Fas Apoptosis Inhibitory Molecule Enhances CD40 Signaling in B Cells and Augments the Plasma Cell Compartment¹

Fas apoptosis inhibitory molecule (FAIM)³ was cloned (1) via differential display from primary B cells, in which sensitivity to Fas-mediated apoptosis had been inducibly down-modulated and, consistent with its origin, expression of exogenous FAIM rendered BAL-17 B cells relatively Fas-resistant (1, 2, 3). In subsequent studies, FAIM has been shown to inhibit death receptor-induced apoptosis in other cell types and other species (4, 5, 6). The Faim gene is located at 9f1 in mouse (and at the syntenic region 3q22 in human) and encodes an ∼1.2 kb transcript that produces a 179 amino acid protein of ∼20 kDa (1, 7). FAIM is highly evolutionarily conserved (from worm to fly to mouse to human) but contains no known effector motifs and the means by which FAIM opposes apoptosis remains uncertain. In keeping with the way in which it was identified, FAIM expression is up-regulated in primary B cells by the combined action of CD40L and anti-Ig (1).

The full-length FAIM gene was originally cloned by screening a murine thymus cDNA library. Subsequent screening of a murine brain cDNA library yielded a novel clone that possessed an insert upstream of the putative FAIM start methionine but that was otherwise identical with the Faim gene first described, resulting in a sequence longer by 66 N-terminal nucleotides (7). This longer FAIM isoform was termed FAIM-long (FAIM-L) and the previously identified shorter FAIM form was renamed FAIM-short (FAIM-S). Characterization of the murine Faim genomic locus revealed that the two FAIM isoforms result from alternative splicing, with FAIM-S lacking exon 2 of six exons, and FAIM-L incorporating all six exons (7). FAIM-L differs dramatically from FAIM-S in terms of its highly restricted expression pattern; whereas FAIM-S is characterized by wide tissue distribution, FAIM-L is expressed almost exclusively in the brain (7).

B cells must be stimulated to up-regulate Fas expression and acquire sensitivity to Fas-mediated apoptosis (2). Because many anti-apoptotic genes are NF-κB responsive (8) and inducible Fas resistance in B cells requires NF-κB (9), we considered the possibility that FAIM might oppose Fas killing by up-regulating NF-κB induction in activated B cells. The studies reported in this paper asked whether FAIM would enhance CD40 induction of NF-κB in B cells (10, 11, 12). It did, and this led to subsequent studies showing that FAIM enhances CD40-mediated IFN regulatory factor (IRF)-4 expression in vitro, that FAIM enhances CD40-mediated BCL-6 down-regulation in vitro, and that, in keeping with enhanced expression of IRF-4 and diminished expression of BCL-6, FAIM augments plasma cell differentiation in vivo.

A20 B lymphoma cells were obtained from the American Type Culture Collection. FAIM-S and FAIM-L were amplified by PCR using LA-Taq (Takara Shuzo). The products were cloned into the MIGW.IRES.GFP vector (13), which was amplified in BOSC packaging cells cultured in DMEM medium containing 10% FCS, 10 mM HEPES, 2 mM l-glutamine, and 0.1 mg/ml penicillin and streptomycin, as previously described (14). A20 cells were transduced with MIGW.FAIMS.IRES.GFP, MIGW.FAIML.IRES.GFP, or empty vector MIGW.IRES.GFP, as described (14). At 10 days following transduction, GFP⁺ A20 cells were sort-purified. At the time of experimentation, transduced A20 cells were >95% GFP-expressing.

The NF-κB-dependent firefly-luciferase vector was obtained from BD Clontech and was cotransfected with pRL-TK (renilla-luciferase) vector using the nucleofector kit V (Amaxa Biosystems) according to the manufacturer’s instructions. At 24 h after transfection, luciferase activity in cell lysates was analyzed by the dual-luciferase reporter assay (SystemPromega) according to the manufacturer’s instructions, using a 20/20n single tube luminometer (Turner BioSystems). Relative luciferase activity was calculated as firefly-luciferase activity/renilla-luciferase activity and the data shown represent the mean fold induction of two to three experiments.

