Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Activation of the transcription factor NF-B has been intensely studied in recent years and has received considerable attention as a paradigmatic signaling pathway. The versatility of the NF-B transcription factors is underscored by the distinct types of NF-B-activating agents as well as the great number of NF-B-regulated genes whose products affect such broad cellular processes as apoptosis, immune response, inflammation, cell adhesion, and the cell cycle (Ghosh et al., 1998). Studies using genetic and biochemical approaches have identified several components of a signaling network that ultimately direct the degradation of NF-B inhibitory proteins (IBs), a prerequisite of NF-B activation. In a majority of cell types NF-B is kept inactive as a cytoplasmic complex with an IB family member until an activating signal allows for the liberation and nuclear translocation of NF-B. However, many instances in which NF-B is constitutively activated have also been described. Such activation of NF-B in the apparent absence of an inducing stimulus is generally attributable to either 1) the deregulation of one or more components in the signaling pathway, or 2) the normal developmental program of a cell. The deregulation of Rel/NF-B or IB protein activity has been implicated in aberrant cell growth, cell death, and oncogenesis (Luque and Gelinas, 1997). It is now known that many lymphoid malignancies as well as solid tumors display constitutively nuclear NF-B activity, although the molecular mechanisms responsible for this activity are not altogether clear. Furthermore, as a part of their transformation process, the Epstein-Barr virus, the human T-cell leukemia virus, and hepatitis B virus constitutively activate NF-B (Mosialos, 1997). Although the transforming abilities of NF-B activity await further clarification, it is clear that activated NF-B can have profound effects on a cell's tendency to undergo apoptosis (Beg and Baltimore, 1996; Van Antwerp et al., 1996; Wang et al., 1996).

Constitutive activity of NF-B may also be established and maintained as a part of normal cellular development. Most notably, both B and T lymphocytes contain constitutively activated NF-B at distinct stages during their development (Sen and Baltimore, 1986). In fact, NF-B was first characterized based on its presence in the nucleus of several unstimulated B-cell lines (Sen and Baltimore, 1986). Within the context of B cells, several roles for constitutive NF-B activity have been suggested, including the demethylation of chromatin surrounding the immunoglobulin  light chain (Ig) locus (Demengeot et al., 1995; Kirillov et al., 1996), the transcription and rearrangement of the Ig gene (Scherer et al., 1996), the temporal regulation of the Oct-2 transcription factor (Bendall et al., 1997), and as a survival mechanism targeted for down-regulation leading to the apoptotic death of self-reactive immature B cells (Schauer et al., 1998).

