M. C. Gordon, U. Dendorfer, K. P. LeClair. 1994. Induction of IL-8 expression in T cells uses the CD28 costimulatory pathway. J. Immunol. 153:2515.
Schulte, R., G. A. Grassl, S. Preger, S. Fessele, C. A. Jacobi, M. Schaller, P. J. Nelson, I. B. Autenrieth. 2000. Yersinia enterocolitica invasin protein triggers IL-8 production in epithelial cells via activation of Rel p65-p65 homodimers. FASEB J. 14:1471.
King, L. B., R. B. Corley. 1990. Lipopolysaccharide and dexamethasone induce mouse mammary tumor proviral gene expression and differentiation in B lymphocytes through distinct regulatory pathways. Mol. Cell. Biol. 8:4211.
Lund, F. E., R. B. Corley. 1991. Regulated expression of mouse mammary tumor proviral genes in cells of the B lineage. J. Exp. Med. 174:1439.
Lund, F. E., T. D. Randall, D. L. Woodland, R. B. Corley. 1993. MHC class II limits the functional expression of endogenous superantigens in B cells. J. Immunol. 150:78.
Li, Z., G. J. Nabel. 1997. A new member of the IB protein family, IB, inhibits RelA (p65)-mediated NF-B transcription. Mol. Cell. Biol. 17:6184.
Grumont, R. J., I. B. Richardson, C. Gaff, S. Gerondakis. 1993. rel/NF-B nuclear complexes that bind B sites in the murine c-rel promoter are required for constitutive c-rel transcription in B-cells. Cell Growth Differ. 4:731.
Ten, R. M., C. V. Paya, N. Israel, O. Le Bail, M. G. Mattei, J. L. Virelizier, P. Kourlsky, A. Israel. 1992. The characterization of the promoter of the gene encoding the p50 subunit of NF-B indicates that it participates in its own regulation. EMBO J. 11:195.
King, L. B., R. B. Corley. 1989. Characterization of a pre-secretory phase in B cell differentiation. Proc. Natl. Acad. Sci. USA 86:2814.
Randall, T. D., F. E. Lund, J. W. Brewer, C. Aldridge, R. Wall, R. B. Corley. 1993. IL-5 and IL-6 define two molecularly distinct pathways of B cell differentiation. Mol. Cell. Biol. 13:3929.
Hoffmann, A., A. Levchenko, M. L. Scott, D. Baltimore. 2002. The IB-NF-B signaling module: temporal control and selective gene activation. Science 298:1241.(Stefan Doerre, Kristin Pe)
The levels and stability of IB have been examined in unstimulated and stimulated splenic B cells and compared with that of IB and IB. Primary murine splenic B cells but not T cells were found to contain high levels of IB protein, equivalent to levels of the abundant IB. Most agents that activate IB and IB degradation do not induce rapid degradation of IB. Interestingly, however, the levels of IB, but not of IB or IB, are dramatically reduced upon the stimulation of B cells both in vivo and in vitro. Since IB exhibits substrate specificity for NF-B Rel homodimers, this suggested the possibility that changes in NF-B-responsive genes might also occur during this transition. Consistent with this hypothesis, we found that a NF-B reporter construct sensitive to p65/RelA homodimers is activated at the time that IB levels decline following B cell stimulation. In IgG+ B cell lines, which contain low levels of IB, this same reporter construct was inactive, suggesting that the increases in Rel homodimer activity that accompany B cell stimulation are transient. However, there are differences in the level of expression of NF-B-responsive genes in these IgG+ B cell lines compared with their IgM+ counterparts. From these data, we conclude that there are transient changes in NF-B activity due to reductions in IB, which might contribute to long-term, persistent changes that accompany B cell differentiation. We propose an important role for IB in the differential regulation of nuclear NF-B activity in stimulated B cells.
Introduction
The NF-B/Rel family of proteins forms homo- and heterodimeric complexes that play a major role in controlling the expression of genes involved in immune, inflammatory and acute phase responses (1, 2, 3, 4, 5). NF-B activity is primarily regulated through nuclear translocation. In their inactive form, the complexes are sequestered in the cytoplasm, bound by members of the IB family of NF-B inhibitor proteins. Activation occurs in response to a diverse array of stimuli that lead to rapid degradation of IB followed by translocation of NF-B to the nucleus. The proteasomal degradation of IB proteins is triggered by the phosphorylation of two highly conserved serine residues within their N-terminal domains conducted by a high molecular mass IB kinase complex (IKK)3 (5, 6, 7, 8). The IB proteins identified to date that are the targets of this pathway include IB, IB, and IB, the members of the IB family that control the transcriptionally active NF-B complexes.
Although NF-B activation is constitutive in B lymphocytes, much of the NF-B remains sequestered in the cytoplasm, allowing for significant increases of nuclear NF-B upon B cell stimulation by agents such as LPS, CD40L, or anti-IgM Abs (9, 10, 11, 12, 13, 14, 15). The genes regulated by NF-B play important roles in B cell development, differentiation, and function (4, 11, 16, 17, 18, 19, 20, 21, 22, 23, 24). However, despite the importance of NF-B to B cells, little is understood about the mechanism(s) of constitutive and inducible NF-B activation in these cells or the role of individual IB proteins in regulating NF-B activity. Although the continual degradation of IB and/or IB is generally believed to be the mechanism for constitutive NF-B activation in IgM+ B cell lines (25, 26, 27, 28), studies from our laboratory have suggested that the activation of NF-B might be a dynamic process incorporating distinct mechanisms, depending on the B cell phenotype. We found that IgG+ B cell lines, which contain nuclear NF-B, do not display accelerated degradation of IB proteins, in striking contrast with their IgM+ counterparts, and have significantly reduced levels of IB (29). IB appears to differ from other IB proteins in two important ways. First, IB functions predominantly in the cytoplasm to sequester p65/RelA and/or cRel homodimers and is relatively inefficient at inhibiting transcription of genes regulated by p50/p65 heterodimers (30, 31). IB has been suggested to play an important role in the transient activation of a subset of genes regulated by Rel homodimers (31). Second, IB does not contain a C-terminal PEST sequence and therefore its expression may be regulated by mechanisms other than by its rapid degradation following activation.