Nuclear extracts were prepared and assayed for NF-κB by EMSA, as previously described (15). NF-Y binding was used to verify that equal amounts of nuclear extract protein were present in each sample. The probe sequences are listed in Table I.

Proteins were extracted from B cell pellets with Nonidet P-40 lysis buffer and equal amounts of protein for each condition were subjected to SDS-PAGE followed by immunoblotting, as previously described (16).

Total RNA was extracted using Ultraspec reagent (Biotecx Laboratories) according to the manufacturer’s instructions, reverse transcribed, amplified by real-time PCR, and normalized to expression of β₂-microglobulin, as previously described (17). The sequences of the primer sets are shown in Table I.

Male BALB/cByJ mice at 8–14 wk of age were obtained from The Jackson Laboratory. Mice were housed at least 1 wk before experimentation. Mice were cared for and handled in accordance with National Institutes of Health and The Feinstein Institute for Medical Research’s guidelines.

Bone marrow chimera mice were generated by MSCV transduction and adoptive transfer, as previously described (14). Briefly, bone marrow cells harvested from the hind legs of 5-fluorouracil-treated BALB/c mice were transduced with FAIM-S-expressing MSCV.FAIMS.IRES.GFP vector. Cells were treated with 100 ng/ml stem cell factor, 100 ng/ml flt3 ligand, and 50 ng/ml thrombopoietin (R&D Systems) during three rounds of transduction over a period of 24 h. MSCV.FAIMS.IRES.GFP-transduced bone marrow cells (1 × 10⁶) were adoptively transferred to lethally irradiated (900 rad) BALB/c recipient mice via the tail vein.

For FACS analysis, cells pretreated with 5 μg/ml 2.4G2 were stained with FITC-, PE-, or PE-Cy5-conjugated mAb in staining buffer (PBS containing 2% FCS and 0.05% NaN₃) on ice for 30 min and washed with staining buffer. Stained cells were analyzed on a FACSCalibur instrument (BD Biosciences) equipped with CellQuest software. To exclude dead cells from the analysis, 7-amino-actinomycin D (2 μg/ml) (BD Pharmingen) was used. For each sample, at least 1 × 10⁴ cells were collected and analyzed.

Splenic B2 cells were obtained by negative selection with anti-Thy1.2 Ab and rabbit complement, as previously described (2). Isolated B2 cells were >95% B220 positive. B cells were cultured in RPMI 1640 medium containing 10% FCS, 10 mM HEPES, 2 mM l-glutamine, 0.1 mg/ml penicillin and streptomycin, and 50 μM 2-ME.

Splenic B2 cells were stimulated with CD40L for 48 h, washed, and then stimulated with anti-Fas Ab (Jo-2; BD Pharmingen) for 4 h. Cells were exposed to Red-DEVD-FMK (MBL International) for 1 h according to the manufacturer’s instructions. Cells were analyzed by flow cytometry on a FACSCalibur instrument.

CD40L was prepared and used as previously described (11, 15, 18). Affinity-purified F(ab′)₂ goat anti-mouse IgM (anti-Ig) was obtained from Jackson ImmunoResearch Laboratories. LPS from Salmonella typhimurium was obtained from Sigma-Aldrich. Monoclonal anti-FcγR Ab 2.4G2, monoclonal anti-Fas Ab Jo-2, FITC-conjugated RA3-6B2 (anti-B220), and PE-conjugated anti-CD80, anti-CD86, and anti-CD138 were obtained from BD Pharmingen. Anti-CD40-PE was obtained from Southern Biotech. Affinity-purified anti-FAIM Ab was obtained from rabbits immunized with CYIKAVSSRKRKEGIIHTLI peptide (located near the C-terminal region of FAIM). Abs against RelB and IκBα were obtained from Cell Signaling Technology. Anti-Lamin B1 and anti-c-Rel Abs were from Santa Cruz Biotechnology. Anti-tubulin Ab was obtained from Calbiochem and anti-actin Ab was obtained from Sigma-Aldrich. Red-DEVD-FMK was obtained from MBL International.