Although many effects of constitutive NF-B activity have been demonstrated in the B-lymphocyte life cycle, the underlying cause of this activity has not been established. We and others have shown previously that unlike most cases of inducible NF-B activity, constitutive NF-B levels in B cells are not affected by proteasome inhibitors (Phillips and Ghosh, 1997; Miyamoto et al., 1998). Conversely, free calcium (Ca²⁺) chelating agents were shown to reduce the amount of DNA binding NF-B in unstimulated B cells but could not inhibit the proteasome-dependent activation of NF-B in either B or non-B cells (Miyamoto et al., 1998). Both upstream and downstream events associated with calcium regulation of NF-B activity in B cells remain to be determined. In the present study, we have examined potential downstream components involved in proteasome-independent IB degradation and NF-B activation in B cells. We found that calmodulin (CaM) inhibitors selectively reduce the active portion of NF-B in B cells, concomitant with a block in IB protein turnover and a marked decrease in the message of a NF-B-dependent target gene. However, the immunosuppressant drug cyclosporin A, which inhibits the Ca²⁺/CaM-dependent phosphatase calcineurin, was unable to reduce NF-B levels in unstimulated cells. Furthermore, inhibition of constitutive NF-B activity through interference of CaM function in the B cells was reversible even after 6 h and did not result in a rapid onset of apoptosis. Our data suggest a unique role for Ca²⁺/CaM activity in the maintenance of a constitutive pool of NF-B in B cells and further suggest that antiapoptotic proteins whose synthesis is regulated by NF-B may have relatively long half-lives in the context of B cells.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture and Reagents. WEHI-231 and 70Z/3 cells were maintained in RPMI 1640 medium (Mediatech, Herndon, VA) supplemented with 10% fetal bovine serum (HyClone Laboratories, Logan, UT), 1250 U/ml penicillin G (Sigma Chemical, St. Louis, MO), 0.5 mg/ml streptomycin sulfate (Sigma Chemical), and 5 × 10⁵ M -mercaptoethanol in a 5% CO₂ humidified incubator. W231.Bcl-X_L cells were generated by retroviral infection of WEHI-231 cells with pLNLCA-flagBcl-X_L followed by selection in 1 mg/ml G418 (Invitrogen, Carlsbad, CA) and were subsequently maintained as described above in the added presence of 0.5 mg/ml G418. For splenocytes, whole spleens were isolated from C57BL/6 female mice approximately 70 days old. Individual cells were manually released, filtered, and red blood cells were removed by density-gradient centrifugation over LymphoPrep (Mediatech). The remaining cells were plated at 1 to 1.5 × 10⁷/ml in the same media described above and allowed to recover for 2 to 3 h before treatment. Dimethyl sulfoxide, tosylphenylalanine chlormethylketone (TPCK), pyrrolidine dithiocarbamate (PDTC), cyclosporin A (CsA), and cycloheximide were obtained from Sigma Chemical. BAPTA-AM; calpain inhibitor I; N-(4-aminobutyl)-2-napthalenesulfonamide, HCl (W12); N-(4-aminobutyl)-5-chloro-2-napthalenesulfonamide, HCl (W13); and calmidazolium were from Calbiochem (San Diego, CA).

Electrophoretic Mobility Shift Assay. Cells were aliquoted in 12-well culture dishes at ~2 × 10⁶/ml and treated at 37°C, 5% CO₂ in a humidified incubator. After treatment the cells were pelleted, washed twice in ice-cold PBS, and stored at 70°C until further processing. Nuclear extract preparation and the conditions for EMSA have been described previously (Miyamoto et al., 1994b). Briefly, 2 to 3 µg of nuclear extract or 4 to 6 µg of whole cell extract was incubated on ice with 0.5 µg of poly dI-dC before addition of ³²P-labeled double-stranded oligonucleotide containing either the B binding site from the Ig intronic enhancer (5'-CTCAACAGAGGGGACTTTCCGAGAGGCCAT-3'), the NF-Y binding site (5'-TTTTCTGATTGGTTCTGGCGAGTTTGG-3'), or the AP-1 binding site (Promega, Madison, WI). Oligonucleotides used for competition assays were either the wild-type B sequence or a mutated B binding site (5'-TCAACAGAGCTCACTTTATGAGAGGCC-3'). The c-Rel antibody 5075 used for supershift analysis has been described previously (Inoue et al., 1992), and others are from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA): p65 (sc-372-G), RelB (sc-226), p50 (sc-114-G), and p52 (sc-297-G). Reactions were separated in a 4% native acrylamide gel, and the gels were dried and exposed to X-ray film.

Apoptosis Assays. For flow cytometric analyses, ~10⁶ cells were first washed in PBS containing 1 mM EDTA and 0.1% BSA, resuspended in 100 µl of the PBS/EDTA/BSA solution plus 900 µl of chilled ethanol, and fixed overnight at 4°C. Cells were washed in phosphate citric acid buffer before being incubated overnight in PBS/EDTA/BSA containing 0.2 mg/ml RNase A and 50 µg/ml propidium iodide (Molecular Probes, Eugene, OR). Samples were processed on FACStar. DNA laddering assays were performed as described previously (Smith et al., 1989).