Given the differences between IB and other IB proteins, we reasoned that IB might be an important regulator of NF-B/Rel activity in B lymphocytes capable of controlling, transiently or persistently, changes in gene expression in these cells. To address this issue, we examined IB levels in normal B cells at different stages of differentiation, determined whether transcriptional or posttranscriptional regulation was responsible for changes in abundance of IB, and examined the consequences of these changes on the expression of NF-B-responsive genes. Our results indicate that the regulation of IB expression and degradation is distinct from that of other IB proteins and suggest an important role for IB in the differential regulation of nuclear NF-B activity in B cells during differentiation.
Materials and Methods
B and T cell separation and culture
Single-cell suspensions were prepared from thymus or spleen of C57BL/6 (B6) mice (purchased from The Jackson Laboratory). B cell-enriched populations were prepared by depletion of T cells by a mixture of cytotoxic Abs, including anti-CD4 (GK1.5), anti-CD8 (53.6.72), and anti-Thy-1.2 (13-4), followed by rabbit complement as described previously (32). The resulting cells were >85% B cells with some monocyte contamination as assessed by staining with anti-B220 and anti-Mac1 Abs (BD Pharmingen). To further enrich B cells, cells from this population (2.5 x 106/ml) were cultured overnight in the presence of IL-4 (25 ng/ml; R&D Systems); IL-4 has been shown to have no effect on nuclear NF-B activity (20). The nonadherent cells were recovered and found to be >95% B220+ B cells and are referred to as "B – selection" (see Fig. 1). B cells were also enriched by positive selection by using anti-B220-coated paramagnetic beads (Miltenyi Biotec) in the presence of Fc block (BD Pharmingen) and then column purified according to the manufacturer’s protocol at 4°C ("B + selection"). This method yielded B cell preparations from normal spleen that were found to be 94–98% B220+, 79–88% IgM+, 9–18% IgG+, and 1% CD69+. T cells were enriched from spleen cells as described using nylon wool filtration (33), resulting in populations that were 90–95% T cells as assessed by staining with anti-CD3 and essentially devoid of B cells as assessed by B220 staining.
FIGURE 1. IB protein is expressed at high levels in primary B cells. A, Left panel, Western blots were prepared from whole cell extracts of murine thymocytes (Thymus), splenocytes (Spleen), T cell depleted splenocytes (T-depleted), or enriched splenic T cells (T). Right panel, Western blots were prepared from whole cell extracts of splenocytes (Spleen), B cells purified by T cell depletion and monocyte adhesion (B –selected), and B cells prepared by positive selection on paramagnetic beads coated with anti-B220 Ab (B + selected). The same extract from splenic T cells and an extract from an IgM+ B cell line (CH27) were used for comparison. Blots were sequentially probed with Abs raised against IB, IB, and calnexin as loading control. B, Left panel, Western blot containing a purified primary B cell extract and 1, 2, or 4 μl of COS extracts after transfection with recombinant FLAG-tagged murine IB (rmIB) or recombinant FLAG-tagged human IB (rhIB) was probed simultaneously with anti-IB and anti-IB (sc-7155) Abs. Right panel, Western blot prepared with 1 μl of the respective COS extracts containing IB or IB was probed with an anti-FLAG Ab.
All cells were cultured in complete DMEM containing 7% FBS (Invitrogen Life Technologies) and 50 μM 2-ME (Sigma-Aldrich) and supplemented as previously described (34). In some experiments, cells (2 x 106/ml) were treated with cycloheximide (CHX, 50 μg/ml; Sigma-Aldrich) or stimulated with LPS (50 μg/ml, Escherichia coli 055:B5; Difco), CD40L, or CD40L plus anti-IgM Ab (10 μg/ml, F(ab')2 goat anti-mouse; Jackson ImmunoResearch Laboratories) for various periods of time as described in Results. CD40L stimulation was conducted using a CD40L-CD8 fusion protein in combination with an anti-CD8 mAb for cross-linking (both reagents kindly provided by A. Marshak-Rothstein, Boston University School of Medicine, Boston, MA) as previously described (35). For studies to determine the half-life of IB proteins, LPS and anti-IgM treatments were initiated 2 h before the addition of CHX, while CD40L was added simultaneously with the CHX, because of the rapid induction of IB degradation following CD40L addition.
Infection of mice with LP-BM5 virus
Mice were inoculated i.p. with 0.1 ml of LP-BM5 murine leukemia virus pools at 6–8 wk of age, and the progression of disease was assessed as previously described (36). Mice were sacrificed at 4, 8, or 12 wk after infection, spleen weights were determined, and protein extracts were prepared as described below. Purified B cells were obtained by positive selection from spleens 4 or 8 wk after infection using paramagnetic beads as described above. The phenotypic characteristics of these B cells are described in Results. Mice were cared for and handled at all times in accordance with National Institutes of Health and institutional guidelines.
Preparation of protein extracts, phosphatase treatment, and immunoblotting
Whole cell extracts were prepared exactly as previously described (29), except in one experiment in which spleen cells from mice 12 wk after BM5 infection were analyzed. Spleen cells from these mice (or uninfected control mice) were prepared by washing splenocytes in 1x PBS (pH 7.4), followed by lysis with 4 pellet volumes of a buffer containing 10 mM HEPES (pH 7.9), 1.5 mM MgCl2, 10 mM KCl, 0.5% Nonidet P-40, 0.5 mM DTT, 30 μg/ml leupeptin, and 1 mM benzamidine. Nuclei were removed by centrifuging for 20 min at 1600 x g at 4°C and supernatants were collected as cytoplasmic extracts. All protein concentrations were determined by Bradford assay (Bio-Rad).
Lysates from equivalent amounts of protein (as indicated in the figure legends) or equivalent cell numbers (for CHX experiments) were fractionated on 10% SDS-PAGE, gels were transferred to nitrocellulose membranes (Micron Separations), and the membranes were stained with Ponceau S (Sigma-Aldrich) to ensure equivalent loading and transfer. Membranes were probed with the appropriate Abs and developed with the Renaissance System (New England Nuclear). Primary Abs included rabbit Abs against IB (sc-371), IB (sc-945), and IB (sc-7155 and sc-7156; identical results were obtained with both Abs), all obtained from Santa Cruz Biotechnology, anti-FLAG (M2; Eastman Kodak), and rabbit anti-calnexin (StressGen Biotechnologies) which served as an additional loading control. Autoradiographs were quantified using a Molecular Dynamics densitometer with ImageQUANT software (Molecular Dynamics).