Το investigate the influence of FAIM on NF-κB activation, we cloned FAIM-S and, separately, FAIM-L into the MSCV retroviral vector MIGW.IRES.GFP (19, 20) and then transduced A20 cells with FAIM-S-expressing MIGW.FAIMS.IRES.GFP, FAIM-L-expressing MIGW.FAIML.IRES.GFP, or empty vector. We then selected transduced A20 cells by fluorescence-activated cell sorting to generate stable transductants. We verified FAIM expression by Western blotting, and thereby demonstrated that FAIM-S-transduced A20 B cells contained 10-fold more FAIM-S than the low endogenous FAIM-S level present in empty vector-transduced cells, and that FAIM-L-transduced A20 cells contained an incalculably greater amount of FAIM-L as compared with vector-transduced A20 cells because the latter did not express detectable levels of FAIM-L (Fig. 1,A). We then obtained nuclear extracts from each population of transduced A20 cells before and after stimulation with CD40L, anti-Ig, or LPS and tested these extracts for κB-binding activity by EMSA with a probe containing a consensus NF-κB binding motif (Fig. 1 B). We found that FAIM-S overexpression produced marked enhancement of CD40L-induced κB-binding activity in A20 cells as compared with vector-transduced A20 cells. There was no difference in CD40 expression between FAIM-S-transduced and empty vector-transduced A20 cells (data not shown). In contrast, FAIM-L was completely inactive. Interestingly, neither FAIM-S nor FAIM-L overexpression enhanced anti-Ig- or LPS-induced κB-binding activity. Thus, FAIM-S specifically enhances CD40L-induced κB-binding activity in murine B cells.

To examine FAIM-enhanced, CD40L-induced NF-κB activity further, we transfected FAIM-S-, FAIM-L-, and vector-transduced A20 B cells with a κB-dependent firefly luciferase reporter plasmid and evaluated luciferase activity before and after stimulation with CD40L. We found that FAIM-S overexpression produced heightened activation of the NF-κB reporter plasmid after stimulation by CD40L as compared with transduction of empty vector (p < 0.05, n = 4) (Fig. 1 C). In contrast, FAIM-L was completely inactive (p > 0.3, n = 4). Similar results were obtained when CD40 was triggered by 1C10 monoclonal rat anti-mouse CD40 Ab. FAIM-S, but not FAIM-L, enhanced NF-κB-dependent reporter activity (data not shown). As with EMSA, these results were specific for CD40 signaling; neither FAIM-S nor FAIM-L transduction enhanced either anti-Ig- or LPS-induced κB-binding activity (data not shown). These results indicate that FAIM-S specifically enhances NF-κB activity in B cells, as detected by reporter gene function and that FAIM-L lacks the NF-κB-enhancing activity of FAIM-S.

There are two NF-κB activation pathways induced by CD40L; the canonical pathway, in which nuclear translocation of p50, p65, and c-Rel is induced though IKKβ activity and IκBα degradation, and the noncanonical pathway, in which translocation of p52 and RelB is induced though NIK and IKKα activity (21, 22). To examine which pathway is enhanced by FAIM-S, we first evaluated IκB degradation by Western blotting. We found that CD40L-induced IκBα degradation was accentuated in FAIM-S A20 transductants as compared with empty vector transductants (Fig. 1,D). FAIM-S enhancement of IκBα degradation was similarly evident when CD40 was triggered by anti-CD40 Ab (data not shown). Thus, enhancement of CD40L-induced NF-κB activity produced by FAIM-S occurs in conjunction with enhanced IκBα degradation, suggesting that FAIM-S acts on the canonical pathway. We then directly tested the role of the noncanonical pathway by examining nuclear accumulation of RelB. We found that CD40L produced an enhanced increase in nuclear RelB, as well as c-Rel, in FAIM-S A20 transductants as compared with empty vector transductants (Fig. 1 E). Thus, FAIM-S enhances CD40-induced NF-κB activation through both canonical and noncanonical mechanisms.