Western and Northern Blotting. Conditions for processing and probing Western blots were as described previously (Miyamoto et al., 1998). Rabbit anti-IB antibody (C21) was purchased from Santa Cruz Biotechnology, Inc., monoclonal anti-flag antibody was from Kodak IBI (New Haven, CT), and anti--tubulin antibody was purchased from Calbiochem. For Northern blots, total cellular RNA was prepared from ~10⁷ cells/sample according to manufacturer's instructions for RNeasy (QIAGEN, Valencia, CA). From each sample 15 µg of total RNA was separated in a 1% formaldehyde-agarose gel, transferred to a GeneScreen nylon membrane (PerkinElmer Life Sciences, Boston, MA), and cross-linked with UV irradiation. DNA probes were labeled with [³²P]dCTP by random priming of cDNA fragments spanning IB (ApaI-Eco72I) or GAPDH (HindIII-PstI).

Semiquantitative Reverse Transcription PCR. Total cellular RNA was prepared from treated or untreated cells as described above for Northern blotting. RT-PCR reactions for each treatment condition were performed with 5, 25, and 125 pg of total RNA with the Access RT-PCR kit according to instructions (Promega) and PCR products were fractionated on 1.5% agarose gels. The primer sequences for PCR amplification of Bcl-2, A1, and -actin have been described previously (Tomayko and Cancro, 1998). Those used for PCR amplification of IB were sense, 5'-CCGCAGGAGGCGCCGCTG-3', and antisense, 5'-GGTATTTCCTCGAAAGTCTCG-3', and generated a product of 285 base pairs.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Inhibitors of NF-B Cause a Rapid Onset of Apoptotic Cell Death. The serine protease inhibitor TPCK as well as the antioxidant PDTC have previously been shown to abrogate the NF-B activity constitutively present in WEHI-231 cells (Miyamoto et al., 1994a; Wu et al., 1996). We have shown that an intracellular Ca²⁺ chelating agent, BAPTA-AM, is also able to effectively inhibit the DNA binding activity of NF-B in these cells by stabilizing the IB inhibitor protein (Miyamoto et al., 1998). Because in the WEHI-231 cell line a loss of constitutive NF-B activity is reported to lead to the rapid onset of apoptosis (Wu et al., 1996), we were interested in determining whether BAPTA-AM could induce apoptosis in these cells as is the case for TPCK and PDTC. The Bcl-2 family member Bcl-X_L protects WEHI-231 cells from apoptosis initiated by TPCK, PDTC, and a number of other apoptotic agents (Fang et al., 1995). Therefore, we introduced epitope-tagged Bcl-X_L into WEHI-231 cells and screened numerous clones for the stable expression of the exogenous Bcl-X_L protein to be included as a control. We selected clone 1.4 (Fig. 1A, lane 5) for our experiments and will refer to this line as W231.Bcl-X_L. To ensure that stable expression of the Bcl-X_L protein does not alter the specificity or composition of constitutive NF-B in these cells, we used nuclear extract from W231.Bcl-X_L cells to perform the competition and supershift experiments shown in Fig. 1B. Consistent with the parental WEHI-231 cells, we were unable to detect p65, RelB, or p52. Rather, as in the WEHI-231 cell line, the upper, diffuse band distinguished by EMSA consists primarily of p50/c-Rel heterodimers as well as c-Rel homodimers, whereas the lower, sharper band is composed of p50 homodimers (Miyamoto et al., 1994b).

View larger version (44K): [in this window] [in a new window]   Fig. 1.   Stable expression of flag-Bcl-XL does not alter NF-B activity in WEHI-231 cells. A, Western blot analysis of parental WEHI-231 (lane 1) and individual WEHI-231 clones stably expressing flag-Bcl-XL (lanes 2-6) probed with anti-flag antibody. NS, nonspecific protein. B, nuclear extract from untreated W231. Bcl-XL cells were used for gel shift analysis in the absence of competing oligonucleotide (lane 1) or in the presence of a 10-fold molar excess wild-type or mutant unlabeled B oligonucleotide (lanes 2 and 3, respectively). In lanes 5 to 9, 200 ng of each of the indicated antibodies was added to the EMSA binding reactions. No antibody was added to lane 4.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Ca2+ chelator BAPTA-AM induces apoptosis independent of NF-B inhibition or Bcl-XL expression. A, parental WEHI-231 cells or W231.Bcl-XL were either left untreated or treated with solvent at 0.1%, TPCK at 25 µM, PDTC at 25 µM, BAPTA-AM at 30 µM, CI-I at 50 µM, or 20 µg/ml cycloheximide for 4 h. A gel shift assay for B-binding activity was performed using equal amounts of nuclear protein extract from each of the indicated treatment conditions. B, WEHI-231, W231. Bcl-XL, or 70Z/3 cells were treated as outlined in B and assayed for sub-G1/G0 DNA content as described under Materials and Methods. The cells with sub-G1/G0 DNA content are represented as a percentage of the total cells assayed over one to three independent experiments. C, WEHI-231 or W231.Bcl-XL cells were treated as described in B and DNA fragmentation was determined by agarose gel electrophoresis.