Whole cell extracts were prepared in the presence of phosphatase inhibitors, 50 mM sodium fluoride and 1 mM sodium orthovanadate. However, in experiments designed to evaluate the phosphorylation status of IB protein, extracts were prepared in the absence of phosphatase inhibitors. Extracts were divided in two and treated with phosphatase (20 U/μl extract; New England Biolabs) for 30 min at 30°C, with or without the addition of sodium fluoride and sodium orthovanadate, according to the manufacturer’s instructions. Reactions were stopped by addition of 2x sample buffer (100 mM Tris (pH 6.8), 4% SDS, 20% glycerol, 0.1% bromphenol blue, and 700 mM 2-ME) and analyzed by SDS-PAGE, along with a sample from untreated cells.
The stability of IB proteins was determined following CHX treatment of cells as described elsewhere (29). Cells were cultured in the presence of CHX (50 μg/ml) for various periods of time, extracts were prepared, and IB expression was determined by Western blot analysis. This method yields results identical to those of conventional pulse-chase analysis (29).
Expression vectors and transfections
The IB expression vector containing a 5' FLAG tag followed by the complete coding region of human IB under the control of a CMV promoter was obtained from Dr. D. Ballard (Vanderbilt University, Nashville, TN) (37). The IB expression vector (pC3 FLAG-IB) was constructed by attaching a 5' FLAG tag sequence (Eastman Kodak) to the 1.1-kb coding region of murine IB via PCR using a plasmid containing the 2.2-kb IB cDNA as template (pC3 IB; obtained from D. Thanos, Columbia University, New York, NY) (30). COS cells were grown in 10-cm dishes in IMDM (Invitrogen Life Technologies) supplemented with 7% FBS, antibiotic/antimycotic solution (Invitrogen Life Technologies), and 2 mM glutamine. Cells were transfected with expression vectors for IB and IB using DEAE-dextran (Amersham/Pharmacia) as previously described (38).
A20 and M12 cells were transfected using DEAE-dextran (Amersham/Pharmacia) with the following luciferase reporter constructs: an ELAM luciferase reporter containing 3 B sites (obtained from D. Golenbock, University of Massachusetts Medical Center, Worcester, MA) (39); an IL-8 promoter construct (generously provided by Dr. K. LeClair, Antigenics, Inc., Woburn, MA), which contains a single B site that responds only to Rel homodimers and not to conventional NF-B heterodimers (40, 41, 42), and a mutant IL-8 construct (mut 2) that has a nonfunctional B site. A -galactosidase reporter construct was used as an internal control for transfection efficiency (Clontech). Whole cell lysates were prepared 24–49 h following transfection as analyzed according to the luciferase detection assay kit (Promega).
RNA analysis
RNA was isolated from cell populations using TRIzol (Invitrogen Life Technologies) according to the manufacturer’s specifications or using Qiagen/lithium chloride/urea/mRNA. Northern blots were performed as previously described (43) and probed with a 1.1-kb IB fragment, comprising the entire coding region (30) (obtained from D. Thanos). Northern blots were also probed with NF-B1 (p50) and c-rel (both described in Ref.38) and Ig (clone sc33; Ref.44) and CHO-B (45) as a loading control. All probes were labeled by random priming.
For semiquantitative RT-PCR, RNA isolated from B cells or B cell lines was isolated using the RNeasy kit according to the manufacturer’s specifications (Qiagen). RNA samples were reversed transcribed and then subjected to varying numbers of cycles (21, 22, 23, 24, 25, 26, 27) of PCR using the Superscript One-Step RT-PCR System (Invitrogen Life Technologies). The products were then resolved on agarose gels and quantitated by densitometry (Ultraviolet Products BioImaging). For this analysis, the following forward and reverse primers (Invitrogen Life Technologies), respectively, were used: for IB, 5'-AGAGTGACTCTGGTTCTGTT-3' and 5'-GGCAGCCGCTTTGGGATG-3' or 5'-GCTATTCTGTTGCTTGGC-3' and 5'-GTACATCAATGTCAGCTC-3'; for IB, 5'-GGCCTGGACTCCATGAAG-3' and 5'-GGTCTGCGTCAAGACTGC-3'; and for calnexin, 5'-CCAAGCCTCTCATTGTTC-3' and 5'-CAGTCATCTGGCTTGACAG3'. The relative abundance of each product was normalized using calnexin as a standardization control.
Results
Splenic B lymphocytes contain high levels of IB
We first evaluated the expression of IB in normal lymphocyte populations, including splenic B cells. As shown in Fig. 1A (left panel), high levels of IB were present in the spleen, in agreement with previous studies (46). Because significantly lower levels of IB were found in the thymus, we compared the expression of IB in enriched populations of splenic B and T cells. IB expression was high in B cell-enriched populations from which T cells were depleted (T depleted) and low when B cells were removed (T). Similar differences were obtained by analyzing the steady-state distribution of IB mRNA in the various lymphocyte subpopulations by Northern blot analysis (data not shown). We also confirmed that the multiple IB bands detected were phosphoisoforms (data not shown), as expected from other studies (31).
To prove that normal splenic B cells had high levels of IB, we subjected these cells to further purification by T cell depletion and the removal of adherent cells (B – selected) or isolated by positive selection using paramagnetic beads (B + selected). Both populations contained >95% B cells as assessed by FACS. As shown in Fig. 1A (right panel), B cells isolated by either method contained significantly higher levels of IB protein than the enriched T cell population and comparable levels of IB to those found in the IgM+ B cell lymphoma CH27. Note that while IB is expressed in higher concentrations in B cells than in T cells, this is not true for other IB proteins, such as IB, which is expressed at similar levels in T and B cells (Fig. 1A). We conclude that B cells are the major source of IB within normal unstimulated splenic lymphocyte populations.
The results in Fig. 1A suggest that IB levels might be very high in B lymphocytes, perhaps as high as IB levels. To obtain a comparative estimate of the relative expression of IB and IB in splenic B cells, we used recombinant FLAG-tagged IB and IB from transfected COS cells to determine whether the Abs raised against IB and IB detected similar levels of proteins. A Western blot was prepared containing 20 μg of whole cell extract from purified splenic B cells and different amounts of extracts from COS cells transiently transfected with the FLAG-tagged IB or IB expression vectors. The blot was probed simultaneously with IB and IB Abs. As shown in Fig. 1B (left panel), the IB and IB signals obtained from the B cell extract corresponded to those obtained from 2 to 4 μl of the respective COS extract. Probing with an anti-FLAG Ab demonstrated that the IB and IB levels in the two COS extracts were similar (Fig. 1B, right panel). These results demonstrate that the level of expression of IB and IB can be compared from the Western blots. From these data, we conclude that IB is highly expressed in normal B cells at levels comparable to those of IB. Therefore, IB is not a "minor" IB protein in splenic B cells, but a major inhibitory protein regulating a significant proportion of NF-B activity.