To further address the physiological activity of FAIM-S-enhanced NF-κB, we analyzed target gene expression of A20, IκBα, CD80, and IL-12 p35 induced by CD40L (23, 24, 25, 26) in A20 cells transduced with FAIM-S or with empty vector. We found that FAIM-S enhanced the CD40L-induced expression of all four NF-κB target genes studied as compared with expression produced by CD40L in vector-transduced A20 cells (Fig. 1 F). Thus, enhancement of CD40L-induced NF-κB activity produced by FAIM-S results in enhanced activation of physiological NF-κB targets. Of note, however, is that enhancement of some genes, such as Cd80 and IL-12 p35, was more substantial and more prolonged than others. These latter genes are regulated by IRF transcription factors as well as by NF-κB (27, 28, 29). All together these results indicate that, by multiple criteria, FAIM-S enhances CD40L-stimulated NF-κB activation.

Our results above showing disproportionate enhancement of some NF-κB target gene responses by FAIM-S raised the possibility that FAIM affects IRF family members as well as NF-κB. The IRF family contains nine members. The activity of IRF1 and IRF4 is regulated by transcription, whereas the activity of other family members, particularly IRF3 and IRF7, is regulated by serine and threonine phosphorylation. To determine whether FAIM enhances IRF activation induced by CD40L, we first evaluated the levels of IRF3 and IRF7 phosphorylation, because IRF3 and IRF7 family members have been implicated in regulating CD80 expression in macrophages (27, 28). We prepared lysates from FAIM-S- and vector-transduced A20 cells before, and at various times after, CD40L stimulation, and then conducted Western blotting with anti-pIRF3 and anti-pIRF7 Abs. We found that phosphorylation of IRF3 and IRF7 at baseline was similar in FAIM-S-transduced A20 cells as compared with vector-transduced A20 cells, and that CD40L failed to stimulate pIRF3 or pIRF7 in either population (data not shown). Next, we analyzed the levels of Irf1 and Irf4 gene expression in separate groups of FAIM-S- and vector-transduced A20 cells. We prepared RNA before, and at various times after, CD40L stimulation of these cells, and then conducted real-time PCR. We found no CD40L-stimulated induction of Irf1 expression above the background observed in empty vector-transduced A20 cells (data not shown). In contrast, CD40L produced up-regulation of Irf4 expression (30, 31) in empty vector-transduced A20 cells and, importantly, we found that this CD40-mediated induction of Irf4 was markedly enhanced by FAIM-S (Fig. 2,A). We further examined IRF4 protein. We prepared lysates from FAIM-S-transduced and vector-transduced A20 cells before, and at two times after, CD40L stimulation, and then conducted Western blotting with anti-IRF4 Ab. We found that FAIM-S greatly enhanced the induction of IRF4 protein produced by CD40L-stimulation of A20 cells (Fig. 2 B). Thus, FAIM-S enhances CD40L-stimulated IRF4 expression, recapitulating the activity of FAIM-S in enhancing CD40L-stimulated NF-κB activation.

CD40 engagement in B cells produces down-regulation of BCL-6 expression through IRF4-mediated transcriptional inhibition (30, 31, 32, 33). In view of our finding that FAIM enhances CD40L-stimulated IRF4, we reasoned that FAIM might augment CD40L-mediated down-regulation of BCL-6. To address this, we stimulated stably transduced A20 cells with CD40L, prepared RNA from stimulated cells at various times, and examined Bcl-6 gene expression by real-time PCR. As predicted by previous work, we found that the level of CD40L-induced down-regulation of Bcl-6 expression was accentuated (i.e., experienced a further decline) in the presence of FAIM-S (Fig. 3). Thus, FAIM-mediated enhancement of CD40L-stimulated NF-κB activation and IRF4 expression in B cells is accompanied by reinforcement of Bcl-6 down-regulation as well.