Inhibition of CaM Activity Results in a Decline of Constitutive NF-B Activity. The observation that BAPTA-AM is a potent inducer of apoptosis in both the W231.Bcl-X_L and the 70Z/3 cell lines raises the possibility that constitutive NF-B activity is not inhibited by calcium chelation per se, but rather as a result of cell death incurred by calcium chelation. Yet, CI-I and cycloheximide induce apoptosis in WEHI-231 cells, and under these conditions, NF-B activity is either unchanged or increased, respectively, indicating that reduction of NF-B activity after BAPTA-AM treatment in WEHI-231 cells is not likely to be a consequence of cell death processes. Therefore, to identify a potential downstream effector(s) of calcium in the regulation of constitutive NF-B activity, we treated WEHI-231 cells with diverse inhibitors and the DNA binding of NF-B was determined by EMSA (data not shown). Among those agents tested, treatment with CaM antagonists consistently led to a reduction in the constitutive activity of NF-B in both the WEHI-231 and W231.Bcl-X_L cell lines.

View larger version (53K): [in this window] [in a new window]   Fig. 3.   W13 acts as a CaM antagonist to inhibit NF-B activity. A, W231.Bcl-XL cells were either left untreated or were treated with increasing concentrations of W13 or W12 for 3 h before termination. Nuclear extracts were prepared, and equal amounts from each sample were used for EMSA of both B and AP-1 DNA binding activity. The diffuse band representing NF-B (p50/c-Rel) is indicated in the top and that representing AP-1 in the bottom. The specificity of the AP-1 band was determined by competition EMSA with unlabeled AP-1 oligonucleotide (data not shown). B, W13 was added at the indicated amounts to samples from the same W231.Bcl-XL cell nuclear extract preparation (lanes 1-4). The reactions were incubated at room temperature for 20 min before addition of radiolabeled B probe and subsequent gel shift analysis. C, freshly extracted splenocytes were cultured in the presence of 0, 10, or 20 µM W13 for a period of 3 h. Total cell extracts were prepared from treated cells and examined by EMSA for DNA binding activity using a B oligonucleotide (top) or NF-Y oligonucleotide (bottom). D, W231.Bcl-XL cells were treated for the indicated times in the presence of 20 µM W13 and total cell extracts were analyzed by EMSA for NF-B DNA binding activity. E, W13 was incubated either with W231.Bcl-XL cells suspended in complete media or with complete media alone. After 6 h the cells were terminated (lane 2) and previously untreated W231.Bcl-XL cells were resuspended with either fresh W13 (lane 4) or with the preincubated W13-containing media (lane 3) and terminated after 3 h. Total cell extracts were analyzed by EMSA.

CaM Inhibition by W13 Does Not Lead to Rapid Cell Death in WEHI-231 Cells. Because W13 treatment of unstimulated W231.Bcl-X_L cells effectively inhibits the constitutive NF-B DNA binding activity, we were next interested in examining whether W13 treatment also results in a rapid induction of apoptosis as is the case for TPCK, PDTC, and BAPTA-AM. To this end, we incubated 70Z/3, WEHI-231, and W231.Bcl-X_L cells in media containing W13 or BAPTA-AM for up to 6 h and assayed for the fragmentation of genomic DNA as an indication of apoptosis. As before, BAPTA-AM treatment led to a rapid onset of apoptosis in WEHI-231 and W231.Bcl-X_L cells (Fig. 4, lanes 6 and 11, respectively). Surprisingly, however, inhibition of CaM activity for up to 6 h by W13 was not able to induce apoptosis in any one of the cell lines tested (Fig. 4, lanes 2, 3, 7, 8, 12, and 13). Therefore, because interfering with CaM function results in a marked decrease in NF-B activity but not apoptosis, we conclude that CaM inhibitors are comparably effective yet overall less toxic than other pharmacological compounds used to inhibit constitutive NF-B activity in WEHI-231 cells. These studies also demonstrate that extensive inhibition of constitutive NF-B activity does not lead to the rapid onset of apoptosis. Additionally, these data suggest that antiapopototic gene products synthesized in a NF-B-dependent manner may have relatively long half-lives in WEHI-231 cells.