Degradation of IB proteins in stimulated B cells
To compare the turnover of IB with that of other IB proteins in B cells, we stimulated primary B cells with agents that activate NF-B, including LPS, CD40L, anti-IgM Ab, or a combination of CD40L and anti-IgM (3, 25) in the presence of CHX. At various times after stimulation, whole cell extracts were prepared and analyzed by Western blot for IB, IB, and IB expression (Fig. 2A). The blots were reprobed with calnexin to monitor loading, the bands were analyzed by densitometry, the signal intensities normalized to the calnexin signal, plotted, and the half-lives were determined (Fig. 2B).
FIGURE 2. Half-lives of IB, IB, and IB in primary B cells after stimulation with B cell mitogens. A, Primary B cells were cultured overnight in IL-4 and subsequently stimulated with LPS, CD40L or a combination of CD40L and anti-IgM. CHX was then added to the cultures (see Materials and Methods) and whole cell extracts were prepared from equivalent cell numbers at 0, 30, 60, and 120 min after CHX addition. Extracts were analyzed by Western blot with the Abs indicated (calx, calnexin). Note that the gel in A, of extracts from cells stimulated with CD40L plus anti-IgM, was not run as far as the other three gels shown and therefore the different phosphoisoforms of IB are not resolved. B, Signals from autoradiographs shown in A were quantitated by densitometry. The intensities of the calnexin signals were used to correct for differences in loading and blotting efficiency. The normalized signal intensities of IB (?), IB (), and IB () (top four panels) were plotted and half-lives were calculated by linear regression; these values are shown in tabular form. The bottom left panel compares the degradation of IB induced by different stimuli.
IB, IB, and IB were relatively stable in untreated B cells with half-lives estimated to be between 90 and 130 min. The slow turnover is not surprising since nuclear NF-B levels are relatively low in resting B cells (11, 14). LPS stimulation led to a significant increase in IB and IB turnover, with half-lives estimated to be 30 and 45 min, respectively. In contrast, the turnover of IB was only slightly faster than in unstimulated B cells. When B cells were activated by CD40L, IB, and IB showed dramatic increases in turnover, with half-lives of only 10–15 min. Surprisingly, however, no increase in turnover of IB was detected in CD40L-stimulated B cells. Treatment with anti-IgM Abs to engage the BCR also had no effect on IB degradation (data not shown). When CD40L stimulation was combined with anti-IgM treatment, accelerated degradation of IB was observed, with a half-life of 45 min. Nevertheless, the turnover of IB following CD40L plus anti-IgM treatment was still 3- to 4-fold slower than the turnover of IB and IB in these same cells. Taken together, these results demonstrate a surprising and unprecedented specificity of NF-B-activating signals in primary B cells for IB and IB and a significantly reduced susceptibility of IB to degradation.
IB levels decrease in stimulated B cells
Although B cell stimuli do not induce rapid degradation of IB, previous studies from our laboratory have shown that levels of IB may be modulated in B cells since IgG+ B cell lines contain very low levels of IB compared with IgM+ B cell lines (29). We therefore sought to determine whether splenic B cells also contained low levels of IB following stimulation. Given the difficulties in obtaining populations of cells containing sufficient numbers of IgG+ cells from normal or immunized mice for Western blot analysis, even after immunization (see, for example, Ref.47), we used an in vivo model of B cell differentiation that makes use of the replication-defective LP-BM5 murine leukemia virus. This virus causes a progressive lymphoproliferation and immunodeficiency, also known as murine AIDS, that is characterized by B cell activation, proliferation, and differentiation, including Ig class switching (48, 49). The activation and differentiation of B cells in this model is dependent on CD4+ T cell stimulation and CD40-CD40L interactions (50, 51), and thus has many of the hallmarks of B cell activation by T-dependent Ags. Spleens of infected mice contain significantly increased numbers of IgG+ B cells (36, 52).
To determine whether levels of IB were lower in these B cells, we first compared the abundance of IB in cytoplasmic extracts from total splenocytes from control uninfected mice (Fig. 3, Ctrl.) and from three individual mice 12 wk after infection with the LP-BM5 virus. As shown in Fig. 3A, levels of IB were significantly lower in extracts from each of the three infected mice compared with uninfected controls. The decrease in IB expression was specific for IB, since IB levels (and IB levels, data not shown) remained largely unchanged. As a measure of B cell differentiation that accompanies infection, we compared levels of IgM and IgG in cell extracts. The decrease in IB was paralleled by an increase in IgG expression while IgM levels in the three mice postinfection were significantly reduced compared with their normal counterparts. Since B cells are the major source of IB in the spleen, these initial results suggested that levels of IB declined in the B cell population.
FIGURE 3. IB expression in primary splenocytes or purified splenic B cells after polyclonal B cell activation caused by LP-BM5 infection. A, Western blots of 20-μg cytoplasmic extracts from pooled splenocytes from normal, uninfected mice (Ctrl.) or from three mice 12 wk after infection with LP-BM5 virus (BM5 #1, 2, and 3) were sequentially probed with anti-IB, anti-IB, anti-IgG, or anti-IgM Abs. B, Western blots of 20-μg whole cell extracts from sorted splenic B cells from uninfected mice (Ctrl.) or from mice 4 or 8 wk after infection with LP-BM5 virus (BM5 4wk, 8wk) were sequentially probed with anti-IB, anti-IB, anti-IgG, or anti-calnexin (calx) Abs. For comparison, Western blots contained equivalent amounts of whole cell extracts from the IgM+ B cell line CH27 and the IgG+ B cell line M12. M12 expresses low levels of IgG (29 ) that is detectable on longer exposures of the blot (data not shown).