IRF4 is a key determinant of plasma cell production (34, 35). The capacity of FAIM-S to enhance IRF4 expression, and to further diminish Bcl-6 expression, suggested a possible role for FAIM in influencing plasma cell development. Moreover, FAIM is up-regulated in germinal center B cells (Ref. 36 and our unpublished observations), which puts FAIM in the right location to influence B cell differentiation. Furthermore, our previous finding that FAIM expression is stimulated in response to BCR and CD40 engagement also supports the notion that FAIM responds to signals involved in late B cell differentiative processes. To address the effect of FAIM-S on B cells in vivo, we transduced bone marrow cells from normal mice with MIGW.FAIMS.IRES.GFP, as described in Materials and Methods. We then adoptively transferred these mixed (retrovirally transduced and nontransduced) bone marrow cells into lethally irradiated recipients, and evaluated and compared GFP+ (retrovirally transduced) lymphocytes with GFP− (nontransduced) lymphocytes obtained from adoptive hosts 8–12 wk later. We first evaluated FAIM-S expression in splenic B cells by Western blotting. We found that FAIM-S expression in GFP+ B cells was much higher (greater than 5-fold higher) than the low endogenous level present in GFP− B cells, as might be expected (Fig. 4,A). As expected, this increased level of FAIM-S eliminated the susceptibility to Fas-mediated apoptosis displayed by primary B cells, as judged by the relative level of activated caspase 3 produced by anti-Fas Ab Jo-2 in CD40L-treated GFP+ and GFP− B cells (Fig. 4 B).

We then analyzed splenic, peritoneal cavity, and bone marrow cell populations. We found no difference between GFP+ and GFP− cells in terms of the fraction of B220+ cells in spleen or bone marrow populations, nor in the fraction of B220+ cells or of B220loCD5+ cells in the peritoneal cavities, of these chimeric mice (data not shown). Neither did we find differences in T cell development or CD4:CD8 ratios (data not shown). Importantly, however, we found a dramatic difference in CD138+ (syndecan+) cells, in that the number of CD138+B220+ cells among GFP+B220+ bone marrow cells was 3-fold higher (n = 5, p < 0.01) than the number of CD138+B220+ cells among GFP−B220+ bone marrow cells (Fig. 4, C and D). In contrast, there was no increase in CD138+B220+ cells among GFP+B220+ bone marrow cells as compared with the number of CD138+B220+ cells among GFP−B220+ bone marrow cells in adoptive recipients of empty vector-transduced bone marrow stem cells (data not shown). These data indicate that FAIM-S enhances plasma cell differentiation during normal physiological processes in vivo.

The chimeric mice presented the opportunity to test whether FAIM-S affects NF-κB target gene expression in primary B cells by evaluating CD80. We stimulated splenic B cells from bone marrow chimera mice with CD40L, anti-Ig, or LPS. After 24 h, we examined stimulated and nonstimulated B cells for CD80 expression, comparing GFP+ and GFP− populations. We found that the increase in CD80 expression produced by CD40L was enhanced in the presence of exogenous FAIM-S (Fig. 4 E). In contrast, CD80 expression induced by anti-Ig and by LPS was little affected by FAIM-S expression (data not shown). As a control, we monitored expression of CD86, induction of which was little changed by FAIM-S under any of the stimulatory conditions (data not shown). Thus, FAIM-S specifically enhances CD40L-induced expression of an NF-κB/IRF4-responsive cell surface marker in primary B cells.

The present results introduce a new player in B cell differentiation. As shown here, FAIM in B cells enhances CD40 signaling for NF-κB activation, IRF4 expression, and BCL-6 down-regulation in vitro, and FAIM overexpression leads to expansion of the plasma cell compartment in vivo. All of these features are connected to, and fully consistent with, current paradigms for the regulation of B cell differentiation and plasma cell generation. What is remarkable is the role of FAIM in acting to promote CD40-driven outcomes.