View larger version (72K): [in this window] [in a new window]   Fig. 4.   W13 treatment does not induce apoptosis. WEHI-231, W231.Bcl-XL, or 70Z/3 cells were left untreated for 6 h or, where indicated, were treated with 20 µM W13, 30 µM BAPTA-AM, or 0.1% dimethyl sulfoxide for either 4 or 6 h. After treatment the cells were assayed for nucleosomal DNA fragmentation by agarose gel electrophoresis.

Calmodulin Inhibition Reversibly Blocks NF-B Target Gene Expression. Although our EMSA data indicate that CaM inhibition leads to a reduction of NF-B DNA binding activity, it is unclear whether NF-B function is equally inhibited. The gene encoding IB is among the target loci whose transcriptional activity is highly increased after NF-B activation. Several B sites have been identified within the promoter region of the IB gene, which provide a direct linkage between NF-B activity and IB mRNA synthesis (Bail et al., 1993; Martin et al., 1993; Chiao et al., 1994). It follows then that one consequence of constitutive NF-B activity in B cells is a relatively large amount of IB transcript (Miyamoto et al., 1994a). The half-life of IB mRNA is very short, approximately 15 to 20 min, which enables us to examine the quantity of IB message as an indication of the NF-B transcriptional activity within the cell. When WEHI-231 cells were treated for increasing times with W13 the total amount of IB transcript dropped (Fig. 5A, middle, lanes 2-4) concomitantly with the decrease in the DNA binding capacity of NF-B (Fig. 5, top, lanes 2-4). Removing W13 from the culture medium after 6 h allowed the WEHI-231 cells to completely recover constitutive NF-B activity (Fig. 5, top, lane 5). Importantly, IB mRNA was resynthesized to a total amount resembling the initial amount of transcript (Fig. 5A, middle, lane 5). RNA loading was relatively constant as indicated by the GAPDH signal (Fig. 5A, bottom).

View larger version (42K): [in this window] [in a new window]   Fig. 5.   W13 treatment causes a reversible decline in the IB gene transcript. A, W231.Bcl-XL cells were either left untreated or were incubated with 20 µM W13 for 2, 4, or 6 h before being terminated. One sample was washed after a 6-h treatment with W13 and allowed to recover in complete media for an additional 6 h (lane 5). Top, EMSA prepared from whole cell extracts. From the same samples total RNA was isolated and 10 µg was used for Northern blot analysis in the bottom two for the determination of IB and GAPDH transcript levels. B, 70Z/3 pre-B cells were either left untreated or were stimulated with 10 µg/ml LPS for 18 h and then further treated with 20 µM W13 for 3 or 6 h. As in A, the top represents an EMSA prepared from whole cell extracts, and in the bottom two 10 µg of total RNA was probed for IB or GAPDH as indicated. C, total RNA was isolated from WEHI-231 cells that were untreated (lanes 1-3), incubated for 6 h with W13 (lanes 4-6), or washed and recovered for 6 h after a 6-h W13 treatment (lanes 7-9). The amount of input RNA per RT-PCR reaction was increased over 25-fold for each sample (1×, 5×, 25×).