To prove that the B cells contained reduced levels of IB, B cells were isolated from the spleens of mice 4 or 8 wk after infection and IB levels in these B cells were compared with levels in cells from control, uninfected mice. FACS analysis confirmed that 94–98% of the isolated cells were B cells, as assessed by B220 staining. As shown in Fig. 3B, IB levels decreased significantly in B cells during the course of infection, such that there was almost as little IB in B cells 8 wk after infection as in the IgG+ B cell line, M12. In contrast, levels of IB were similar in all B cells. Interestingly, the decrease in IB levels preceded the increase in total IgG levels as determined by Western blot analysis (Fig. 3B) and surface IgG staining. FACS analysis revealed that IgG+ B cells increased from 3 to 8% in B cells from uninfected animals to 28–36% of the B cells from individual mice 8 wk after infection and continued to increase up to 12 wk when 30–54% of splenic B cells were IgG+. These data indicate that there appears to be a significant difference in the regulation of IB expression in differentiating B cells compared with primary splenic IgM+ B cells. However, it is not known from these data whether the changes in IB occur soon after activation of responding IgM+ B cells or at later times during their differentiation.
IgG+ B cells do not contain increased levels of functional Rel homodimers
Unlike other IB proteins, IB demonstrates substrate specificity for NF-B in vivo, in that it selectively retains Rel (p65/RelA and c-Rel) homodimers but not conventional p50/p65 heterodimers (Refs.30, 31 and our unpublished results). Given that IgG+ B cell lines have low levels of IB, it was possible that higher levels of transcriptionally active Rel homodimers might be present in these cells. To test this, we transfected the IgG+ B cell lines M12 and A20 with two different NF-B luciferase reporter constructs, one driven by the ELAM promoter containing three conventional B sites (39) and the second under the control of an IL-8 promoter containing a single B site. The IL-8 reporter provides a specific monitor for the presence of functional nuclear RelA homodimers (40, 41, 42). Cells were transfected with one of the NF-B reporter constructs along with a -galactosidase reporter construct as an internal control for transfection efficiency and normalization and luciferase activity was measured 24 h later. As shown in Fig. 4, the IL-8 promoter was not functional in either cell line. This was not due to the fact that the IL-8 promoter contains only a single B site and might therefore be insensitive to activation (see below). In contrast with the IL-8 promoter, the ELAM promoter was fully active in both IgG+ B cell lines, confirming the presence of functional NF-B. These results indicate that while conventional p50/p65 heterodimers are present in the nucleus of these cells, Rel homodimers are not, despite the low levels of IB. Consequently, if reductions in IB lead to increased Rel homodimer activity, these increases are likely to be transient rather than long-term, stable increases in this subset of NF-B complexes.
FIGURE 4. NF-B reporter activity in two IgG+ (M12 and A20) B cell lines. Cells (107) were transiently transfected with either the ELAM-luciferase or the IL-8-luciferase reporter constructs along with a -galactosidase reporter construct as a control for transfection efficiency. Whole cell lysates were prepared, promoter activity was measured (Promega) with a luminometer, and corrected for transfection efficiency.
Reductions in IB and increased Rel homodimer activity after B cell stimulation
The data in Fig. 3 suggested that IB levels decline before differentiation of stimulated B cells into IgG+ B cells. To determine when IB might decline during B cell differentiation, we monitored IB protein levels in purified splenic B cells at various times after stimulation with LPS or CD40L and anti-IgM. As shown in Fig. 5, IB levels declined precipitously within the first 24 h after stimulation with LPS. Levels of IB were calculated by densitometry and found to be only 2% of initial levels at this time. In contrast, IB levels were still 60% of control levels after 24 h of stimulation and never declined to the same extent as IB. Similar results were obtained in cells stimulated with CD40L and anti-IgM. In these cells, IB protein levels were reduced by 70 and 85% at 24 and 48 h of stimulation, respectively, while levels of IB were not reduced in the first 24 h and exhibited only a modest 30% reduction at 48 h of stimulation. Therefore, despite the fact that IB is not degraded nearly as rapidly as IB following stimulation (Fig. 2), levels of this protein decline more dramatically following stimulation, perhaps as the result of transcriptional changes. The sharp declines in IB levels suggest that selective and transient increases in nuclear Rel homodimer activity might be generated during this period of time.
FIGURE 5. Steady-state levels of IB and IB after stimulation of primary B cells. Purified splenic B cells were stimulated with either LPS (A) or CD40L and anti-IgM (B) for 72 h. Whole cell lysates were prepared at 0, 24, 48, and 72 h. Western blots of 15 μg of whole cell lysates were sequentially probed with anti-IB, anti-IB, or anti-calnexin (calx) Abs.
To investigate this possibility, we used the inducible IgM+ B cell line CH12-LBK, which serves as a useful model for the differentiation of IgM+ B cells (34, 53, 54). LPS stimulation resulted in significant reductions in steady-state levels of IB over a 48-h period, with reductions in both IB and a predominant phosphoisoform showing similar reductions of 50% in the first 24 h and almost 90% by 48 h after stimulation (Fig. 6, A and B). In contrast, IB levels remained essentially unchanged over the same period of time. To determine whether these changes led to increased in functional Rel homodimer activity, CH12 cells were transiently transfected with one of several NF-B reporter constructs, and reporter activity in unstimulated and LPS-stimulated cells was compared 24 h later. A representative example of eight independent experiments is shown in Fig. 6C. IL-8 reporter activity was significantly increased in LPS-stimulated compared with unstimulated cells, even though the overall NF-B activity was essentially unchanged, as monitored by the ELAM promoter. The mut 2 construct, in which the single IL-8 B site has been mutated (41), showed only limited activity, indicating that most of the increase on the IL-8 promoter was due to activation by NF-B, specifically Rel homodimers. As expected, cotransfection of an IB expression vector with the IL-8 reporter construct into LPS-stimulated B cells resulted in the total inhibition of luciferase activity (data not shown). This confirms that IB is capable of retaining Rel homodimers and inhibiting the activation of the IL-8 promoter in these cells. Although overexpression studies must be interpreted with caution, these results are consistent with the suggestion that the decline in IB levels following B cell stimulation is responsible for increases in nuclear Rel homodimer activity.
FIGURE 6. Increase in Rel homodimer activity accompanies decreases in IB following LPS stimulation. A, Western blots of whole cell lysates from the IgM+ B cell line CH12 LBK after stimulation with LPS. Blots were sequentially probed with anti-IB, anti-IB, or anti-calnexin (calx) Abs. B, Signals from autoradiographs shown in A were quantitated by densitometry. The intensities of the calnexin signals were used to correct for differences in loading and blotting efficiency. These values were then plotted on a log scale (, IB; , IB; ?, IB-p). C, Normalized luciferase reporter activity in CH12LBK IgM+ B cells. Cells were stimulated with LPS for 24 h () or left untreated (). The cells were then transiently transfected with either the ELAM-luciferase (with three B sites) or IL-8-luciferase (with a single B site) reporter constructs. mut 2 is a mutant IL-8 construct in which the B site has been mutated. Cells were also cotransfected with a -galactosidase reporter construct as a control for transfection efficiency. Whole cell lysates were prepared 24 h following transfection and luciferase activity was measured.