Induction of plasma cell generation revolves around a cascade of four sequentially inhibiting transcription factors (reviewed in Ref. 37): B cell leukemia/lymphoma-6 (BCL-6), B lymphocyte-induced maturation protein 1 (BLIMP-1), paired box gene 5 (PAX-5) and X-box binding protein 1 (XBP-1). In brief, BCL-6 represses BLIMP-1 (which can, conversely, repress BCL-6, resulting in relatively stable states of BCL-6-dominant or BLIMP-1-dominant expression); BLIMP-1 represses PAX-5; and PAX-5 represses XBP-1. Whereas BCL-6 is required for germinal center development (38, 39), BLIMP-1 is considered the master regulator of plasma cell differentiation (40, 41). Among many other activities, up-regulation of BLIMP-1 represses PAX-5, a gene product that maintains B cell identity (42, 43). Repression of PAX-5, in turn, leads to, among other effects, up-regulation of XBP-1, which expands elements of the secretory pathway, enhances the unfolded protein response, and promotes plasma cell differentiation (44, 45, 46). B cells express abundant levels of BCL-6 and PAX-5, but little BLIMP-1 or XBP-1, whereas plasma cells express abundant amounts of BLIMP-1 and XBP-1, but little BCL-6 or PAX-5. At the top of this cascade lies BCL-6, and loss of BCL-6 is the key event that initiates transformation of the BCL-6^high/PAX-5^high phenotype to the BLIMP-1^high/XBP-1^high phenotype, leading to plasma cell differentiation. In keeping with this, BCL-6 expression is tightly regulated by several distinct mechanisms (30, 32, 33, 47, 48). Most prominently, and directly related to our results with FAIM, BCL-6 is transcriptionally down-regulated as a result of CD40 engagement (30, 32, 33).

The transcription factor cascade leading to plasma cell differentiation takes place within germinal center (GC) B cells (reviewed in Ref. 49), which experience CD40 signaling, as evidenced by the CD40 transcriptional signature in a subset of GC centrocytes in situ (30). This result is supported by additional studies indicating that CD40 signaling plays a key role in centrocyte survival and differentiation (30, 50, 51, 52). It has been shown that CD40 engagement induces NF-κB activation in B cells (10, 11, 12). In several landmark studies, Dalla-Favera and colleagues elucidated the functional relationships among CD40, NF-κB, and BCL-6, by identifying IRF4, known to be induced by CD40 (30) and known to be required for plasma cell generation (34), as the key intermediary (31). They demonstrated that NF-κB binds to the IRF4 promoter and induces IRF4 transcription and that IRF4 binds to the BCL-6 promoter and suppresses BCL-6 expression. To confirm the physiological relevance of these results, Dalla-Favera and colleagues examined primary GC B cells and found that, following CD40L stimulation, induction of nuclear NF-κB is accompanied by up-regulation of IRF4 and that up-regulation of IRF4 transcripts and IRF4 protein is accompanied by down-regulation of BCL-6 transcripts and BCL-6 protein. Finally, Saito et al. showed by in situ staining that GC centrocytes that express nuclear NF-κB lack BCL-6 expression, and vice versa (31). All together, these results indicate that CD40 signaling produces activation of NF-κB, which then stimulates IRF4 expression, which in turn represses BCL-6 transcription, and thereby promotes terminal differentiation of GC B cells.

FAIM expression is relatively enriched in GC B cells (Ref. 36 and our unpublished observations) and FAIM enhances each of the CD40-triggered events noted above that are connected to each other and to plasma cell differentiation: NF-κB activation, IRF4 expression, and BCL-6 down-regulation. This strongly suggests that the positive influence of FAIM on CD40 signaling produces, perforce, a positive increase in the plasma cell compartment. Recent work by Sciammas et al. demonstrating that the level of IRF4 determines the level of plasma cell differentiation (35) supports the notion that more CD40 signaling in B cells equals more plasma cell generation, fully in keeping with the effects of FAIM noted here. Thus, work reported by Dalla-Favera and colleagues (30, 31) provides a foundation for our results with FAIM. FAIM-S does not stimulate on its own but rather acts as a force multiplier for a series of events that begins with CD40 engagement and ends with plasma cell differentiation. However, it cannot be said with certainty that the signaling enhancement produced by FAIM-S in A20 cells occurs in primary B cells, nor can it be said at this time whether the level of FAIM-S produced by transduction of A20 cells or the level present in primary B cells derived from FAIM-S-transduced bone marrow stem cells approximates the level that occurs physiologically in GC B cells. In addition, we cannot rule out the possibility that apoptosis protection contributes to increased plasma cell generation in the presence of FAIM-S, although the normal development of B cells up to the plasma cell stage weighs against this, nor that constitutive (and thus potentially early) FAIM-S expression plays a role.