Rapid Turnover of IB in WEHI-231 Cells Requires CaM Activity. Because the degradation of IB almost invariably precedes the activation of NF-B, including the constitutively active NF-B in WEHI-231 cells (Miyamoto et al., 1994a), we wanted next to determine the effect that W13 treatment has on the stability of the IB protein. Previously, others and ourselves have demonstrated that IB is unusually short-lived in WEHI-231 cells relative to non-B and other cell types (Rice and Ernst, 1993; Miyamoto et al., 1998; Doerre and Corley, 1999). The half-life of IB was tested by blocking protein synthesis with cycloheximide and terminating cells at successive time points. Both IB and -tubulin as a loading control were probed against in the same blot. In the absence of W13 and consistent with previous studies, the IB protein is rapidly degraded in the WEHI-231 cells with an approximate half-life of under 1 h (Fig. 6A, lanes 1-4). However, the addition of W13 completely blocked the degradation of IB even up to 3 h after treatment (Fig. 6A, lanes 5-7). When an equal concentration of W12 was added in addition to cycloheximide over 3 h, a very weak effect on the turnover of IB was observed (Fig. 6A, compare lanes 4 and 8), consistent with its relatively weak CaM inhibitory activity. The basal turnover of IB in the 70Z/3 pre-B cell line was not affected by W13 treatment (data not shown). Moreover, signal-dependent degradation of IB is also not affected with W13 or W12 treatment (Fig. 6B, compare lane 2 with lanes 4 and 6). Therefore, CaM activity is critical for the accelerated degradation of IB and the maintenance of constitutive NF-B function in WEHI-231 cells.

View larger version (50K): [in this window] [in a new window]   Fig. 6.   W13 prevents the constitutive but not the inducible degradation of IB in W231.Bcl-XL cells. A, W231.Bcl-XL cells were incubated with 25 µg/ml cycloheximide (CHX) alone (lanes 1-4), or in addition to either W13 (lanes 5-7) or W12 (lane 8), each at 20 µM concentration. The cells were incubated for approximately 15 min to allow for drug action before terminating the reference sample (lane 1). Other samples were terminated at the indicated times. Whole cell extract (20 µg) was used for Western blotting against the IB and -tubulin proteins. B, W231. Bcl-XL cells were left untreated or were incubated over a period of 3 h with 20 µM W12 or 20 µM W13. After this pretreatment the cells were either left unstimulated or were stimulated with 20 µg/ml LPS for 20 min. Top, Western blot performed on 20 µg of total protein and probed for IB and -tubulin. Bottom, EMSA demonstrating B DNA binding activity that was performed on the same whole cell extracts described for the top. The p50/c-Rel homodimeric complex is indicated as NF-B.

Constitutive NF-B Activity Requires Calmodulin but Not Calcineurin Activity. A dependence on Ca²⁺ for NF-B activation been demonstrated for various inducers of NF-B, most notably the T-cell receptor (TCR). In this case an intracellular Ca²⁺ increase after stimulation is a requisite for activating the Ca²⁺/CaM-dependent phosphatase calcineurin (CaN), which in turn is necessary for activating NF-B (Mattila et al., 1990; Frantz et al., 1994; Steffan et al., 1995). Because we have shown a Ca²⁺/CaM dependence of constitutive NF-B activity in a B-cell line, we were interested in assessing whether this activity was also CaN-mediated as is the case in T cells after TCR stimulation. Signaling through the TCR can be simulated through the synergistic effect of Ca²⁺ ionophore and phorbol ester added simultaneously. Treatment of the W231.Bcl-X_L cell line with phorbol-12-myristate-13-acetate (PMA) was able to activate NF-B (Fig. 7A). Surprisingly, however, addition of the Ca²⁺ ionophore inonomycin at concentrations ranging from 0.01 to 10 µg/ml was unable to further activate NF-B (data not shown). Nevertheless, CaN seems to be required for the induction of NF-B by PMA because this activation was sensitive to both W13 and the CaN inhibitor CsA. And as expected, these agents had no effect on the induction of NF-B after LPS treatment of W231.Bcl-X_L cells. To determine, then, whether CaN plays a role in the constitutive maintenance of NF-B we incubated cells with either W13 or CsA and measured NF-B activity over 3 h. Although W13 was effective, we were unable to detect a reduction of NF-B DNA binding activity in cells treated with CsA (Fig. 7B). We conclude from these studies that a CaN requiring event is unlikely to be the Ca²⁺/CaM-dependent step for maintaining constitutively active NF-B.