Selective reductions in IB mRNA and changes in gene expression following B cell stimulation
We next investigated the basis for the selective reductions in IB expression. Although IB does not degrade as rapidly as IB in response to B cell stimulation (Fig. 2), IB levels are reduced much more dramatically after stimulation than IB (Figs. 3 and 5). Although IB expression, like that of IB, has been reported to be regulated by NF-B (31), we considered the possibility that IB might be regulated at the RNA level. RNA from the inducible B cell line CH12-LBK or from primary B cells were subjected to semiquantitative RT-PCR. As shown in Fig. 7A, levels of IB mRNA declined by 50% following stimulation of CH12-LBK cells. These data were corroborated by results of DNA microarray data (Affymetrix GeneChip), which consistently showed a 50% reduction in IB levels following LPS stimulation (data not shown). In contrast, IB levels remained steady over the course of the experiment. Similar results were obtained using RNA from primary B cells stimulated with LPS (Fig. 7B). IB mRNA levels declined by at least 50% within 48 h after stimulation and fell by 80% or more by 72 h after stimulation. In contrast, IB levels increased during the first 24–48 h and then returned to baseline levels.
FIGURE 7. Selective decrease in IB mRNA levels in stimulated B cells. A, Semiquantitative RT-PCR of total RNA from the IgM+ B cell line CH12-LBK after LPS stimulation. RNA was isolated at various time points following LPS stimulation and subjected to increasing rounds of RT-PCR (1, 2, and 3 represent 21, 24, and 27 cycles, respectively). B, Semiquantitative RT-PCR of total RNA from primary B cells. RNA was isolated at various time points following LPS stimulation and subjected to increasing rounds of RT-PCR as in A. C, Northern blot analysis of NF-B-dependent genes from two IgM+ (WEHI231 and CH27) and two IgG+ (A20 and M12) B cell lines. Twenty micrograms of total RNA was separated on 1.2% denaturing agarose gels, transferred to nitrocellulose membrane, and sequentially probed with 32P-labeled cDNA probes for IB, NF-B1 (p50), c-rel, Ig, and CHO B as a loading control.
To determine whether reductions in IB expression in stimulated B cells might result in selective changes in the expression of a subset of NF-B-regulated genes, mRNA expression levels of four genes known to be regulated by NF-B were compared in IgM+ (WEHI231 and CH27) and IgG+ (A20 and M12) B cell lines. These included c-Rel (55), NF-B1 (56), and Ig L chain (10) in addition to IB. As shown in Fig. 7C, IB mRNA was abundant in IgM+ B cells, but present at much lower levels in the two IgG+ B cells. However, not all NF-B target genes showed similar reductions. The levels of c-Rel and Ig were not reduced in IgG+ B cells relative to their IgM+ counterparts, while NF-B1 levels were reduced, albeit to a lesser degree than IB. These data indicate that the decreased expression of IB in IgG+ B cells is not simply a reflection of overall reduced NF-B transactivation potential, but instead may reflect a selective reduction in a specific subset of NF-B-regulated genes.
Discussion
Our data are consistent with a role for IB in the differential modulation of NF-B-responsive genes during B cell differentiation. IB is expressed at very high levels in IgM+ B cells, levels comparable to IB. Levels of IB decline significantly following B cell activation to levels much lower than IB. However, in contrast to IB, whose levels remain high at all stages of B cell differentiation, IB levels may remain low, since little IB (mRNA or protein) is expressed in IgG+ B cell lines, and IB levels are low in chronically stimulated splenic B cells. This reduction may have the potential for profoundly altering the nature of the NF-B target genes activated following B cell stimulation.
Hoffmann et al. (57) demonstrated that IB (and IB) function to normalize NF-B activity in responding cells that would otherwise be extinguished by the re-expression of IB. These investigators also predicted that this would result in changes in specific gene expression in cells in which NF-B was transiently activated. Whiteside et al. (31) also suggested that IB might regulate a distinct subset of NF-B-responsive genes, based in part on its substrate specificity. IB, like IB and IB, retains NF-B complexes containing c-Rel and RelA, the transcriptionally active NF-B/Rel proteins in the cytoplasm (30, 31, 46). However, unlike IB and IB, IB shows substrate specificity when expressed at normal levels in vivo and has strong binding affinity for c-Rel and RelA homodimers (Refs.30, 31 ; K. M. Daley and R. B. Corley, unpublished results), IB does not inhibit the more common heterodimeric complexes formed with NF-B p50 except at high concentrations (30). Rel homodimers are known to have specificity for certain promoters (40, 41, 42). During the differentiation of B cells, at a time that IB levels first decline (Figs. 5 and 6), promoters sensitive to Rel homodimers would expected to be stimulated, as demonstrated by activity on the IL-8 promoter (Fig. 6). Interestingly, in IgG+ B cells, there is no increase in Rel homodimer activity (Fig. 4), suggesting that this change is transient and results in a new steady-state level of NF-B. It is possible that this initial transition is required for the changes in the levels of NF-B-responsive genes that emerge in IgM+ and IgG+ B cells (Fig. 7), but additional studies linking these events will be required. Nevertheless, we conclude from these data that, in addition to the transient changes in NF-B activity observed within the first 1–2 days after B cell stimulation, long-term, persistent changes also occur, as evidenced by the decreased expression of some NF-B-responsive genes, including IB itself.
Although our current results do not establish a cause and effect relationship between the decline in IB and persistent changes in NF-B activity, the results of Memet et al. (58) are consistent with this hypothesis. They generated IB-deficient mice using a lacZ reporter that was included within a targeting construct 3' of the IB 5' regulatory sequences. By analyzing -galactosidase activity in cells from mutant mice, these authors found that very few B cells were positive, while T cells were positive. They concluded that IB was not highly expressed in B cells. However, direct analysis of IB expression shows that it is highly expressed in primary IgM+ B cells, not only based on the results of the current studies, but also from previous studies of IgM+ B cell lines (29, 46) and from gene expression profiling analysis on peripheral B cells (59). Thus, we suggest another interpretation of the results of Memet et al. (58): in the absence of IB, a different steady-state level of NF-B-responsive genes is expressed in B cells, one that does not induce high levels of IB transcription, but maintains high levels of IB, among other NF-B-responsive genes.