We found that the specificity of FAIM-S in enhancing CD40 signaling in B cells, but not signaling produced by other stimuli, is matched by the specificity embodied in the failure of FAIM-L to recapitulate the effects of FAIM-S. This inversion of FAIM-S/FAIM-L activities, which has been noted in other systems (4, 53), is unexpected, inasmuch as FAIM-L includes the entire sequence of FAIM-S, differing only in an additional 22 N-terminal amino acids. This suggests that the N-terminal domain in general, or the nonspliced additional N-terminal 22 amino acids of FAIM-L in particular, are regulatory in nature and influence the activity of a more C-terminal effector motif. Because the evolutionarily conserved FAIM-S sequence contains no known effector motifs, and because the FAIM-S structure expresses a unique fold that has no known homologues (54), it is highly likely that, when defined, the FAIM-S effector domain will be found to be a currently unrecognized and novel sequence. The mechanism by which FAIM-S effects enhancement of CD40 signaling in B cells remains unknown; however, inasmuch as NF-κB stimulates IRF4 expression and IRF4 inhibits BCL-6 expression, it appears that the various facets of FAIM activity are determined by its amplification of NF-κB activation. Because the levels of NF-κB produced by other NF-κB-inducing stimuli are not affected by FAIM-S, elucidation of the molecular details by which FAIM acts might properly focus on upstream events surrounding CD40 triggering.

As expected from our early work (1), FAIM-S opposed Fas-mediated apoptosis in primary B cells. Our new finding that FAIM-S enhances CD40-mediated NF-κB activation provides information that may be relevant to Fas-resistance. Many anti-apoptotic genes are NF-κB-dependent (8), and we previously showed that inducible Fas-resistance in B cells requires NF-κB (9). Further, we found that CD40-induced expression of BCL-xL and FLIP were enhanced by FAIM-S (data not shown). Note that although FAIM-S affects CD40 signaling, FAIM-S itself is induced not by CD40L alone, but by anti-Ig stimulation of CD40L-treated B cells, and in our previous experiments, CD40L was present when anti-Ig was added to induce Fas-resistance (1). Thus, it seems likely that the mechanism responsible for the anti-apoptotic effect of FAIM relates to its ability to enhance NF-κB activation, shown here, and that the FAIM-S features of signaling enhancement and anti-apoptosis would seem to be of one piece, with the latter dependent on the former.

Sole et al. evaluated the function of FAIM in neuronal tissue by studying PC12 cells and superior cervical ganglion neurons and found that FAIM-S (but not FAIM-L) enhanced nerve growth factor (NGF)-induced neurite outgrowth (53). They further showed that NF-κB activation induced by NGF, which is critical for neurite outgrowth, was enhanced by FAIM-S (53). Together with the present results, it is clear that FAIM-S is capable of enhancing NF-κB activation in two very distinct tissues (B cells and neural cells) in conjunction with triggering of two independent receptors (CD40 and NGFR) that, however, are members of the same overall family (55, 56). TNFR is a member of the same family of cysteine-rich pseudorepeat receptors and in preliminary experiments we found that FAIM-S (but not FAIM-L) enhanced TNF-induced NF-κB activation in 3T3 cells (our unpublished observations). In view of these results, and considering the evolutionary conservation of FAIM and of NGFR family members (57), it may be speculated that these molecules have been functionally interrelated for some time, and that FAIM can effect enhancement of signaling following engagement of multiple NGFR/TNFR family members even beyond CD40, NGFR, and TNFR.

We are grateful to our colleagues for helpful discussions and technical assistance throughout the course of this study.

The authors have no financial conflict of interest.