View larger version (59K): [in this window] [in a new window]   Fig. 7.   CaN inhibition is unable to block constitutive NF-B. A, W231.Bcl-XL cells were either left untreated (lanes 1, 2, and 5) or were incubated for 30 min with 20 µM W13 (lanes 3 and 6) or 4 µM CsA (lanes 4 and 7). After this pretreatment, cells were stimulated with either 10 nM PMA (lanes 2-4) or 20 µg/ml LPS (lanes 5-7) for an additional 30 min. Whole cell extracts were analyzed by EMSA for B binding activity (top) or NF-Y binding activity (bottom). B, W231.Bcl-XL cells were treated with 20 µM W13 (lanes 2 and 3) or 4 µM CsA (lanes 4 and 5) and were terminated at the times indicated. Whole cell extracts (untreated cells are included in lane 1) were analyzed as in A.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

We have identified previously a novel proteolytic inactivation of the NF-B inhibitory protein IB. Whereas IB is typically degraded by the ubiquitin-dependent 26S proteasome, in WEHI-231 cells the ongoing turnover of IB is not blocked by excess concentrations of proteasome inhibitors. However, these inhibitors effectively block the LPS-induced degradation of IB in these cells. Conversely, the intra- and extracellular Ca²⁺ chelators BAPTA-AM and EGTA, respectively, are able to block the constitutive turnover of IB but not that of the LPS-induced turnover of IB (Miyamoto et al., 1998; Fields et al., 2000). We further provided evidence that this calcium-dependent degradation of IB is preceded by a proteasome-dependent mechanism during differentiation of B cells in vitro (Fields et al., 2000). However, both upstream and downstream components required for rapid IB degradation and the maintenance of constitutive NF-B activity in B cells remain undefined. In this study, we show that the activity of the calcium-sensing protein CaM is necessary for these processes both in the WEHI-231 B cell line and in primary splenocytes.

The WEHI-231 line is an immature B-cell lymphoma (IgM^hi/IgD^lo) that undergoes apoptosis after cross-linking of surface IgM antigen receptors, and therefore has been widely studied as a model for the clonal deletion of self-reactive B lymphocytes. It has been shown using WEHI-231 cells that, in the absence of costimulation, surface IgM cross-linking results in a transient increase of NF-B activity followed by a complete loss of the constitutive NF-B activity at approximately 4 to 6 h after treatment (Lee et al., 1995). Detectable signs of apoptosis under these conditions have been detected at 18 to 24 h after treatment (Benhamou et al., 1990; Gottschalk et al., 1994). Furthermore, when an inducible dominant negative form of IB was used to inhibit the constitutive NF-B activity in an Epstein-Barr virus-transformed cell line, appreciable apoptosis was observed at 24 h after a reduction in NF-B activity (Cahir-McFarland et al., 2000). This lapse in time between a reduction in NF-B activity and the onset of apoptosis is not seen when TPCK, PDTC, or BAPTA-AM is used to inhibit NF-B (Fig. 1; Wu et al., 1996). Perhaps, then, a drop in NF-B function alone may not be sufficient to cause a rapid onset of apoptosis in these cells. Additional events influenced by these agents are probably causing rapid apoptotic cell death. Thus, our data suggest that the use of these pharmacological agents to study NF-B-dependent antiapoptotic processes in B cells should be viewed with caution. Moreover, our data suggest that any antiapoptotic gene product(s) produced in a NF-B-dependent manner probably has a relatively long half-life, making a sustained drop in NF-B activity a requirement for the onset of apoptosis.