In B cells, IB is not subject to rapid degradation by stimuli that otherwise result in the degradation of IB and IB (Fig. 2). The differential regulation of IB protein degradation is not unprecedented. Like IB, IB has also been reported to be susceptible to a subset of activating signals in certain cells (60). In previous studies, IB was found to be of intermediate susceptibility to proteasomal degradation in a monocytic cell line in response to NF-B-activating signals such as LPS, being degraded faster than IB but less rapidly than IB (31). Surprisingly, IB is not degraded nearly as rapidly as IB in B cells in response to LPS (Fig. 2). This suggests that the stimulation of degradation of the different IB proteins not only depends on the stimulus, but exhibits cell-type specificity as well. Thus, the degree to which IB proteins are susceptible to degradation is not solely an intrinsic property of these proteins, but can be regulated depending on factors modified by the signal(s) given or on factors unique to particular cells. This difference could take place at the level of the IKK complexes (5, 61), for example, through the presence or absence of adaptor molecules or at the level of the IB, for example by a specific phosphorylation event that would render it either unsuitable as IKK substrate or that would prevent proteasomal degradation. In any event, our results suggest that IB is an ideal inhibitory protein for differentially regulating nuclear NF-B activity in B cells. In B cells, unlike IB and IB, IB is not subject to rapid degradation by stimuli that otherwise result in the degradation of IB and IB (Fig. 2). However, IB protein levels do decline after stimulation (Figs. 3, 5, and 6), but this might result from a combination of slow degradation and reductions in IB transcript levels. The slow degradation rates of IB following LPS stimulation are still sufficient to account for the significant reductions in IB levels 24 h later, assuming the abundance of IB mRNA is reduced as well.
Changes in IB levels results in the release of functional Rel homodimers to the nucleus, as evidenced by the activation of the IL-8 promoter within 48 h of stimulation of IgM+ B cells. This change in Rel homodimer activity appears to be a transient event because this same promoter is not active in IgG+ B cells, in which IB levels are low. Whether the genes activated following declines in the levels of IB contribute to isotype switching is unknown, but they are clearly insufficient based on the results of Strober and colleagues (62). They found that NF-B activated via CD40-mediated induction of IB degradation by the IKK complex was absolutely essential for isotype switching. Nevertheless, the transient promoter activation initiated by reductions in IB suggests that whatever subset of genes might be activated in response to Rel homodimers are likely to be induced only briefly. These results therefore support the prediction of Israel and colleagues (31) who suggested an important role for IB in the transient regulation of a subset of late genes. The transient activation of this set of genes is likely to result in the stable expression of a new subset of genes in activated B cells, including those regulated by NF-B.
Acknowledgments
We thank Lia Luus for technical assistance; Dr. Herbert Morse (National Institute of Allergy and Infectious Diseases, Bethesda, MD) for LP-BM5-infected mice; Dr. Ann Marshak-Rothstein (Boston University School of Medicine, Boston, MA) for the reagents for T cell depletion and the CD40L-CD8 fusion protein and anti-CD8 Ab; Dr. Dimitris Thanos (Columbia University, New York, NY) for the IB plasmid; Dr. Dean Ballard (Vanderbilt University, Nashville, TN) for the FLAG-tagged IB expression construct; D. Golenbock for the ELAM reporter construct; and Dr. Kenneth LeClair (Antigenics, Inc., Woburn, MA) for the IL-8-luciferase reporter constructs.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants AI31209 and CA36642 (to R.B.C.).
2 Address correspondence and reprint requests to Dr. Ronald B. Corley, Department of Microbiology, 715 Albany Street, L504, Boston University School of Medicine, Boston, MA 02118. E-mail address: rbcorley@bu.edu
3 Abbreviations used in this paper: IKK, IB kinase complex; CHX, cycloheximide; CHO, Chinese hamster ovary.
Received for publication November 14, 2002. Accepted for publication November 5, 2004.
References
Baeuerle, P. A., T. Henkel. 1994. Function and activation of NF-B in the immune system. Annu. Rev. Immunol. 12:141
Baldwin, A. S., Jr. 1996. The NF-B and IB proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649.
Ghosh, S., M. J. May, E. B. Kopp. 1998. NF-B and rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16:225.
Sha, W. C.. 1998. Regulation of immune responses by NF-B/Rel transcription factors. J. Exp. Med. 187:143
Karin, M., Y. Ben-Neriah. 2000. Phosphorylation meets ubiquitination: the control of NF-B activity. Annu. Rev. Immunol. 18:621.
Stancovski, I., D. Baltimore. 1997. NF-B activation: the IB kinase revealed?. Cell 91:299.
May, M. J., S. Ghosh. 1999. IB kinases: kinsmen with different crafts. Science 284:271.
Israel, A.. 2000. The IKK complex: an integrator of all signals that activate NF-B?. Trends Cell Biol. 10:129.
Sen, R., D. Baltimore. 1986. Inducibility of immunoglobulin enhancer binding protein NF-B by a posttranslational mechanism. Cell 47:921.
Sen, R., D. Baltimore. 1986. Multiple nuclear factors interact with the immunoglobulin enhancer sequences. Cell 46:705.
Liou, H.-C., D. Baltimore. 1993. Regulation of the NF-B/rel transcription factor and IB inhibitor system. Curr. Opin. Cell Biol. 5:477.
Miyamoto, S., M. J. Schmitt, I. M. Verma. 1994. Qualitative changes in the subunit composition of B-binding complexes during murine B-cell differentiation. Proc. Natl. Acad. Sci. USA 91:5056.
Grumont, R. J., S. Gerondakis. 1994. The subunit composition of NF-B complexes changes during B-cell development. Cell Growth Differ. 5:1321.
Francis, D. A., J. G. Karras, X.-y. Ke, R. Sen, T. L. Rothstein. 1995. Induction of the transcription factors NF-B, AP-1 and NF-AT during B cell stimulation through the CD40 receptor. Int. Immunol. 7:151.
Neumann, M., G. Wohlleben, S. Chuvpilo, B. Kistler, T. Wirth, E. Serfling, A. Schimpl. 1996. CD40, but not lipopolysaccharide and anti-IgM stimulation of primary B lymphocytes, leads to persistent nuclear accumulation of RelB. J. Immunol. 157:4862.
Grilli, M., J. J. S. Chiu, M. J. Lenardo. 1993. NF-B and Rel: participants in a multiform transcriptional regulatory system. Int. Rev. Cytol. 143:1.
Kopp, E. B., S. Ghosh. 1995. NF-B and rel proteins in innate immunity. Adv. Immunol. 58:1.
Schauer, S. L., Z. Wang, G. E. Sonenshein, T. L. Rothstein. 1996. Maintenance of nuclear factor-B/Rel and c-myc expression during CD40 ligand rescue of WEHI 231 early B cells from receptor-mediated apoptosis through modulation of IB proteins. J. Immunol. 157:81.
Schauer, S. L., R. E. Bellas, G. E. Sonenshein. 1998. Dominant signals leading to inhibitor B protein degradation mediate CD40 ligand rescue of WEHI231 immature B cells from receptor-mediated apoptosis. J. Immunol. 160:4398.
Phillips, R. J., S. Ghosh. 1997. Regulation of IB in WEHI 231 mature B cells. Mol. Cell. Biol. 17:4390.
Miyamoto, S., B. J. Seufzer, S. D. Shumway. 1998. Novel IB proteolytic pathway in WEHI231 immature B cells. Mol. Cell. Biol. 18:19.
Fields, E. R., B. J. Seufzer, E. M. Oltz, S. Miyamoto. 2000. A switch in distinct IB degradation mechanisms mediates constitutive NF-B activation in mature B cells. J. Immunol. 164:4762.
Doerre, S., R. B. Corley. 1999. Constitutive nuclear translocation of NF-B in B cells in the absence of IB degradation. J. Immunol. 163:267.
Simeonidis, S., S. Liang, G. Chen, D. Thanos. 1997. Cloning and functional characterization of mouse IB. Proc. Natl. Acad. Sci. USA 94:14372.
Whiteside, S. T., J.-C. Epinat, N. R. Rice, A. Israel. 1997. IB, a novel member of the IB family, controls RelA and cRel NF-B activity. EMBO J. 16:1413.
Rifkin, I. R., E. A. Leadbetter, B. C. Beaudette, C. Kiani, M. Monestier, M. J. Shlomchik, A. Marshak-Rothstein. 2000. Immune complexes present in the sera of autoimmune mice activate rheumatoid factor B cells. J. Immunol. 165:1626
Corley, R. B.. 1982. The regulation of T cell growth: requirements for the activation and replication of antigen-specific interleukin 2 producing T cells. Immunobiology 163:63.
Ovnic, M., R. B. Corley. 1987. Quantitation of cell surface molecules on a differentiating Ly-1+ B cell lymphoma. J. Immunol. 138:3075.
Rothstein, T. L., J. K. Wang, D. J. Panka, L. C. Foote, Z. Wang, B. Stanger, H. Cui, S. T. Ju, A. Marshak-Rothstein. 1995. Protection against Fas-dependent Th1-mediated apoptosis by antigen receptor engagement in B cells. Nature 374:163.
Hartley, J. W., T. N. Fredrickson, R. A. Yetter, M. Makino, H. C. Morse, 3d. 1989. Retrovirus-induced murine acquired immunodeficiency syndrome: natural history of infection and differing susceptibility of inbred mouse strains. J. Virol. 63:1223.
Brockman, J. A., D. C. Scherer, T. A. McKinsey, S. M. hall, X. Qi, W. Y. Lee, D. W. Ballard. 1995. Coupling of a signal response domain in IB to multiple pathways for NF-B activation. Mol. Cell. Biol. 15:2809.
Doerre, S., P. Sista, S.-C. Sun, D. W. Ballard, W. C. Greene. 1993. The c-rel protooncogene product represses NF-B p65-mediated transcriptional activation of the long terminal repeat of type 1 human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 90:1023.
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. 274:10689.
Kunsch, C., C. A. Rosen. 1993. NF-B subunit-specific regulation of the interleukin-8 promoter. Mol. Cell. Biol. 13:6137.
Wechsler, A. S., M. C. Gordon, U. Dendorfer, K. P. LeClair. 1994. Induction of IL-8 expression in T cells uses the CD28 costimulatory pathway. J. Immunol. 153:2515.
Schulte, R., G. A. Grassl, S. Preger, S. Fessele, C. A. Jacobi, M. Schaller, P. J. Nelson, I. B. Autenrieth. 2000. Yersinia enterocolitica invasin protein triggers IL-8 production in epithelial cells via activation of Rel p65-p65 homodimers. FASEB J. 14:1471.
King, L. B., R. B. Corley. 1990. Lipopolysaccharide and dexamethasone induce mouse mammary tumor proviral gene expression and differentiation in B lymphocytes through distinct regulatory pathways. Mol. Cell. Biol. 8:4211.
Lund, F. E., R. B. Corley. 1991. Regulated expression of mouse mammary tumor proviral genes in cells of the B lineage. J. Exp. Med. 174:1439.
Lund, F. E., T. D. Randall, D. L. Woodland, R. B. Corley. 1993. MHC class II limits the functional expression of endogenous superantigens in B cells. J. Immunol. 150:78.
Li, Z., G. J. Nabel. 1997. A new member of the IB protein family, IB, inhibits RelA (p65)-mediated NF-B transcription. Mol. Cell. Biol. 17:6184.
Grumont, R. J., I. B. Richardson, C. Gaff, S. Gerondakis. 1993. rel/NF-B nuclear complexes that bind B sites in the murine c-rel promoter are required for constitutive c-rel transcription in B-cells. Cell Growth Differ. 4:731.
Ten, R. M., C. V. Paya, N. Israel, O. Le Bail, M. G. Mattei, J. L. Virelizier, P. Kourlsky, A. Israel. 1992. The characterization of the promoter of the gene encoding the p50 subunit of NF-B indicates that it participates in its own regulation. EMBO J. 11:195.
King, L. B., R. B. Corley. 1989. Characterization of a pre-secretory phase in B cell differentiation. Proc. Natl. Acad. Sci. USA 86:2814.
Randall, T. D., F. E. Lund, J. W. Brewer, C. Aldridge, R. Wall, R. B. Corley. 1993. IL-5 and IL-6 define two molecularly distinct pathways of B cell differentiation. Mol. Cell. Biol. 13:3929.
Hoffmann, A., A. Levchenko, M. L. Scott, D. Baltimore. 2002. The IB-NF-B signaling module: temporal control and selective gene activation. Science 298:1241.(Stefan Doerre, Kristin Pe)