How long is a drop in NF-B activity required for WEHI231 B cells to commit to apoptotic cell death? We have failed to detect either apoptosis or loss of cell viability in WEHI-231 cells after treatment with W13 for indefinite periods. Surprisingly, time course experiments revealed that the NF-B inhibitory effect of W13 was completely lost at approximately 8 to 10 h after its addition to the culture medium (Fig. 3D). Incubating W13 in media alone for 6 h abrogated its ability to inhibit NF-B, suggesting that the loss of a W13 effect was due to instability of W13 in the medium. Therefore, the cytotoxic consequences of W13-mediated inhibition of NF-B function could not be directly assessed at time points greater than 8 h after treatment. Nevertheless, because W13 is able to inhibit NF-B activity with similar kinetics as for TPCK, PDTC, and BAPTA-AM, yet does not induce apoptosis up to 8 h, we propose that NF-B activity needs to be down-regulated for a time greater than 8 h for proapoptotic events to be able to overcome the antiapoptotic effects of NF-B. There are a number of antiapoptotic genes implicated as direct targets of NF-B, such as Bfl-1/Al, IAP1/2, and Bcl-X_L. Thus, it is likely that one or a combination of these antiapoptotic genes is regulated by NF-B in WEHI-231 cells and their in vivo half-lives are such that at least 8 h or more is required to lose their antiapoptotic effects. A role for NF-B-dependent expression of A1 in WEHI-231 cells is supported by our finding that A1 mRNA levels drop in the presence of W13 but return after its removal (Fig. 5C).

Treatment of T lymphocytes with PMA in conjunction with ionomycin mimics cross-linking of surface TCRs. One result of this is the activation of diverse transcription factors such as nuclear factor of activated T cells, AP-1, and NF-B. Using PMA plus ionomycin to treat T cells, it has been shown that the Ca²⁺/CaM-dependent phosphatase calcineurin is required for activation of NF-B (Mattila et al., 1990; Frantz et al., 1994; Steffan et al., 1995). And although a requirement for CaM activity has been implicated in this model, it was not until recently that a necessary role for Ca²⁺/CaM activity upstream of IB phosphorylation and degradation was more directly supported (Hughes et al., 1998). Hughes et al. (1998) conducted studies that demonstrate that in T cells NF-B activation, as well as IB phosphorylation and degradation, are sensitive to an array of CaM antagonists after treatment with either PMA alone or PMA plus ionomycin. Because CsA could not inhibit NF-B activation by PMA alone in T cells, Hughes et al. (1998) conclude that CaM is involved in CaN-independent as well as CaN-dependent NF-B activation pathways. In support of this, our results show that although CaM antagonists block the constitutive activity of NF-B in the W231.Bcl-X_L cell line, inhibition of CaN does not (Fig. 7). Furthermore, a slower migrating phosphorylated form of IB was not detectable in W231.Bcl-X_L cells in which IB degradation was blocked by W13 (Fig. 6A; data not shown). Therefore, the step in the constitutive IB proteolytic pathway in W231.Bcl-X_L cells at which CaM inhibitors intercede seems to be distinct from that in which CaM inhibitors block the inducible IB proteolytic pathway in T cells. The nature of the Ca²⁺/CaM-dependent step in the constitutive, signal-independent IB degradative pathway is currently under investigation.

Although we investigated potential downstream effectors of Ca²⁺ that are critical for the regulation of constitutive NF-B in B cells, we did not address potential upstream regulators of the Ca²⁺ levels required for constitutive NF-B activity in B cells. In our recent study, we have obtained several lines of evidence that 1,4-dihydropyridine-sensitive calcium channels may be involved in the maintenance of Ca²⁺ levels in B cells (C. M. Berchtold, S. D. Shumway, S. Miyamoto, M. N. Gould, manuscript in preparation). Together, these studies suggest that there is a signaling system in B cells that begins with 1,4-dihydropyridine-sensitive calcium channels to maintain Ca²⁺/CaM activity that then leads to downstream events, ultimately causing the degradation of IB in a proteasome-independent manner followed by the release of NF-B into the nucleus. Because NF-B (p50/cRel) can stimulate the synthesis of p50, c-Rel, and IB in this system (Miyamoto et al., 1994b), these events lead to the maintenance of constitutive NF-B activity to continue their viability in vitro and in vivo. Thus, important future investigations include the identification of an IB protease responsible for its Ca²⁺/CaM-dependent degradation and an "endogenous signal" that maintains this process independent of an added exogenous stimulus.